Agroforestry systems for sustainable livelihoods and

improved land management

in the East Usambara Mountains, Tanzania

Teija Reyes

Academic Dissertation

To be presented, with the permission of the Faculty of Agriculture and Forestry of the
University of Helsinki, for public discussion in Auditorium XII of the University of Helsinki
Main Building, Unioninkatu 34, on Friday 29 February at 12 oclock noon.

Helsinki 2008
Supervisors: Professor Olavi Luukkanen
Director
Viikki Tropical Resources Institute (VITRI)
Department of Forest Ecology
University of Helsinki
Helsinki
Finland

Dr. Roberto Quiroz

Leader
Natural Resources Management Division
International Potato Center (CIP)
Lima
Peru

Reviewers: Dr. Markku Kanninen

Director
Environmental Services and Sustainable Use of Forests
Center for International Forestry Research (CIFOR)
Bogor
Indonesia

Dr. Stig Johansson

Regional Director
Natural Heritage Services, Southern Finland
Metshallitus
Vantaa
Finland

Opponent: Dr. Bjrn Lundgren

Docent
Swedish University of Agricultural Sciences
Faculty of Forest Sciences
Uppsala
Sweden

2
ABSTRACT

Forest destruction for agriculture continues to be a major threat to the rich biological diversity
in the East Usambara Mountains in the northeastern corner of Tanzania. The highest ratio of
endemic plant and animal species found on 100 km2 anywhere in the world is here depending
on the remaining natural forests. Like elsewhere, forests are vitally important for the local
population in many different ways, and nationally they are an important source of water and
hydroelectricity. These sub-montane rain forests have been recognized globally as a priority
ecoregion. As public forests become all the time more seriously depleted, the pressure
towards forests assigned as conservation areas will increase.

The soils in the Usambaras are Ferrasols, of low fertility and strongly acidic, mainly having
nutrients only in the easily lost topsoil. Forests are sustained by a very tight cycling of
nutrients between vegetation and soil. After clear cutting, the soil is soon getting poor while
the fertile topsoil is leaching away. Still the human population is depending on unsustainable
forms of shifting cultivation for food production. Another harmful practice is the cash crop
cultivation, normally cardamom production, under the forest canopy, which causes chronic
degradation of the natural forests.

The general aim was to study the possibilities to develop new profitable and sustainable
agroforestry systems for the benefit of the local people that could contribute to relieving the
pressure on the natural forests in the East Usambara Mountains.

Socio-economic data were collected on the local cultivation methods (Study I) using a formal
survey and Participatory Learning and Action (PLA) methods in four villages located in two
different agroecological zones, highlands (>850 m) and lowlands (<850 m), which were using
different cultivation methods. These villages were located at the border of the Amani Nature
Reserve. The sample represented 10% of the total household heads in these villages. The data
were analysed with univariate and multivariate analysis of variance for clarifying the
environmental and socio-economic aspects of cardamom cultivation (Studies II and III), and
10% of the cardamom farmers in the buffer zones of two conservation areas, the Amani
Nature Reserve and the proposed Derema Wildlife Corridor, were interviewed and their fields
were observed. In Study III farmers were grouped according to their quantitative responses,
using cluster analysis of the variables. For an agroecological study (IV), a field trial was

3
established at the end of 2000 to clarify how two popular cash crops, cardamom (Elettaria
cardamomum (L.) Maton.) and black pepper (Piper nigrum L.), normally grown under the
natural forest, would perform in an intensive agroforestry system in combination with two
multipurpose farm trees, Grevillea robusta A.Cunn. and the nitrogen fixing Gliricidia sepium
Jacq.

With improved agroforestry methods (Study I) the households obtained twice as much of
annual gross income and 13 times the net income as compared to incomes from traditional
practices. Even if the incomes were higher with improved systems, which also provided a
better food security, the per-capita income remained still below the national average. Cash
flows were mainly derived from cash crops, livestock and off-farm activities. The highland
farmers cultivated more cash crops and kept more cattle than the lowland farmers, and
received also a better income.

About 60% of the highland farmers in the Usambaras grew cardamom and obtained more than
half of the total cash-crop income from it (Study II). Cardamom cultivation in the forest
caused forest deterioration and poor forest regeneration. Fragmentation of the forest was
considered a serious threat to many endemic species. Problems similar to those in the East
Usambaras were also noticed in other cardamom cultivation areas in the world, in the Western
Ghats in India and Alta Verapaz in Guatemala. Socio-economic differences were large
between these countries. Cardamom growers in India cultivated private lands, formed an
upper social class, were well educated, organized and supported, and had good networks and
social services, whereas in Guatemala and Tanzania they were poor, uneducated and
confronting high risks due to unstable conditions. Tanzania can compete neither with the
volume nor with prices in the global cardamom markets.

A further analysis clustered the cardamom farmers associated with the two conservation areas
(Study III). Villages in the buffer zone of the Derema Wildlife Corridor were more dedicated
to cardamom cultivation, had a larger land area allocated for it, managed their fields better,
and produced cardamom of better quality than the villages around the Amani Nature Reserve.
The villagers in Derema had also increased the cardamom cultivation in the homegardens,
since conservation activities had reduced the farmers cardamom cultivation areas in the
forest. Still most farmers supported forest conservation. Negative aspects towards
conservation had provoked an unclear and prolonged compensation procedure in the proposed
conservation area. A majority of the farmers reported a decline in yields and changes in the

4
local climate as a result of environmental changes due to deforestation. Yet, most farmers
around both conservation areas were not ready to change their cultivation methods.

Results from the spice crops agroforestry trial (Study IV) showed clear advantage for
intercropping cardamom and pepper especially with grevillea: pepper produced 3.9 times
more, whereas cardamom intercropped with grevillea and pepper produced 2.3 times more
than in monoculture. In the fourth year, cardamom yielded 5.5 times more with grevillea than
in the natural forest. Gliricidia improved the nitrogen and organic matter content of the soil
over levels found in the natural forest. Soil acidity was, however, preventing the crop plants
from using the available mineral nutrients more effectively.

There are sustainable and profitable ways to cultivate cash crops in well-managed
homegardens and to raise the standard of living of the local population by promoting nature-
based sustainable spice-businesses and supporting farmers in them. It is relatively easy to
intensify the cash crop production on the farms if stable markets for the products are obtained.
Farmers need access to credit and comprehensive extension. The soil fertility depletion should
be reversed and an enabling policy environment for the smallholder-farming sector provided.
Strong farmers organisations and equal rights to resources and decision-making are needed.
For developing the East Usambara Mountains, improving peoples livelihoods and conserving
the biodiversity is a matter of functioning teamwork by government officers, the local
population, extension staff, traders, NGOs and donors. Participatory forest management could
have an important role in natural resource management here. What is, however, needed most
for a change is a political will of government that understands the importance of agricultural
and forestry development in poverty reduction.

5
PREFACE

The study was carried out at the beginning in the Viikki Tropical Resources Institute (VITRI),
Helsinki, and since 2004 in the International Potato Center (CIP) in Lima, Peru. The research
has been initiated with financing from the Finnish Biodiversity Research programme (FIBRE)
between 2000 and 2002. The University of Helsinki financed a part (from 2002 to 2003) and
the Finnish Graduate School of Forest Science another from 2004 to 2007. The Finnish
Society for Forest Science has financed one field trip to Tanzania. University of Helsinki
travel grants and the Finnish Graduate School of Forest Science financed conference travel to
USA and to Tanzania irrespectively.

I am grateful to my supervisor Professor Olavi Luukkanen for his continuous support and
help, for encouraging me and keeping me motivated throughout the whole research. Dr.
Roberto Quiroz, Research Scientist and Leader of Natural Resources Management Division
helped me at CIP with data analysing and supporting me during the research. Both
supervisors helped me with their great experience. Thank you also for my pre-reviewers, Dr.
Stig Johansson and Dr. Markku Kanninen for their valuable comments. I also give my special
thanks to Docent Wending Huang who helped me in planning the field experiment in the
beginning. The Conservator of Amani Nature Reserve, Mr. Sawe, greatly facilitated the
research activities in the area and gave much advice and information. In the beginning of my
research I also got assistance from the EUCAMP project, working in the area until the end of
year 2002. Thank you, Dr. Veli Pohjonen and Ms Taina Veltheim! I am always thankful to
CIPs Statistical Advisor Felipe de Mendiburu for patiently helping me to solve the statistical
problems that the huge amount of ecological data caused, and to Cecilia Ferreyra in the
library of CIP for searching valuable articles for my research from the other side of the world
using the functioning CGIAR library links.

Special thanks to all my colleagues in VITRI and at CIP Natural Resources Management
Division, especially to Dr. Minna Hares, Dr. Eshetu Yirdaw, Dr. Mark Appiah, Dr. Vesa
Kaarakka and Kourosh Kabiri in VITRI, and Dr. Victor Mares, Ivonne Valdizan, Dr. Mariana
Cruz, Alberto Garca, Diana Torres and Javier Osorio in CIP. Kiitos rakkaat ystvt Irmeli
Mustalahti and Mari Pennanen-Kok for every kind of support during the research in Tanzania
and for hosting me for so many times in Dar es Salaam and in Tanga. Also thank you, all my

6
other friends everywhere in this world, for helping me with little messages and keeping in
touch where ever I have been.

Special thanks also go to my loyal and committed field assistant, Mr. Abduel Kajiru, whose
seven years of reliable and constant commitment was necessary for my work. Without his
excellent work my research would not have been possible. He was an excellent research
assistant, talented guide, translator, friend and local expert in culture and nature. Asante sana
Bwana Kajiru! Also I would like to thank my soil scientist Mr. Godson Urassa, who assisted
me three times in soil analysis and whose working institute, Mlingano Soil Research Institute,
the nearest possible soil institute close the Usambaras, was implementing the soil analysis.
Nakushukuru sana Bwana Urassa!

Many thanks to my parents, my family and my husbands family for encouraging me and
helping me with children and housework. Thanks to my brother Jari who also accompanied
me to the field as my driver and field assistant (he is also a forest engineer, like my father),
and to my sister Kirsi who many times gave me technical computer assistance and positive
motivation to continue. Thanks to my husband Renato, for support in every phase of my
research, even by assisting me in planting seedlings (quite exotic for a lawyer) when
establishing the experiment. And to my children, Kimi and Renja, who patiently waited for
me coming back home from my long field trips to Tanzania.

7
LIST OF ORIGINAL PAPERS

This thesis is based on the following original articles:

Study I
Reyes, T., Quiroz, R., Msikula, S. 2005. Socio-economic comparison between
traditional and improved cultivation methods in the East Usambara Mountains,
Tanzania. Environmental Management 36(5): 682-690.
(http://www.springerlink.com/content/l87j328470070622/fulltext.html).

Study II
Reyes, T., Luukkanen, O., Quiroz, R. 2006. Small cardamom precious for people,
harmful for mountain forests: Possibilities for sustainable cultivation in the East
Usambaras, Tanzania. Mountain Research and Development 26(2):131-137.
(http://www.mrd-journal.org/abstracts.asp?Issue_ID=47#868).

Study III
Reyes, T., Luukkanen, O., Quiroz, R. 2009. Conservation and Cardamom Cultivation
in Nature Reserve Buffer Zones in the East Usambara Mountains, Tanzania. Journal
of Sustainable Forestry 29(1-2). (Accepted).

Study IV
Reyes, T., Quiroz, R. Luukkanen, O., de Mendiburu, F. Growth and yield analisis of
spice-agroforestry systems in the East Usambaras, Tanzania. Agroforestry Systems.
(Submitted in July 2007).

In studies I IV T. Reyes introduced the research idea, organized the experimental

arrangements, collected and analysed the data and prepared the manuscripts, which
were revised with R. Quiroz and O. Luukkanen. R. Quiroz was involved in statistical
design. In study I, S. Msikula assisted in data collecting and in study IV, F. de
Mendiburu helped in statistical analysis.

8
TABLE OF CONTENT

ABSTRACT................................................................................................................ 3

PREFACE .................................................................................................................. 6

LIST OF ORIGINAL PAPERS.................................................................................... 8

TABLE OF CONTENT ............................................................................................... 9

LIST OF ABBREVIATIONS ..................................................................................... 12

1. INTRODUCTION .................................................................................................. 13

1.1. Background...................................................................................................................... 13

1.2. Forest loss and soil degradation in the tropics ............................................................. 14

1.3. Importance of mountain forest ecosystems in the world ............................................. 16

1.4. The aim of the study........................................................................................................ 18

1.5. Hypotheses ....................................................................................................................... 19

2. THEORETICAL FRAMEWORK AND LITERATURE REVIEW........................... 20

2.1. Special characteristics of East Usambara Mountain forests....................................... 20

2.2. Forest degradation in the East Usambara Mountains................................................. 22

2.3. Previous Finnish activities in the East Usambara Mountains..................................... 26

2.3.1. Project activities in the East Usambaras .................................................................... 26
2.3.2. Previous studies in the East Usambaras ..................................................................... 27

2.4. Improved agroforestry systems as a tool in sustainable development ....................... 28

2.4.1. Intercropping in Tanzania .......................................................................................... 28
2.4.2. The role of agroforestry in the conservation of mountain ecosystems ...................... 30
2.4.3. Improved agroforestry systems .................................................................................. 32

2.5. Some constraints for adopting better land-use management in Tanzania ................ 34
2.5.1. Short-term thinking .................................................................................................... 34
2.5.2. Lack of resources ....................................................................................................... 35
2.5.3. Low education level and inadequate extension.......................................................... 36
2.5.4. Governance shortfalls................................................................................................. 37

2.6. Background to the agroforestry species studied........................................................... 39

2.6.1. Queen of spices, cardamom (Elettaria cardamomum Maton) ................................... 39
2.6.2. King of the spices, black pepper (Piper nigrum L.)................................................... 42
2.6.3. Grevillea robusta, the multipurpose agroforestry tree................................................ 46
2.6.4. Gliricidia sepium, the N-fixing agroforestry tree....................................................... 46

9
3. MATERIAL AND METHODS ............................................................................... 48

3.1 Study area ......................................................................................................................... 48

3.1.1. The East Usambara Mountains .................................................................................. 48
3.1.2. Climate and soil characteristics.................................................................................. 52
3.1.3. The Amani Nature reserve and the Derema Wildlife Corridor .................................. 55

3.2. Socio-economic data analysis (Study I) ......................................................................... 58

3.3. Harmful cardamom cultivation practices (Study II) ................................................... 59

3.4. Conservation and sustainable cultivation (Study III).................................................. 60

3.5. Experimental study on the spice-crop agroforestry systems (Study IV).................... 60

3.6. Ecological analysis of spice-agroforestry system (Study IV)....................................... 62

3.6.1. Growth analysis.......................................................................................................... 63
3.6.2. Treatment comparison................................................................................................ 63
3.6.3. Light conditions.......................................................................................................... 63
3.6.4. Soil analyses............................................................................................................... 64
3.6.5. Yields vs. Land Equivalent Ratio (LER) ................................................................... 64
3.6.6. Manure test for cardamom ......................................................................................... 64

4. RESULTS............................................................................................................. 66

4.1. Socio-economic aspects of improved agroforestry systems (Study I)......................... 66

4.1.1. Improved agroforestry vs. traditional cultivation....................................................... 66
4.1.2. Cultivation in highlands vs. lowlands ........................................................................ 68

4.2. Precious cardamom harmful for the mountain forests (Study II)........................... 69
4.2.1. Harmful cardamom cultivation methods.................................................................... 69
4.2.2. Socio-economic comparison with cardamom growers in India and Guatemala ........ 70

4.3. How to link conservation and sustainable livelihood in the buffer zone of a nature
reserve (Study III) .................................................................................................................. 71
4.3.1. Differences between villages in Derema and Amani Nature Reserve ....................... 71
4.3.2. Conservation and sustainability ................................................................................. 73
4.3.3. Factors hindering farming intensification .................................................................. 74

4.4. Spice-agoforestry system (Study IV) ............................................................................. 76

4.4.1. Growth and yield........................................................................................................ 76
4.4.2. Light conditions.......................................................................................................... 80
4.4.3. Soil ............................................................................................................................. 81
4.4.4. Land Equivalent Ratio................................................................................................ 83
4.4.5. Manure test for cardamom ......................................................................................... 83

5. DISCUSSION ....................................................................................................... 85

5.1. Biodiversity conservation and sustainability ................................................................ 85

5.2. Better land use ................................................................................................................. 90

5.3. Improved agroforestry as a sustainable cultivation system ........................................ 92

10
5.4. Spice-crop agroforestry systems .................................................................................... 95

5.5. Improved soil quality .................................................................................................... 102

5.6. Organic products and marketing................................................................................. 109

5.7. Better extension ............................................................................................................. 114

5.8. Poverty reduction and incentives for the local population........................................ 117

5.9. Participatory forest management ................................................................................ 122

5.10. Government support ................................................................................................... 126

6. CONCLUSIONS AND RECOMMENDATIONS ...................................................133

REFERENCES ........................................................................................................137

ANNEXES ...............................................................................................................159

Annex 1. Soil profiles ........................................................................................................... 159

Annex II. Questionnaire for cardamom farmers in the East Usambara Mountains..... 163

11
LIST OF ABBREVIATIONS

AFTP Agroforestry tree products

ANR Amani Nature Reserve
CEPF Critical Ecosystem Partnership Fund
CGIAR Consultative Group on International Agricultural Research
CIDA Canadian International Development Agency
CIP International Potato Center
EUCADEP East Usambara Conservation and Agricultural Development Project
EUCAMP East Usambara Conservation Area Management Programme
EUCFP East Usambara Catchment Forest Project
EUM East Usambara Mountains
EUTCO East Usambaran Tea Company
Derema Derema Forest and Wildlife Corridor
FAO Food and Agriculture Organization of the United Nations
GDP Gross Domestic Product
ICRAF World Agroforestry Centre
IUCN World Conservation Union
LER Land Equivalent Ratio
NFTA Nitrogen Fixing Tree Association
NGO Non-governmental organization
NFTA Nitrogen Fixing Tree Association
NTFP Non-timber-forest-products
PASS Private Agribusiness Sector Support
PES Payments for Environmental Services
PLA Participatory Learning and Action
PRA Participatory Rural Appraisal
PRSP Poverty Reduction Strategy Paper
TFCG Tanzania Forest Conservation Group
TZS Tanzania shilling
UNCTAD United Nations Conference on Trade and Development
UNICEF United Nations Children's Fund
UNDP United Nations Development Programme
URT United Republic of Tanzania
WWF World Wildlife Fund

12
1. INTRODUCTION

1.1. Background

The conversion of natural forest for cardamom (Elettaria cardamomum (L.) Maton.)
cultivation is one of the main problems threatening the East Usambaran forest and its species
(Newmark 2002; CEPF 2005). Cardamom cultivation is estimated to cover about 30% of the
forested area and it has been claimed to be harmful for the rich biodiversity in the East
Usambara Mountains of Tanzania (Johansson and Sandy 1996). Here local extinction of
endemic species dependent on dense primary forests has been reported by Newmark (2000),
and a decrease in air humidity at higher altitudes has also been associated with forest
clearance (Hamilton and Bensted-Smith 1989).

Cardamom is the third most expensive spice in the world, and as such it is a very important
income source for local farmers. About 60% of the highland farmers in the Usambaras grow it
as a cash crop. The contribution of cardamom to the average household income is about 30%.
As cardamom needs shade, farmers generally cultivate it under the forest canopy after
clearing the undergrowth vegetation, leaving only some top canopy trees for ideal shading.
However, in the East Usambaras, with an increasing population and a diminishing forested
area suitable for cardamom cultivation, this method is no longer sustainable.

In the East Usambaras, formal conservation measures have come into conflict with cardamom
cultivation. Farmers have experienced economic losses due to conservation measures, as
payments in compensation for the loss of access to the forest resources and cultivated land
areas were seen as inadequate. Cardamom cultivation was frequent in the forest in the past,
and it is still mainly practiced in the public forests, illegally also in the forest conservation
reserves. The local population has very limited access to forest reserves, and poor farmers are
not yet benefiting from conservation. To the contrary, strict conservation rules have forced
them many times to recur to illegal activities in order to obtain necessary goods that used to
be free for them.

13
1.2. Forest loss and soil degradation in the tropics

Tropical forests cover less than 6% of the Earths land area but contain the vast majority of
the worlds plant and animal genetic resources. It is estimated that the tropical rain forest may
contain 30 million plant and animal species. The annual deforestation rate is about 11.3
million ha (Mastrantonio and Francis 2004). People depend on forests and trees in the
developing countries in many different ways (Dubois 2003):

- of the worlds poor depend directly on forests for their livelihood;

- 350 million people live in or adjacent to dense forests and rely on them;
- At least 2 billion people rely on biomass fuels (mainly fuelwood) for cooking and heating;
- Forestry provides employment for more than 10 million people;
- Natural products from forests are the only source of medicine for 75-90% of people in the
world.

Shifting cultivation 1 is believed to have originated around 7000 BC, and this system is still
common in the mountainous areas of tropical Asia, and in Africa and Latin America. It was
predominantly sustainable in the past due to a low population pressure and the availability of
large forest areas. Today, shifting cultivation contributes to excessive soil erosion and to land
degradation (Steppler and Nair 1987). It is estimated that 500 million farmers in developing
countries still use shifting cultivation systems (ASB undated in Scherr 1999). Most of them
cultivate their land in marginal areas with poor soil quality, or on steep slopes. Shortening
fallow periods and widespread burning to control weeds and pests further contribute to land
degradation. Large areas have already been abandoned due to nutrient and organic matter
depletion or invasive weeds (Scherr 1999). Soils on steep slopes have commonly completely
lost their productivity due to soil erosion. This causes serious local food shortages (Sah
1996).

Productivity has declined 16% on the African agricultural lands in the past 50 years. Of the
degraded soils, 58% are in drylands and 42% in humid areas. The most common reason for
declining productivity is water erosion. Other reasons are wind erosion, chemical soil
degradation (loss of nutrients, salinization, pollution, and acidification) and physical soil
degradation (loss of organic matter, water logging, and compaction sealing or crusting; Scherr

1
Farming systems where farmers move from one place to another when the land becomes exhausted.

14
1999). The extent and effect of water erosion depend on the soil erosivity, which is the power
of the rain to cause erosion; and erodibility, which is the ability of the soil to resist the rain
(Hellin 2006). Water erosion is problematic in the tropics because of heavier rain showers, as
compared to other regions. Erosion is caused in the tropics due to uncovered soil, absence of
windbreaks, lack of organic matter in the soil, and monocultures in farming (Glover 2005).

The highest erosion rates in Africa have been calculated in the Maghreb region, East African
highlands (including the East Usambara Mountains), eastern Madagascar, and parts of
Southern Africa (Scherr 1999). In Tanzania forest and woodlands support 87% of the rural
poor population (Milledge at al. 2007). Tanzania has 38.8 million ha of forests, about 41% of
the total land area. These resources are under pressure caused by human settlements and
activities such as clearing for agriculture, illegal harvesting, fires and mining. These pressures
lead to deforestation, estimated to be 91 000 ha per annum in the country (Meghji 2003). At
least ten million ha of forest land was lost in Tanzania between 1970 and 1998 (Milledge at
al. 2007).

Forest soils in the tropics, as also in the East Usambara Mountains, maintain their fertility due
a tight cycling of nutrients between vegetation and soil; if this cycle is broken through forest
destruction, nutrients are likely to be rapidly lost, which results in an impoverished soil
(Hamilton and Bensted-Smith 1989). Soil degradation is a major contributor to nutrient
losses, because most of the scarce soil nutrients in the tropics are in the top 5-10 cm of the soil
(Nkonya et al. 2004). The soils have a low water holding capacity due to a low content of
small soil particles. High temperatures favour rapid decomposition of organic residues; thus
organic inputs are needed to avoid erosion. Steep lands are more sensitive to rapid soil
degradation through runoff (Hellin 2006).

Soil fertility depletion on smallholder farms in Africa is already considered as a biophysical

limiting factor affecting food production (Sanchez et al. 1997). This soil degradation affects
more the rural poor, because they are more dependent on annual agricultural crops that also
cause more degradation than the other crops. They also rely more on common-property lands,
which often are most seriously degraded. During the present field work local farmers in the
East Usambaras pointed out that many times they can survive only by overexploitation of
natural resources. Land degradation has thus become a social, economic, political and
technical problem (Hellin 2006; Figure 1).

15
 Population Shortening
Decline in Low crop Shortage of
increase of fallows
fertility yields food or
income
Lack of
Shortage Low
capital
of land inputs

Sloping Shortage of
Soil Decline in Low crop
land food or
erosion fertility yields
income

Intensive
Lack of
land use
conservation

Figure 1. Chains of cause and effect linked to decline in soil fertility (UNDP 1995).

1.3. Importance of mountain forest ecosystems in the world

Mountain regions cover 20% of the Earth's land surface and are one of the most fragile
ecosystems together with deserts, polar caps and coral reefs. They are very sensitive to effects
caused by the climate change, an increasing human population; as well as pollution, loss of
genetic diversity or habitat loss. Mountain ecosystems contain a considerable biodiversity,
and they are the home to many endangered species, because they include a rich variety of
ecological and climatic systems.

Over half of the world's population is affected by mountain ecology and the degradation of
watershed areas, but relatively little attention has been paid to this fact (UN, Agenda 21
1992). Mountain forests are significant globally, nationally and locally for providing various
irreplaceable products and services, but they are under threat due to low management
capacity, overuse of resources, and agricultural encroachment. This importance of mountains
has only recently been globally recognized; many mountain countries, however, have
recognized the crucial interrelationships between mountain ecosystems and the people who
inhabit them (Mishra 2002).

16
About 10% of the global population relies on the resources of high mountain slopes for their
survival, and there is considerable poverty among mountain inhabitants that also contributes
to environmental degradation (UN, Agenda 21 1992). Many of the tropical mountain forest
areas are highly productive but have come under increasing pressure from outside economic
forces and the human population growth (Price and Kohler 2000).

Poor mountain peoples have become the guardians of irreplaceable global resources, their
homeland serving as storehouses of timber, water, minerals, meat, hydroelectric power
(Investing in Mountains 1997) and a rich biodiversity. Forest biodiversity is threatened in
mountain areas where forests have been cleared for agriculture or where plantation forests
have replaced the natural forests. Mountains are securing water for over half of the human
population. By helping to capture and store rainfall and moisture, maintaining the water
quality, regulating streamflows and reducing erosion, mountain forests are vitally important
for protection against natural hazards, such as landslides, avalanches, rockfalls and floods.
Forests also protect the very scarce but vitally important infrastructure in the mountain areas.
This infrastructure allows information and product flows and the movement of the people. In
many parts of the world, mountains are also key tourism destinations (Price and Kohler 2000).

Mountains are characterised by specific development challenges: difficult access, economic

and political alienation, outmigration or high immigration, environmental sensitivity, and a
diversity of livelihoods and cultures (Mountains of the World 2002). Multifunctionality has
become a key concept in mountain forest management. This concept adopts a long-term
perspective: mountain forests may be destroyed quickly but often need decades to recover
(Price and Kohler 2000).

In subtropical montane cloud forests, such as the East Usambaran forests in Tanzania, a low
transpiration rate and canopy interception result in the addition of water to the hydrological
system. In general, the vegetation has an ability to capture and store water (Aldrich 2000), and
about 15-20% of the rainwater is held in the canopy of closed forests (Wadsworth 2004).
Mountain cloud forests thus play a vital role in sustaining the livelihoods by providing water
and non-timber forest products. They also often include endemic species and yet unknown
species. As a whole they have a high socio-economic value for the local population.

The main problem in the East Usambaras is how to link biodiversity conservation of these
unique forests and ensure livelihoods to the poor local population living in the conservation

17
areas and adjacent to it. Cardamom is a valuable cash crop for the farmers, and the farmers
wish to continue cultivating it, but the cultivation methods in the natural forests are harmful
for the biodiversity. Without the possibility of expanding cultivated areas and with limited
access to forest resources, a livelihood-improvement option that now particularly needs
attention is diversification and intensification of the existing production systems on a small
scale, so as to include both timber products and non-timber forest products such as cardamom
and black pepper.

1.4. The aim of the study

The general aim was to study the possibilities to develop new profitable and sustainable
agroforestry systems for the benefit of the local people that could contribute to relieving the
pressure on the natural forests in the East Usambara Mountains.

An overall aim of the study was also to survey and analyse traditional and improved
agroforestry practices in relation to the farmers' socio-economic situation and expectations.
An assumption was made that by improving the traditional agroforestry systems better socio-
economic benefits for the people could be achieved. It was further assumed that scientifically
based new information would contribute to new alternatives for the local farmers to adopt
better land management practices that then could spearhead the development towards better
overall land management. For this purpose, a decision was made to conduct the field research
in close cooperation with forest-adjacent farmers and using on-farm trials in their fields.

The first specific aim of the present work was to describe the cultivation methods in the East
Usambara Mountains in relation to the socio-economic situation of the farmers (Study I).
Another specific aim was to clarify the environmental aspects of cardamom cultivation and
their relationships to the socio-economic situation of the local people; for this purpose, a
global comparison was also made (Study II). An additional part of the work (Study III) dealt
with issues related to the local and national agricultural trade and its linkages with rural
development policies for instance, on questions how conservation activities were affected
by cardamom cultivation, what the farmers responses to these conservation measures were,
and, finally, what possible solutions could be found for linking the national conservation
policies to the livelihoods of the rural people in the East Usambaras context.

18
A separate, experimental part (Study IV) focused on the performance of improved
agroforestry systems in the East Usambaras especially in relation to soil characteristic as a
result of soil improvement interventions, with two common tree species in the area, Grevillea
robusta and Gliricidia sepium; and spices, cardamom (Elettaria cardamomum) and black
pepper (Piper nigrum) as system components and using growth and yield data as indicators.
The ultimate aim was to produce the necessary information needed for designing systems with
improved performance under the specific conditions of the study area.

In order to facilitate the research for the last aim above, a review of the existing agroforestry
literature relevant for the area in question was also conducted.

1.5. Hypotheses

In the present study, the following hypotheses were set:

1. By using improved agroforestry methods farmers are able to get a higher

income without additional risks;
2. Intercropping with nitrogen-fixing trees leads to improved soil fertility;
3. Agroforestry is a useful tool for achieving sustainable spice cultivation in the
East Usambara Mountains; and,
4. Organic spices cultivation for exportation in intensive agroforestry systems is a
viable option for rural development and biodiversity conservation in the East
Usambara Mountains.

19
2. THEORETICAL FRAMEWORK AND LITERATURE REVIEW

2.1. Special characteristics of East Usambara Mountain forests

The East Usambara Mountains, at 900-1 000 m elevation in NE Tanzania, belong to the chain
of isolated mountains called the Eastern Arc. The Eastern Arc chain of low mountains was
formed about 100 million years ago. It is thought that these mountains have been isolated
from each other for the last 20 million years, maintaining uniform climatic regimes during
this time (Sayer et al. 1992). This has resulted in the highest ratio of endemic flora and fauna
per 100 km2 of all biodiversity hotspots in the world (Myers et al. 2000; Table 1).

Table 1. Diversity and endemism in the Eastern Arc. As many as 13% of the total plant
species found in Africa are found in only 0.1% of its land area (Conservation International
2001).

Taxonomic group Species Endemic species % endemism

Plants 4000 1500 35.0
Mammals 183 16 8.7
Birds 585 22 3.8
Reptiles 188 50 26.6
Amphibians 63 33 52.4

Newmark (1992) has mentioned that most probably the East Usambara Mountains are the
most important area biologically in all of Eastern and Central Africa. They are considered as
belonging to the top three most important blocks of the Eastern Arc concerning the endemic
animal and plant species. Among all the 13 mountains belonging to the Eastern Arc, the East
Usambaras have the largest number of endemic tree species per km2 and they are placed as
second in the case of endemic vertebrates (Burgess et al. 2007b). They have been recognized
globally as a Biodiversity Hotspot by Conservation International, an Endemic Bird Area by
BirdLife International, a Centre of Plant Diversity by WWF and IUCN, and a Globally
Important Ecoregion by WWF (Burgess et al. 2007a). The process has also initiated the
inclusion of the Eastern Arc in the list of Natural World Heritage Sites by UNESCO (Burgess
et al. 2007b). New species are continuously found, even a new primate species was found
there in 2006 (Zimmer 2007).

20
The high degree of endemism is explained by isolation and a uniform climate. A speciality in
the case of the East Usambaras is also their proximity to sea. Another explanation for such a
high concentration of endemism is the theory that the Eastern Arc is a centre of speciation,
with the effect that various plant and animal groups exhibit remarkable local diversity
(Burgess et al. 2007b). Such areas are particularly found in mountain cloud forests. Since the
rich flora and fauna of the East Usambaras have concentrated on a small geographic area,
there is a high risk of rapid and irreversible loss of biodiversity. This is the reason why the
East Usambaras are considered a top-priority area for biodiversity conservation. The
extinction risk to the fauna and flora has been constantly increasing, and concerted quick
actions are needed to conserve these areas. It is noteworthy that in comparison to the
Amazonian forests, where most species are widely distributed, the East Usambara area is
more threatened by irreversible species loss (Fjelds and Lovett 1997).

Existing research results (cf. Iversen 1991; Newmark 1992; Johansson et al. 1998; and Azeria
et al. 2007) in the area have shown that this endemism is largely confined to undisturbed or
fragmented natural forests. Many of the endemic species are forest specialists depending on
diverse and dense primary forests in a given altitudinal range (Azeria et al. 2007). Any
process, such as cardamom cultivation, that causes large-scale forest disturbance, has a strong
effect on the conservation value of the forests. Thus the protected forests in the East
Usambaras should remain as undisturbed as possible; otherwise their specific value is lost
(Table 2).

Table 2. Summary of biodiversity of species groups in Amani Nature Reserve (Doody et al.
2001).

Species Total no. % forest No. of No. of No. of No. of

of species dependent non-forest endemic near- forest
species species endemic dependent
species endemics
and near-
endemics
Trees and 631 43 22 19 49 53
shrubs
Mammals 59 15.3 6 0 3 2
Birds 65 33.8 15 2 3 3
Reptiles 49 46.7 6 3 15 17
Amphibians 27 66.6 0 2 14 16
Butterflies 112 20.5 4 1 10 9
Total 943 n/a 53 27 94 100

21
It is estimated that the annual economic worth of the Eastern Arc forests is at least $620
million. These forests provide drinking water for at least 60% of the Tanzanian urban
population and over 90% of the nations hydroelectricity generation capacity. These estimates
include the biodiversity values (Burgess et al. 2007b). It is also estimated that forest products
there are worth annually at least $100 per rural inhabitant in nutrition and medicinal value,
which means that the forests may be generating an additional $150 million to these people
(Pfliegner and Burgess 2005).

The East Usambara forests secure the water supply for Tanga's 300 000 inhabitants, and they
are vitally important for the local people (about 135 000) in many different ways. For the
local population the forests often do not have much direct commercial importance, but they
are important for the daily subsistence of the people. Forests provide fuelwood (from an
estimated 33 tree species), poles (35 species), ropes (43 species), food (28 species), medicines
(185 species), and household utensils (83 species). Apart from these material functions, the
local people value the forests for their environmental or religious (e.g. traditional rainmaking
ceremonies) values (Kessy 1998). Forests provide the long-term sustainability for both
smallholder agriculture and the industrial plantations, such as the tea estates. Forest genetic
resources in the Usambaras have economic value at the global scale (Woodcock 1995).

2.2. Forest degradation in the East Usambara Mountains

The main threat to the East Usambaran forests is the biodiversity loss. The causes for the
biodiversity loss are forest cover decrease and forest fragmentation, physical alteration of
forest structure, over-exploitation of the forest resources, and the introduction of alien species.
The situation is aggravated by a high human population density, inadequate environmental
policies, lack of scientific knowledge, and weak institutions. Forest fires cause significant
damages in the Usambaras which exceed the combined effects of pitsawing, fuelwood
collection and pole harvesting (Lulanda 1998).

According to several authors (e.g. Hamilton and Smith 1989; Sayer 1991; Stocking and
Perkin 1992; Newmark 2002; and CEFP 2005), uncontrollable land use in the form of
cardamom cultivation and farmland encroachment is the main threat to the East Usambaran
forests and their species. Cardamom cultivation in Tanzania is still mostly practiced in the
forest after clearing totally the understory and the middle canopy layer, and selectively

22
thinning the top canopy. The impacts can appear innocuous and small, but if the disturbance
continues, they can have more widespread consequences than logging or land clearance alone.
Continuous removal of the forest groundstory may, over time, affect the whole forest
succession (Ashton et al. 2001). There is a risk that the mountain rainforest ecosystem will
not recover in fallowed cardamom fields. Cardamom cultivation in the forests is accelerating
the total forest clearance, also because the Usambaran farmers consider that the forest areas
already converted to cardamom are easier to clear completely afterwards for annual crops.

Hamunen (1998) in his study in the East Usambaras found that a serious effect of cardamom
cultivation was the risk of poor forest regeneration on fallow land. The assessment was made
on sites abandoned 2-20 years earlier. The length of the fallow period did not much affect the
number of seedlings or the understory vegetation cover percentage. A total of 24% of the 177
abandoned cardamom sites and 36% of 194 totally cleared sites did not have any tree
seedlings. Abandoned cardamom sites were, however, distinctly more diverse than the
completely cleared sites. On fallows, the number of tree species found was 74, whereas in
open areas it was only 45. The species composition appearing on the abandoned cultivation
sites depended on the intensity of man-made disturbance. Weeds easily invaded the disturbed
areas and prevented the establishment and survival of tree seedlings (Hamunen 1998).

As also described by Hamunen (1998), normally after cardamom cultivation, the site was
totally cleared and converted to farmland for annual crops (sugarcane, Saccharum officinarum
L.; maize, Zea mays L.; and cassava, Manihot esculenta Crantz) until the soil nutrients were
totally depleted and only a dense undergrowth thicket remained, mainly consisting of Lantana
camara L., Clidemia hirta L. and Psidium cattleianum Sabine; thus the forest regeneration on
these most degraded sites seemed almost impossible (cf. Stocking and Perkin 1992; Figure 2).
The weed species mentioned above are alien and potentially extremely invasive, and they thus
pose an additional risk to the biological diversity. Lantana can colonise natural forest gaps
very quickly (as observed in the present study, local people anyhow claimed that lantana is
quite easy to clear from the site). Clidemia hirta can invade the understorey of forests with a
relatively intact canopy and impair the forest regeneration process significantly. The most
serious weed is, however, the wild guava species (Psidium cattleianum) also known as
Chinese or strawberry guava. It can colonise intact-canopy forests very densely and prevent
the regeneration of other tree species. Also cultivated guava (Psidium guajava L.) invades
easily from the gardens to the open sites. All these species require at least mechanical control
(Fowler and Nyambo 1996).

23
Figure 2. A typical pattern of forest degradation following the planting of cardamom in the
East Usambara Mountains (modified after Stocking and Perkin 1992).

For the case of the East Usambaras it has been assumed that 26% of the forest area is under
crop cultivation, mainly for cardamom in the highlands (Johansson and Sandy 1996). In order
to predict how much cultivation actually exists in the forest is difficult. The cardamom areas
are not apparent from aerial photographs, as most of the cardamom cultivation occurs under
the tree canopy, and, since this activity is illegal, the farmers are not very willing to provide
this information. Sah (1996) assumed that there is about 17 000 ha of cultivations in the
forests, and Johansson and Sandy (1996) gave an estimation of about 11 000 ha (Figure 3).

Forest fragmentation is a serious problem from the viewpoint of endemic species which
depend on the natural forests of the East Usambaras. Forest destruction continues to be a
major threat to biological diversity. Already 70 percent of the original forest cover is gone
(Zimmer 2007), whereas the whole mountain range was earlier covered with dense forest
(Figure 4). There are estimations that already over 30% of the Eastern Arc Mountain forest
species have become extinct or are in danger of future extinction (Newmark 1998). The
extinction of understorey birds was claimed by Newmark (2000) to be due to removal of
understorey from forest cultivations. A decrease of air humidity has been associated with
forest clearance (Hamilton and Bensted-Smith 1989); this was also claimed by the local
people in the present study.

24
Figure 3. A local farmer picking mature cardamom capsules in the East Usambaras.
(Photograph: Teija Reyes).

Figure 4. Cardamom cultivation in natural forest after selective thinning of the tree canopy in
the East Usambaras, Tanzania. (Photograph: Teija Reyes).

25
2.3. Previous Finnish activities in the East Usambara Mountains

2.3.1. Project activities in the East Usambaras

Finland was for the first time involved in the East Usambaras in 1977-1978 by supporting a
forest inventory. Development of forest industries was then an aim of the development co-
operation, and forests in this area were mainly considered as raw material sources for the
plywood and sawmilling industries. The Sikh Saw Mills project co-financed by the Finnish
Government started to harvest trees in natural forests following a management plan based on
the forest inventory. It was believed then that the sub-montane rain forests could sustain
heavy logging methods.

In the mid 1980's national and international concerns rose on the effects of commercial
logging on the East Usambara forests and led to the termination of logging activities. Finland
and Norway financed in 1986-1987 the Amani Forest Inventory and Management Plan
Project, which concluded that the timber harvesting had not been at a sustainable level.
Threats that were observed included the following: a decrease in biodiversity value, extinction
of endemic species, and general destruction of the forests. A new management plan focused
on nature conservation and gentle management of watershed areas.

This plan served as a basis for the East Usambara Catchment Forest Project (EUCFP),
implemented from 1990 to 1998. The project aimed at establishing the Amani Nature Reserve
(ANR), protecting water sources, establishing and protecting of other forest reserves, ensuring
local community benefits from the forest, and rehabilitating the Amani Botanical Garden. The
project was implemented by the Forestry and Beekeeping Division of the Tanzanian Ministry
of Natural Resources and Tourism with financial support from the Government of Finland
and implementation support from Metshallitus (formerly known as the Finnish Forest and
Park Service). The Amani Nature Reserve was gazetted on 9 May 1997. Thereafter the main
task of the EUCFP was to prepare a management plan for ANR (Eastern Arc Mountains
Information Source 2001). The project continued as the East Usambara Conservation Area
Management Programme (EUCAMP), with planning and implementation of the ANR
Management Plan and conservation activities in the East Usambaras, until the end of year
2002. The only difference as compared to the former phase was the support from the
European Union, mainly to implement compensation for the proposed Derema Forest and
Wildlife Corridor.

26
EUCFP and EUCAMP resulted in comprehensive biodiversity surveys, a substantial decrease
in illegal harvesting, and better control over forest fires in the area. They also provided
capacity building to government staff, supporting professional education and in-service
training. The projects also promoted the establishment of tree nurseries and fish ponds,
beekeeping, soil conservation, and the use of wood-saving stoves. Ecotourism guides were
trained to escort tourists wishing acquaintance with Usambaran nature. Participatory forest
management and joint forest activities were initiated, with Mpanga being the first village
forest reserve, gazetted in 1995. Projects also engaged district councils in the work of
participatory forest management, which proved to be important for the continuity of the
process (White and Mustalahti 2005).

2.3.2. Previous studies in the East Usambaras

Earlier studies in the East Usambaras have used the concept of "functional groups" for
classifying plant species in different production systems, including agroforestry, and
specifically, for quantifying the effects of different production systems on biodiversity
conservation in adjacent natural forests (Huang et al. 2002). In particular, production system
components were determined as having conservation, livelihood or ecological functions.
Conservation functions were determined based on the degree of reduction in the pressure on
the natural forest for obtaining goods such as fuelwood or timber. Livelihood functions
included the provision of subsistence commodities or income from cash crops. Ecological
functions were related, for instance, to erosion rates caused by each production system.

For the case of conservation functions in the East Usambaras, Huang et al. (2002) found that
agroforestry systems were more efficient than tree plantations. Agroforestry systems
consisting of trees and agricultural crops were also ranked high by their livelihood and
ecological functions, as compared to monocropping systems. A final conclusion from this
earlier research was that farmers in the East Usambaras who earlier much relied on the natural
forest for their livelihoods, must have alternative, new sources for their subsistence needs and
cash income that can reduce the dependence on the natural forest. New, improved
agroforestry systems were presented as a viable solution.

27
Two other earlier studies, also carried out by researchers at VITRI, University of Helsinki,
and conducted in the East Usambara Mountains in 2000, were used as a starting point of the
present research. One was a study on black pepper cultivation methods and the other an
investigation on biodiversity in agroforestry systems. Differences in the condition of black
pepper depending on the support tree species were noticed, especially at low altitude. Of the
75 tree species tested, the nitrogen-fixing Gliricidia sepium was graded as one of the best
support trees for black pepper at low altitude and as the second most popular tree on the
farms; it was recommended to extend the cultivation of gliricidia to higher elevations in the
Usambaras (Vihemki 2001).

In the other study, farmers were found to cultivate 104 different woody plant species, of
which the most common was Grevillea robusta. Most of the farm production was for the
households own use, and the most common products were firewood, vegetables, bananas and
cassava. Products grown for markets were most often spices (Risnen 2001). Grevillea was
ranked eighth as far as the black pepper support trees were concerned and mostly used at high
altitude (above 850 m; cf. Vihemki 2001).

In the whole African continent, spice cultivation has not been much studied, even if there
exist the so-called spice islands of Zanzibar or Madagascar. ICRAF (the World Agroforestry
Centre) in Nairobi has confirmed the lack of spice-crop agroforestry research in Africa 2.
Studies on black pepper and cardamom cultivation have been made, however, in India,
especially by the Spices Research Center and the Indian Cardamom Research Institute, but
their research results are not readily available.

2.4. Improved agroforestry systems as a tool in sustainable development

2.4.1. Intercropping in Tanzania

Intercropping is one of the major technological inventions of Africa. African agriculture is the
result of 4 000 years of evolution, and home gardens are the oldest kind of permanent
production systems in tropical Africa (Okigbo 1990). In pre-colonial Tanzania, intercropping
was the main agricultural method; several different species and different varieties of the same
species were planted simultaneously on the same plot. The attacks of pests and diseases were

2
Pers. com. with Richard Coe, Head of Research Support Unit, ICRAF, 7 April, 2005.

28
minimized with this method. In traditional intercropping, manure and compost were used
whereby the soil moisture was maintained for a longer time. A cyclic rotation of crops was
also practised. A new cultivation cycle was started with legumes (groundnuts - Arachis
hypogaea L. or beans) and tuber crops (sweet potato - Ipomoea batatas L.) because of their
positive effects on soil fertility. After them, crops needing nutrient-rich soil (maize Zea
mays L., pearl millet Pennisetum typhoides (Burm.) Stapf and Hubb. and sorghum -
Sorghum bicolor (L.) Moench, were cultivated for a few years. When the fertility again
declined, less-demanding crops (such as cassava - Manihot esculenta Crantz) were grown
before the land was left to fallow (Koponen 1988). It is worth mentioning that, of the above
crops, groundnut, sweet potato, maize and cassava are all introduced crops in Africa.

In the Usambaras the pre-colonial cultivation system was based on bananas and plantains
(Musa spp.), but it changed to maize and rice (Oryza sativa L.), both being crops emphasized
in the colonial era. The introduced food crops often replaced the traditional ones and made the
cultivation methods simpler but also less sustainable. The same crops that were grown in the
lowlands were also grown in the highlands, but in different cycles. Irrigation systems were
constructed by using watercourses, canals and ditches. The Usambaran farmers were said to
be excellent hydrotechnicians, who were able to skilfully use the topography of the land for
conducting water to the fields (Koponen 1988).

The Chagga people in Kilimanjaro have usually been given the highest praise in managing
and irrigating upland gardens in Tanzania. They developed the agroforestry homegardens,
which comprise an integrated crop, tree and livestock system that has remained productive
and sustained high population densities (500-700 persons/km2, with an annual population
growth reaching at least 3%) with little degradation. Native forest species that provide fodder,
fruits, fuel and timber have been retained and the less useful ones have been replaced by
exotic trees and crops. The main cash crop in Kilimanjaro is coffee, but these homegardens
consist of over 100 different species growing in four to five distinct canopy layers. Some crop
failures have occurred every 3-4 years, but they never result in total failure. Honey from
stingless bees is a valuable addition to the farmers income (Fernandes 2000).

Many farmers in Tanzania still feel that trees are something they have always harvested
freely. They do not want nor need to plant them to their fields to compete with agricultural
crops, as they have worked hard to remove them from the site when converting forest to
farmland. Poor farmers strive to rewards for their work during the next growing season; they

29
do not want to wait up to ten years to harvest a tree. In the highlands, farmers have, however,
grown shade trees over coffee, and bananas have been always intermixed with annual food
crops, such as maize and beans. Another common feature has been that farmers have worked
in partnership with the government in industrial tree plantations. The farmer cleared the land,
grew annual or short-term food crops, and planted tree seedlings between the food crops. As
the trees grew larger, the space for food crops became limited, and the farmer was given a
new piece of land (Keswani and Ndunguru 1982). This is the way that many tree plantations
have been established in Tanzania 3. As forest resources become scarcer trees outside forests 4
are becoming more and more important for people.

Homegardens in many countries have been time-tested to be both productive and sustainable.
These multispecies agroforestry systems provide goods and services to people, especially in
the humid tropics. They have similarities to natural forest ecosystems, which acts as a
safeguard against pests and diseases. Homegarden products are considered to have a quality
superior to other products, since no chemicals are normally used in their cultivation, and they
are an important complement in nutrition, particularity for children. However, they seldom
provide the entire food supply of the family (Kumar and Nair 2004).

2.4.2. The role of agroforestry in the conservation of mountain ecosystems

New farms cleared in upland forests are difficult to maintain productive because of climatic,
topographic, and socio-economic conditions that easily lead to over-exploitation of the
resources and to site degradation. The unsustainability of land use systems in the uplands is
associated with an increasing population of subsistence farm families cultivating infertile
soils, accelerating land degradation, and soil erosion. Adding mineral fertilizers to the soil is
not possible for poor farmers. More productive and sustainable land-use systems must be
developed under these conditions of severe social, environmental and economic constraints. It
has been suggested that the most viable alternative for maintaining the soil productivity in the
tropics is the application of low-cost agroforestry systems that promote soil conservation and
minimize nutrient losses, especially on marginal lands. Agroforestry seems to be among the
most promising sustainable alternatives to slash-and-burn farming or shifting cultivation
(Garrity 1993).

3
The system is widely known as taungya, or (in East Africa) shamba forestry.
4
FAO (2000) is defining them as trees found on non-forest and non-wood lands such as agricultural lands,
urban and settlement areas, roadsides, homegardens, hedgerows, pasture/rangelands and scattered in the
landscape (FAO 2000).

30
Uplands have the potential to destroy the productivity of the lowlands if the former are
improperly managed. Differences in the ecological requirements between annual and
perennial crops could minimize this mutual competition and allow their cultivation in
intercropping systems. Fast-growing annual crops mature quickly and can be harvested
shortly after planting, but their continuous management can cause soil erosion on steep
slopes. In contrast, the slow-growing perennials need a long time from planting to harvest, but
they normally require little maintenance and could minimize soil erosion. In intercropping of
annuals and perennials the soil surface is well covered, which decreases the erosion caused by
rain. In addition, the intercropping system constantly adds organic matter to the soil. It is
recommended that annual crops should not be grown on badly deteriorated areas, and steep
slopes should never be without vegetative cover (Gomez and Gomez 1983).

The only way to reduce soil erosion on hilly land is to decrease the rate at which water moves
down the slope. Slowing the water movement on slopes by regulating the vegetative cover
ensures that water is available for plants for a longer time, the vegetation becomes denser and,
with time, the soil properties for crop growth improve. The role of trees in maintaining and
improving the soil productivity is considered central to the sustainability of many agroforestry
systems. Much of the evidence related to soil quality is indirect, but it indicates more stable
structure, better water holding capacity, increased resistance to erosion, and good overall soil
fertility (Nair 1984; Young 1987; Sanchez 1987).

Agroforestry is recommended for mountainous terrain in the tropics and subtropics, because it
has always two functions: production (fuelwood, timber, fodder and other non-timber forest
products) and services (increase in soil fertility, erosion control; Beets 1990). The
multifunctional nature of agroforestry can solve several problems simultaneously (Lundgren
and Raintree 1983). Evans (1982) has listed some advantages of growing crops under tree
canopies in the tropics as follows:

1. Climatic damages are reduced under shade;

2. Soil erosion is reduced;
3. Less undergrowth;
4. Often some reduction in pests and diseases;
5. More pleasant working conditions under shade.

31
2.4.3. Improved agroforestry systems

Tree growing on farms has been practiced already for centuries in the tropical regions. The
ancient cropping practices, including agricultural crops mixed with tree species, obtained the
new collective name, agroforestry, among scientists only in 1977 (Lundgren and Raintree
1983). This is a complex research area because of a broad range of disciplines involved, from
social sciences and economics to agronomy. Each agroforestry system is unique, combining
the experience and knowledge of forestry, tropical agriculture, ecology, soil science and rural
socio-economics (Lundgren and Nair 1985). Traditional practices have often become less
diversified, unproductive and ecologically damaging. Improved agroforestry tries to take the
best out of the traditional agroforestry methods and combine them with new scientific
findings and inventions.

To improve the traditional farming systems, emphasis has been given to measures promoting
sustainability and serving as intermediate steps towards more permanent cultivation. This
includes various soil conservation, crop rotation and intercropping techniques, and it also
commonly means intercropping particularly with nitrogen-fixing leguminous tree species.
Since subsistence farming systems in developing countries are ecologically, and often also
economically, becoming less sustainable, there is an increasing interest towards more
productive agroforestry techniques as a practical alternative to the traditional systems.
Agroforestry has shown its potentials as a land management alternative for maintaining the
soil fertility and productivity in the tropics. It is known that the soil organic matter and other
biotic factors are important in maintaining the productivity especially in the fragile humid
tropical uplands (Kang 1993). Home gardens can serve as models for the design of improved
agroforestry practices (Kumar and Nair 2004).

Interaction between components of an agroforestry system means that a plant and its
environment would modify each other to the extent that the environment causes a response in
plant function and growth, and the plant then has an effect on the environment by changing
one or more of its factors (Huang 1998). The nature of the interactions within and between
species concern the ways in which a plant can influence its coexistence gain by changing its
environment, either directly by addition or subtraction (e.g. of water and nutrients), or
indirectly (e.g. by favouring or hindering pests and diseases). Competition in an intercropping
situation is defined as an interaction between two individual plants that reduces the fitness of
one or both of them. When trees are added to a cropping system there are a number of
possible outcomes of the interactions between trees and crops (Andersson and Sinclair 1993).

32
Huang (1998), in his coexistence gain research in China, confirmed that deliberately designed
agroforestry systems can have a positive interaction that contributes to sharing of the available
resources and to increasing the total land productivity.

Intensification of land use needs improvement of husbandry by investing in more labour or

other inputs, such as fertilizers, into existing cropping systems. It can also involve a new
pattern of planting, pruning or harvesting the component plants. However, in many situations
farmers do not seem to be willing to put in extra labour. Often they do not have any extra
resources available, and sometimes they are unaware of the benefits of more intensive land
use (Beets 1990). Farmers need to perceive the concrete results from new cultivation methods
before they can adopt them. They must also be able to adopt a new time scale and consider the
long-term impacts of new methods. Sustainability means an ability to produce a stable annual
yield of the desired crop over a long period of time (Sanchez et al. 1997). Sustainable
development should meet the various needs of the present and future generations. These needs
can be social, economical, cultural or spiritual. Livelihoods are considered sustainable when
they can provide protection against unusual incidents and maintain their capabilities currently
and in the future. This multi-dimensional approach to well-being recognises as key factors the
income, health, education and vulnerability (Mery et al. 2005).

In natural forests the nutrients are efficiently cycled with small gains and losses in the system.
In most agricultural systems the opposite happens, and between these two extremes there are
the agroforestry systems (Sanchez et al. 1997). Trees recycle nutrients, protect soils, and
provide various products (fruit, vegetables, fodder, spices, medicines, oils, nuts, fibres,
fuelwood, and timber) which can generate income especially for smallholders. Moreover,
with more trees growing on farms, the pressure on the tropical forests can be released. Deep-
rooted nitrogen-fixing, multipurpose trees growing together with agricultural crops offer
benefits in terms of soil fertility. They suit the poor farmers who are not able to buy fertilizers.
With trees it becomes easier to attain sustainable development, which is also the uttermost
goal of agroforestry research (Land and People 2003).

33
2.5. Some constraints for adopting better land-use management in Tanzania

2.5.1. Short-term thinking

There have been various conservation projects in Tanzania to change the land management
towards more sustainable practices, but the results from adopting new soil conservation
technology are disappointing. In many projects, incentives have been used to encourage
farmers in the first years when tangible results are not yet clearly seen. In general, incentives
can be direct (cash payment for labour, grants, food, agricultural materials) or indirect (secure
access to land, tax concessions). In theory, when farmers see the benefits of a new technology,
the incentives can be phased out. Often, however, long-term changes in attitudes and values
are not attained; farmers commonly apply new technology only as long as the incentives are
provided (Hellin 2006). Often the donor-financed projects are too short in duration, lasting
only three to four years.

Improved cultivation systems, e.g. agroforestry, require medium to long-term investments

(Angelsen and Kaimowitz 2004). The land tenure system in Tanzania has placed constraints
on the long-term investment in land that would be vital for increasing the agricultural
productivity (Masayanyika 1995; Sah 1996; Edwards et al. 1997; Msikula 2003). About 30%
of the farmers in the East Usambaras are tenants on leased land, and thus farmers have no
long-term commitment to land management. When the government owns the resources, the
local communities also have greater incentives for illegal exploitation of the forest resources,
because there are no individual costs involved.

Farmers in Sub-Saharan Africa may have limited interest in long-term investments also due to
the prevailing product prices, climate risks, or a lack of stable markets. An improved access to
markets may encourage the farmers to practice unsustainable land management for gaining
short-term profits. The more farmers sell products from their farms, the more severe the soil
nutrient depletion becomes. In Uganda, commercially oriented farmers have experienced
worse nutrient depletion than the subsistence farmers. Farmers with better market access
tended to produce non-food crops, which contributed to increased soil nutrient depletion when
the use of fertilizers was insufficient. Participation in extension programmes increased the
crop production and household income significantly, but it also led to higher soil nutrient
depletion (Nkonya et al. 2004).

34
There are also new constrains in the world leading to farmers short-term thinking.
Commodity prices have until recently fallen in the open international markets, and the
HIV/AIDS epidemic is affecting households socially and economically. Investments in
training and some important skills can be lost with HIV/AIDS victims (White and Mustalahti
2005).

The climate change is assumed to exacerbate the effects of varying weather conditions. In the
East Usambaras, the farmers used to have two annual growing seasons for crops, but in recent
years they only have had one. Farmers here also claim that a decrease in humidity and higher
temperatures occur in the mountain areas. Mango and citrus trees are now common in the
mountains in places where they could not survive before (Hamilton and Bensted-Smith 1989).

2.5.2. Lack of resources

About 70% of the farmers in Tanzania cultivate less than one hectare of land (Uliwa and
Fished 2004). The custom to distribute the land among the children in a family has resulted in
land holding fragmentation and economically inefficient farms (Kaoneka and Solberg 1994).
Long distances between parcels belonging to the same farmer hinder better land management.
Farmers are less likely to care for soil fertility using manure on distant farmlots and more
likely to use slash-and-burn for land preparation, which leads to increased nutrient loss and
erosion (Nkonya et al. 2004). Cultivation is normally done with family labour, and labour
shortage is often a reason why improved practices are not adopted (Angelsen and Kaimowitz
2004).

Inadequate access to capital and credit is commonly considered as a major constraint for
increasing the household production and income, e.g. to invest in more efficient land use. It is
common in Tanzania and in other developing countries that local money-lenders charge very
high private discount and interest rates (Pannell 1999). Consequently future benefits become
less important than the daily struggle for survival, and investments for agriculture become
associated with high risks. Farmers willingness to invest in soil improvement is closely
associated with the overall economic profitability of farming (Scherr 1999; Angelsen and
Kaimowitz 2004). Off-farm activities can lead to more investment in soil and water
conservation (Nkonya et al. 2004) and to hiring labour, because a lack of manpower often
prevents any improvement of the cultivation methods.

35
For any improvement of the farming practices, the essential public goods funded by the state,
such as education, health care, agricultural research and extension, and the maintenance of
rural roads should be in place (Hazell et al. 2007). Transportation of better planting material,
fertilizers, tools, and extension services to remote fields requires roads that are maintained in
good condition. The remoteness of health services consumes time and money in the case of
illness and decreases the availability of a workforce for farming.

Densely populated marginal lands need increased nutrient inputs and improved nutrient use
efficiency. The fertilizer use has not increased among Tanzanian smallholder farmers since
the early 1980s. In 1994 the amount of fertilizers used was about 10 kg/ha in Tanzania, 18
kg/ha on average in Africa, and 94 kg/ha in the world. While improved seeds are usually
adopted rapidly, the use of chemical fertilizers is not adopted until farmers have gathered
knowledge about this practice. It has been concluded that farmers adopt new land
management practices when they realize the effect of land degradation on crop production
(Current and Scherr 1995). In other studies it has been found that farmers will not shift to
alternative systems unless there is a good prospect for monetary gain either by increasing the
output or by lowering the production costs (UNDP 1995).

2.5.3. Low education level and inadequate extension

A low education level can be a barrier for agricultural development, since education normally
has a significant influence on a households income strategies, land management and labour
use (Nkonya et al. 2004). There is lack of knowledge of effective means for soil
improvements, like basic information about the farmyard manure application and compost
preparation. On many farms in Tanzania, the weeded grass is still removed from the cropland
and destroyed by burning. There is a direct relationship between soil loss and productivity,
but many times farmers are not identifying soil erosion as a reason for soil degradation.
Farmers are just complaining that the yields decline.

In Tanzania the extension services are commonly less than adequate. Crop cultivation skills
are based on the farmers indigenous knowledge, without external support to the farmers
initiatives. The National Forest Policy of Tanzania of 1998 proposes strengthening of the
extension services, but they are lacking resources and fragmented among different sectors

36
(URT 1998). Forestry and agricultural extension are working separately, which is a
considerable constraint for the agroforestry extension. The question remains, who should
provide extension in agroforestry management and spice cultivation (some spices are
agricultural crops, while others are trees)?

Agricultural training and extension programmes have been found to improve the agricultural
productivity in Uganda. It was concluded that they have to be intensive enough for the soil-
fertility restoration and soil conservation technologies to become adopted. Inadequate
extension services may contribute to unsustainable land management practices if the farmers
adopt improved crop varieties without the soil fertility management measures needed for the
cultivation of such high-yielding crop varieties. Partial adoption of improved measures leads
to short-term gains only and can accelerate the land degradation caused by intensive farming
(Nkonya et al. 2004).

In Peru in the 1970s and 1980s the government policies supported the use of chemical
fertilizers without proper extension; this led to serious land degradation in many villages in
the Andean mountains. Peruvian farmers have used chemical fertilizers and pesticides also on
steep slopes and in other vulnerable areas, which has exacerbated the erosion situation,
decreased the soil fertility, and caused contamination of watercourses. Now many projects try
to revitalize the traditional Andean sustainable agriculture techniques 5. Very few of these
vulnerable mountain soils can be managed sustainably with chemical fertilizers.

2.5.4. Governance shortfalls

Governance shortfalls, including corruption, are major factors limiting development in

Tanzania. Improper practices of government officers and village leaders (involved in illegal
timber harvesting) have led to a strong mistrust between villagers and authorities. The reasons
for corruption to be associated with natural resource utilisation come from history. Natural
resources have for a long time been considered as a common and free property, and that is
why governance shortfalls in their use were not taken seriously. Unpaid and generally low
salaries and poor living conditions encourage people to resort to self-interest actions

5
Pers. com. with Leena Hokkanen, Director of Diaconia, Lutheran Church Aid, Peru, 15 May, 2005.

37
(Milledge et al. 2007). In 1998, Tanzania was ranked as the fourth most corrupt country in the
world (Transparency International 2006).

The central government and the districts in Tanzania loose revenue in the form of under-
collected natural forest product royalties to a total value of US$ 58 million (about 45 million
euros) per year (Milledge at al. 2007). In Tanga Region 9.6-12.6% of the royalties from the
estimated revenues were collected (Kobb 1998). An example from Kilwa District, one of the
poorest areas in Tanzania, shows that it would have increased its total annual budget four-
fold, if all potential timber revenues had been collected. Only about 4% of the royalties were
collected, due to the prevalence of illegal logging (Milledge at al. 2007).

Less than 3% of the rural households in Tanzania are affiliated with active producer
organizations. In these organisations there have been problems with poor management,
undemocratic practices, and especially a failure of the cooperative unions established by the
government (Uliwa and Fisher 2004). During the past twenty-five years, there have been no
active business associations in Tanzania, as all efforts have been directed towards supporting
the public sector. Private-sector business associations were ignored and left to wither
(Edwards et al. 1997).

Some experts think that it is not realistic to continue to prioritise agriculture in poor countries
at all. In Tanzania the sector is much behind the rest of the world in productivity, and there is
an increasing competition in the trade in agricultural commodities in the world market.
However, many times there is no alternative to farming as an activity capable of creating jobs
and raising incomes in remote rural places. Some analysts even have an opinion that there is
no hope for small farms in the actual world and all the emphasis should be put on developing
larger farms with high-value products (Hazell et al. 2007). However, comparisons made
across many developing countries show that increases in agricultural productivity lead to
poverty reduction (Oksanen et al. 2003). Unfortunately current policy preferences are in other
sectors than in agricultural and forestry development. Some donors have shifted towards
direct budget support linked to a poverty reduction support program, where low priority has
been given to agricultural investments. This can be a serious threat to agriculture. Technical
weakness in the Ministry of Agriculture has failed in promoting agricultural issues as national
priorities, in contrast to better-organized ministries such as those for health and education
(Hazell et al. 2007).

38
The traditional cultivation and land-use methods in Tanzania ought to change also because
land is continuously getting scarcer. It is estimated that within the period of 1995-2020 the
population in Africa will almost double. The average per-capita landholding will decrease in
size from 0.3 ha to 0.1-0.2 ha (Scherr 1999). Any long-term solution to deforestation and land
degradation should take into account all key aspects related to farmers livelihoods: the need
for more efficient and more productive agriculture on suitable farmland; better forestry
practices, including those needed for tree plantations; and the establishment of forest reserves
to protect species and ecosystems (Mastrantonio and Francis 2004).

This conclusion is valid also for the forest area of the East Usambaras. Here forests in
conservation areas are protected, and some forest plantations (mainly with teak or eucalyptus)
are found, but more effective and profitable farming practices are urgently called for.
Measures for biodiversity conservation should also consider that the highest densities of
human population are found near sites that are most valuable for biodiversity management
(Burgess et al. 2007a), and that the people are poor. It is clear that if the unprotected forests
become completely depleted (mainly because of timber and charcoal utilisation), the pressure
towards reserved forests will further increase.

2.6. Background to the agroforestry species studied

2.6.1. Queen of spices, cardamom (Elettaria cardamomum Maton)

Small cardamom (Elettaria cardamomum Maton; Figure 5), one of the worlds most ancient
spices, is already mentioned in Sanskrit texts ca. 3000 BC in India (Ravindran and
Madhusoodanan 2002). This Queen of Spices belongs to the ginger family (Zingiberaceae)
and is the third most expensive spice in the world, after saffron (Crocus sativus L.) and
vanilla (Vanilla planifolia, V. fragrans or V. vanilla Jacks.). It grows native in the southern
Indian forests of the Western Ghats. Being sensitive to wind, drought and waterlogging, it
reaches the optimum yield on warm (10-35C) and humid (with >1 500 mm of well
distributed rainfall) mountain slopes at 600-1 500 m elevation, under a canopy of evergreen
trees. Cardamom has already been commercially cultivated in the Western Ghats for 150
years, and India has had a virtual monopoly of its trade until recently (Nair and Kutty 2004).

39
Spices were also the primary reason why the sea route from Europe to the Far East was
opened. Before actual plantations had been established, cardamom was harvested as a forest
product. It was known that felling few large trees encouraged the growth of the crop. The
largest producers of true cardamom are presently Guatemala and India, and smaller producers
include Tanzania, Sri Lanka, Papua New Guinea, El Salvador, Laos and Vietnam. Cheaper
substitutes to real cardamom (from Amomum spp. and Aframomum spp.) are grown and used
in some Asian countries (Nair and Kutty 2004). Their aroma and flavour are, however,
inferior if compared with true cardamom (Purseglove et al. 1981).

India and Saudi Arabia consume more than half of the worlds total cardamom production. In
the Arab countries and India, cardamom is a common flavouring ingredient for coffee and tea.
In the Scandinavian countries, as well as in Germany and Russia, it is used to flavour cakes,
pastries and sausages. It is popular in South and Southeast Asian cooking and used to make
spice blends, such as curries and garam masala. Chewing cardamom after a meal is
recommended to aid digestion and to clean the teeth in India. In Eastern medicinal practices it
is used for curing such ailments as flu, infections, asthma, bronchitis, cardiac disorders,
diarrhea, nausea and cataracts, and for strengthening the nervous system. It is also said to
have a cooling effect in a hot climate. Ancient Greeks and Romans already used its delicate
aroma to make perfume (Ravindran and Madhusoodanan 2002).

The German settlers introduced cardamom in the East Usambaras in 1890s. In 1954 the
Amani Botanical Garden distributed ten seedlings to farmers who vegetatively propagated the
material for other farmers. After a decade the cardamom business started to flourish. The local
Washambaa people and immigrants from the nearby highlands (West Usambaras, Pare
Mountains, Iringa, Moshi and Kilimanjaro) are presently the main cardamom growers in the
area. The cardamom stock in the East Usambaras belongs to three varieties (Thwarites,
Malabar and Mysore) or to their hybrids (Sah 1996). In many parts of the East Usambaran
highlands more than 90% of the households depend on cardamom as a source of cash income
(Sah 1996).

After two and a half years, small cardamom is ready for harvest, and the fourth year normally
gives the highest yield, about 125 kg/ha. Farmers maintain their one-hectare cardamom farms
by weeding and trashing (removing the old and dry shoots). Only creamy/white cardamom is
produced in the area nowadays; it is obtained by drying fresh cardamom fruits in the sun.
First-grade green cardamom, which is fire-dried, is no longer produced, due to lack of

40
firewood. From 1950-1970, fire was the dominant method for drying the seeds, also perhaps
because the climate in the mountains was moister than now. Normally the capsules are dried
on mats near a road, thus increasing the amount of grit among the seeds.

Figure 5. Cardamom (Elettaria cardamomum) in the Amani Botanical Garden. (Photograph:

Teija Reyes).

The whole production is normally exported from the country. The average annual cardamom
yield in the East Usambaras is 80 kg/ha (Masayanyika 1995); nonetheless, there is empirical
evidence that under good land management and by adding manure it could easily reach about
one kg per stump and year, giving an average of 300 kg/ha. In the mid-1970s, the East
Usambaras were still producing about 760 t of cardamom per annum (equalling 20% of the
total world production), which made Tanzania the third largest producer after India and
Guatemala. Guatemala took the leading position from India in 1980. In 1998 Guatemala
already produced 64% of the total cardamom traded globally, while India supplied 30% and
the other cardamom producing countries 6% (Ravindran and Madhusoodanan 2002). In 2002,

41
 6
Tanzania exported 560 t which is very little compared with Guatemalas exportation of 19
000 t that year.

Cultivation of cardamom has often been observed to be unsustainable. Cardamom is mostly

grown in the fragile higher parts of the mountains (above 850 m), and it is expanding to less
suitable sites. Cultivation is done on the basis of indigenous knowledge, wherein farmers have
designed their own crop husbandry techniques based on trial and error over time. Cardamom
cultivation is still mostly practiced in the forest after clearing totally the understory and the
middle layer, and selectively thinning the top canopy. Farmers leave 75-100 trees/ha standing
to provide shade, to increase the soil moisture and to ensure a variety of tree products.
Beneficial for cardamom and providing shade and increasing the moisture are e.g.
Allanblackia stuhlmanii Engl. (yielding oil nuts), and Newtonia buchananii (Baker) Gilb.
(producing good timber). Unfavourable for cardamom are trees that contain toxic substances
in seeds or leaves or which tend to dry up the soil, such as Pterocarpus tinctorius Jacq. (with
poisonous seeds), Afrosersalisia cerasifera (Welw.) Pennington (drying up the soil), and
Maesopsis eminii Engl. (Masayanyika 1995).

The cultivation becomes unprofitable after about seven years (Sah 1996), or at some sites
even after one production year (three years after planting), while the experience from other
countries shows that the cardamom crop remains productive until an age of 10 to 15 years
(Nair and Kutty 2004). Farmers response to low cardamom production has been a conversion
of new forest areas to cardamom cultivation while the old area is cleared completely and
converted to annual crops. This practice of shifting cultivation aggravates the problem of
depletion of valuable forest resources. Continuous removal of the forest ground layer leads to
chronic degradation of the forest (Ashton et al. 2001).

2.6.2. King of the spices, black pepper (Piper nigrum L.)

Black pepper (Figure 6) is one of the oldest and the worlds most important spice. Black
pepper is unique in the spice world, because its fruits can be marketed in four different
versions. They can be processed to give black, white, green and pink/red peppercorns.
Pink/red pepper from Piper nigrum is distinct from the more common dried "pink
peppercorns" or ros pepper, which are the fruits of a plant from a different family, the

6
Pers.com. with the Board of External Trade, Tanzania (22 May 2004).

42
Peruvian pepper tree, Schinus molle L. By choosing the time of harvest and the processing
method carefully, all four types could, in principle, be produced from the same pepper plant.
Pepper is native to the Malabar Coast of south-western India. It was one of the first articles of
trade between the East and Europe. Black pepper was called also black gold, because in the
past it was ranked equal with gold and used as barter money. In ancient times only rulers and
the highest classes of society were allowed to possess it. Spices, especially pepper, attracted
the western world to India. India still has approximately 48% of the pepper cultivation area of
the world, but its share in the production is only 27%. The productivity of pepper is highest in
Thailand, Malaysia, Indonesia, China, Vietnam and Brazil (Ravindran 2000).

Dried ground pepper is one of the most common spices in world cuisine, having been known
and prized since antiquity for both its flavour and its use as a medicine. The essential oil of
pepper is used for aching muscles, arthritis, constipation, muscle cramps, poor circulation and
sluggish digestion (Lawless 1995).

Piper is a large genus, with over 1 000 species, in the family Piperaceae. Piper nigrum is a
woody climber to 10 m or more in height. It is essentially a crop of the wet tropics. As pepper
is a climbing vine, it needs a support. The simplest and oldest form of cultivation used in
India is to plant the pepper cuttings at the foot of a tree near a house. Pepper may also be
grown as a secondary crop in tree stands. The production of the crop increases with time and
can continue up to 40-60 years, having a peak at about 15 years of age (Purseglove et al.
1981).

Black pepper grows successfully up to 20 degrees North and South of the Equator and from
the sea level up to 1 500 m elevation. It is a plant of the humid tropics, requiring 2 000-3 000
mm of rainfall, a tropical temperature (ideal 23-30 degrees) and a high relative humidity
(ideal 60-95%) with little variation in day length throughout the year. It does not tolerate
excessive heat or dryness. Black pepper vines require support for their establishment. An ideal
living standard should be straight, have a tap root system, and not compete with black pepper
for water, nutrients and solar radiation. It should have a strong trunk with a rough surface and
be able to withstand regular pruning; it should also have economic value after the life span of
black pepper (Ravindran 2000).

43
Figure 6. Black pepper (Piper nigrum). (Photograph: Teija Reyes).

As a vine, pepper is one of the possible crops to be grown in multi-storey or inter-cropping

agroforestry systems that are recommended to prevent soil erosion. Black pepper cultivation
was promoted in the area by the East Usambara Conservation and Agricultural Development
Project (EUCADEP), which was implemented by IUCN at the end of the 1980s, and now it is
one of the most important cash crops cultivated by farmers in agroforestry systems. Almost
60% of the farmers in the research area are growing black pepper as a cash crop. The first
black pepper plants were introduced to the area already in the early colonial period, like
cardamom, when several cash crops were experimented with. Black pepper and cardamom
are, however, newcomers on the farms of the East Usambara Mountains. Only few farmers
indicated that they had cultivated black pepper for more than 30 years (Vihemki 2001). The
most common support tree for black pepper in the East Usambaras has been Cedrela odorata
L. (Owen 1992; Ellman 1996). Tanzania exported 3 300 t of black pepper in 2002 7.

7
Pers.com. with the Board of External Trade, Tanzania (22 May 2004).

44
Sustainable intercropping of black pepper in Sri Lanka is associated with mature coconut trees
(Cocos nucifera L.). The combinations of coconut-coffee-black pepper or coconut-cacao
(Theobroma cacao L.)-coffee-black pepper are perceived as very effective and sustainable,
particularly if extra organic matter is added, because coconut does not add much organic
matter. These intercropping systems are economically advantageous, and the other species do
not normally much affect the growth of coconut, which would normally grow in monocultures
(Liyanage et al. 1989).

Generally the first harvest of pepper is obtained during the third year after planting. The black
pepper berries are mature when they easily separate from the spikes upon rubbing by hand or
when one or more berries turn red on a spike. The number and frequency of harvests depend
on crop vigour, season, level and distribution of the rainfall, and on the solar radiation. During
the last harvest all the spikes should be collected to ensure even and uniform bearing during
the next season. The yield of black pepper varies greatly in different areas. It depends on the
elevation, temperature, rainfall, soil fertility, cultural practices, age of the crop, and on the
climatic conditions during flowering, fruit set and development. Black pepper is well adapted
to grow as an under-crop, mixed crop, or intercrop with plantation crops. Weeds are normally
the major problem in black pepper plantations, whereby weeding is essential for the first
years. Production technologies adopted at a particular location may not be suitable for other
ones, and there is generally not enough information on the various cultivation techniques
(Sivaraman et al. 1999).

The markets of black pepper are stable, because there is no replacement for pepper products
(also including insecticides, cosmetic products, perfumes, beverage, paintings, medicines, and
fungicides). Synthetic colour or aroma is not preferred in pepper spice. USA is the main
importer of pepper, followed by Germany, the Netherlands, France and Russia. The world
pepper demand is increasing at the rate of 2.7% per year. In 50 years the producing countries
should double their productivity to meet the global demand. White pepper has demand in
Japan, Europe and Indonesia. The more developed black pepper producing countries, such as
Malaysia, Thailand and Brazil, are moving away from labour-demanding pepper production,
whereby the future pepper production and trade countries will most likely be India, China
Vietnam and Indonesia. As the land availability decreases, the production per unit area should
increase. Hence it is necessary to develop high-quality and high-yielding cultivars (Ravindran
2000). Organically grown black pepper fetches premium prices at the international market
(Sivaraman et al. 1999).

45
2.6.3. Grevillea robusta, the multipurpose agroforestry tree

Grevillea robusta A. Cunn. ex R. Br. is a tree native to the subtropical eastern Australia and
belongs to the family Proteaceae. As the largest species in the genus, it reaches a maximum
height of nearly 40 m and a diameter approaching 100 cm (Harwood 1989). It is a fast-
growing tree in suitable environments. In Kenya at the altitude of 1400 m grevillea reaches a
maximum height of about 33 m at an age of 50-60 years, while the diameter is 50-70 cm. It is
planted for soil improvement due to the shading and mulch it provides (Muchiri 2001).

Grevillea became a very common boundary tree to claim land rights in Tanzania after the
villagization program in the 1970s. About 15 years later the farmers built new houses with the
good quality timber from these boundary plantations. In the 1980s tree planting was promoted
with free seedling distribution to the households. Most of the farmers chose to plant grevillea
(Yasu 1999), because it is fast-growing and easy to propagate and it provides economically
viable products. Pests and diseases do not significantly affect it. The leaves are used
occasionally as animal fodder in East Africa (Harwood 1989). Grevillea has been one of the
most common shade and support trees for black pepper and tea in India (Purseglove et al.
1981). Possibly it was introduced also in the East Usambaras primarily as a shade tree for
coffee and tea plantations, as it is used in India and Sri Lanka. It is also said to compete less
with the adjacent agricultural crops when compared to other trees available for farmers, due to
its deep rooting system. As it tolerates intensive pruning it provides fuelwood and fodder for
households all year round, and it has a good form (Muchiri 2001). The properties of the
grevillea timber are well known. It is used in many ways, and also for pulp and firewood. Its
attractive foliage and orange floral display have made it popular as a shade, beekeeping and
ornamental tree in many countries. It performs best on reasonably fertile, deep soils, and does
not tolerate waterlogging. The most suitable climate for it has a temperature of 15-18 C and a
mean annual rainfall of 1 000-2 000 mm (Harwood 1989).

2.6.4. Gliricidia sepium, the N-fixing agroforestry tree

Gliricidia sepium Jacq. Kunth ex Walph. (Madre de cacao, Quickstick), native to Mexico,
is arguably the most widely cultivated multipurpose tree in the tropics after Leucaena
leucocephala (Lam.) de Wit. It is mentioned as one of the five most important agroforestry
trees in the humid tropics. Gliricidia has been described as one of the most promising woody

46
species for alley cropping and alley farming (NFTA 1987). It is flexible in management and
adapted to dry lowlands as well as to humid mountain areas up to 1 000 m. It can grow a
height of up to 15 m. It is the most common live fence species in Central America and in
other tropical areas because of the easy rooting of large stem cuttings and its multiple uses, as
food (flowers), green manure, fuelwood, rat poison and medicine, and as forage due to a high
crude protein content (25%) of the leaves (Gholz 1987). It is an important fuelwood species in
the sub-humid tropics, because of the quality of the wood and a quick growth after pruning
and lopping. The wood is hard, durable, and termite- (Agroforestree Database 2006) and fire-
resistant (Mbuya et al. 1994). It is used as a boundary tree, a shade tree for perennial crops,
and as a support tree for black pepper.

Gliricidia improves the soil through its rhizobial nitrogen fixing. It has been shown to
nodulate with fast-growing rhizobia. Estimates of its nitrogen fixation based on the nodule
biomass and the rates of nitrogenase activity are approximately 13 kg/ha/a in the conditions
prevailing in Mexico (Steppler and Nair 1987). It is thus understandable that it has been used
as green manure. In the East Usambaras as well it is one of the most popular tree species used
in agroforestry cultivation. In Indonesia and Sri Lanka gliricidia is commonly grown with
black pepper, coffee and garden crops, and also with cacao and coconut (NFTA 1989).

47
3. MATERIAL AND METHODS

3.1 Study area

3.1.1. The East Usambara Mountains

The East Usambara Mountains are located in Korogwe and Muheza districts, in Tanga region
in northeastern Tanzania. They form an area of 1 300 km2 (located between 4 48 and 5
13S latitude and 38 32 and 38 48E longitude) and belong to the chain of isolated
mountains called the Eastern Arc. The other mountains belonging to the Eastern Arc are:
North Pare, South Pare, West Usambara, Nguu, Nguru, Uluguru, Ukaguru, Rubeho,
Malundwe, Udzungwa and Mahenge (Figure 7; Hamilton and Bensted-Smith 1989).

Figure 7. The location of the East Usambara Mountains in the Eastern Arc (The Bugwood
Network 2001).

48
The East Usambara forests are vitally important for the local people in many different ways
(Figure 8); however, forest destruction continues to be a major threat to the biological
diversity. Forests are cleared mostly for cultivation purposes. The soil remains fertile for a
few years after clear-cutting, but is then getting poorer while the topsoil is leaching away, and
a new plot is cleared (Eastern Arc Mountains Information Source 2001).

Figure 8. The East Usambaran forests are an important water catchments area and vitally
important for the local population. Local farmer and research assistant, Mr. Abduel Kajiru,
whets his panga; the important tool in the forest. (Photograph: Teija Reyes).

About 75% of the Tanzanian population is concentrated on about 15% of the land area of the
country. The densities are high in the northern moist highlands (Sengoe 1994). There are
many reasons why some areas are more densely populated than others in Tanzania. Perhaps
the most important factor has been a favourable climate for cultivation and absence of tsetse
flies. However, historical factors, such as tribal wars, invasion of colonialists, and the slave
trade, have also forced the agriculturist tribes to settle in the mountains. From the beginning
of the nineteenth century, employment opportunities on tea estates have tempted people to
immigrate to the East Usambaras (Lundgren and Lundgren 1982). The highlands of Tanzania
have better potentials for crop production than any other area in the country (Lulanda 1998).
In 1994 the population density was 180 persons per km2 in the East Usambaras, when it was
41 persons per km2 on average in the whole country. The natural increase in Tanzania since
the mid-1970s has been about 2.6% per annum (EUTCO 1994), and some estimates in other

49
highlands (Uluguru Mountains) indicate annual growth rates of up to 6.5% (Lulanda 1998). In
addition, a half of the population is younger than 18 years, indicating continuous population
pressure (EUTCO 1994).

Traditional cultivation practices are based on slash and burn, and nowadays with very short
fallows. The result is low agricultural productivity with low income levels. People are very
poor in the East Usambaras. Major annual crops grown in the highland villages in
monocultures or intercropped with banana, are maize, cassava, yams and sugarcane, none of
which produce high yields at altitudes close to 1 000 m on acidic and infertile soils. Land is
inefficiently cultivated and poorly managed, and even very steep slopes are often completely
cleared. Fertilizers, improved seeds, and pesticides are generally not used (Kajembe and
Mwaseba 1994).

All land belongs to the state, and the management of forests is mainly under the government
control (Kessy 1998). Aims of forestry activities were originally focused on national interests,
emphasising timber production for economic development, or water supplies for industrial
and urban use. During the 1980s until to date a strong interest has been put on the
conservation value of the East Usambaran forests, especially by strengthening biodiversity
aspects in their management (Katigula 1999). Natural Resource Committees have been
established and agreements have been made to allow the communities around the Amani
Nature Reserve to collect fuelwood in the local use zone (EUCAMP 2001). Households in the
East Usambaras consist, on average, of six persons, and the land is normally managed with
family labour.

There is a dual system of land tenure in Tanzania. Even if all land officially belongs to the
state, the right of occupancy can be granted by the Commissioner for Lands or acquired
through customs and tradition. In the Usambaras, the land is mainly inherited through a
customary tenure system based on patrilineal kinship structures. As part of customary land
ownership, the land can be leased by the state to individual farmers, and the leases can be
inherited, sold or bought in an open market. Forest encroachment on public land is still very
common in the study area; 40% of the farming land is obtained by clearing the forest. The
average farm size per household is about 3.5 ha, of which about 0.8 ha is set aside as fallow or
rented to new couples or to landless immigrants. This so-called Kinyange farming system is
still a common practice in the study area. There is also a registered land ownership for tea
estates, forest reserves and a few large farms. In contrast to private land, the so-called Ujamaa

50
land is property of the whole community and does not have private ownership rights on it. It
is a remnant of the socialist experiment on ownership of land resources carried out by Julius
Nyerere at the end of the 1960s. All members of a community worked on this land in turns,
and the proceeds were distributed to the entire community (Sah 1996).

Before colonial times, even if some settlements existed on steep-sloping escarpments, soil
erosion was minimal due to banana gardens. Also the Washambaa people had and still have a
good knowledge about their environment, and they have always respected the limitation of the
land (Kaoneka et al. 2000). At the end of the nineteenth century the forests were for the first
time cut on a large scale, so as to create coffee and tea plantations (Iversen 1991). German
settlers changed the pattern of agriculture and the types of crops also grown in the Usambaras,
introduced exotic crop plants, and started commercial exploitation of the timber resources.
They also introduced various rules with the aim to conserve the forests. The British
introduced sawmill practices in the mountain area but also initiated conservation activities
(Kaoneka et al. 2000), since serious forest deterioration was already seen (Lundgren and
Lundgren 1982).

After the independence, exploitation of the forest reserves for immediate short-term benefits
started. Half of the forest at Amani was cut for agriculture from 1954 to 1976. In the mid-
1970s soil erosion became a common concern, but the centralised forest management did not
have resources to reduce the impact of human activities (Kaoneka et al. 2000). Selective
harvesting techniques, mechanical logging, and road construction have been destructive. Pit
sawing has accelerated in the area as people have migrated into the Usambaras (Sengoe
1994). During 1980-1984 the amount of logging reached 16 000 m3 annually (Hamilton and
Bensted-Smith 1989). Today, the vegetation in the areas once covered by forest consists of
remnant patches of forest, most of which are modified by man; estates of tea, sisal and other
plantation crops; plantations of hardwoods; small farms and secondary grassland; as well as
shrub and woodland (East Usambara Mountains 1988).

Some of the forests have been declared as traditional forest reserves to be used for ritual and
cultural practices. These patches of natural forest have not been exploited since they are
considered sacred places.

The major threats to the forests are fire, illegal logging, agricultural enroachment, lack of
awareness, and poverty. Fire is considered as the single most serious threat (Burgess and

51
Kilahama 2005). In the East Usambaras, the fires have decreased in recent years due to the
awareness raising activities carried out by the EUCAMP programme. Local people are
becoming more careful when dealing with fire, and many farmers have participated in fire-
fighting on a number of occasions.

The oldest form of land use in the Usambaras has been a combination of shifting agriculture
with livestock management. The main source of income is trading in cash crops, working on
tea estates, involvement in petty business, dairy farming and agricultural trade. The main cash
crops are sugarcane (Saccharum officinarum L.), black pepper, cardamom, cloves (Syzygium
aromaticum L.), cinnamon (Cinnamomum verum J. Presl or C. zeylanicum Bl.), coffee
(Coffea spp.), groundnut and beans. The main subsistence crops are maize, cassava, beans and
cocoyams (Colocasia esculenta L.). The most commonly grown fruits are bananas, coconuts,
oranges (Citrus sinensis L. or C. aurantium L.) and avocados (Persea americana Mill.); cocoa
is also produced. The most common other tree species planted on the farms are typically
Grevillea robusta, Eucalyptus saligna, Tectona grandis L.f. (teak) and Cedrela odorata (Sah
1996). Thousands of the local people are employed on the tea estates around Amani. In 5 of
the 18 villages around Amani, up to 75% of the inhabitants work on the tea estates, providing
necessary cash income; this indirectly also relieves some of the pressure on the Reserve
(EUCAMP 2001) but also attracts new immigrants to the area. Tea estates buy fuelwood
from the local people and give incentives for villagers to plant trees for the estates and
rewards for tending them (Masayanyika 1995).

3.1.2. Climate and soil characteristics

Amani is situated at about 1 000 meters above sea level. The mean annual rainfall is 1 918
mm and the mean annual temperature is 20.6 C. The occurrence of many species of mountain
forests at unusually low altitudes in the East Usambaras is related to an unusually cool
climate: mean maximum and minimum temperatures at high altitudes in the East Usambaras
do not show extreme values (Figure 9). Lowest temperatures are found in relatively wet
places. There are distinct local microclimates, which are referred to as exceptional mistiness
and general wetness. Proximity to the Indian Ocean has a cooling effect, and temperatures in
the East Usambaras are 4-5 C lower at 700 m, and 2-3 C lower at 500 m than those in other
parts of Tanzania at similar altitudes. The forest itself creates a remarkably cool microclimate,
and the relative humidity is high throughout the year (Lovett and Wasser 1993).

52
 25,0
Temperature C 20,0

15,0 Emau Hill

10,0 Amani

5,0

0,0

O er

ec er

r
ne

ov e r
ch
Fe ary

ri l

Au ly
ay
ry

be
Se g us
Ju

b
Ap
ua

b
Ju
M
ar

em

em

em
nu

o
M

ct
br
Ja

pt

D
Month

Figure 9. Average temperatures in Amani 1956-75 (National Institute for Medical Research,
Amani) and in the trial area in 2003-2005 (Catholic Church in Emau Hill, Shebomeza).

In the East Usambaras there used to be two growing seasons in the past, but due to changes in
climate there is nowadays only one growing season. Hamilton (1989) has found reliable
evidence for a decreased annual rainfall since 1960. People who have lived all their life in the
Usambaras claim that mistiness has decreased and the climate is hotter than before. They can
nowadays produce citrus fruits, mangoes (Mangifera indica L.) and coconuts that used to be
absent in this area before. From June to September people earlier used to keep fires to warm
their houses, but nowadays it is not anymore necessary. Malaria was also unknown in the area
before, but it is now very common in the Usambaras as well (Hamilton and Bensted-Smith
1989). Forest clearance is mentioned among possible causes of this climate change (East
Usambara Mountains 1988).

Rainfall in the East Usambaras is very variable and the microclimate changes distinctly from
place to place. The mean annual rainfall is highest near the southern escarpment (2 300 mm)
and lower in the north (1 650 mm; Hamilton and Bensted-Smith 1989). In Emau Hill, where
the agroforestry trial was established, the mean annual rainfall during 2003-2006 was 1 619
mm, while in Amani (situated about 5 km from Emau Hill at the same altitude) it was 1 882
mm during 1901-1997 and in Mlesa (about 15 km from Emau Hill at the same altitude) 1 780
mm during 1967-2007 (Figure 10).

53
 400
350
300
250 Emau Hill
mm

200 Amani
150 Mlesa
100
50
0

O er

ec er

r
ov be r
ne
ch
Fe ary

ril

Au ly
ay
M y

p t st

be
r

Ju

D mb
Se g u
Ap
ua

Ju
M
ar

em

em
nu

o
ct
br

e
Ja

N
Month

Figure 10. Average rainfall in Amani in 1901-1997 (National Institute for Medical Research,
Amani), in Mlesa (Marikitanda Tea Research Station) in 1967-2007, and at the trial site in
Emau Hill in 2003-2006 (Catholic Church in Emau Hill, Shebomeza).

From January to March the climate is hot and dry. During the long rains (March-May), the
temperature falls and the mountains receive 45% of the total annual rainfall. From June to
September the climate is rather dry and cool until the short rains start (normally in October to
December) and the temperature increases (Hamilton and Bensted-Smith 1989).

The high-altitude soils are not very suitable for long-term agriculture, except for crops
preferring very acidic soils, such as tea (Hamilton and Bensted-Smith 1989). The soil is
generally very deep (over 150 cm), well drained, and strongly acidic (pH 3.5-5.5). The colour
of the soil varies from the topsoils dark reddish brown to the subsoils yellowish reddish or
red brown. There is often a clay layer in the subsoil. Calcium, magnesium, potassium and
phosphorus levels are normally low or very low. When the soils are acidic there is also a
danger of aluminium and manganese toxicity. Steep slopes cause soil erosion and can initiate
landslides (Masayanyika 1995). Ferrasols, which have a low supply of available nutrients,
strong acidity, limited labile topsoil organic matter (which accounts for the main nutrient
storing capacity of the soil), exhibit low resilience and moderate sensitivity to water erosion.
Even with good soil cover, yields decline rapidly without measures to control the soil erosion
(Scherr 1999).

Soils in the East Usambaras are acidic at high altitudes and neutral to alkaline near the
foothills (Sah 1996). There is a rapid change in soil acidity from pH 6.5 at 850 m to 5.0 at

54
900 m and pH 4 at 1 050 m, (particularly if compared to the change from pH 7 at 300 m to pH
6.5 at 850 m). The altitude of 850 m marks a rapid climatic transition. Low clouds frequently
envelop the highest slopes, commonly with a lower boundary at just this altitude. There is
therefore a causal relationship between the common occurrence of clouds above 850 m and
the predominance of acidic soils (Hamilton and Bensted-Smith 1989). This is also the
biologically richest elevation zone (Newmark 1992; Rodgers 1993). Higher altitudes typically
have a thick (up to 10 cm) upper layer of mor-type humus underneath the litter. This is absent
from the lowland forest soils. Hamilton postulates that a higher acidity in the submontane
soils is due to effects of climate. No correlation was found, however, between soil acidity and
the inclination of the slope (cf. Hamilton and Bensted-Smith 1989).

For the present study, an initial soil analysis was done by the Mlingano Agricultural Research
Institute at the end of October 2000, before establishing the field trial. The trial is located on a
slope which was divided into four parts: Near valley part (NV1), Lower part (UL1), Middle
part (MS1) and Upper part (US1). From each level, 12 samples (0-20 cm stratum) were
analysed. Soils were classified as Haplic Acrisols in the upper section and Haplic Lixisols at
the bottom. In the valley bordering the experimental site the soils were Eutric Gleysols.
Analytical results showed that the soil conditions of the study area were poor and followed
the general soil characteristics in the area (Annex I).

3.1.3. The Amani Nature reserve and the Derema Wildlife Corridor

The Government of Tanzania gazetted the Amani Nature Reserve in the East Usambara
Mountains in 1997 for the protection of the biodiversity and the endemic plants and animals.
The main goal of the Amani Nature Reserve, as stated in the ANR General Management Plan,
is the conservation of the areas biodiversity as untouched as possible for further generations.
It is also stated that the goal should be reached in such a way that the living conditions of the
local people are ensured and their activities supported by the reserve staff and other sources
have a sustainable basis (Amani Nature Reserve General Management Plan 1998). At the
establishment of the Amani Nature Reserve buffer zone, hundreds of farmers had to give up
their lands, and the compensations paid were not regarded as sufficient (URT 2006).

The Amani Nature Reserve covers the largest block of forest in the East Usambara
Mountains. It has a total area of 6 880 ha consisting of several forest reserves, namely Amani-

55
Sigi Forest Reserve, Kwamsambia Forest Reserve, Kwamkoro Forest Reserve, Mnyusi-Scarp
Forest Reserve, Amani West Forest Reserve, and Amani East Forest Reserve. The remaining
area of 1 640 ha comprises enlargement areas, land owned by tea estate companies, and land
owned by the local people under customary law (Masilingi 1992).

The East Usambara Mountains are essentially forest islands. About 33 000 ha is within the
central government forest and nature reserves, scattered in 13 forest blocks (EUCAMP 2002).
The Government of Tanzania recommended in the early 1990s reconnecting of the nine
largest forest blocks with a system of wildlife corridors (Newmark 1992). Amani Nature
Reserve, being the largest forest block in the East Usambara Mountains, has a lower risk of
plant and animal population extinction than the smaller, highly fragmented forest reserves.
From the Amani Nature Reserve animal and plant species might be capable of migrating to
and re-colonising also other East Usambaran forests. It is, however, largely isolated from
other forest reserves, and the only forested connection that could link the Amani Nature
Reserve to other forest reserves is the Derema Corridor.

The protection of the Derema forest area is very important for conservation aims, as this area
was considered one of the largest unprotected forest areas in the East Usambara Mountains
(Newmark 2002). However, various social factors have had to be considered in this proposed
conservation area (Doody et al. 2001). The Derema Corridor was created out of 1 300 ha,
including about 800 ha of village land where the former main land-use was cardamom
cultivation in the forest (Figure 11). It has been estimated that about 50% of this area was
used for cardamom cultivation. In fact, forests have here been saved from clear-cutting due to
the cardamom cultivation. In 2001 the farmers handed over their lands during a process of
boundary demarcation (TFCG 2007a).

Those farmers who had their crops slashed in the 3-meter-wide reserve boundaries were
compensated in 2002. The first financial compensations were so high that some farmers
started to plant more cardamom to previously uncultivated areas in order to get more
compensation 8. The compensation process was in 2007 still going on, and the budgets for this
purpose have increased unexpectedly, due to the implications of the new land act and delays

8
Pers. com. with researcher Heini Vihemki, University of Helsinki, 14 December 2007.

56
Figure 11. The location of the research site and study villages (figure courtesy of Juhana
Nieminen).

in the process. The total compensation amount has increased from the first estimation of 9.3
million shillings to 2 700 million, which equals US$ 2.1 million. In the compensation process,
the declared yields were based on the farmers own estimations, and clear overestimation
occurred. No physical relocation of residence was required, but farmers would permanently
loose the income from this forest area where cash crops, firewood and other forest products

57
had been harvested. The new law is demanding compensations for lost profits, disturbance,
and the value of lost production until new crops can be established to produce the same
quantity of the product as had been lost. This is a very difficult task, since the rural
production, marketing and inputs are not recorded by the farmers, and they vary considerably
according to the management pattern and other conditions. Cardamom plots in the forest,
which accounted for 74% of the value of all compensations, cannot really be replaced at all,
since there are no sites to which the lost forest cultivation could be transferred. The Ministry
of Lands has allocated some land from planned sisal plantations to the Derema farmers, but
this land is not suitable for cardamom production. Derema actually belongs to an ungazetted
public forest reserve surrounded by five villages (URT 2006).

3.2. Socio-economic data analysis (Study I)

Data were collected from four villages located in the buffer zone of ANR. Two separate agro-
ecological zones were included: Shebomeza and Mlesa in the highlands (900-1 000 m), and
Mashewa and Kimbo in the lowlands (800 m). From the four sampled villages, 10% (154) of
the total number of household heads (all together 1 516) was randomly selected using the
village register books. Two approaches were used: a formal survey and a Participatory
Learning and Action (PLA) approach. PLA is also known as PRA (Participatory Rural
Appraisal), but the name has been changed because the method has been successfully used
also in urban areas. PLA methods 9 were used for socio-economic data collection, because the
aim was to incorporate the knowledge and opinions of local people to the study. The PLA
methods used in the study were mapping, a seasonal calendar, transect walks, and a time line,
also based on group discussions. The sampling unit was a single household where cultivation
practices (mapping, seasonal calendar, transect walk), productivity, costs and revenues (cash-
flow time line), and pros and cons of living in the highlands were clarified. The information
included socio-economic data such as occupation, education, household composition, land use
activity, agricultural crops, livestock types and productivity, cost of production, market prices,
and sources of income. Physical observations were carried out to classify farm practices as
improved or traditional agroforestry. The improved farms were more organized, cultivated the
crops densely and manage their fields better with manure. (Table 3). A market survey was

9
PLA usually engages the commitment and analysis of local people, enables the expression and sharing of their
diverse and complex realities, gives insights into their values, needs and priorities, and can also lead on into
participatory action (Chambers 1997).

58
conducted at local markets in the study villages. A number of traders were also interviewed to
discuss the market constraints and opportunities.

Table 3. Criteria describing the selection of farming system.

Traditional cultivation methods:

Based on slash and burn and/or short fallow farms;
Land inefficiently planted with annual crops in monoculture or with banana;
Poor management of crops;
Few trees (generally fruit trees), but not planted/retained for soil conservation
functions;
No/few cash crops, if available, poorly managed.

Improved agroforestry systems:

Normally based on a traditional cultivation system;
Land-use well planned, multipurpose trees intercropped, e.g. in alley cropping,
windbreaks or shelterbelts;
Farms with various annual and perennial crops, densely cultivated;
Farms enriched with well-managed cash crops;
Manure used if available.

Out of the 92 variables elicited, 27 continuous variables related to household revenues were
used for an analysis of variance, using either agroforestry system (traditional or improved) or
agroecosystem (highland or lowland) as a source of variation. The univariate analysis for each
continuous variable was further confirmed by a multivariate analysis of variance, to test the
hypotheses whether the agroforestry systems or agro-ecosystems were different, when all the
variables were jointly analysed. Only significantly different variables (p<0.1) were presented
in the results. A multiple linear analysis was run to determine the influence of the components
of the households income on the net income. Cash flow analyses were carried out to assess
the farmers liquidity situations during different times of the year.

3.3. Harmful cardamom cultivation practices (Study II)

The effects of small cardamom cultivation on the African natural mountain rainforest
ecosystems and on the economy of the local communities in the East Usambaras were
described in this particular study. Possibilities for sustainable and profitable cultivation with
agroforestry methods were analysed. The data were based on study III; observations of
cultivated areas, informal interviews, and questionnaires in the buffer zones of the Amani

59
Nature Reserve and the Derema Corridor. An agroforestry field trial was established in the
buffer zone of the Amani Nature Reserve in 2000 to assess possible intensive cardamom
cultivation methods. In addition, a socio-economic and environmental comparison, based on
literature review, was made with the cardamom-growing highlands in India and Guatemala.

3.4. Conservation and sustainable cultivation (Study III)

A sample of 75 cardamom-cultivating households in nine villages in the buffer zones of

Amani Nature Reserve and the planned Derema Wildlife Corridor were interviewed. Many of
them also cultivated black pepper and other spices. It was estimated that the sample
corresponded to about 10% of the cardamom-cultivating households in the buffer zone. The
questionnaire surveys (Annex II) were conducted in the following villages: Shebomeza,
Mlesa, Chemka, Kisiwani and Amani around Amani Nature Reserve; and in IBC Msasa,
Antakae, Makanya and Kwezitu around the Derema Corridor. About 10% (75) of the
cardamom-producing households in the buffer zone of the conservation area were interviewed
and their fields were observed. Informal interviews and questionnaire surveys were conducted
with the farmers during October 2005 and May 2006 on the local cardamom cultivation
practices, on how conservation has affected farming activity, and on what environmental
changes farmers had observed in the East Usambaras. Observations on cultivated cardamom
areas included cardamom quality, the species cardamom was intercropped with, years of
cardamom cultivation, cultivation land area, and the management practices.

3.5. Experimental study on the spice-crop agroforestry systems (Study IV)

A field trial plot of 0.8 ha was established in mid-November 2000, during the short rains, in
Emau Hill, close to the village of Shebomeza. The trial was located about 500 m from the
boundary of Amani Nature Reserve. A local farmer, Mr Bruno Mallya, was the landowner.
The previous land use management of the experiment site was as follows: in 1991 the trees
were cut down; in 1994 the area was cleared for cultivation purposes; from 1995 to 1997
maize, cassava, yams, sugarcane and banana were grown, then the land was abandoned and
only some banana and sugarcane plants had remained in the area.

60
Two tree species, Grevillea robusta and Gliricidia sepium, were intercropped with the two
spice crops with a high value in the area, black pepper and cardamom. There were eight
treatments, including a control plot of each species in a pure stand, and four replications of
each treatment. Plot size was 16m x 16m and there were altogether 32 plots (each 256 m2).
The planting density was 3m x 3m for trees and pepper and 2m x 2m for cardamom (sole
crops). Three black pepper seedlings were planted per each support tree, following a common
practice (Figure 12).

Consequently, the treatments were:

(T1) grevillea + pepper;
(T2) grevillea + pepper + cardamom;
(T3) monoculture of grevillea;
(T4) monoculture of cardamom,
(T5) monoculture of gliricidia;
(T6) gliricidia + pepper;
(T7) gliricidia + pepper + cardamom; and
(T8) monoculture of pepper (= with dead wooden poles).

31
3
32
Alt: 997 m 6 8 30
8 13 4 14 29
26 27 28
5 7 1 2
15 25
12 5 1
2
24 23 22
3 4 7 21
11 16 17
4 6 7 5
18 19 20
1 6 2
10 9 8
3 1 8
6 5 4
6 7 5
plot
number 3
7
treatment 3
2
1 8
2 4 Alt: 987 m

Figure 12. Trial plan for spice crop agroforestry in East Usambara, Tanzania.
61
All planting material was obtained from nearby villages. Grevillea robusta seeds were
obtained from the National Tree Seed Program (NTSP) from Lushoto, West Usambaras. They
were sown in Tongwe village, on March 2000, by direct sowing, and 6-month-old seedlings
were planted in the trial. Gliricidia cuttings were locally collected from Tongwe village from
trees planted in August 1990. Black pepper seedlings and cardamom tillers were obtained
from Kisiwani village. Rhizome multiplication was done for cardamom; this is a simple
method and facilitates earlier production as compared to seedlings raised from seeds. Root
clumps with one tiller and a growing young shoot were planted in the trial.

The experiment was realized without irrigation at the beginning of the rainy season. No
fertilizers, chemical pesticides or organic manure was applied, except or the organic material
naturally originating from the plants. No pruning, branching or thinning was applied. The
weeding occurred twice a year and the trashing of cardamom (removing old and dry leaves)
once a year. Dead poles in the pepper monoculture were changed when required. The
direction of the slope was north (down) - south (up).

3.6. Ecological analysis of spice-agroforestry system (Study IV)

The nine sampling trees and associated crops were systemically measured at the middle of
each plot every second month. The measuring included tree height, diameter at ground level
and in breast height (dbh), canopy width, and canopy height. For pepper, height and diameter
at ground level were measured; for cardamom, the number of tillers, average height and
average diameter of the tillers were recorded. The production was weighed per plot. The fresh
weight was converted to dry weight by measuring and drying sample of crop production.

The stem volume for grevillea trees was obtained as a product of height and diameter with the
model: v= h x g x f, where v is volume, h height, g the basal area (r2) and f a tree form factor
(f = 0.5 for grevillea). Dry wood biomass was determined as a product of volume and wood
density (469 kg/m3).

The stem volume for gliricidia was derived from the regression between fresh weight and
height for each sample tree. The dry matter content was 0.21% in this experiment. The
volume was obtained with fresh weight and wood density (750 kg/m3). All yields were

62
converted to per-hectare values for easier comparison. Root biomass could not be measured,
as in most forestry trials, as the methods would have required considerable effort and costs.

3.6.1. Growth analysis

The growth rate of crops and trees was estimated using a linear model of height as a function
of time. We compared the height and diameter growth between treatments with a Bonferroni
test. All data were analysed with R, which is an open-source Linux-based language and
environment for statistical computing. We used agricolae, which is a statistical procedure
for agricultural research, R-package version 1.0-3 created by CIP.

3.6.2. Treatment comparison

Treatment differences were analysed with the Waller-Duncan K-ratio t-test. The test was
selected due to its suitability for analysing large samples (3 000-5 000 units in this case).
The variables were:
- for trees: height, ground diameter, breast-height diameter, canopy height, height of the
lowest branch, canopy width, canopy cover, volume per tree, volume/ha, dry
weight/ha, economic value/ha,
- for crops: height, ground diameter, dry yield in kg/ha in 3rd, 4th , 5th and 6th year (for
black pepper only in 5th and 6th year). In addition, for cardamom, the average numbers
of tillers, panicles and fruits were recorded.

3.6.3. Light conditions

Light was measured in September 2002 every hour during daytime for three different days
with a quantum sensor (Li-189, LI-COR, Lincolin, USA). It measures the photosynthetically
active photon flux density (wavelength between 400-700 nm) in mol/m2/s. The light was
measured by placing the sensor in the middle of the cardamom crop and for pepper at breast
height (1.3 m altitude). The light intensity was compared to that measured in the open, giving
a percentage from full light intensity above the canopy.

63
3.6.4. Soil analyses

All laboratory analyses were carried out at the Mlingano Agricultural Research Institute soil
laboratory in Tanga region, Tanzania. Field soil analyses were done at the site before planting
the experiment in October 2000, in October 2002, and in October 2006. A total of 20
composite topsoil (0-30 cm) samples were collected from each treatment. The soil samples
were air-dried and sieved. Measurements included the soil pH (H2O and KCl), electrical
conductivity (EC), the exchangeable magnesium (Mg), calcium (Ca), potassium (K) and
sodium (Na) contents, and the cation exchange capacity (CEC) of the soil. In addition, organic
carbon (C %), total nitrogen (N%) and available phosphorus (P) were determined and the base
saturation ratio (BS) measured. Soils at the experimental site were determined as having low
fertility in year 2000 when the field experiments were started.

3.6.5. Yields vs. Land Equivalent Ratio (LER)

The Land Equivalent Ratio (LER) compares the yields from growing two or more crops
together with yields from growing the same crops in monoculture (pure stand). To calculate
the LER, the intercropped yield of one crop is divided by the yield of the pure stand and
added to the intercrop yield of the next crop, and so on:

LER = intercrop1/pure crop1 + intercrop2/purecrop2 + etc.

LER is thus a ratio that indicates the area of land needed to grow two crops together as
compared to the land area needed to grow pure stands of each. If the ratio is greater than 1.0,
intercropping is advantageous, and if it is less than 1.0, a disadvantage is indicated.

3.6.6. Manure test for cardamom

About 16.5-18 kg of cattle manure was applied on August 2006 to each cardamom plant in
three plots (plot numbers 1, 2 and 5) representing three treatments with cardamom: T2 (with
grevillea and black pepper); T4 (monoculture) and T7 (with gliricidia and black pepper). The
manure was placed around the plant at a distance of 30 cm and covered with soil. Analytical

64
data on the applied farmyard manure indicated that this fertilizer was of good quality (C/N 12)
and had a very high content of total nitrogen (40-56%), phosphorus (30-47 mg/kg), potassium
(1.06-2.38 Cmol(+)/kg) and organic carbon (4.82%), and a high content of calcium 10.80-
11.16 Cmol(+)/kg).

65
4. RESULTS

4.1. Socio-economic aspects of improved agroforestry systems (Study I)

The study convinced that with improved agroforestry methods the farmers are able to raise
incomes from their crop cultivation significantly. The households obtained twice as high an
annual gross income with improved agroforestry systems as when using traditional practices,
and a net income around 13 times the income from traditional practices. This was possible
without great changes in cultivation systems and achieved by adding some high-value cash
crops and cultivating the fields more intensively in combination with multipurpose tree
species. The income from traditional cultivation methods could be doubled by just adopting
high-value cash crops and by better management. Even if the incomes were higher with
improved systems, the per-capita income was still below the national average. Poor extension
in the study area was mentioned as one of the main factors hindering adoption of new farming
technology by 67% of the respondents. Most of the respondents (87%) claimed they had very
little access to capital and credit facilities.

4.1.1. Improved agroforestry vs. traditional cultivation

Improved agroforestry systems proved to be more sustainable and viable than the traditional
agroforestry systems in the East Usambaras, indicating a doubled income for the households
analysed. Farmers with improved agroforestry systems also had better access to land and
demonstrated higher productivity for all commodities except maize and rice (Table 4).

Cash crop revenues, especially from cardamom cultivation, were significantly higher for the
improved cultivation methods. The total income was doubled (US$ 157 vs. US$ 74), but also
the total costs were higher (US$ 473 vs. US$ 380) for the improved methods. The net income
was anyhow 13 times higher in the improved systems (US$ 387 vs. US$ 29). The incomes
from cash crops represented 30% of the total income in both systems. Livestock products
played a significant role in the annual household income, especially for the case of improved
systems (23% vs. 9%; Table 5).

66
Table 4. Indicators for production in traditional and improved systems (per household/farm)

Variable, unit Improved Traditional Ratio (Imp/Trad)

Maize, kg *) 396 511 0.8
Rice, 50 kg sacks *) 0.4 1.2 0.3
Land size, hectares 4.1 3.0 1.4
Cassava, 70 kg sacks 6.8 3.7 1.8
Beans, kg *) 21.1 9.0 2.3
Bananas, 30-50 kg bunches 67 33 2
Cardamom, kg 8.9 1.6 5.6
Pepper, kg 40.6 17.4 2.3
Cardamom revenue, US$ 25 5 5.4
Cardamom+pepper revenue, US$ 60 28 2.2
Dairy revenue, US$ 139 50 2.8
Goat revenue, US$ 27 5 5.2
Food crop revenue, US$ 165 127 1.3
Cash crop revenue, US$ 268 139 1.9
Total income, US$ 861 409 2.1
Total costs, US$ 473 380 1.2
N= 16
p < 0.05, except in three cases where p < 0.1, marked with *) in the table.
Currency exchange rate during the study in November 2001: US$ 1 = TZS 876.4.

Table 5. Farming system financial statement (per household/farm).

Farm income category Improved systems Traditional systems

TZS US$ TZS US$

Food crops 136 798 156 116 009 132
Cash crops 235 299 268 121 930 139
Livestock products 171 235 195 32 605 37
Local poultry production 20 237 23 13 342 15
Forest products 68 538 78 30 738 35
Off-farm activities 122 089 139 44 244 50
Sub total/ Farm income 754 196 861 358 868 409
Variable costs 64 339 73 30 738 35
Operating costs (overheads)
Living expenses 316 702 361 298 200 340
Taxes and fees 33 547 38 4 449 5
Sub total costs 414 588 473 333 387 380
Net operating profit 339 608 387 25 481 29

Currency rate during the study in November 2001 US$ 1 = TZS 876.4

The net income was mainly influenced by land size, land tenure, cash crops, dairy farming,
and off-farm income (Table 6). Improved agroforestry secured enough food throughout the

67
year for 40% of the households, while traditional practices did it only for 18% of the
households, even if they were mainly cultivating food crops.

Table 6. Socio-economic factors influencing the household net income.

Predictor Coefficient Std. deviation t-ratio Probability

Constant 147442 75844.7 1.944 0.054
Age -46464 12327.9 -3.769 0.000
Family size -58983.5 4457.6 -13.232 0.000
Land size 24526 5000.2 4.905 0.000
Tenure 16050.4 8452.0 1.899 0.060
Education -23752 30.2 -786 0.433
Cash crops 0.900 0.1 11.356 0.000
Dairy farming 0.699 0.1 11.501 0.000
Marital status -10932 13267.0 -0.824 0.412
Off-farm 1.021 0.1 19.6 0.000
2
R =72.5 %. Dependent variable: household total net income.

4.1.2. Cultivation in highlands vs. lowlands

The strategy used by the farmers in the lowlands (<850 m altitude) differed from that in the
highlands (>850 m). The highland farmers cultivated more cash crops and kept more cattle
than the lowland farmers, and also received a better income (US$ 664 vs. US$ 323). About
70% of the farmers in the lowlands were dependent only on the income from farming, mainly
based on traditional farming methods, while in the highlands the income came from improved
agroforestry systems together with that from livestock and off-farm activities. Maize and rice
(both food crops) were cultivated in the lowlands in traditional ways, while all the cash crops
(cardamom, cinnamon, cloves and sugarcane) were cultivated in the highlands with improved
agroforestry systems. The only exception was black pepper, which was grown more in the
lowlands than in the highlands. The income from cash crops was three times higher in the
highlands (US$ 207 vs. US$ 77 for the lowlands). The greater share of the off-farm income in
the highlands is probably explained by the income from tea picking. Goats were kept in the
lowlands and dairy cattle in the highlands. Milk sales contributed 25% of the total household
income in the highlands (Table 7). The milk cooperative also operated in the highlands.

68
Table 7. Indicators for production in the highlands and in the lowlands.

Variable, unit Highland Lowland Ratio

(Imp/Trad)
Land size, hectares *) 3.2 3.9 0.8
Maize, kg 232 674 0.3
Rice, 50 kg sacks 0.08 1.5 0.1
Pepper, kg 14 44 0.3
Pepper revenue, US$ 15 42 0.4
Goat revenue, US$ 0.4 32 0.01
Dairy revenue, US$ 169 19 8.9
Yams, 70 kg sacks 1.8 0.25 7.2
Potatoes, 70 kg sacks 0.5 0.06 8.3
Cardamom revenue, US$ 29 0
Cardamom, kg 10.5 0
Cloves, kg 22 0
Cinnamon, kg 36 0.3 120.0
Sugarcane, pcs. 573 5 114.6
Cash crop revenue, US$ 207 77 2.7
Off-farm income, US$ 112 360 43 500 2.6
Total income, US$ 664 323 2
N= 16
p < 0.05, except in one case where p < 0.1, marked with *) in the table.
Currency rate during the study in November 2001 US$ 1 = TZS 876.4.

4.2. Precious cardamom harmful for the mountain forests (Study II)

4.2.1. Harmful cardamom cultivation methods

In this analysis, apart from Tanzania, also India and Guatemala were included, based on a
literature study. All three countries, each in different continents, cultivated cardamom in some
of the worlds most valuable forests in terms of biodiversity and had noticed environmental
deterioration due to this activity. Based on the literature, similar problems as those in the East
Usambaras were noticed in the other cardamom producing biodiversity hotspots as well, in
the Western Ghats in India and in Alta Verapaz in Guatemala. These regions were more
intensively managed using chemical fertilizers and pesticides and by constantly pruning and
planting fast-growing exotic trees. A large number of the rain forest fragments in the
Anamalai Hills and in the Western Ghats in India that were under cardamom cultivation are in
the long run doomed to disappear, due to the absence of any forest regeneration (Umapathy
and Kumar 2000). The deforestation rates in Guatemala are also alarming, especially in Alta
Verapaz, due to commercial cultivation, mainly for cardamom. In earlier studies it had been

69
estimated that if the actual deforestation rates continue, there would be no forests left in this
part of Guatemala in 40 years (Melgar 2003).

4.2.2. Socio-economic comparison with cardamom growers in India and Guatemala

Socio-economic differences were also great among the three countries. Cardamom growers in
India were of upper social class, well educated, organized and supported, and had good
networks and social services. In contrast, in Tanzania and Guatemala, they were poor,
uneducated and confronting high risks due to unstable conditions. About 60% of the highland
farmers in the Usambaras grew cardamom. The contribution of cardamom to the average
household income was about 30%, and it gave more than half of the total cash-crop income.
Cardamom-growing farmers were better off than the other semi-subsistence farmers, but their
income was still far below the national per-capita average. In Tanzania no chemicals or
irrigation were used, and improved planting material did not exist. There were no producer
associations, nor any quality-control mechanisms.

Cardamom is the most important spice that India produces and trades in, but only two per cent
of the production was found to be exported, because of a large domestic demand. In the
cardamom producing areas in India, this crop provided the livelihood to the majority of the
population. About 70% of the farms were privately owned. The produce was sold through
auction houses registered with the Spices Board, which also controlled the quality. Recently,
the yields in India have declined due to forest exploitation and a longer dry season. India
traditionally had a near monopoly of the global cardamom markets.

Suitable conditions (a well-distributed annual rainfall and a sufficiently cool climate year
round) and a rapidly expanding production have made Guatemala the worlds foremost
exporter of cardamom since 1980. In Guatemala cardamom can grow under those most
suitable conditions also on open land or having very little shade, which gives a high yield.
Cardamom is Guatemalas fifth largest export product, and the whole production is exported.
It was concluded, based on the literature review that the constantly renewed and developed
production technology, a large drying capacity, and effective export companies in this country
aimed at good quality with the lowest prices. The production cost was reported to be only half
of that in India, but the quality was still claimed to be superior in India. About 70% of the
cardamom in Guatemala was grown by 200 000 farmers, most of them living in poverty. The

70
department of Alta Verapaz, for example, had the lowest development indicators in
Guatemala.

The average yield in Guatemala was 366 kg/ha, which was almost 2.5 times higher than that
obtained in India (150 kg/ha; Ravindran and Madhusoodanan 2002). In contrast, the average
annual cardamom yield in the East Usambaras was 80 kg/ha as earlier reported by
Masayanyika (1995) and only about 62 kg/ha as reported by farmers during the present
interviews in 2005. Tanzania can thus compete neither with the amount nor with the prices in
the global cardamom markets. For Tanzania it can thus be recommended to focus on high-
quality organic spices produced in well-managed homegardens by organized farmers (Study
II).

4.3. How to link conservation and sustainable livelihood in the buffer zone of a nature
reserve (Study III)

4.3.1. Differences between villages in Derema and Amani Nature Reserve

A comparison of the villages around Amani Nature Reserve ANR and Derema, using 13
variables, showed a clear difference between these two areas (Table 8).

Table 8. Comparison of cardamom cultivation variables (per household/farm) between

villages in the Amani Nature Reserve (ANR) and the Derema Corridor (* = significant
difference < P = 0.05; NS = non-significant difference)

Variables ANR Derema Significance

Mean (standard Mean (standard
deviation) deviation)
Conservation has diminished the
cardamom cultivation area (1=no,
2=yes) 1.1 (1.0-1.2) 1.5 (1.4-1.7) *
Distance from farm to nature
reserve, km 1.5 (0.9-2.1) 3.5 (2.8-4.2) *
Total land, ha 2.6 (1.5-3.6) 4.1 (2.7-5.5) NS
Land allocated for cardamom, ha 0.6 (0.3-1) 2.1 (1.1-3.1) *
Cardamom production, kg/year 35 (20-51) 138 (99-177) *
Total cardamom income, US$ per
year 37 (19-55) 193 (134-252) *
Cardamom price, kg/US$ 1.4 (1.1-1.7) 1.4 (1.2-1.6) NS
Cardamom cultivation years 15 (10-19) 23 (19-27) *
Importance of cardamom income
(1=necessary, 2=very important,
3=good extra income,
4=not important) 2.4 (2.2-2.6) 1.7 (1.5-1.9) *
Manure used (1=used, 2=not used) 1.6 (1.4-1.7) 1.4 (1.3-1.6) NS
N=36 in ANR and N=35 in Derema

71
In comparison to the ANR farmers, the villagers in the buffer zone of the Derema Corridor
were more dedicated to cardamom cultivation, had a larger land area allocated for it, and had
more knowledge to manage their fields (using manure, trashing and replanting the crops with
few years' intervals), which was clearly seen as a better quality of the crop as well. The
average yield per farm in the ANR was 58 kg/ha, while in Derema it was 66 kg/ha. The
Derema farmers produced cardamom both in homegardens and in forest cultivation. There
was no significant difference between the yields of cardamom grown in the forest and in
homegardens, but farmers who used both cultivation strategies simultaneously claimed that
they obtained about 40% higher yields (75 kg/ha by using both methods vs. 47 kg/ha with just
one) and a higher income (US$ 102 vs. US$ 64; US$ 1 = TZS 1 253.8 in 2006; Figure 13).

60 54
50
42 40
% of farmers

40 35
31 ANR
29
30 27 Derema
23
20 Total
20

10

0
Forest Homegarden Both
Cultivation practice

Figure 13. Share of farmers implementing two different cardamom cultivation systems or
both systems simultaneously in the buffer zone of Amani Nature Reserve and in the Derema
Wildlife Corridor.

Most farmers around the Derema Corridor reported the cardamom income as continuing to be
the most important income source for their household. All interviewed farmers in Makanya
and 60% of those in Antakae stated that it was still necessary for survival, and it was
mentioned as very important for almost all (88%) villagers in Derema. Cardamom income
was still very important for more than three quarters of the households in both areas. When
asked how the conservation activities had affected cardamom cultivation, 31% of the farmers
(96% in Derema) identified a reduced cultivation area; about 20% (almost equal share in both
areas) reported that they were not able to expand their cultivation further because of

72
conservation; and three farmers (4%), one in ANR and two in Derema, mentioned that there
were more crop-destroying animals due to the enlarged conservation area (Figure 14).

Amani (A)

Kwezitu (D)

Antakae (D)

Kisiwani (A)

Makanya (D)

IBC (D)

0% 20% 40% 60% 80% 100%

Figure 14. Proportion of Amani Nature Reserve (A) and Derema (D) farmers reporting
reduction in the cardamom cultivation area due to conservation activities (only villages
reporting reduction in the cultivation area included).

4.3.2. Conservation and sustainability

Only 6% of the farmers interviewed reported that they had not seen harmful environmental
impacts in the East Usambaras. Changes in climate (less humidity and rains, higher
temperatures) and declines in yields were reported by 65% of the farmers in both areas. Most
of the cardamom farmers (93% in Derema and 76% in ANR) reported a decline in cardamom
yields in recent years. In Derema the majority (83%) of farmers stated a decline in crop yields
but only less than half (48%) of them did it in ANR, even though erosion was more distinctly
observed in ANR in comparison to Derema. This can be explained by a longer cultivation
period in Derema (23 vs. 15 years; Table 8). However, only 19% of the farmers identified
cardamom as a potential cause of degradation of the environment. In contrast, most farmers
suggested that the cardamom markets should be improved and the cultivation encouraged,
because cardamom cultivation was an environmentally more friendly option than the total
forest clearance that is necessary for most other cash crops.

Most of the farmers associated the changes in climate with forest exploitation and were
particularly worried about the expanding tea and sugarcane cultivation. Sugarcane cultivation
had already led to environmental degradation on steep slopes in the area. A few farmers

73
(10%) mentioned fire or gold mining also having severe effects for environment. Most of the
farmers interviewed understood and supported forest conservation. Some negative aspects
towards conservation had provoked an unresolved compensation procedure in the proposed
Derema conservation area. The farmers who had the highest cardamom cultivation areas in
the proposed Derema conservation area were from IBC Msasa, Makanya, and Antakae.
However, many of these farmers had already started to grow cardamom in their homegardens.

Cardamom cultivation in the forest was considered to be the more profitable and sustainable
cultivation method by about 60% of the farmers, since a suitable natural microclimate helps to
maintain the moisture even during the dry seasons. On the other hand, homegardens were
considered a more sustainable system by 38% of the farmers, mainly because it was
considered easier to add manure to fields that were closer to the home, and protection against
thieves become easier. Few farmers said that both methods could be profitable and
sustainable, depending on how the cultivation was managed. Some admitted that
homegardens might be the only available cultivation method in the future because of the
expansion of conservation activities and the lack of suitable public areas for forest cultivation.

In the homegardens cardamom was typically intercropped with banana (in 90% of all
homegardens studied), cloves (46%) and cinnamon (42%). Among the other species
mentioned were fruit trees such as avocado, jackfruit (Artocarpus heterophyllus Lam.), Malay
apple (Syzygium malaccensis L.), guava (Psidium guajava L.), papaya (Carica papaya L.) and
mango (Mangifera indica L.); cash crops, including black pepper (Piper nigrum), chili pepper
(Capsicum annuum L., sugarcane (Saccharum spp.), coffee (Coffea spp.); exotic tree species
(Grevillea robusta, Cedrela odorata L. and Gliricidia sepium); and indigenous tree species
(Milicia excelsa Welw., Allanblackia stuhlmannii Engl. (for nut oil), Cephalosphera
usambarensis Warb., Newtonia paucijuga (Harms) Brenan. Altogether, 25 different species
intercropped with cardamom were mentioned.

4.3.3. Factors hindering farming intensification

Most farmers around both conservation areas were not ready to change their cultivation
practices. Farmers indicated that they were sensitive to risks and uncertainty and had a lack of
capital for investments and of knowledge about other cultivation alternatives. Every farmer

74
mentioned either declining prices or the insecure markets not conservation as the greatest
problem in cardamom cultivation.

Only half of the households applied manure in their cultivation, although cattle manure was
available in all villages. Farmers claimed that the soil fertility was still adequate without
manure and that there was a lack of manpower. With manure application, farms attained at
least a 15% higher cardamom yield.

Farmers suggested more extension and training in cultivation of spices in homegardens, in

quality improvement and in the use of manure (Table 9). Extension services were available
for food crops and cattle farming, but not for intercropping of trees with agricultural crops.
Farmers expressed the desire to have better markets and more land to improve the cardamom
business in the area. They also mentioned that more trees had to be planted on farms, and the
farmers should work in groups. Buyers complained about a low quality of the products
(poorly-dried, with impurities, or with empty cardamom capsules due to too early harvesting)
and about farmers who were not keeping their promises to have the harvest ready for the
buyers on agreed dates.

Table 9. Factors identified by farmers as hindrances to the adoption of more intensified

farming methods.

% of farmers
Factor agreeing
Lack of credit 87
Lack of markets 37
Lack of training 37
Lack of knowledge 37
Lack of manure 33
Lack of manpower 28
Lack of land for expanding cultivation 21

Comments pointing to a lack of social capital were recurrent. Farmers were disappointed with
unreliable cooperative leaders in the past and had strong mistrust for government authorities,
which many times were accused for being involved in illegal logging.

75
4.4. Spice-agoforestry system (Study IV)

4.4.1. Growth and yield

The survival rates for plants, one year after planting, were 97% for grevillea, 77% for
gliricidia, 59% for pepper and 87% for cardamom. The steepness of the slope at the
experiment site varied between 6 and 26 but it did not affect the survival rate. However, if
the support trees for pepper died, pepper plants also gradually withered.

Grevillea in monoculture (T3) had the highest values for all variables studied (Table 10).
Treatment 2 (grevillea with two crops) ranked second, and Treatment 1 (grevillea with
pepper) followed. Only the height and canopy height did not significantly differ (p> 0.1181)
among the treatments. The mean annual volume increment varied from 3 to 11 m3/ha/a.
Grevillea was a suitable black pepper support tree as its rough bark allows the vine to climb
and attach well to the trunk. Cardamom did not reduce the growth parameters of grevillea; on
the contrary, it improved them. It could be concluded that intercropping did not significantly
affect the growth of grevillea for the first 6 years. The annual height growth was 1.64-1.84
m/a, and the trees reached 10 m in height at the age of five years (Figure 15a).

Gliricidia grew clearly best (p< 0.0005) in monoculture, reaching the height of 7.3 m at the
end of the 6th year; the second best growth occurred in intercropping with pepper (T6) and the
slowest one was found in multiple intercropping with pepper and cardamom (T7; Figure 15b).
Gliricidia in monoculture (T5) had the highest values in all variables studied. The second best
treatment was intercropping of gliricidia with pepper (T6), and the third one was
intercropping with two species (T7). The height growth rates of the tree varied from 86-97
cm/a (Table 10).

76
 Grevillea Black pepper

T1
14 T3 400 T2
12 350
T2 T1
10 300 T7

Height, cm
T1
Height, m

250 T2
8 T6
T1 200
6 T8 T6
150
4 T2 T7
100
2 T3 50 T8
0 0
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7

Year Year

a c
Gliricidia Cardam om

8 T5 250
7 T6
200
6 T7 Height, cm
T2
Height, m

5 150 T7
T4
4 T2
T5 100
3
T6 T4
2 50
T7 T7
1
0
0
0 1 2 3 4 5 6 7
0 1 2 3 4 5 6 7
Year
Year

b d

Figure 15a-d: Regression lines describing the growth of Grevillea robusta, Gliricidia
sepium, Piper nigrum, and growth curves for Elettaria cardamomum.

Treatments: T1: Grevillea + pepper, T2: Grevillea + pepper + cardamom, T3: Grevillea,
T4: Cardamom, T5: Gliricidia, T6: Gliricidia + pepper, T7: Gliricidia + pepper + cardamom,
T8: Pepper.

Black pepper grew best (having the greatest height and fastest growth rate) and had the best
survival rate and vitality when intercropped with grevillea (T1 and T2: 350-360 cm; 65-70
cm/year; survival rate 72%). With gliricidia the growth was quite slow and the survival was
poor (T6 and T7: 243-249 cm; 46-49 cm/year; survival rate 45%). However, the slowest
growth was observed in monoculture with dead support trees (T8: 201cm; 40 cm/year). The
difference (p<0.0001) was clearly observed in height; as for the diameter, only Treatment 1
(grevillea with pepper) differed p<0.0002) from the other treatments in which pepper was
included (Figure 15c). In the second production year, the best pepper intercropping with

77
grevillea (T1) gave on average 124 (48-226) kg of dry pepper/ha, which equalled about four
times that in the other treatments and about 0.33 (0.12-0.61 kg) kg per plant (Table 10).

Cardamom started to produce two years after planting, whereas for pepper the onset of
production took five years. In general, cardamom and pepper production followed the trend of
plant growth. It means that good plant growth was correlated with high production. However,
in the case of cardamom, a high number of tillers was not necessarily correlated with a good
yield.

Cardamom intercropped with grevillea and pepper (T2) exceeded all the other treatments in
all aspects studied, except for the number of tillers, which was the lowest one of all treatments
comprising cardamom. In the first year the yield was tripled and in the two following years
doubled in comparison to the other treatments. In the first year, the dry-weight yield of
cardamom produced by treatment T2 averaged 560 kg/ha (370-710 kg/ha). The yield and the
number of panicles were correlated (p<0.001). Good growth of the tillers (for both height and
diameter) and a high number of panicles had a positive effect on production (Table 10).
Cardamom tillers grew best with grevillea and pepper (T2). The survival rate (92%) and plant
vigour were also the best in this treatment. Cardamom tillers had the highest growth at 3.6-3.9
years, depending on the treatment. The tillers grew shorter after that, probably due to
increasing shade and competition. The fastest growth of tillers correlated with the best
production time. The best treatment with grevillea (T2) was the one with an early fast growth
rate (Figure 15d).

Height functions and the time for reaching maximum height in cardamom were as follows:
T2: Height (cm) = 110.1702 Year - 15.5077 Year 2; Year < 7; R = 96.5%, max h 195.67 cm in 3.55 years
T4: Height (cm) = 88.9520 Year - 11.8893 Year 2; Year < 7; R = 94.2%, max h 166.36 cm in 3.74 years
T7: Height (cm) = 82.6874 Year - 10.5093 Year 2; Year < 7; R = 95.1%, max h 162.64 cm in 3.93 years

78
Table 10. Performance of grevillea, gliricidia, pepper and cardamom in 2006 (year 6 from
establishment) and pepper and cardamom yields in years 3 to 6. Cardamom growth rate was
analysed before this crop reached its maximum height. Rows within species with different
superscripts represent statistical differences between treatments (p<0.05).

Treatments: T1: Grevillea + pepper, T2: Grevillea + pepper + cardamom, T3: Grevillea,
T4: Cardamom, T5: Gliricidia, T6: Gliricidia + pepper, T7: Gliricidia + pepper + cardamom,
T8: Pepper.

Variable Grevillea robusta Gliricidia sepium

Mean Treatment Mean Treatment
T1 T2 T3 T5 T6 T7
Growth rate, cm/a 175 164.1 177.0 183.9 92.3 97.1 93.6 86.1
Height, cm 1120 1050a 1147a 1164a 654 726a 644b 614b
Ground diameter, cm 10.4 9.5b 10.5ab 11.2a 8 9.8a 8.2b 6.7c
b
DBH, cm 7 6.2 7.1ab 7.9a 4.7 5.8 a
4.6b 4.0c
Canopy height, cm 922 854a 939a 976a 614 690a 604b 571b
a
H lowest branch, cm 198 196 208a 188a 39.8 35.2 a
29.7a 42.8a
Canopy width, cm 290 297ab 266b 307a 330 393a 356a 265b
Canopy cover, m2 4.5 4.6ab 4.2b 4.7a 5.1 6.1a 5.5a 4.2b
b
Dry weight, 1000 13.7 9.8 13.8ab 17.7a 14.2 15.7 a
14.0b 13.3b
kg/ha
Volume, m3/tree 0.03 0.019b 0.027ab 0.034a 0.017 0.019a 0.017b 0.016b
Volume, m3/ha 29.2 21.0b 29.5ab 37.7a 18.9 21.0a 18.6b 17.8b
Value, US$/ha 77.7 48.7b 75.3ab 111.0a 30.2 33.6a 29.8b 28.4b

Variable Pepper Cardamom

Mean Treatment Mean Treatment
T1 T2 T6 T7 T8 T2 T4 T7
Growth rate, cm/a 53.9 69.5 65.2 45.6 48.6 40.5 31.1 50,5 33.8 28.3
Height, cm 287 360a 351a 249 b
243b 201c 133 135 a
131a 133a
Ground diameter, cm 1.26 1.50a 1.26b 1.17b 1.20b 1.10b 1.55 1.67a 1.51ab 1.46b
Yield (dry), year 3, kg/ha 0 0 0 0 0 0 74 143a 52ab 25b
Yield (dry), year 4, kg/ha 0 0 0 0 0 0 314 563a 216b 163b
Yield (dry), year 5, kg/ha 32 78a 22b 17 b
11b 31ab 309 436a 279a 213a
Yield (dry), year 6, kg/ha 51 124a 32b 42 b
25b 32b 85 135 a
59a 62a
Number of tillers 23.4 20.6b 21.2b 28.2a
Number of panicles 58 104a 22b 49ab
Number of fruits 1160 2457a 278b 748ab

79
4.4.2. Light conditions

The light intensity, measured as photosynthetically active radiation, under grevillea trees two
years after planting, was on average 29% for cardamom and 24% for pepper, compared to the
light intensity in the open. For cardamom cultivation in the natural forest it was 30%.
Corresponding values under gliricidia were 30% and 45% for pepper and for cardamom,
respectively (Figures 16 and 17).

2500

2000

1500
mol/m2/s
/s
2
mol/m

1000

500

Cardamom+gliricidia Cardamom+grevillea Forest cultivation Open

Hour (6 to 18h)

Figure 16. Examples of the daily course of solar radiation (PAR) in the East Usambara
Mountains.

1000

900

800

700

600
umol/m/s
/s
2
mol/m

500

400

300

200

100

Pepper+grevillea Pepper+gliricidia Open

Hour (6 to 18h)

Figure 17. Examples of the daily course of solar radiation (PAR) in different pepper
cultivation systems in the East Usambara Mountains

80
4.4.3. Soil

All other soil characteristics improved from year 2002 to 2006, except for the pH water and
the Ca content. Most distinct changes were noticed in the cultivation systems including
gliricidia. The levels were higher than in other systems in: N%, C/N, exchangeable Na and
Ca, and Electrical Conductivity. The soil organic carbon increased with time from medium to
very high values in every cultivation system, and most distinctly in systems with gliricidia.
Total soil nitrogen also improved distinctly with gliricidia as the tree component. Soils
remained very acidic in all cultivation patterns. Many soil characteristics (pH KCl, available
P, K and Mg; B.S. % and CEC) showed a decreasing trend from year 2000 to year 2002 but
rose again until 2006. The organic carbon and nitrogen contents increased steadily (Table 11).

Table 11. Comparison of changes in soil characteristics between 2000 and 2006 with
Bonferroni test. An asterisk denotes significant change (p<0.05) from 2000 to 2006 when
comparing each plot separately.

Variable Means Ratio treatment in 2006 / treatment in 2000

2000 2002 2006 T1 T2 T3 T4 T5 T6 T7 T8
pH H2O 1:2.5 5.90 4.91 4.69 -0.4 -0.6 -0.25 -0.75 -0.6 -1.4 -0.8 -0.25
pH KCl 1:2.5 4.30 3.88 4.18 0.25 0.1 0.2 0.15 0.2 -0.3 0.05 0.15
Org C% 2.59 3.58 3.65 -0.24 0.165 -0.39 0.49 -0.365 2.325* 1.865* 0.03
N% 0.16 0.27 0.29 0.015 -0.045 0.01 -0.005 0.065* 0.26* 0.175* 0.01
C/N 16.00 13.50 12.69 -1 2.5 -2.5 1 -4.58* -5.5* -1.5 -1
Av. P mg/kg 2.05 1.25 5.54 2.8 3.595 4.205 4.66 2.92 3.665 4.245 5.29
Na Me/100g 0.21 0.09 0.10 -0.02 -0.015 0.005 -0.05 -0.035 -0.16 -0.105 0.04
K Me/100g 1.28 0.03 0.08 -0.58 -0.57 0.055 -0.59 -0.56 -1.18 -0.635 0.04
Ca Me/100g 3.73 1.51 1.39 -0.42 -0.71 -0.65 -2.13* -0.49 -2.7* -2.76* -0.69
Mg Me/100g 2.33 0.29 0.31 -0.95 -0.96 0.16 -1.34 -1.04 -2.24 -2.01 0.01
EC Ms/cm 0.04 0.07 0.16 0.065 0.085 0.065 0.125 0.115 0.175 0.12 0.07
B.S.% 64.30 37.25 50.94 6 2 19.5* 0 1.5 -15.5 -8 16*
CEC Me/100g 11.70 4.78 9.45 0.78 1.345 2.17 -1.415 1.335 -2.82 1.75 3.635

Gliricidia improved the soil conditions more than did the other plants in the trial. There was a
clear contrast between intercropping with gliricidia and the other treatments without gliricidia
in terms of N (p<0.0001), C (p<0.05), C/N (p<0.005), Ca (p<0.06) and B.S. (p<0.06). The
contrast was highest between Treatment 6 (gliricidia with pepper) and Treatment 7 (gliricidia
with pepper and cardamom), and the contrast was strongest against crop monocultures. Total
soil nitrogen increased in the presence of gliricidia from a low (0.16) to a medium (0.42) level
(Figure 18).

81
 0.4
Nitrogen %

0.3
0.2
0.1
0
T1 T2 T3 T4 T5 T6 T7 T8
Treatm ent

Figure 18. Soil nitrogen content (%) at year 6 by treatment.

The soil organic C content increased from medium (2.5) to very high (3.7), especially in
cultivation systems with gliricidia (up to 4.7; cf. Figure 19). The values of organic C and N
were superior to those in the natural forest. The C/N ratios indicated that the organic matter in
the soil had moderate to good levels, with the value ranging between 10 and 17.

4
Carbon %

0
T1 T2 T3 T4 T5 T6 T7 T8

Treatm ent

Figure 19. Soil organic carbon (%) at year 6 by treatment.

Soils remained very acidic (pH 4.0-5.2) in all present cultivation systems. Highest values
were recorded in systems consisting of crop monocultures (T4 and T8). The available P rose,
on average, from 2 to 5.5, ranging from 4.05-7.02 mg/kg. Notwithstanding this improvement,
the levels were still below critical ones. Maize, banana and grass around the experiment had
also a purplish colour indicating phosphorus deficiency. Magnesium (0.16-0.44 me/100g) and
potassium (0.05-0.13 me/100g) levels remained very low, even though improvements were
observed in the plots. Exchangeable calcium (0.04-2.28 mg/kg) and sodium (0.06-0.15
me/100g) levels were low. Conductivity results (0.09-0.26 mS/cm) showed that the salinity in
the experimental soils was low. CEC (6.5-14.4 me/100g) was low and B.S. (46-56%) was
moderate.

82
4.4.4. Land Equivalent Ratio

Intercropping was highly advantageous for pepper and cardamom, e.g. pepper intercropped
with grevillea produced 3.9 times more than in monoculture, and cardamom intercropped with
both grevillea and pepper yielded 2.3 times more than in monoculture (Table 12).

Table 12. Yields (kg/ha) and the Land Equivalent Ratio (LER) in different agroforestry
components.

Yield LER
T1 T2 T6 T7 Monoculture
GR+BP GR+BP+C GS+BP GS+BP+C (M) T1/M T2/M T6/M T7/M
Grevillea 9827 13816 17684 0.56 0.78
Gliricidia 13968 13321 15736 0.89 0.85
Pepper 123.57 31.91 42.12 24.68 31.54 3.92 1.01 1.34 0.78
Cardamom 135.11 61.76 58.98 2.29 1.05
LER 4.47 4.08 2.22 2.68

The Land Equivalent Ratio was first calculated for each species separately, e.g. as the yield of intercropped
grevillea/yield of monoculture of grevillea (T1/M= 0.56). In other words, grevillea intercropped with pepper
yielded only 56% of the yield of the pure stand. Then the LER was calculated for each treatment. For instance:
T1 = intercropped grevillea/pure grevillea + intercropped pepper/pure pepper = 4.5.

The LER for treatment T1 (grevillea with pepper) was 4.5. That means that 4.5 times more
land would be needed to produce the combined yields of the two crops in pure stands. LER
ratios in treatments with grevillea, T1 (grevillea with pepper) and T2 (grevillea with pepper
and cardamom), showed very high biological efficiency (with LER values 4.5 and 4.1,
respectively), yielding at least 4 times more than the corresponding monocrop systems.
Treatments with gliricidia, i.e. T6 (gliricidia with pepper) and T7 (gliricidia with pepper and
cardamom) resulted in higher than double yields (LER 2.2 and 2.7) as compared to the
monocrop systems. In intercropping the timber yield was always lower, and the yield of the
spice crop higher, meaning that the economic advantage of intercropping is even higher than
its biological efficiency, due to the higher economic value of the spice crops.

4.4.5. Manure test for cardamom

Manure application in August 2006 to cardamom resulted, one year after the treatment in
September 2007, in double yields compared to plots without manure application. The yields
increased most with gliricidia (Table 13).

83
Table 13. Effect of manure on cardamom yield one year after manure application in the
agroforestry trial.

Treatment Yield increase, % Ratio

Without manure With manure with manure/without manure
T2 (GR+BP+C) 16 31 1.94
T4 (mono) 95 186 1.96
T7 (GS+BP+C) 78 183 2.35

84
5. DISCUSSION

5.1. Biodiversity conservation and sustainability

There is a conflict in Africa between people and biodiversity. People need to get their
livelihood from the forest, and high-biodiversity areas should be protected. A study on three
African ecoregions (the Eastern Arc, the Albertine Rift and the Cameroon-Nigeria mountains)
shows that high biodiversity values of tropical mountains are correlated with a high human
population density. High endemism is correlated with a higher prevalence of agriculture. This
is logical, because human populations have been concentrated in biologically rich areas where
the climate is also wetter and more reliable (Balmford et al. 2001; Burgess et al. 2007a).
Communities in the East Usambaras continuously change the forests surrounding them by
different activities (Kessy 1998). A high conservation value is often linked with rapid habitat
clearance or a high human population density (Balmford et al. 2001). More than 1.1 billion
people live within the worlds 25 biodiversity hotspots, and the population growth in these
areas remains high, at about 3.1%, which is almost twice the global average (Mery et al.
2005).

Buffer zones are areas that form an interface between human and conservation activities.
They are thus an integral part of the protected area concept. In the East Usambaras, a
sustainable supply of products is made available in the buffer zones to the local people, who
reduce their dependence on products from protected areas. The intention is that farmers
should feel that they have a stake in the reserves so as to motivate the protection of these
reserves (Kajembe and Mwaseba 1994). The villages situated close to the border of the Amani
Nature Reserve (at a distance of about 3 km) constitute the buffer zone. An ANR management
guide is suggesting agroforestry systems as a more sustainable land use practice in buffer
zone management 10.

Communities in the East Usambaras depend on the natural forests mostly when they need
building material and fuelwood. Even the house sizes were found to be highly dependent on
the distance to the forest; if the forest where building material was collected was far away, the
houses built were small-sized. About 79% of the poles originated from natural forest species,

10
Pers. com. with Mr. Corodius Thomas Sawe, conservator of ANR, 30 September, 2005.

85
of these 14% were endemic. Species from the farms (mostly Tectona grandis, Eucalyptus
camaldulensis, Senna siamea and Grevillea robusta) comprised 21% of the building material.
Guava (Psidium guajava) has also been used as building material in areas where there is not
better species available. The quantities of other products collected from the forests were so
small that they hardly comprise a threat to the environment and biodiversity. However, if the
plant has to be harvested totally, or if the roots of a plant are used, this can also cause
biodiversity loss. People are cultivating many crops under forest canopy also due to the more
humid growing conditions there. The edges of conservations areas are the most commonly
utilised areas (Kessy 1998).

Woodcock (1995), however, states that in Tanzania the forest-adjacent communities do not
depend entirely upon forest products from the reserves for their subsistence. When forest
resources become inaccessible or unavailable, the local people have proven to be highly
adaptive and able to manage other wild resources derived from their farms or common land
(Woodcock 1995). Local farmers have developed their farming systems by trial and error, and
they have a strong connection to the land. Many of them have seen the effects of
unsustainable land use in the West Usambaras and the other highlands. Actually, many
immigrants in the East Usambaras come from the nearby highlands that are already highly
eroded and where the land has become poor. Some of these people have learned to cultivate
crops in a more intensive and sustainable manner due to their own trials and due to the
facilitation of many development projects. They use farmyard manure and keep the same
piece of land productive year after year. But there are also always those farmers who have
immigrated to the Usambara area recently, without roots in the area, and hence are less
concerned with the restrictions and controls of land and forest use that generally are
associated with traditional settlements (Howard 1996).

Biodiversity is valued differently among the stakeholders. Conservationists value the

scientific aspects in biodiversity, whereas the local people are more thinking about utilitarian
diversity, i.e. the value of various benefits and uses of forests (Kessy 1998). Protected areas in
Tanzania lack much of the basic biodiversity knowledge that would help in conservation
management. Comprehensive inventories of flora and fauna are thus needed (Johansson et al.
1998). The global concern of the ecological situation in the East Usambaras has continuously
increased, and there have been conservation projects in this area since 1985. The Eastern Arc
Mountains have been identified as a Priority Ecoregion (WWF), an Endemic Bird Area
(BirdLife International) and a Biodiversity Hotspot (Conservation International; Burgess et al.

86
2007a). This helps in obtaining more resources from various donors for conservation area
management.

An expansion of the conservation area limits the farmers access to the forests, and many of
them had to leave their cardamom cultivation sites in the proposed Derema Corridor area
(Figure 20). The farmers have been compensated with the value of the crops (mainly
consisting of cardamom), and in some cases with a new farm site. Farmers have complained
that the amount of compensation hardly covered 20% of the real crop value, and the new land
areas were far away and not suitable for cardamom cultivation. Additionally, the
compensations were paid late: farmers had to leave their cultivation sites in 2001, but the
compensations were still in progress in December 2007. There were also information gaps
between villagers and the forest authorities during the process, which has been in
contradiction with the transparency and participatory process originally envisaged. Still most
of the interviewed farmers supported the conservation activities. They wanted to preserve the
forests to maintain the rains and the microclimate that are suitable for their cultivation. They
were worried about further soil degradation, unpredictable rains and a decrease of yields. In
particular, the unclear compensation process has created negative attitudes towards
conservation in many villages. A sound compensation scheme would have the potential to
improve peoples daily lives and support the conservation status of the protected area, but the
poorly implemented compensation scheme seems to create the opposite effect.

Figure 20. Cardamom cultivation in the East Usambaran forest under the canopy of endemic
Cephalosphaera usambarensis. Other species at the site are Anthoclesita grandiflora (Gilg.),
Macaranga capensis (Bail.), with a few Maesopsis eminii (Engl.). (Photograph: Teija Reyes).

87
Local peoples rights over natural resources provide a basis for compensating them for the
environmental services their resources provide, because the whole landscape, not only the
core protected areas, should be more biodiversity-friendly. Further conservation activities
must acknowledge that local people, with stakes in the forest resources or land, will have to
benefit from and become part of conservation management, whether the forests are within
conservation areas, in forest reserves, or on unprotected public land. Insecure land rights are
often claimed as a key cause for deforestation, and a barrier for adopting better management
and tree planting initiatives (Katila 2008). Interestingly, in some cases land insecurity has also
boosted tree planting on farms. In some parts of Tanzania (Yasu 1999) and Sumatra
(Angelsen and Kaimowitz 2004) tree planting has been a strategy to claim land rights.

Improvements in the productivity and sustainability of existing cultivations are the most
effective ways for tackling the problems of deforestation in the East Usambaras. In general
terms, sustainable agroforestry systems contribute to conserving the plant and animal
biodiversity by reducing the further clearance of tropical forests. Farmers can benefit from
already cleared lands by obtaining cash income and various products from improved
agroforestry systems on them (Lundgren 1985). Each successful agroforestry hectare has been
estimated to save five hectares of tropical forests (Sanchez and Benites 1987). Globally, the
establishment of 1 million ha/year of agroforestry systems was later estimated to save up to
7.5 million ha of forest annually (Schroeder 1993 in Sanchez et al. 1997). These figures have
been, however, criticized as too optimistic and general by Angelsen and Kaimowitz (2004)
because they were based only on a single study. Although agroforestry may not entirely put
on end to deforestation, in many cases it can reduce the pressure on natural ecosystems.

Agroforestry contributes to conservation by reducing the pressure on protected areas, by

increasing the biological diversity of production systems, and by forming habitat-rich mosaics
around protected areas (Swallow et al. 2006b). Agroforestry can help in connecting important
conservation areas, as the Derema case has shown. Another example on the potentials of
sustainable cardamom cultivation is found in Nepal, where wildlife corridors are formed by
Alnus nepalensis D. Don grown in a cardamom agroforestry system for connecting the
Makalu Barun National Park with the Conservation Area of Eastern Nepal (Swallow et al.
2006a).

88
Agroforestry systems are more diverse than crop or forest plantation monocultures, and
sometimes, e.g. in Southeast Asia, almost as diverse as the natural forests. In mature rubber
agroforestry systems in Sumatra, Indonesia, the species diversity can be more than 70% (300
species/ha) of that in natural forests (420 species/ha; Sanchez et al. 1997; cf. also Ihalainen
2006). In the East Usambaras, the plant species diversity at some agroforestry sites reaches up
to 50 to 80% of that in natural forests and provides most of the products needed by the local
families (Huang et al. 2002). These agroforestry systems have been reported to maintain half
of the richness of the original bird species and almost all of the richness of mammal species.
The below-ground biodiversity is also higher in agroforestry systems than in crop
monoculture, and in the long term, agroforestry systems can sequester carbon in their tree and
underground biomass for decades (Sanchez et al. 1997). Agroforestry systems have a higher
carbon sequestration potential than degraded forests. The above-ground biomass
accumulation in an agroforestry system in Central Himalaya was found to be almost four-hold
(3.9 t/ha/a) compared to degraded forest lands (1.1 t/ha/a; Pandey 2007).

Forests may also maintain the stability of climate. It is assumed that climate change will
worsen the food security situation especially in Sub-Saharan Africa, with reduced yields,
increased pest and disease attacks and extreme natural phenomena, such as floods and
droughts as a result (Kaimowitz 2003). There are already some symptoms seen which can be
understood as effects of climate change in the East Usambaras. In any case it is clear that rural
Africa has limited capacity to adapt to any possible climate change. That is also why the
natural forest cover should be maintained for ensuring at least stable local microclimates.
Farmers have to broaden the genetic base of their planting material to face the challenges of
the future. The planting stock of a tree species may on a particular farm originate from one
single tree. Many agroforestry systems now contribute to keeping the soil in place by forming
biological terraces; thus with agroforestry it is also possible to rehabilitate already degraded
and abandoned lands in the mountains areas (Sanchez et al.1997). For rehabilitating
abandoned cardamom cultivation areas in the East Usambaras, planting of large-sized
seedlings is probably the only way to accelerate succession on sites covered with dense weeds
(Hamunen 1998).

A combination of improved agroforestry with protected forests areas, and intensive forest
plantations, could thus contribute both to improved livelihoods and maintain the biodiversity
in the East Usambaras.

89
5.2. Better land use

Widespread forest cultivation, especially for cardamom, has caused a rapid decline in the East
Usambaran forest cover and been the first step in a land degradation process. Sugarcane
cultivation has led to environmental degradation on steep slopes in the area. The remaining
biodiversity-rich forests in the East Usambaras can only be saved if the land-use is better
planned. Specific areas with high biodiversity values should be identified and protected
completely, while other areas could be developed to become community use zones (Johansson
et al. 1998).

Further land expansion for cultivation is already impossible in many East Usambaran villages.
In the cultivated areas, a transition from short-term land exploitation to sustainable land use
should be encouraged. The aim is to develop a farming system that is sustainable and cheap,
and which can be adopted by farmers on densely populated marginal lands. The aim should
also be to achieve and maintain optimal soil conditions (Hellin 2006).

It is commonly assumed that a higher income will automatically provide prerequisites for
higher agricultural yields, especially when the income is obtained from off-farm activities, as
is the case in the East Usambaras. It was found in the present study that off-farm activities
contribute about 16% of the income for farmers practicing improved agroforestry systems in
the study area. Off-farm activities can be associated with more investment in soil and water
conservation (Nkonya et al. 2004). However, Masayanyika (1995) came to the conclusion that
in the East Usambaras the best performers were full-time farmers who drew their livelihood
mainly from the farm. When the livelihood security was obtained from on-farm activities,
farmers tended to put more efforts to conserve the soil and water resources. The higher the
proportion of off-farm income was in the overall household income, the lower was the on-
farm income and the households effort to protect the land. Off-farm activities can also further
reduce the availability of on-farm labour, because normally men are working in the off-farm
sector, and the management of the farm is left to an already overworking wife and to children.

It has been claimed that small-scale farmers do not have a future, and they should shift
completely to working on larger farms or in off-farm activities. However, small farms have
been found to be more efficient when considering the production per unit of land area. This is
normally explained by higher self-motivation. Small farms use the labour more efficiently and

90
are more cost-efficient, due to low supervision costs. Small farms have also other important
non-monetary values. They form an important social safety net for the rural poor, even if the
farms are too small to become commercially viable. New technologies may require capital
and higher levels of education, which may disadvantage the small farms, but with regard to
equity and poverty reduction, the development of small farms should be preferred over that
for the large ones (Hazell et al. 2007).

Some soil types cannot support the continuous cultivation of annual crops, and steep slopes
cannot support sustainable agriculture. Sometimes developing a simple policy for land use
can be important for decreasing the erosion rates. In Japan, gently sloping fields (5% or less
slope) are used for the production of vegetables; middle slopes (8 to 15%) for fruit
production; steep slopes (16 to 30%) for uncultivated grasslands, and beyond 30% slope the
land is reserved for forest trees (Gomez and Gomez 1983). In Tanzania, special attention
should be given to the management of the mostly degraded public lands which cover about
54% of the natural forests and woodlands. Most of this area is without proper management
(Luukkanen et al. 2006; Katila 2008) and a half of these forests has already been encroached
by farmers (Kessy 1998). These lands are, however particularly important for the landless and
the poorest families (Mery et al. 2005).

The Tanzanian National Soil Service (1989) has given recommendations for sustainable land
use in the East Usambara Mountains which are still valid:

- Replant all treeless slopes of over 30% with trees, also with leguminous ones;
- Only lands with slopes of less than 30% should be considered for agriculture;
- Use multipurpose agroforestry;
- Strongly discourage the cultivation of cardamom in the natural forests;
- After tree cutting, replanting should be done as soon as possible;
- Planting of fast-growing tree species for commercial plantations is not
recommended;
- Tillage should be kept to at a minimum level;
- All annual and perennial crops need to be supplied with added nutrients (e.g. by using
manure or leguminous trees);
- Organic household wastes should be converted into compost and used in
homegardens.

91
Land degradation is often related to an increasing human population pressure, but this is not
always the case. There are examples, e.g. in Kenya, Burundi and Uganda, where the number
of trees on farms is expanding with an increasing population. Experience from complex
agroforestry systems in Southeast Asia also shows that a sustainable, biologically diverse and
economically viable intensification of land use is possible (Sanchez et al.1997). Complex
forest gardens or agroforestry systems, dominated by perennial plants, are ecologically
sustainable but labour-intensive and thereby requiring skilful management.

5.3. Improved agroforestry as a sustainable cultivation system

It is widely believed that agroforestry has a considerable potential as a land management

alternative for maintaining the soil fertility and productivity in the tropics. In addition, low
input costs in many agroforestry systems render them available also for poorer farmers, who
normally suffer most from the effects of erosion. Agroforestry is applicable to a wide range
land types and has a potential for erosion control through the soil cover provided by the tree
canopy and litter. Field crop production mainly affects the upper soil strata, while agroforestry
exploits the entire soil profile (Young 1989). The aim of agroforestry is to optimize the
positive interactions of all components included and the physical environment (Lundgren and
Raintree 1983). Improved agroforestry systems can provide for a landowner different
alternatives for short- and long-term investments and allow the spreading of financial risks
through diversification (Hoekstra 1987). This approach can reduce both the ecological and the
economic risks of the farmer (Angelsen and Kaimowitz 2004).

Simple ways to fight erosion are a higher plant density and mixed cropping, which both
provide better soil cover than monoculture. Trees provide a double protection of the soil
surface with the litter layer and with the leaf canopy. It has been found that a litter cover can
reduce the erosion rate by 95% as compared to bare soil, and the maintenance of a ground
vegetation cover of 60% or more provides a good way to reduce the erosion (Young 1989).

A change from traditional systems to improved agroforestry requires more time for planting,
and often there are limited means to hire extra labour. However, a mixed cropping system also
allows flexibililty in the time of planting, e.g. for trees and food crops in the first year, food
crops and spices between the tree rows in the second year, and food crops between the trees
and spices in the third year. For adequate yields, all planting has to be done at the onset of the

92
rains, so as to allow the crops to successfully compete with weeds. All possible land
preparations should be done before the rains. Digging of holes for seedlings and some seeding
can also be done prior to the onset of the rains (Beets 1990). Research results from Kenya
show that when the change to a new recommended system call for five times more effort for
planting, the adoption of the new practice was not very rapid, even though the yields
increased substantially (Keswani and Ndunguru 1982).

Organic matter management is critical for protecting the physical structure of soils, but it is
not sufficient as the only practice. Some soil types cannot support continuous cultivation of
annual crops, which is a reason why economically attractive perennials and cover crops must
be incorporated into the cultivation system (Scherr 1999). Annual crops should not be grown
in badly deteriorated areas, and steep slopes should never be left bare (Gomez and Gomez
1983). There are small, simple things one can do to better manage sloping lands. In a case
reported by Young (1989) a farmer in the Usambara Mountains maintained a good ground
cover, allowed the weeds to grow, cut them, and left them at the site as mulch; he did the
same with the maize residues and never burnt them. As a result, the soil loss on a 25-degree
slope was only 0.01 t/ha/a, when normally it is in the excess of 10 t/ha/a on the steep slopes of
the Usambara Mountains (Young 1989).

Some researchers estimate that the yields of any given crop in Tanzania are only 20-40% of
the potential (Uliwa and Fischer 2004). However, for more profitable cultivation the use of
external inputs should be limited to the essential. That is done by optimising the use of
internal resources, by using efficiently what already is available. There is no need for
agriculture to expand into uncultivated lands, as the existing farmlands already contain a
potential that is overlooked. There is no need for expensive commercial fertilises and
pesticides when there are natural ones available.

The present study resulted in a number of characteristics for the two groups of farmers which
were compared in the present study (with traditional or improved cultivation methods; cf.
Table 3). Most significantly, the farmers who used improved methods (74%) also kept cross-
bred cattle, which implies that the Dutch dairy project had a great influence on the incomes
and the livelihood development in the area. These farmers also had 30% larger land holdings
and more off-farm income. However, there were also farmers who did not own cattle, or
possess large farms, but who were efficiently using the existing land area, with mixed
cropping systems and available organic manure, and thus obtaining a higher income than their

93
neighbours. There were also farmers with large land holdings and improved cattle who did not
cultivate their land efficiently or use farmyard manure, and thus remaining with low income.

According to the farmers interviewed, one of the most important reasons for adopting an
improved cultivation method is the knowledge of it. Many farmers lacked information and
encouragement to change their traditional cultivation systems. There are gaps in knowledge,
and often research findings, projects and extension services fail to reach the majority of the
farmers. It can also be speculated that some farmers are more innovative than others in using
new methods. For example, most of the farmers practicing improved agroforestry also used
improved stoves.

The location was an important factor characterising the farmers who used either improved or
traditional cultivation methods. As regards adoption of improved cultivation methods, it
seemed in Study I that the highland farmers had better opportunities to develop effective
farming systems than those in the lowland. It was observed that 67% of the highland farmers
used improved farming methods, compared to only 23% of the lowland farmers. This was
most probably due to better road connections, higher off-farm income (mainly from tea
estates), larger land holding size, and a more secure land tenure. Highlands have until recently
had more public forest areas available, which explains the larger land holdings there. As an
additional factor, highlands, which are more fragile ecologically, have been the focus of
various development projects in the area (EUCADEP, EUCFP and EUCAMP), and the
information on new income generating activities has better reached the highland farmers.
These facts can at least partly explain the considerable difference in income and cultivation
practices between highland and lowland farmers.

As discussed above, the present study could not precisely or separately analyse all the factors
underlying the observed differences in crop yields and income between the two farmer groups
(i.e. those with traditional or improved cultivation methods). The differences may also partly
be explained by the qualitative criteria used when forming the two groups (traditional,
improved). These may provide a categorization into well resourced and less resourced
farmers. All the information on cultivation practices and their resulting effects was based on
reporting by the farmers. As it is known, the farmers in the East Usambaras are not normally
keeping any records concerning the yields or income; thus all the quantitative information can
be considered only as an estimate.

94
There are actually as many farming combinations and systems as there are farms. On the
farms studied in the East Usambaras, the number of economically utilized plant species varied
from 7 to 30 on an individual farm and totalled more than 200 (Risnen 2001); over 10% of
the domesticated species of forest origin were endemic or near-endemic (Kessy 1998).
Farmers need, however, professional assistance to find the most suitable combinations in
respect of their family needs, the soils, and market and labour limitations. The main task is to
choose the most suitable tree species for system.

In a study where a complex artificial system (100 species per plot) was compared to natural
succession, it was found that the natural system was always ahead in terms of floristic
complexity, canopy development and productivity. Both systems retained the nutrients well,
the natural system being only moderately better than the mimicked one. It was concluded that
the longevity of the components, as well as diversity, is a key to nutrient retention in the
humid tropics, and that it is important to be prepared to accept lower but biologically
sustainable yields (Ewel 1999). Long-term commitment normally brings about better
management practices and yields for farmers, as was noticed in the present study in the
comparison which indicated that the cardamom farmers in the Derema Corridor obtained 13%
higher yields than those in the Amani Nature Reserve.

5.4. Spice-crop agroforestry systems

Documented agroforestry systems with spices including cardamom and pepper are found at
least in Kerala, India; Kilimanjaro, Tanzania (Fernandes et al. 1989); in the Kandyan gardens
of Sri Lanka (Jacob and Alles 1989); and in Guatemala (Urquijo 2004). Subsistence-oriented
homegardens are gradually changing so as to also produce high-value cash crops. This seems
to be a global phenomenon observed in many countries (e.g. India: Peyre et al. 2006;
Tanzania in Kilimanjaro: Soini 2005; and the East Usambaras: Reyes, personal data;
Guatemala: Urquijo 2004; Kenya: Nkonya, et al. 2004). At least in some parts of the
highlands this is already necessary due to limited possibilities to expand the area under
cultivation.

It is recommendable to encourage the farmers to plant cash crops in the upper parts of the East
Usambaras for several reasons. Cash crops can generate the same income on a smaller land
area as compared with food crops, and high prices may serve as incentives to farmers to adopt

95
soil conservation practices (Brown 1981, Hamilton 1989). Generally, however, cash crop
cultivation should shift from the forest to homegardens.

In the East Usambaras, people living adjacent to the forests are still highly dependent on them
as a source of timber and non-timber products (Katigula 1999), but they have also showed
very high adaptability to new conditions (Woodcock 1995). Farmers who had the largest
cultivation areas, especially commercial cardamom cultivation in the proposed Derema
conservation area, were from IBC Msasa, Makanya, and Antakae. Many of these farmers,
who had to leave their forest cultivations, subsequently started to cultivate cardamom in their
homegardens (Figure 21). Many farmers admitted that cardamom is producing better with the
help of manure and with other management improvements, even though in the forest the
moisture and temperature conditions were better throughout the year.

Figure 21. Homegarden of Mrs Yasinta Stephen. Cardamom is intercropped with guava,
papaya, banana, among others. (Photograph: Teija Reyes).

The present on-farm experiment showed a high potential especially for cardamom production
in agroforestry systems. The Land Equivalent Ratio was highly advantageous for mixed
cropping of cardamom and black pepper, especially with Grevillea robusta, in the East
Usambaran highlands. Both spices grew better with shade trees than alone. The growing of
Gliricidia sepium also improved the soil, especially nitrogen and organic matter contents. In
comparison, the ideal solar radiation level for most medicinal and aromatic plants, including

96
cardamom and pepper, in Karnataka, India is 30% to 50% (Rao et al. 2004), and in the
Western Ghats in India, 40-60% of the light intensity in the open (Kumar et al. 1995). In the
present study, trees allowed 24-45% of the incident light to reach the understory, and the light
conditions under these trees were considered suitable for spice production.

Farmers in Derema claimed that the soil in the cardamom cultivations areas in the forest
maintains its fertility without manure application. However, this was observed only in some
gaps in the forest, usually in small valleys, which had a very moist microclimate. Farmers in
the study area normally thought that the soil fertility is adequate, even when they complained
that the yields were decreasing. The average cardamom yields decreased by 22% in 10 years
in the East Usambaras, from 80 kg/ha as reported by Masayanyika in 1995 to 62 kg/ha in the
present study. On the other hand, the yields were reported by farmers who do not usually keep
records, and they can be considered as rough estimates. Soil analyses showed that
intercropping of trees and spice crops can improve some soil characteristics, but the addition
of organic manure is still needed to maintain the soil productive in the long term. Five years
after establishing the plantation the production of cardamom rapidly starts to decline. With
fertilizers it could probably be maintained at a profitable level for a longer time. Recent
preliminary new data from the study area support this finding. Application of farmyard
manure clearly increases the cardamom yields, especially in combination with nitrogen-fixing
trees.

The dry weight yield of cardamom (560 kg/ha in average) in year four with grevillea and
pepper in Treatment 2 (T2) could be considered as an optimum without manure application
and improved planting material (Figure 22). The normal average yield in India is about 150
kg/ha/a (Korikanthimanth 2001), but with improved varieties and good management
cardamom can have much higher yields. Thus, intercropped with Robusta coffee (Coffea
robusta L.) cardamom yielded 1 400 kg/ha four years after planting and, on average, 672
kg/ha over seven cropping seasons. In this case organic and commercial fertilizers and
irrigation were also applied and plant protection measures implemented (Korikanthimanth et
al. 1998). The second-best treatment at the present experimental site, surprisingly, was a
monoculture of cardamom (T4). The results were, however, biased by a high production on
one plot (more than 570 kg/a) of the four replications; the yield in the three other plots was
160, 60 and 24 kg/ha, respectively.

97
Figure 22. Grevillea, cardamom and black pepper (Treatment 2) in the field in November
2005. (Photograph: Teija Reyes).

Farmers in the East Usambaras grow the pepper vines one by one around suitable trees and
stones near their houses. The estimated average annual yield per plant among 37 farmers was
0.92 kg. There are no comparable hectare yield data from the East Usambaras for comparison
with yields in other black pepper producing countries. The average production of black
pepper in India is 300 kg/ha, but the countrys productivity figures vary much, ranging from
91 kg/ha in Pondicherr to 1 406 kg/ha in Karnataka. In the same country, one hectare of
monocrop pepper regularly fertilized and irrigated gave an average of 600-800 kg/year, which
is estimated to equal to 0.375-0.5 kg/plant (Tejwani 1994). Worldwide, pepper yields have
reported to range from 270 kg/ha in India to 3 445 kg/ha in Thailand (Black Pepper
Cultivation 2007) and 1 742 kg/ha in Vietnam (Thankamani et al. 2007). The considerable
yield variations are partly explained by differences in agroecological conditions, ages of
pepper plants, and the number and frequency of harvests. Production technologies adopted at
a particular location may not be suitable for other ones, because there is not enough
information on cultivation techniques at different locations. The first pepper harvest in India
occurs after three years of cultivation (Sivavaram et al. 1999), but in the present trials the first
harvest was obtained five years after planting.

98
In Tanzania, black pepper is not regularly pruned. The pepper production could be, however,
increased with regular pruning of the main stems, which could induce the development of new
lateral fruiting branches. In India the vines are pruned 7-8 times until they reach about 3 m in
height in about three years. When this has been achieved the terminal shoot is pinched out
periodically to prevent pepper plants from growing taller and to facilitate harvesting
(Purseglove et al. 1981). We planted the support trees and black pepper seedlings at the same
time, as is common in Indonesia. Planting the support trees earlier and letting them root well
before planting the pepper crop could give better results. In Sri Lanka gliricidia cuttings are
planted six months before the pepper vines (NFTA 1989), and in India living standards are
planted up to 3-4 years in advance to allow them to attain a sufficient height at the time of the
planting of black pepper. It is also important to weed around pepper plants regularly and to
protect the roots from heat with mulch (Sivavaram et al. 1999).

In our study, grevillea performed very well in intercropping with spices, which explains its
success as the most popular farm tree in the East Usambara Mountains. Harwood (1989)
claimed that grevillea does not perform well in monoculture plantations. In addition, grevillea
can have a positive effect on the growth of associated crops. Intercropping grevillea at 300
trees/ha with banana and beans increased the crop yields by 70-75% in the highlands of
Burundi (Akyeampong et al. 1999). Small sawlogs of grevillea are often harvested on farms
in Africa only 15-20 years after planting, but some agroforestry designs advocate rotations as
short as six years (Harwood 1989). The present trial did not support that recommendation. In
6 years only 40% of the trees exceeded the breast height diameter of 11 cm, which is the limit
for firewood becoming a timber tree and doubling its value per m3. A rotation of 15-20 years
is more reasonable in the East Usambaran highlands.

Grevillea can produce more when it is regularly pruned. Removing grevillea branches
provides fodder and firewood and more sawn timber, improves the timber quality, and
reduces excessive shading. In Kenya it has been estimated that pollarding ten trees was
enough to provide fuelwood for a family of eight people on a sustained basis (Muchiri 2001).
Kiriinya (1999) observed that light to moderate pruning (removing 1/3 to half of the crown)
induces the tree to increase its diameter in the upper part of the stem, making the stem more
cylindrical and thus increasing its value. Side pruning and chopping off side roots are also
common practices on small East African farms (Harwood 1989).

99
In the East Usambaras, gliricidia was mentioned as one of the best pepper support tree when
the yield and viability of pepper were concerned, and it was the second most popular tree on
lowland farms. Its cultivation at higher elevations was also recommended (Vihemki 2001).
The growth of gliricidia was very modest in this trial, which was at an altitude of 1 000 m.
This is probably explained why this species is not common at the higher altitudes (above 850
m) in the East Usambaras. The ecological altitude limit of gliricidia is, however, claimed to be
1 200 m (Agroforestree Database 2006). The reasons for poor results with gliricidia can also
be a late planting of cuttings, disease attacks on some plots, or an altitude too high for this
particular variety (cuttings were collected from 600 m elevation). These reasons could have
increased the mortality of gliricidia cuttings during the first two years, together with the delay
of rains that year. However, the survival rate observed (77%) was superior to that (58%)
reported for Gliricidia sepium in Sri Lanka (Gunathilake and Wasanthe 2004).

The mosaic virus transmitted by banana aphids attacked gliricidia in the experiment in
September 2002. The most seriously affected plots were plots number 6, 15, 19 and 32. They
were sprayed with natural, locally prepared neem extract, as follows: One kg of neem
(Azadirachta indica L.) leaves was soaked over night in water. The next day the leaves were
mashed and the extract was filtered. Affected plants were sprayed at the end of October 2002,
and the treatment was repeated after the rains. In four months the affected trees had recovered.

The height of gliricidia varied from 2.4-9.5 m at six years after planting. Normally, the height
of gliricidia varies much; for instance, it ranged from 2-15 m among provenances grown in
similar conditions in Sierra Leone (Amara and Kamara 1998). The wood production of
gliricidia in Guatemala was found to be from 5 to 6.8 kg/tree at 36 months (NRI 2005). In the
present trial the dry weight at the same age was only about 2-3 kg/tree, and after six years it
was, on average, 12.8 kg/tree. Pruning can accelerate the growth, which means that actually
the potential biomass production of gliricidia in the East Usambaras is not known. Gliricidia
can be cut every four months for fodder or green manure, but this practice was not followed in
this experiment, because a study of the functioning of the whole system was the main aim.
The steepness of the slope at the experimental site varied between 6 and 26 degrees, but this
did not affect the survival rate or production of any trial species.

Species selection in agroforestry systems depends on the primary goal of that particular
agroforesty system. In the East Usambaras, where biodiversity conservation of the natural
forest is the priority aim, domestication of wild forest plants used by the local population

100
could be promoted in the agroforestry systems. In the East Usambaras, 10% of the
domesticated forest species were found to be endemic or near-endemic ones (Kessy 1998).
Separate protected areas are needed, and agroforestry systems could provide the farmers with
sustainable and profitable products (Pandey 2007).

It may be criticised why exotic tree species were chosen for the present study. There is,
however, a strong belief that cardamom is only growing well under the shade provided by
indigenous trees (EUCAMP 2001), and that is why popular exotic trees of farmers
homegardens were now selected for an analysis. Some farmers mentioned that it was good to
know how cardamom is growing under other than indigenous trees. Furthermore, Grevillea
robusta and Gliricidia sepium are included in the list of principal trees and shrubs
recommended for soil improvement in this region (Young 2004). Farmers reported many
indigenous trees suitable for growing together with cardamom, e.g. Allanblackia stuhlmannii,
Cephalosphera usambarensis and Newtonia paucijuga. Almost all farmers intercropped
cardamom with banana, and these two crops performed well together. Popular homegarden
tree species also included other spices (mostly clove and cinnamon) and fruit trees. The trial
was established on land belonging to a local farmer who voluntarily offered that piece of land
for research. As remuneration he got all the production from the experiment, and, naturally,
that piece of land was returned to him after the research was completed.

The propagation material for spice plants is quite expensive in the East Usambaras. The initial
investment for cardamom is higher than for other crops; i.e. (in 2001) US$ 2.3 (TZS 2000)
per ha, as compared to beans, US$ 0.006 (TZS 5) per ha, or maize US$ 0.009 (TZS 8) per ha
(currency rate during the study in November 2001: US$ 1 = TZS 876.4). Harvesting of
cardamom is also labour-demanding, because mature capsules have to be collected manually,
one by one, every three weeks for six months. The farmers could, however, plant few
cardamom plants and let them spread and then propagate the plants from tillers. Gliricidia is
propagated with cuttings, which is a very quick and cheap method. The tree has also great
possibilities in apiculture. Grevillea can be propagated from seeds in a seedbed next to the
field, and the seedlings can then be planted where needed.

Advances in biotechnology can make little difference for poor farmers who know the use and
cultivation practices of the locally available resources so well (Pretty et al. 1996). New good-
yielding varieties of exotic timber trees and spices would, however, be welcome. This could
include introducing pepper varieties that could be harvested at a younger age. There is a

101
considerable potential to increase the yields with better planting material, as evidenced by the
two-fold increase in cardamom yield and five-fold increase in pepper yield obtained in India
with improved material (Korikanthimath et al. 1998). Only a very limited spice plant material
was introduced to the present area for experimental purposes during the colonial era.
Certainly there would be other suitable varieties for growing in the East Usambaras. In a
study conducted in Brazil, the growth of the local grevillea variety was inferior to that in
newly introduced provenances, but on poorer sites the local land races seemed to perform
better (Shimizu et al. 2002). This indicates that in some cases the local varieties have been
better adapted to the prevailing conditions. In the case of the Usambaras in Tanzania, it is
necessary, however, to introduce new propagation material to widen the genetic base, and to
provide plants more resistant to drought, soil acidity and climate change. As far as food crops
are concerned, farmers in the study area have indicated that they are not interested in trying
improved varieties, due to their different taste, but for spices and trees such ways for
obtaining better yields are easier to adopt.

5.5. Improved soil quality

A secured tenure and a right to resources and adequate livelihoods derived from family
farming are prerequisites for good land husbandry and management. When people have
secured property rights, there is often greater incentive and motivation to manage the
resources sustainably (Investing in Mountains 1997; Katila 2008). A general uncertainty of
land tenure in Tanzania has had a negative effect on soil fertility, because there are few
incentives to apply long-term soil fertility improvement measures on lands leased perhaps for
one season only. If the yields fall below the average, farmers normally decide to put the fields
under fallow. Farmers normally pay more attention to soil fertility management on their
homesteads, by regularly applying organic household waste and cattle manure.

In the present study an analysis of cattle manure showed that it had very high contents of
nitrogen, phosphorus, potassium, and organic carbon, and a high content of calcium which
would improve the pH of the soil. But only one farmer out of ten used any organic fertilizer in
the East Usambaras, although most of them kept livestock and had farmyard manure
available. Households applying manure used, on average, 150 kg/year of dry manure on their
3-ha cultivated land out of the 2-4 t/year produced by its approximately three heads of cattle.
Thus less than 10% of the available manure was used, mainly because of high transport costs

102
(cf. Boesen and Friis-Hansen 2001). The production of cardamom in the forest decreased after
a few years due a decline in soil fertility. Farmers normally did not apply any manure to the
forest cultivations. In addition, they many times still removed or burned the grasses weeded
from the site (Sengoe 1994).

Spreading the manure requires less work and has a residual effect for 4-5 years, compared to
the application of manure in planting pits which has an effect that only lasts up to one year.
However, in rural Tanzania, the most common way to add soil nutrients is to burn the crop
residues and add the ash to the field, for facilitating the cultivation and avoiding termites
(Boesen and Friis-Hansen 2001). The burning of residues, however, results in huge losses of
N (up to 80%), P (25%), K (21%) and S (4-60%), depriving the soils of organic matter, which
is a threat to sustainability (Mandal et al. 2004).

Low crop density and diversity accelerates the nutrient losses. Research results from Uganda
have suggested that crop biodiversity (in intercropping systems) contributes to a favourable
nutrient balance. For households which cultivated diversified perennial-annual crop systems,
it was common to recycle mulch or to compost the plant residuals (Nkonya et al. 2004).
Covering the soil with growing plants or plant residues has been shown around the world to
provide important erosion protection, especially on sloping land, but the amount of mulch
needs to be sufficient. Covering the soil with 4-6 t of mulch residues per hectare can protect
the soil from erosion as much as would a secondary forest about 30 m in height (Glover
2005). Slow decomposition of residues on the soil surface increases the organic carbon and
total N contents in the top 5-15 cm soil stratum while protecting the soil from erosion.
Retention of residues on the surface has been found to increase the soil nitrate (NO3)
concentration by 46%, the N uptake by 29%, and the yield by 37%, as compared to burning
(Mandal et al. 2004).

Nitrogen and phosphorus are the main limiting nutrients on smallholder farms in Africa.
About half of the N, P and K loss in Africa is due to runoff, erosion and leaching. Organic
cycling involves materials grown in situ and producing litterfall, dead roots, green manure and
crop residues. They do not add much to the total amount of nutrients in the system, except for
the possible additional N-fixation and nutrient pumping through the roots. Real organic inputs
are created with biomass transfers and with compost and manure which are produced outside
the field (Sanchez et al. 1997). In the present trial only in situ organic materials were applied;
all the weeds and trashed cardamom shoots were left as mulch around the plants, and the

103
natural green manure originated from the plants in the trial. A greater amount of mulch and a
higher crop diversity could be reasons why grevillea grew better with cardamom and pepper
than with pepper only. Pepper also grew better with gliricidia and cardamom than with
gliricidia alone.

Farmers can improve their sustainable yields by using more organic mulch in the fields,
which improves the moisture retention, promotes root symbioses, suppresses weeds,
decreases erosion, and prevents pest and disease infestation. If there is a mulch layer on the
soil, it is commonly found that the majority of the roots are growing in the mulch layer. This
facilitates the internal nutrient cycling and limits the leaching losses. Adding organic mulch to
the soil also increases the cation exchange capacity of the soil and increases the availability of
phosphorus, which is normally a limiting factor in acidic soils. Mulching keeps the soil in
place and prevents nutrient and organic matter losses. In the Philippines the erosion rate was
found to decrease by 95% with mulch, while sole vegetative barriers reduced it by 65%.
Mulch suppressed the weeds in India by 73% in comparison to an unmulched control; in the
same study pests and diseases (including fungal plant pathogens) declined since the microbial
respiration was five times faster in the mulch than in the soil litter layer. Mulch also promotes
root symbioses, and gliricidia has been found to increase the nodulation of beans (Rosemeyer
2001). With permanent soil cover, zero tillage and a rotation of crops with green manure, the
yields in Brazil improved by a factor of 5.5 in just two years (Hellin 2006).

Agroforestry systems can limit the nutrient losses to minimal amounts and even increase
some nutrients in the soil. Gliricidia improves the organic carbon and nitrogen contents of the
soil. For green manure purposes, gliricidia (leaves and branches) is excellent because its
foliage contains about 4% of nitrogen (Gunathilake and Wasanthe 2004). Loppings from
gliricidia also had the highest rate of decomposition among six hedgerow species studied in a
tea plantation in Sri Lanka (De Costa et al. 2005). Results from Indias Himalayan Region on
the N-fixing Alnus nepalensis D. Don. (Betulaceae) with cardamom indicated that the
presence of a N2-fixing species helped in maintaining the soil organic carbon levels. The total
soil N and organic C levels and the soil moisture content were high in a near-forest situation,
but the effect decreased with intensification of agriculture (Sharma et al. 1997).

There are thus great potentials to improve the soil properties with gliricidia branches and
leaves: 1 t dry weight of leaves is equivalent to 27 kg N (Patil 1989). Field observations have
indicated that G. sepium produces more foliage biomass than any of the other ten species
studied in Malawi, i.e. Leucaena leucocephala, L. diversifolia, Albizzia caribaea, A.

104
guachepele, Enterolobium cyclocarpum, Crescenta alata, Senna atomaria, Guazuma
ulmifolia, Acacia farnesiana and A. deami (Sitaubi 1990). High-quality organic materials,
such as green manure from gliricidia, are readily mineralised with rapid decomposition rates
(Sanchez et al. 1997). In Nigeria, gliricidia could produce up to 15 t/ha/year of leaf biomass
on good soils, providing 40 kg N/ha/year (Kang and Mulongoy 1987). MacDicken (1994)
reported the nitrogen fixation rate in gliricidia as about13 kg/ha/a. Nitrogen fixation depends
on the age of the plantation. The nitrogen fixation in Alnus nepalensis in the Himalayas (with
Frankia as the microbial agent) increased over a period of 15 years and then started to decline
(Pandey 2007). Using rhizobium inoculants, the nitrogen fixation in gliricidia could probably
be made more efficient (Dommergues 1995). Farmers are well aware that legumes, normally
beans and leguminous trees, and some other plants such as Alnus spp., improve the crop
growth, even though they do not necessarily know the mechanism of nitrogen fixation.

It is clear that farmers are not growing trees only to improve the soil fertility. A survey in the
West Usambaras showed that farmers were mainly interested in planting fast-growing timber
species such as eucalypts and Grevillea robusta. Only 18% grew trees to increase the soil
fertility. Most of the trees were planted along field boundaries, because from the farmers
point of view, this causes less competition with crops (Kerkhof 1990).

Agroforestry systems can decrease the nutrient losses to minimal amounts and even increase
some nutrients in the soil with time, but this might not eliminate the need for P fertilizers.
Woody species are able to transform the inorganic phosphorus into more readily available
organic forms and to supply potassium via litterfall. Phosphorus deficiencies can also limit the
N fixation and the growth of N-fixing trees (Sanchez et al. 1997). That was also seen in the
present manure application test. When manure was applied to cardamom, the production was
doubled, as compared to the treatments without manure application. The production increased
most, being 2.4 times higher, when manure was used in an intercropping system with the
nitrogen-fixing gliricidia.

There are some non-leguminous shrubs that also have a high nutrient content, such as
Tithonia diversifolia (Mexican sunflower). Its leaves have higher P and K levels than most
legumes used in agroforestry (Sanchez et al. 1997). Fresh leaves of T. diversifolia contain
3.5% N, 0.3% P and 3.8% K. They also decompose rapidly. In Kenya, the application of
green manure from this shrub has resulted in higher crop yields than those obtained with
inorganic fertilizers, and the residual effects were longer (Jama et al. 2006). This plant (alizeti

105
mwitu in Kiswahili) is widely available in the Tanzanian highlands, and it has been found to
be efficient in improving the soil nutrients status, especially when mixed with Minjingu rock
phosphate (Ikerra et al. 2006). In Argentina, the P content in the surface soil was found to
increase when perennial crops were grown with elephant grass (also called Napier grass,
Pennisetum purpureum Schumach) mulch (Rosemeyer 2001). In The East Usambaras, other
shrubs listed by farmers as improving the soil fertility were Vernonia subligera O. Hoffn.
(tughutu in Kiswahili), Vernonia amyridiantha (mhasha), Albizia schiniperiana (mshai),
Ficus vallis-choudae Del. (mkuyu), Kalanchoe crinata Andrew (sopolwa), Justicia glabra
Koenig (tundashozi) and Bothriocline tementosa S. Moore (boho; Wickama and Mowo 2001).

Alley cropping based on multiple-use leguminous trees has been promoted for over two
decades in tropical countries. Grass strips have been used for erosion control in different parts
of the world, e.g. Guatemala grass (Tripsacum laxum Scrib & Merr.) in the Usambara
Mountains. However a high biomass production of grasses caused competition with the
adjacent food crops, particularly when the grasses are used for fodder and their nutrients are
removed from the system. Grasses also shade the adjacent crops if not slashed. A better option
could be narrow contour strips of the field that are left unploughed and allowed to develop
naturally. They provide an excellent erosion control, with negligible competition with annual
crops and with low establishment and maintenance costs. In time they capture sediment and
develop into terraces. The system of natural vegetative strips is based on Asian farmers
knowledge and practice (Garrity 1993).

In the present trial the soils remained very acidic. For better production in the East
Usambaras, there is a need to reduce the soil acidity and increase the soil phosphorus levels. If
sufficient improvement is not possible with farmyard manure or grass mulch, it is necessary
to consider applying some rock phosphate, which is also available in organic form. Minjingu
rock phosphate is of guano sedimentary origin, and is considered as an organic fertilizer,
having the following nutrient content: P2O5 28 to 30%, CaO 41.7%, MgO 3.2%, Al2O3 1.2%,
K2O 0.78%, Fe2O3 0.89%, and other micronutrients, like Zn and Mn 11. Its cost in the nearest
town, Muheza, in June 2007 was US$ 7.8 for a 50 kg bag. It is natural, produced in the
country, and cheap, as compared to the usual chemical fertilizers, which are at least four times
more expensive. The fertilizer price would be even cheaper if the farmers could organize
themselves and buy directly larger amounts from the processing plant. In any case, extension

11
Pers. com. with Mr. Achary, Minjingu Mines and Fertiliser Ltd, 3 July 2007.

106
is needed to sensitise the farmers to use rock phosphate with the already available organic
fertilizers in their fields.

In Kenya, rock phosphate has contributed to maize yields higher than those obtained with
applying urea. In acidic soils, like those occurring in the East Usambaras, there is a need to
add organic material that may help in making P available. Mixing rock phosphate with
compost also increased the soil P availability in experiments in Burkina Faso and Tanzania
(Sanchez et al. 1997). Jama et al. (1997) recommended application of large amounts of rock
phosphate, because the larger the initial application is, the longer will the residual effect be in
the soil. In western Kenya the residual effect of P lasted for 5-10 years, but the cumulative
maize yields after five years were almost the same, regardless of fertilizer application as a
large one-time portion (250 kg P/ha) or annually (50 kg P/ha/a; Jama et al. 2006).

The Chagga homegardens on the slopes of Kilimanjaro have always been an example of one
of the most successful African agroforestry systems. The system has supported a population
density of 500 persons per km2 with an average size of homegardens of 0.68 ha, and included
cattle, with very little land degradation. The banana-based homegardens had at least 15 types
of banana grown for food, brewing and fodder. In addition, cardamom, Grevillea robusta and
coffee (Coffea arabica) were grown as cash crops. This famous system has been sustainable,
because of the following factors (Fernandes et al. 1989):

- The soils are fertile, of volcanic origin;

- The biological and structural diversity minimize the biophysical (pests and drought)
and economic (volatile markets) risks;
- There is dense vegetation cover on the soil surface;
- The system yields products of high economic value (coffee) with a low total biomass,
which minimizes nutrient exports in harvests;
- N-fixing crops and tree species are used;
- Crop residues and animal manure are returned to the cropped land, and
- There are well functioning irrigation canals.

Although the system has been sustainable at least for a century, it is now under pressure due
to a rapid population growth, diminishing land resources, and a change in dietary habits
(maize replacing bananas as the staple food and as cultivation base). Young people migrate to
urban areas, which leads to labour shortage on the farms (Fernandes et al. 1989). The whole

107
system was built up with the high-value coffee, and falling coffee prices at the beginning of
the 1990s have led to a breakdown of this traditional farming system. There have been also
problems with declining yields and losses of biodiversity (Fernandes 2000). Soini (2005) has
suggested increasing the use of improved fallows, and borderline or contour plants, so as to
produce more fodder and to control the erosion. In addition she has suggested the following
measures: adoption of more diverse cash crops which would encourage cash crop marketing;
changing coffee cultivation for tea estates; adopting goat keeping instead of cattle herding;
and increasing collective action to improve the access to credits, marketing and processing of
products.

In contrast to the situation in Kilimanjaro, the soils in the East Usambaras are not volcanic but
of low fertility. However, most of the suggested actions can also be recommended for the
latter area. A well functioning milk association is providing a stable income for the highland
farmers and would diminish their interest to change dairy cows to goat keeping. Tea is already
more commonly cultivated than coffee, and it has proved to be a more sustainable crop,
because it protects the soil from erosion by effectively covering the soil surface. Tea,
however, also demand clear-cutting of the forest for establishing the estate, and the curing tea
requires considerably amounts of fuelwood, which the tea companies tree plantations cannot
supply.

Cardamom needs shade and a suitable microclimate (a surrounding forest) to thrive, and thus
the farmers do not cut more trees than is necessary for profitable cultivation. Because of the
changing cardamom prices and markets, farmers have started to convert uneconomical
cardamom cultivations in the forest to tea, especially in Kwezitu village. Tea companies are
willing to buy the cardamom shade trees and provide tea seedlings for farmers. Chopping
down of the cardamom shade trees destroys increasingly larger forest areas and abolishes the
opportunity to rehabilitate these extensive forest cultivation areas. The most committed spice
farmers have, however, diversified their cash crop cultivation for diminishing the market
risks.

108
5.6. Organic products and marketing

Positive development of non-timber-forest-products (NTFPs) gives the rural farmers

opportunities for increasing the income by diversifying of enterprises. A majority (about
80%) of the people in the developing world uses NTFPs as part of their nutrition and for
medicinal purposes. The estimated total value of NTFPs in the world trade is approximately
US$ 1 100 million (1999), and it has grown by nearly 20% annually over the last several
years (Wilkinson and Elevitch 2004). To generate income for poor farmers, high-value crops
should be grown at least on small-scale, when low-value crops are grown on a large-scale
(Sanchez et al. 1997). Monetary gains could be reached with high-value cash crops which
have functioning markets and stable prices. Cash crop cultivation is usually better organized
than subsistence farming, because cash cropping offers more incentives for the maintenance
of land productivity, including the use of fertilizers (Kaoneka and Solberg 1994).

The value of medicinal and aromatic plants, including spices, has increased in recent years, as
they are increasingly used in the production of nutrition supplements, cosmetics and
perfumes. However, poorly controlled extraction of medicinal and aromatic plants from
tropical natural forests is threatening the existence of many valuable plant species. As an
alternative, it has been recommended that such species be cultivated outside their natural
habitats. Only few such species are actually cultivated, because their management and
propagation practices are not known (Rao et al. 2004).

There are thousands of small farmers already with knowledge and experience about spice
cultivation in the East Usambaras. Thus the possibilities to increase the spice cultivation in
this area in a short time are good. Tanzania earns a substantial part of its foreign exchange
from spice exports. The Business Times (2003) in Dar es Salaam stated that the spice industry
presents a major opportunity for Tanzania to reap economic benefits in a relatively short term
with only a nominal input in investments. The data available on the spice industry sector
indicate that it has been growing in Tanzania by more than 10% per annum in value terms
since 1997. The export value was 11 million US dollars in 2001 (Business Times 2003).
Encouraging sustainable cultivation of spices would be very relevant for the improvement of
the economic situation of the rural people in the country. Export markets for most spices are
already available. Transport problems are also forcing the farmers to specialize in spices in
remote mountain areas, since these products are valuable per unit weight and easy to transport

109
to the village from the field. The East Usambara Mountains are situated only one and half
hours drive from Tanga harbour, which is facilitating the export of the products.

The entire world cardamom production is consumed within a year from harvest. In the last 15
years, the worlds cardamom consumption and production has increased almost 2.5 times, and
it appears that cardamom has a bright future and shows a steady increase in both demand and
supply. Prices for the product are also expected to remain stable or even increase (Ravindran
and Madhusoodanan 2002). As there is no replacement for black pepper products and the
world demand for this commodity is increasing at the rate of 2.7% per year (Ravindran 2000),
future markets should also be secured for pepper.

Yet the present study confirmed that the Usambaran farmers suffered from low prices and
price fluctuations. Prices could suddenly drop to half of the level of the previous month. The
average cardamom prices in the East Usambaras have dropped by about 60% in 5 years in US
dollar terms. In 2001, the average price for dried cardamom sold in the East Usambaran
villages was US$ 2.9 (US$ 1=TZS 876.4, November 2001) per kg; in 2006 it was as low as
US$ 1.8/kg (US$ 1 = TZS 1 253.8). The prices were higher when the producer was able to
travel to the town to sell the product there. The downward trend of prices was most likely due
to a huge production increases in Guatemala. Cardamom farmers still relied on cardamom as a
cash crop in spite of the low market prices, and only two farmers suggested it would be better
to change to another cash crop.

The Usambaran farmers can compete neither with the volume nor with the price in the global
spice markets, but they could compete by supplying organic, good quality and sustainably
produced spices and thus secure a satisfactory income for their living. Globally, most of the
spices today are cultivated with chemical fertilizers and pesticides. In the Usambaras, where
smallholder farmers do not use mineral fertilizers or agricultural chemicals, the promotion of
organic spice cultivation could potentially raise the value of the product in the world markets
by as much as 30-40%.

Intensive cash-crop production by smallholder farmers is generally developed through

contract farming (Hazell et al. 2007). Organic spice buyers make contracts with farmers and
provide extension and a commitment to buy a farmers whole organic production. Farms that
are not yet certified as organic must go through a process lasting 1-3 years, including
extension and control visits to obtain the certificate for organic production. Since the

110
Usambaran farmers do not use chemicals, the aim should be to improve the sustainability,
production and quality through better cultivation methods, organic manure application and
better post-harvest processing methods. Better quality and processing would raise the price,
and manure application would give better yields. Organic product buyers would estimate the
amount and quality of the produce, and farmers could pledge themselves to sell the whole
production for these buyers.

The buyers of organic spices, however, prefer contacts with organized farmers groups, not
with individual farmers, in matters related to quantities produced or quality standards, and for
assuring the commitment of farmers. These have been the most problematic issues hindering
the business development with the Usambaran farmers. The commitment would be stronger if
a farmers group/association controls the quality and production activities. Farmers
organizations could also help in developing some kind of a business strategy, as far as
designing a proper product label or promoting the products in organic trade shows is
concerned. The existing farmers association has to date only had a priority in milk
cooperation, but its involvement is urgently needed in other sectors as well.

The certification of organic spices has already started in some of the villages of the present
study. The demand is higher than the supply, and the markets in the USA and Europe are
growing. Spices are normally exported as dry, non-processed products. Further processing of
spice products should also be considered, for instance, for the production of cardamom-based
essential oils, which would increase the revenues considerably. The profit in the Usambaras
could then be up to six times higher as compared to selling dry cardamom, since the essential
oil from cardamom seeds gives a market value of more than US$ 15 per 5 ml. Processing of
white pepper from black pepper would increase the price per kg threefold, and harvesting
cinnamon using an improved harvesting method would provide the farmers a four-fold price
per kg. If these products could then be sold as organic ones, the profit would be multiple.
Organic black and white pepper fetches premium prices at international markets (Sivaraman
et al. 1999).

Some organic spices have already been exported from the East Usambaras to Europe (by the
Tanzania Zanzibar Organic Spices Producers), which indicates that functioning market links
already exist. The buyers claim that the demand is high. Tanzania is already producing
organic cotton as the first country in Sub-Saharan Africa and exporting the highest quality in
the world of this commodity (Financial Times 2005; The Guardian 2006). Organic coffee and

111
cloves are also exported by Zanzibar, but the potentials in the entire country would allow for
much larger business. Tanzania has only 0.01% (5 155 ha) under certified organic crops of
the total agricultural area in the country. This is the lowest percentage among 14 developing
countries studied in Asia, Africa and Latin America. If we compare this share, for instance,
with Uganda having 1.39% of agricultural land certified for organic production, we can
understand that the potential in Tanzania is considerable (UNCTAD 2004).

To diminish the risks of organic spice production, diversification is recommendable, but the
product quality should not be compromised. It is found that high-value products managed by
specialized producers generally provide higher revenues than those obtained by less
specialized producers (Ruiz-Perez et al. 2004). Specialized products with good quality and
stable markets are then more justified than a diversity of many low-quality products. In the
current situation, for individual rural farmers, it would be wisest to find a flexible balance
between diversification and specialization. This gives a powerful tool for a farmer for
confronting an unknown future and for rapid change of the farming regimes towards a needed
direction.

Market-oriented domestication of forest products requires that agroforestry producers work

closely with the marketing companies. There are already good examples of large companies
focusing on smallholder farmers who produce in their multi-strata agroforestry systems
various raw materials, e.g. wooden decorations for Daimler-Benz (for C-class Mercedes-
Benz cars), for Masterfoods (cocoa), and for Distell Corporation (marula, Sclerocarya birrea
(A. Rich.) Hochst. for Amarula liqueur; Leakey et al. 2006). These agroforestry tree
products (AFTPs), extracted from agroforestry systems outside forests, were for the first
time distinguished by Simons and Leakey (2004) from those timber and non-timber forest
products (NTFPs) that are extracted from natural forests.

Without cooperatives, farmers have problems in identifying the best crops for growing and in
accessing extension and marketing services (Uliwa and Fisher 2004). NTFP producers
commonly lack price awareness (FAO 1995). A strong farmers association could play a key
role, for instance, in buying and selling bulk products, in standardizing the production, and in
improving the marketing infrastructure, storage and drying techniques, and the rural transport
facilities. In rural Tanzania, producers organizations link members to new markets, and give
access to financial services, new technologies and collective crop marketing. The support
provided by an organisation and better price awareness increase the farmers bargaining

112
power. There are producers organizations in Tanzania, about 50 in each district, but they are
not functioning efficiently, since they lack good management and business skills (Uliwa and
Fisher 2004). In the village of Antakae, a kind of cooperative among cardamom farmers was
already initiated, and the farmers involved sold the produce with a fixed price that was 12.5%
higher than the average price in the area. Farmers said they could easily intensify their
cultivation if they had better markets. The available market for processed milk was also the
main factor contributing to the success of a long-term Dutch-supported livestock programme,
which focused on the highland villages in the East Usambaras. The project also organized
training and intensive extension for those farmers who were interested in dairy farming. Only
when enough milk was produced, the milk processing and the marketing to Tanga and Dar es
Salaam were initiated (Swai et al. 1992; Figure 23).

Technically demanding crops, like cardamom, need strong extension services. Now the
cardamom crops are often harvested too early, which results in empty capsules and low-
quality products. For motivation to improve the land husbandry and quality, farmers should
receive a grater share of the market prices. That could also be possible through farmers
associations improving the marketing systems and providing credits to invest, e.g. in storage
capacity, which is a considerable problem in the prevailing humid conditions. The
effectiveness of a cooperative depends, however, on governance and management. Strong
farmers organizations also facilitate advocacy and service delivery, since with them there are
more possibilities to provide services for an affordable cost to a large number of farmers
(Hazell et al. 2007). As of now, the middlemen buyers come to the village to buy the product.
Farmers indicated they did not have other alternative than to sell their produce at the price the
buyers offered. No price differentials were offered for the farmers for good-quality products.
High-quality products were commonly mixed with lower-quality products, whereby the total
quality remained low and good-quality producers lost their price premiums. If this is the case
also in the future, there will be no incentives for farmers to raise the quality of their products.

113
Figure 23. A good example of a functioning farmers association is the East Usambaran
farmers milk association, UWAMA, in Amani. Ushirika wa wafugaji ngombe wamaziwa=
Milk producers association, Maziwa ni chakula bora = Milk is great food.

5.7. Better extension

The Usambaran farmers called for better extension and support to develop their agribusiness,
and especially for organizing themselves. The role of the local government is crucial for
providing better extension services to the rural areas in Tanzania. Government extension staff
members are normally well-trained, but they many times lack motivation, since they are
underpaid. Extension is unfortunately the most resource-poor area in the government service.
However, the extension officers need regular training on new technologies and findings,
including the importance of genetic diversity (Atta-Krah et al. 2004), and on the
domestication of wild forest species known by the local population, for use in agroforestry
systems (Kessy 1998). In addition, forestry and agricultural extension are still working
separately in Tanzania, which is a great problem in spice cultivation in particular (some spices
are agricultural crops while others are trees).

Using many different dissemination methods at the same time has been proven to lead to
better adaptation of a new technology. Individual farm visits are preferred, but they are the
most expensive and ineffective methods for scattered farms. There is also a risk that some
farms are visited more due to personal relationships (Place et al. 2005). Regular contacts with

114
extension agents were also found to be very important. Results from Sudan have concluded
that the adoption rate of new technologies is high if farmers are regularly supported by
extension (Glover 2005). Regular extension activities on crop residue application have
considerably reduced the nutrient losses on farms in Uganda (Nkonya et al. 2004).

Extension methods for large groups are cost-effective but can cause unintentional
discrimination against farmers of lower social status and against women. The best way is to
use a variety of methods in disseminating information. It is often advisable to form separate
womens groups, because they have generally worked well (Place et al. 2005). Female
extension agents should work with womens groups (World Bank 1995), because women can
be reached and mobilized easily only by a fellow woman (Binagi 1992). According to a study
in Kenya, only 12% of the farmers had adopted at least one component of soil fertility
improving technologies. The low adoption rate implies that a farmer-to-farmer information
diffusion strategy is not enough. It should be complemented by other dissemination methods
(Mose et al. 2000).

Despite a high level of literacy in the East Usambaras, 93% of the information there is
delivered orally. Thus the most effective methods are group methods (meetings,
demonstrations, field trips, and demonstration plots), mass methods (audio-visual methods),
and campaigns (Kajembe and Mwasemba 1994). Binagi (1992) has recommended organizing
of public activities, such as the Forest Conservation Day, with programmes ranging from
speeches to cultural shows touching on the conservation subject. During such campaigns the
local ecotourism guides could also instruct the villagers on environmental awareness. Football
games between district authorities and villagers could also improve the relationships between
these two groups. After a game also the important environmental issues could be discussed in
a more relaxed atmosphere.

The author has experience from rural development projects (RIPS) in southern Tanzania
where the schools have organised singing and dance (ngoma) competitions related to the
management of cashewnut trees. Successful information methods were also the village radio
programmes and video shows on natural resources. The technical equipment in the village
radio consisted of a simple tape recorder. The village radio was also used as an important
information channel. The video showed an area in Tanzania first with abundant nature
resources and afterwards when it was highly eroded. Past mistakes were thoroughly
discussed. People in the village bordering the Angai Forest (later gazetted as a village forest

115
reserve managed by 13 villages) admitted realizing the importance of good forest
management. 12 Similarly to this experience, videos could also be made on soil improvement
practices for agroforestry systems.

Practical demonstrations of new technology and farmer-to-farmer extension are convincing

ways of information dissemination, since the willingness to adopt a particular innovation
decreases with increasing distance from the information source. The new technology should
be profitable for the farmers and cover the establishment and maintenance costs (Pannell
1999). The new method should also be sustainable and socially acceptable (Lundgren and
Raintree 1983).

Extension related to community development is still needed in the present study area to
support the livelihoods and conservation aims. For instance, the rural communities need
information on alternative house construction materials and on improved, fuel-saving stoves
which have been introduced by the EUCFP and EUCAMP activities (Kessy 1998); these
stoves can save 50-70% of the firewood previously used (White and Mustalahti 2005). The
information disseminated by EUCFP/EUCAMP on forest fire control has most probably been
the reason for a low incidence of fire in Muheza District where the East Usambaras area is
situated (Wurster and Burgess 2005).

Communication between individuals is generally more effective when they have similar
customs and beliefs. In a study on 540 villages in Tanzania it was found that a similar ethnic
background and shared social activities contributed to faster adoption of a decision on
fertilizer use (Isham 2002). Members of associations, co-operatives, or farmers groups are
normally the first adopters of new technology skills. Apart from new technology, producer
organizations especially need extension in agribusiness, including the financial systems, and
in lobbying, networking and marketing (Uliwa and Fisher 2004).

The lack of social capital 13 makes the market development difficult in Africa (The World
Bank 2005). If capital, quality control and markets are needed for small-scale farm-derived

12
Based on the authors own experiences in 1997-1998 in the Rural Integrated Project Support (RIPS)
Programme implemented through cooperation of the Tanzanian and Finnish Governments in Mtwara and Lindi
Regions in Tanzania.
13
Social capital is generally referred to as the set of trust, institutions, social norms, social networks, and
organizations that shape the interactions of actors within a society and are an asset for the individual and
collective production of well-being. At the macro level, social capital can affect the economic performance and
the processes of economic growth and development (Social Capital Gateway 2006).

116
products, it is clear that this is not possible without proper organization. Farmers have
difficulties in organizing themselves within their villages due to negative experiences from
the past with cooperative leaders. Extension work for building local capacity for networking,
communication, and marketing could have a strong impact on the farmers ability to organize
themselves and to reach better markets.

5.8. Poverty reduction and incentives for the local population

About 60% of the rural households in Sub-Saharan Africa are living below the poverty line
(Place and Prudencio 2006). It has been calculated that a 1% farm yield increase is associated
with a 1% drop in the number of people living on less than US$ 1 a day. Nearly three fourths
of them live in the rural areas of developing countries (Garrity 2004). Agriculture has always
had an important role in boosting economic development in Tanzania, as illustrated by the
fact that rural growth has had more impacts on poverty reduction than the urban growth
(Koning 2002). Agriculture forms the basis of Tanzanias overall development, since
economically it is contributing about 80% to the nations GDP. Improved social development
can be attained through improved farm incomes and better food security (URT 2002a).

Boosting of the export-oriented agriculture is claimed to be the most effective way for
increasing the income levels in developing countries. Trade in agricultural products has also
fuelled the economic growth of such industrial countries as South Korea and Taiwan (Koning
2002). The links between agricultural development and the overall economic development are
strong; with the estimated expenditure multipliers equalling 2 or 3. This means that US$ 1 of
new household income from sales of export crops in a remote area can lead to a further US$
2-3 worth of additional local employment in the production of non-tradable goods and
services (World Bank Country Study 2000).

It is widely recognised that forests can contribute to poverty alleviation and rural
development. It is estimated that about 60 million indigenous people depend on natural
forests, and approximately 1.6 billion people rely on them for at least part of their basic
livelihoods. Forest products are especially crucial during periods of hardship. They can
provide up to 60% of the farmers total cash income in rural Tanzania (Mery et al. 2005). The
Millenium Ecosystem Assessment (2005) points out that protected areas may increase the
poverty in situations where the rural people are excluded from resources that have

117
traditionally supported their well-being. The protected areas can, on the other hand, contribute
to improved livelihoods when they are managed to benefit the local people. Forest-based
poverty alleviation must be part of an overall rural development strategy. It cannot be carried
out in isolation (FAO 2003).

The social capital has proved to affect positively on the household income. It has facilitated,
for instance, pressure for improving government services, dissemination of information, the
provision of credit, and the management of common property (World Bank Country Study
2000). Sustainable exploitation of at least some forest products in the East Usambaras would
be possible and could benefit the local people. There are, for example, already 40 000
cultivated varieties of African violet (Saintpaulia spp.), which form the basis of a US$ 100
million global houseplant trade; these varieties are all derived from a few species found in the
East Usambara Mountains (Biodiversity Hotspots 2007). The Finnish Saintpaulia Society has
introduced the African Violet as a flagship plant for delivering information on the East
Usambaran environment and for financing the conservation of the endangered remaining sites
where the different species and varieties of this plant grow. A further objective is to integrate
Saintpaulia with local ecotourism initiatives for the benefit of the local communities. Also t-
shirts, postcards, and African Violet plants are sold to finance conservation activities in the
East Usambara Mountains (Finnish Saintpaulia Society 1997).

The local communities would be motivated to conserve the forests by obtaining some benefits
of this activity. The lowland actors (at least in the private industrial sector) could create
financial incentives through Payments for Environmental Services (PES) schemes, for the
benefits that the mountain watersheds and the people living there provide (cf. Garrity 2004).
There are four types of environmental services: carbon sequestration, biodiversity, aesthetic
values and watershed protection (Mery et al. 2005). In Costa Rica, farmers have received
subsidies for managing tree plantations and agroforestry systems. Funding has come, for
instance, from a special tax on gasoline, public water and hydroelectric companies, and from
international carbon trading derived from public protected areas (Montagnini et al. 2003).
Many urban water supply systems in Tanzania depend on the water originating from the
Eastern Arc forests, and the East Usambara Mountains alone secure the clean water for half a
million people. A part of this revenue could be used to develop the economies of the
respective communities.

118
There have been diverse income-generating activities in the present study area directly offered
to interested individual villagers. The Tanga Forest Catchment Project, together with UNDP,
ICRAF and Unilever UK, has been encouraging the farmers to grow Allanblackia stuhlmannii
trees in the villages. The large seeds of this species provide oil for cooking and soap-making.
At the time of the present study, 650 farmers were involved in this activity. Butterfly farming
has been introduced by the Tanzania Conservation Group together with Conservation
International (through the Critical Ecosystems Partnership Fund), World Vision, and the
McKnight Foundation (URT 2006). The Amani Butterfly Project captures a few female
butterflies from the wild and places them into net cages with proper host plants. Butterfly
pupae are then sold and delivered to the United States and Europe for butterfly gardens.
Butterfly farming has meant an average annual income gain of 15% to 25% for the 250
households participating in the project. A total of 65% of the earnings goes directly to the
butterfly farmers, 7% to a farmer-controlled community development fund, and 28% for the
project running costs (Amani Butterfly Project 2007). The Dutch-supported zero-grazing
dairy project was providing about 20% of the income of its highland farmers. Other
introduced programmes in the area include beekeeping, fish ponds and improved-building
projects that use less timber in construction (URT 2006).

In the Amani Nature Reserve, the measures to generate revenues for conservation include
allowing some universities to organize tropical forest courses in the area, the entrance and
research fees, and the possible profit from two guesthouses. Eighteen communities
surrounding the Amani Nature Reserve receive a 20% share of the entrance and research fees.
In 2002, this provided about US$ 100 per village (Vihemki 2005). Community members
have a right to collect fuelwood and medicinal plants from the local use zone (6% of the area)
twice a week for eight years. The cutting of poles is also permitted, but subject to approval by
the ANR Conservator. In return, villagers are supposed to protect the nature reserve
(EUCAMP 2001). At the end of the eight-year period, the villages are assumed to be self-
sufficient in fuelwood production (EUCAMP 2002).

This goal has not yet been achieved. The Amani Nature Reserve Management Plan is now
under revision. The communities rights most probably will continue as they are 14. Logging
has been allowed by permit only on public land, but the license fees for reserved species are
higher than what farmers can afford. Some licenses should be obtained from local offices
accessible to the rural farmers. Logging licenses for the public land in the East Usambaras in

14
Pers com. with Mr. Corodius Thomas Sawe, Conservator of ANR, 15 November 2006).

119
2005 varied from TZS 7 000-27 000, which equalled US$ 6-23.5, per m3 (1US$ = 1150 TZS,
24.10.05) 15.

There is a growing recognition that tourism can be one alternative for adding incomes in the
rural mountain areas, if some of this revenue could be directed to the rural communities. For
example, it is claimed that one mountain tourist in Nepal provides employment for five
people, mostly for the poorest ones and for women, in remote and isolated mountain regions
(Mishra 2002). In the whole of Tanzania, wildlife-based tourism provided US$ 360 million in
2002 (Kaimowitz 2003). There are now some efforts to redirect some tourists from the large
game reserves also to visit less-known places, such as the East Usambara Mountains.
Ecotourism in the East Usambaras is a very small-scale business, despite the fact that the
facilities are rather good. However, it has already benefited nine trained ecotourism guides
and their families. The guides are divided into three groups, which each work ten days per
month, since the number of tourists in the East Usambara is still low. One guide can earn
about TZS 70 000 (US$ 62.33; 1US$ = TZS 1123, 26.11.2007) per month. Each non-citizen
tourist pays US$ 25 for a guiding fee, of which 60% is for the guide, 20% for the village
development fund and 20% for a conservation fund 16. There is potential for much more, and
some innovations (e.g. organic farming tourism or agritourism) could bring more income for a
larger number of families. Ecotourism activities in the area could be linked to organic farming
products, especially spices. The tourists could be brought there to buy some organic spices
and other products directly from the farm. The product could be packed and labelled nicely,
which would increase the price per kg quickly and easily.

In Uganda, investing in education beyond the secondary level has been found to contribute
significantly to a higher household income from off-farm activities (Nkoya 2004). In Sudan it
was also found that the adoption of new technologies increased with the education level of the
farm household heads (Glover 2005). In the present study area, 84% of the farmers had passed
the primary education, 2.6% the secondary education level, and 14% did not have any formal
education. In the present case, such a low level of formal education did not seem to affect to
the household income.

The government of Tanzania abolished the school fees in 2001 as one of the first countries in
Africa. Eliminating fees led clearly to a higher enrolment, especially in the case of girls

15
Pers com. with Mr. Corodius Thomas Sawe, Conservator of ANR, 30 October, 2005).
16
Pers com. with Mr. Abduel Kajiru, ecotourism guide in ANR, 15 November, 2007).

120
(UNICEF 2006). In 2005, the school enrolment was already 95%, and 49% of the students
were girls (CIDA 2006). Investing in womens education beyond the primary school has been
shown to be a way for long-term change to improve the family's health and nutrition status,
and for economic empowerment. Research conducted by the World Bank (Abu Ghaida and
Klasen 2004) and the International Food Policy Research Institute (Quisumbing 2003; Smith
et al. 2003) indicated that increasing of girls secondary schooling by one percentage point
will raise the nations income by 0.3%. In addition, keeping the girls in school reduces child
mortality and malnutrition, improves family health, delays the age of first marriage, lowers
the fertility rates, enhances womens domestic role and their political participation in society,
improves their functioning in the wage labour force, and strengthens a familys survival
strategies. A new report on the girls position shows that, globally, the girls are still a heavily
discriminated and marginalised group (Plan 2007).

For improving the general livelihood of the rural communities, planners and funding agencies
need to take into account intra-household differences in resource use and control. Study
results from Sub-Saharan Africa have showed that women have a tendency to use their
normally scarce incomes to benefit the whole household, but men do not distribute their
higher incomes evenly among all members of the family. Thus women's spending patterns are
better indicators of the family welfare (Manuh 1998).

In the rural extension in Sub-Saharan Africa, most (95%) of the resources and technical
assistance (improved seeds and tools) have been channelled to men, although women are
responsible for 80 per cent of the work, and their labour inputs particularly in food production
normally exceed those of men by 12-13 hours a week,. Women also work for men's export
crop cultivations, even if they do not receive much direct benefit from them (Manuh 1998).
Studies in Sub-Saharan Africa further indicate that the agricultural productivity would
increase by up to 10-20%, if the gap in capital and inputs between men and women could be
reduced (Quisumbing 2003).

Women in Tanzania have traditionally, according to customary law, not been allowed to
control the use of such resources as land, cash crops and animals. Women have controlled
only the use of crops with low monetary value. Social exclusion of women has kept them in
poverty traps. Women have hesitated to plant trees in fear that they will not be able to harvest
anything in case of divorce or death of their husband. Women consider the issue of land
ownership as being beyond their power to change (Shelutete 1996).

121
In Tanzania, new, more equal land laws came into operation on 1 May 2001. The Land Act
deals with public lands which are outside villages or reserved areas; and the Village Land Act
deals with village land (Alden Wily 2003). The Land Act states: The right of every woman
to acquire, hold, use and deal with land shall to the same extent and subject to the same
restriction be treated as the right of any man. Women also in rural villages can now own
land, regardless of customary and religious restrictions (URT 1999a). The Village Land Act
(URT 1999b) states that a customary right of occupancy is in very respect of equal status and
effect to a granted right of occupancy, but such formulations can be sometimes very
ambiguous and will most probably raise some disputes (Sundet 2005). During the present
field work, the poorest households visited in the East Usambaras were headed by a woman.
Nowadays the off-farm jobs in towns and abroad, as well as the HIV/AIDS epidemic, are
changing the situation also in the rural villages: there will be more single-parent households.
The livelihoods in rural Tanzania would become more stable if the women had equal land
rights, but even then the problems would not be solved overnight.

5.9. Participatory forest management

Forests and people have evolved together over thousands of years; forests are part of the
human landscape, and humans are as essential part of the biodiversity. People have also
profoundly influenced the biodiversity (McNeely 2004). For protecting and managing high-
biodiversity areas, people have to be part of conservation, but they need help in their
adaptation to changing conditions.

Cultivating private land is more motivating than cultivating communal land. This is also
clearly seen in India, where about 70% of the cardamom farms are privately owned. That
motivates people to invest more on their land and to cultivate more intensively and
sustainably. If private land ownership is not possible, the communal property rights can,
however, be almost as effective as the private property rights (Investing in Mountains 1997).
Participatory forest management approaches include joint forest management, which divides
the forest management responsibility and returns in national forest reserves and local
government forest reserves between the government and the local communities; and
community-based forest management, where registered village land is managed by a village
council (Blomley and Ramadhani 2006). Participatory forest management is a necessity in

122
Tanzania, since the government does not have enough resources to protect and manage the
forests. It is also possible in Tanzania, because of the long history of strong village
governance. Effective social institutions are a prerequisite for successful community-based
forest management (Alden Wily 2001).

The government of Tanzania has provided an enabling legal and policy environment for the
village level, and participatory forest management is actually the countrys central forest
management strategy (Blomley and Ramadhani 2006; Katila 2008). Devolution of forest
management to the local authorities and communities was recommended and joint forest
management provided in the new National Forest Policy (URT 1998). In 2002, the New
Forest Act 7/02 (URT 2002c) was approved, which defined the legal framework for the
promotion of private and community-based forest management and enabled an effective
implementation of the policy (FAO 2004). The Tanzanian National Forest Policy identified
the following problems on public forest lands: unclear and unequal land tenure; lack of
incentives and funds; poor extension services; gender inequality in decision making; poor
communication and infrastructure; and inadequate communication with government
authorities. For solutions it was advocating the allocation of forests and their management
responsibility to villages and private individuals. Communities would be given rights to some
products and services derived from community-managed forests. Forestry activities by local
communities and farmers would be enabled through extension services, technical assistance
and establishment of appropriate financial incentives (URT 1998).

The strongest pressure is threatening the forest areas outside the conservation reserves.
Governments have realised that the local communities cannot be excluded from the forests.
However, in general, governments are not willing to hand over the total ownership to them;
even if they are recognizing the local people as capable managers of national forest assets
(Rodgers et al. 2002; Katila 2008). Also in Tanzania, collaborative forest management, in
which stakeholders jointly participate in the resource management and decision-making, is
assumed to lead to sustainable forest management. This recommendation is based on the
premise that the Washambaa have the capacity and indigenous knowledge to manage the
forests (Kaoneka et al. 2000), which they depend on for survival. They adapt in many ways to
the changing environment. Wild-resource management by the local people needs to be taken
into account both in agricultural planning and in forest management. In shifting agriculture,
the distribution between forest and farmland, and wild land and cultivated land, is often
unclear, as the management of fallows and fields is closely integrated (Woodcock 1995).

123
In many countries conflicts between fulfilling local and national forest-related needs persist,
but it is clear that forest management plans could be successfully achieved only if the
villagers themselves participate in managing the natural resources (Glover 2005). A National
Forest Programme is a good start to better management of the natural resources, but also more
detailed local plans are needed. Tanzanian villages are allowed to allocate village forest lands
and make management plans for their use (URT 1998). The Village Land Act allows
registration of village land, and the Local Government Act gives the village councils
executive rights and a possibility to establish village by-laws (Blomley and Ramadhani 2006).
Tanzania is considered to be the most progressive country of the whole of East Africa in
promoting participation in forest management (FAO 2004; cf. Katila 2008). Participatory
forest management covers about 3 million ha of forests in over 1 200 villages out of the more
than 11 000 that exist in Tanzania (Blomley and Ramadhani 2006). Participatory forest
management has also been initiated in the East Usambaras in the EUCFP/EUCAMP projects.
Mtai and Manga forest reserves were selected as joint forest management pilot areas, and
Mpanga was the first (in 1994) formally and officially gazetted village forest reserve in
Tanzania. A continuation of the joint forest management process has been planned for the
Derema Corridor (Hokkanen 2002). There are, however, problems in the administration of the
joint forest management areas in the government catchment forest reserves, because of the
fact that they generate few financial returns to villagers (Blomley and Ramadhani 2006). A
further general problem is that the villages do not have resources (time or energy) to invest in
forest surveying and managing activities. Villagers expect payments for these labour-intensive
tasks (Veltheim and Kijazi 2002; Katila 2008) and they many times feel that they have been
only used as a free labour force (Mustalahti 2007).

A positive example on successful participatory forest management is, however, found in the
miombo forests in Tanzania. The Suledo village forest reserve (shared among nine villages in
Kiteto District, Manyara Region) and the Angai village forest reserve (shared among 13
villages in Liwale District, Lindi Region) have generated high potential revenues to these
villages (Blomley and Ramadhani 2006). In the Duru-Haitemba forest in Arusha Region,
northern Tanzania, it has been experienced that a degraded national forest reserve was
recovering without costs to government when the people were motivated for managing it
(Alden Wily 2001). In this case the wildlife also returned to some degraded forest areas to the
extent that it caused problems to the farmers by damaging the fields and threating life and
property (Blomlay and Ramadhani 2006).

124
The fact remains, however, that the process of establishing community forest management
areas has been very slow in Tanzania. For establishing a community-managed forest reserve,
the assigned area needs to have clearly defined and marked boundaries and the title deeds and
by-laws legally approved. These processes are very time-consuming. In Liwale District in
Lindi Region in southern Tanzania the process of establishing the Angai Village Land Forest
Reserve took 11 years from 1994-2005 (White and Mustalahti 2005). Regional forest officers
were accusing the central government for the delays, and the government officers stated that
the reason was a lack of regional forest authorities who would be working (or preferred
working) with the issue. Some authorities did not want to loose their additional income and
power (Vihemki 2005).

Corruption makes natural resource management even more difficult. Logging permits are
difficult to obtain and so expensive that farmers cannot afford them. Instead, people prefer to
bribe the forest officers for permission to cut trees. Low salaries and the irregularity of salary
payments are tempting government officers to break rules and join illegal logging activities,
even if the Tanzanian government has started to take measures for ending the corruption
(Prevention of Corruption Act 2002; URT 2002b). These power relations are hindering the
development of participatory forest management and resulting in only limited benefits for
local communities. This is again an issue of political will - if it does not exist, then the
improvements would not take place.

In order to secure the sustainability of participatory forest management, more effort has been
put on conservation and economic benefits derived from the forest for the communities.
Emphasize has been given to strengthening and reintroducing indigenous knowledge and
practices for managing and protecting the forests (Meghji 2003). These steps toward
decentralization and devolution suggest that it is already possible to implement highly
productive small village forest reserves. These forest reserves could also be partly established
for rehabilitating unproductive, abandoned land areas (e.g. the old cardamom cultivation
areas, by planting large tree seedlings capable of competing with weeds), so as to solve
problems such as the lack of poles, timber, fuelwood and manure (by planting trees such as
Gliricidia sepium for green manure). Training and extension would also be cheaper to arrange
for the whole community simultaneously in the village forest reserves.

125
5.10. Government support

The human population and the degradation of natural resources are still both increasing in the
East Usambara Mountains. Recent recommendations have confirmed that new attitudes have
to be taken to abolish the corruption and the illegal use of natural resources. Until now the
income from natural resources has mostly benefited only a narrow elite. With strong new
measures, positive development should be seen (Milledge at al. 2007).

Over-regulation in forestry and a limited accountability among public officials encourage

corruption and illegal activities. The collection of royalties and taxes and the issuing of
logging licenses are not well organized in Tanzania, and that makes illegal forest product
extraction possible (Kobb 1998). In the present study, cardamom buyers also complained that
they often had to pay double taxes. Villagers were supposed to pay taxes to village
governments based on their sales, but they seldom did. As a consequence, village
governments were turning to buyers to make them pay the sales tax separately, even though it
was already incorporated into the price paid to a farmer. Regulations on the use of natural
resources should be fewer and simpler and they should require less paperwork. That would
also facilitate their control.

The government and the communities must work together for rural development, but the
government authorities have to regain their trusted position first before the farmers can
cooperate with them. The communication between government authorities and the rural
people should be improved by arranging more social interaction, for instance, in the form of
sports events or environmental campaigns, so as to facilitate discussion on important issues.
Information on new laws and regulations should be provided to all citizens, but in reality the
contents of the new forest policies and laws from 1998 were found to be unknown to 91% of
the villagers in the East Usambaras. In addition, in the present case, the local government
officers did not have copies of such documents. A majority of the community members (70%)
were not aware of the East Usambaras belonging to the Eastern Arc Mountain chain, or of
their biodiversity values (95%). In contrast, the role of forests in water catchments was known
by a clear majority (79%) of people (cf. Nyagawa and Kahamela 2005).

The laws and procedures of participatory forest management (possibilities for adopting village
forest management or community-based forest management) are not generally understood in
the villages (Blomley and Ramadhani 2006), even if EUCAMP has facilitated the preparation

126
of simplified step-by step instructions for participatory forest management guidelines with
translations into Kiswahili (White and Mustalahti 2005). The community awareness has to be
improved and linked with the extension services. The Tanzania Forest Conservation Group
(TFCG), in collaboration with CARE-Tanzania, has established a Community Forest
Conservation Network Unit, which provides information, training and various activities among
community groups involved in forest management (TFCG 2007b).

Even if there always exists political pressure for utilising funds for more concrete
development activities, such as building of schools or clinics, more funds and other resources
have to be allocated in the East Usambaras for nature conservation, forest management and
extension services. This is possible through compensations and environmental payments and
by eliminating the illegal extraction of forest products. The government of Tanzania invests
on the management of the Eastern Arc Forests less than 0.3% of the value of the benefits
these forests provide, measured in terms of water, hydropower and non-timber forest products
(Pfliegner and Burgess 2005). This is too little. Tanzania cannot expect the international
donors to finance all the conservation activities. The global concern is real, however, and it
can help in obtaining additional financing from international donors for sustainable
management and conservation activities. Only about 10% of the potential revenue from
timber export is collected annually in Tanzania, and it normally remains in the nations
treasury, which is not helping forest management activities (Milledge et al. 2007). The fund
raising could be linked to more effective collection of royalties, dissemination of information
on the conservation values in the area, promotion of ecotourism activities, and organizing of
training courses, such as those coordinated every year in this area by the Tropical Biology
Association for its more than 40 member institutions.

Specialized local-level support is required, and the capacity of the local government staff has
to be strengthened, so as to enable participatory forest management. It has been found that
small local-level changes can result in significant improvement in the livelihoods of the
forest-adjacent communities. This has to be done alongside with the national-level support,
e.g. to agricultural research. Better cooperation and coordination is needed among the
government, donors, NGOs and the farmers organizations. Comprehensive long-term
strategies and programmes should be developed by using participatory processes across
different sectors (FAO 2006). The main opportunities come from farmers' innovation,
growing markets and policies favourable to local communities (Kaimowitz 2003).

127
An encouraging policy environment would be important for the farmers in order to facilitate
commercialisation, reduce price risks, increase access to infrastructure, provide land security,
and promote technical innovations (Hazell et al. 2007). Some countries in Asia and Latin
America have made significant policy revisions that enable the small farmers to improve their
farm productivity. This has facilitated commercialisation and created better livelihoods among
smallholders in the rural areas (Tomich et al. 1995).

Current policy preferences in Tanzania are, however, in other sectors than in agriculture and
forestry, even if experience has shown that improved agricultural productivity reduces
poverty. The poverty alleviation is more efficient if small farms are preferred instead of the
large ones (Hazell et al. 2007). For supporting the agribusiness of smallholders, the
government should also assist in creating an internal quality control system that secures a
continuous supply of good-quality products and offers price differentials to farmers for
products of different quality. The introduction of new propagation material for spices and
trees could increase the yields considerably and widen their genetic base.

Most decision makers do not realize how important forest and tree resources really are for the
rural people (Kaimowitz 2003). Non-monetary values are difficult to estimate, but the value
of forest products in the Eastern Arc has been estimated to be at least $ 100 per person
annually (Pfliegner and Burgess 2005), and logging and the timber trade, most often illegal, in
some areas of rural Tanzania still bring benefits to about 60% of the households seasonally
(Milledge at al. 2007). There is, however, an urgent need for more equitable benefit
distribution (Mery et al. 2005).

Equal rights to resources are a necessity for improving the production and for changing the
cultivation to a more sustainable direction. That would provide security, motivation and
commitment to farmers activities. A process of economic improvement accompanied by
widening inequalities is generally unstable (Sharma 1998). The Land Act (URT 1999a) and
Village Land Act from year 1999 (URT 1999b) came into force in 2001, and they will
obviously contribute to more equal land use and natural resource management in Tanzania.
The existing patterns of land ownership have a strong influence on who benefits and who
does not. The present analysis (Study II) suggests that, in India, private land holdings and
well-organized farmers and buyers, together with a good infrastructure, have been able to
bring about better livelihoods for cardamom farmers. Strong farmers organizations should be
supported, because they are a necessity for developing the agribusiness.

128
A small elite in the villages is not representing the opinion and the conditions of the whole
village. It is critically important for policy makers to listen to and work also with the poorest
and most alienated groups, such as women. As women have an important role in rural
development, they also need equal rights to the natural resources, and decision-making and
financial support. Women need legal counselling to be able to fully claim those legal rights
that already exist for them.

In times of family emergencies and crop failures farmers reduce their livestock herds, but this
is not normally done for investing in agriculture or for better land management. The interest
rates of loans for the rural population should be reduced for enabling them to invest for their
future. Policies are needed to organize and finance investments also for soil improvement,
since it is always easier and cheaper to improve something that has not yet been totally
destroyed.

For remote mountain areas, low-cost sources of plant nutrients must be introduced. Extension
should support low-input farming and especially high-value perennials that encourage
sustainable cultivation of the same land area with continuous production (Scherr 1999).
Possible immediate field-level conservation activities include planting on the contour; using
mixed cropping and agroforestry; keeping the soil continuously covered; use of mulching; and
good crop rotations. The right timing in planting is also important (Beets 1990). There is a
great need of reorienting the extension staff and project officials by using participatory
workshops, which have proved to be efficient in exchanging information. In these workshops
everyone can and has something to contribute to the mutual learning process (Kajembe and
Mwaseba 1994). In a comparative study on the impacts of participatory forest management in
Tanzania, Mozambique, Laos and Vietnam, it was found that the attitude change of local
leaders and elite was very important for the adaptation of a participatory process. When the
village leaders were active in promoting tree planting, forest protection and fire management,
the villagers were also motivated to participate in forest management activities (Mustalahti
2007).

Local government authorities, however, have not functioned as expected in Tanzania. Most
villages do not even have clear inventories of their resources. Local governments would need
strengthening in planning, identifying and making sustainable use of their resources for the
benefit of the villagers. Village governments have the power to collect tax revenue from their

129
own residents and plan its use to fulfil community needs. All adult community members have
a right to participate in making decisions concerning their village and elect the responsible
village government members. When individuals or communities feel they have power to
affect the community affairs, they also become more creative, recognizing activities they have
not tried before. Decentralisation and empowerment of the rural people is the way to attain a
sustainable development which responds to rural peoples needs (Freling 1998). Specific
village land-use and management plans made according to the local concerns would be
particularly useful.

For successful rural development, demand-driven, interdisciplinary and participatory research

is needed that is market-oriented and can integrate local technologies, policies and
institutions. Examples on successful participatory research where new scientific information
has been distributed through strong technical support are found in Brazil, China and India
(Place and Prudencio 2006). Training and extension should provide the farmers with
important knowledge and skills, but also self-confidence and hope. A long history of donor
aid has left its marks: people in rural Tanzania are not using their own initiative to create a
change; instead, they may expect that a donor will come to implement a new project and give
solutions to their problems. Locally available resources should be used with locally available
technologies (Dolan 2006), and extensionists should also help the farmers in keeping simple
records on their crop yields. In rural Tanzania, farmers do not normally keep yield records,
and thus they are not aware of any benefits from investments they make (Place and Prudencio
2006).

All villages should have access to trained extensionists, also female ones. That would allow
continuous, regular and long-term support for farmers. Extensionists residing in the villages
would not have transport problems, and they would be ready to help when needed. They
would also know the local circumstances and special problems better than an outsider.
Decisions on their remuneration could be made in village meetings. Village extensionists need
to be regularly trained by the government extension staff.

A poor road network in Tanzania is blocking farmers from getting better prices and markets
for their products. In the East Usambaras, a much higher price for cardamom was obtained
from Kenyan businessmen coming to the closest town Muheza to buy cardamom; they could
pay TZS 300-500 more per kg than the other buyers. Access to an improved rural road system
would substantially improve the incomes of the rural poor. In the East Usambaras, better road

130
connections in the highlands have already resulted in higher producer prices, improved crop
management and a better overall income for the farmers.

The benefits that a smallholder farmer can get from lower transaction costs are multiple: more
traders would come to buy the products, and that would increase the farmers bargaining
power. The information exchange would also be more efficient. On the other hand, better
roads may also increase the immigration to the area and thus increase the pressure on the
available resources (Ehardt et al. 2005). A good relationship with product buyers is very
important and can reduce the transaction costs by half, as has been found in a study in Peru
(Escobal 2001). In the Himachal Pradesh mountain area in India, introduction of new crops
for farmers needed market promotion and better infrastructure; on the other hand, farmers
needed technical know-how and subsidised prices before any changes could be expected
(Sharma 1998).

It is known that the technical capacity is poor in the ministries of agriculture in Africa. What
is, however, needed most for a change is a political will of the government. In China the
economic reform was initiated in the agricultural sector because of thinking that the
stagnation of agricultural productivity was a bottleneck hindering further development of the
overall economy. Public interventions in land policies, marketing, and support services, as
well as research and development were initiated. The main focus was on commercially
oriented small farms (Hazell et al. 2007). Undeveloped agriculture is also hindering Africas
development. Long-term investments to raise the productive capacity are lacking, or they are
left to donors, who normally are involved for only a short period of time.

In most developing countries the potential of forestry is undervalued economically and

ecologically, and the same is true for agroforestry. Extension, education, fund allocation, laws
and research are still divided between agriculture and forestry (Lundgren 1985). The
development of rural Tanzania to the right direction needs, however, an interdisciplinary
perspective and cooperation between professionals and the farmers (Lundgren and Raintree
1983).

The Tanzanian Government has begun to be more active in developing agribusiness and
supporting farmers in it. In the Third Progress Report in Poverty Reduction Strategy (URT
2004) it is stated that a programme called Private Agribusiness Sector Support (PASS) was
implemented and, as a result, 77 farmers groups were organized. They received assistance on

131
contract farming, input supply, producer-price negotiations and advisory services. Also 75
credit guarantees were issued for agriculture investments. The social capacity was found to be
very important; creating an enabling and conducive environment has been mentioned as the
first one in the objectives related to agriculture in the Poverty Reduction Strategy Paper (URT
2004). The priorities of the PRSP are support to extension services, an increase in production,
and strengthening of commercial agriculture. Oksanen et al. (2003) noticed, however, that the
forest sector has not yet been specifically mentioned in the Tanzanian PRSP. This poor
understanding of the forest sectors role in poverty alleviation has been a problem in other
countries as well (Mery et al. 2005).

Tanzania has recently implemented policy changes which can be considered as investments
for a better future, such as the abolishment of school fees in 2001. The new Cooperation
Development Policy of 1997 (URT 1997b) already eliminated the governments role in
cooperative management, which has encouraged the farmers to form new cooperatives. The
Prevention of Corruption Act of 2002 (URT 2002b) started the fight against corruption at all
levels in Tanzania. Some improvement in this respect has obviously been made, as in 2006
there were as many as 65 countries more corrupt than Tanzania, while in 1998 Tanzania was
ranked fourth among the most corrupt countries in the world (Transparency International
2006). The National Policy on Women in Development (URT 1992) has highlighted equal
rights, job opportunities, and decision-making opportunities for the women in Tanzania. The
Village Land Act (URT 1999b) has given women equal rights to land. The National Trade
Policy (URT 2003) and The Agricultural and Livestock Policy (URT 1997a) recognize spices
as high-potential export products. Fertilizer subsidies have recently been tripled in order to
increase the agricultural productivity. It remains to be seen in the future how these changes
will help in improving the rural livelihoods in Tanzania.

132
6. CONCLUSIONS AND RECOMMENDATIONS

A primary conservation challenge in the East Usambara Mountains is to find solutions for the
livelihood needs of a dense rural population which suffers from poverty and is dependent on
unsustainable forms of agriculture and forest product extraction. Clearing the forest for
agriculture and establishing cultivation in the forests, especially for cardamom production, is
changing both the tree species composition and the structure of the forests radically. This is
clearly harmful for the biodiversity of these forests. Separate conservation areas are thus
required for maintaining habitats with a dense primary forest cover, which the many endemic
species depend on. With improved agroforestry systems, the local people will have an
opportunity to carry out sustainable farming outside the protected areas.

Cultivation of spices in agroforestry systems is possible, and organic fertilizer application

would allow an even more sustainable and profitable cultivation. Cardamom and other spices
are shade-demanding perennial crops yielding products with high economic value but low
biomass; this minimizes the nutrient losses with harvests. The diversity of the Usambaran
agroforestry systems is high as compared to other production systems, and with agroforestry it
is also possible to re-establish rich tree-based ecosystems that satisfy the farmers needs.

Rural poverty and environmental degradation in the East Usambaras should be addressed
simultaneously, since forest conservation can be successful only when the livelihoods of local
people are improved. It is a challenge to find a balance between ecosystem protection, forest
and tree management, and agricultural crop production, while maintaining or enhancing the
flow of forest benefits. Villages in this area need land-use planning that takes into
consideration these different aspects, regardless of whether it is promoted under the name of
integrated natural resources management, or an ecosystem, landscape or sustainable
livelihood approach. In a country such as Tanzania, where the government does not have
enough resources, participatory forest management can play an important role in forest
conservation and utilisation, on the condition that long-term support will be provided for the
communities. Thus the central government should also devolve its authority. Productive small
village forest reserves could be implemented by rehabilitating unproductive land using
agroforestry. A decentralization of forest management would empower the villages to plan
and realize their activities in a way which best responds to community needs.

133
Promotion of new suitable cash crops can contribute to increasing the family income while
preserving the forest. There are underutilised possibilities in rural Tanzania to develop the
homegardens. A combination of commercial and intensified small-scale agroforestry with
sustainable forest and land use can result in a better balance between a sufficient food security
and improved ecosystem conservation. With farmers participation it is possible to develop
new technical and economic practices that contribute to both poverty alleviation and
environmental protection. A basic assumption is a balanced interaction that includes, apart
from government officials, also the farmers and the traders and their respective organisations.

When farmers grow spices outside the forest using agroforestry systems, their crops could be
certified as being sustainably produced. This would increase the product marketability and
price, and encourage other cultivators to move their production out from the forest. Obtaining
organic certification for selected high-quality products is also possible in the East Usambaras.
Buyers of organic spices in Tanzania are already facilitating farmers in matters related to the
transport, processing, training and marketing of products. Spice agroforestry is a viable option
under the condition that the farmers organise themselves for ensuring the crop quantity and
quality targets. Strong farmers organisations are a necessity in developing this agribusiness,
and this is also a sector on which the government and development organizations should focus
in the East Usambaras and in rural Tanzania in general.

Land Equivalent Ratios examined in the present study indicated that intercropping of
cardamom or black pepper with grevillea or gliricidia trees is advantageous, especially in the
case of grevillea (grevillea with pepper or grevillea with pepper and cardamom). A
comparison of other response variables also showed the superiority of these cultivation
systems. Cardamom yields were 5-7 times higher than the average yields of this crop in
intercropping with grevillea; and black pepper yields were four times higher with grevillea
than in other cultivation systems. Black pepper yields gradually increase with time for several
years, and the data in the present study obtained for the last year in the best system, with
grevillea, still showed a five-fold increase as compared to the previous year. In black pepper
production, regular pruning of the main stem should be introduced. The quality of both spices
could be improved considerably with better harvesting and post-harvesting technology.

Grevillea does not interfere much with associated crops, and also crops do not affect much the
growth of grevillea. The growth of gliricidia in the East Usambaras was modest, but this tree
species proved its ability to improve the organic matter content and the total nitrogen of the

134
soil to levels that exceed those found in the natural forest. This would ensure a high and long-
term production for the spice crops. Both tree species produced more biomass when they were
regularly pruned. Pruned gliricidia also provides excellent green manure.

Soil analyses showed that intercropping of trees and spice crops can improve some soil
characteristics, but addition of organic manure is still needed to maintain the soil fertile in the
long term. Farmers urgently need training on how to apply the available organic fertilizers
more efficiently. It is highly recommended to use soil-improving plants (leguminous
agricultural crops and trees, or trees with Frankia) and to apply green manure (e.g. branches
and leaves of gliricidia) to the fields, especially to those located far away from the
homesteads. Green manure would raise the yields and maintain the soil fertility without
causing high transporting costs. Steep slopes are not suitable for agriculture and should never
be without a vegetative cover. Natural vegetative strips can improve the erosion control, with
negligible competition with annual crops and low installation and maintenance costs. In time
they develop into terraces, which would form a good base for agroforestry and more
sustainable cultivation.

Five years after the establishment of a cardamom cultivation system its production rapidly
starts to decline. With fertilizers the production could be maintained at a profitable level for a
longer time, as also indicated by preliminary results obtained in research after the present
study period. Farmyard manure application clearly increases the cardamom yields, especially
in combination with nitrogen-fixing trees. If a manure application is not sufficient for
improving the soil nutrient and pH status, it is also recommendable to use rock phosphate.
Organic rock phosphate is available, which allows organic cultivation. The introduction of
new propagation material for spices and trees could also increase the yields, and in addition,
broaden the genetic base of the planting stock.

In Tanzania the policy environment has rapidly changed in a way that facilitates agribusiness
and rural development. Equal rights for men and women in natural resource management and
clear land tenure arrangements would further motivate the rural communities for more
sustainable cultivation practices. Access to low-interest loans would decrease the future risks
of investments. Extension services related to spice growing and agroforestry should be
intensive, long-term and comprehensive, using various methods. Small local-level changes,
supported by an enabling policy environment for the smallholder farming sector, would
provide the steps necessary for improved rural development in the East Usambaras. Farmers

135
changed their forest spice cultivation for intensive homegarden management when the
conservation area was enlarged and diminished the availability of cultivation land inside the
forest. It can thus be concluded that intensification of farming systems supports forest and
biodiversity conservation.

Rural poverty in Tanzania can be alleviated with a long-term strategy: by improving the soil
fertility, by intensifying the sustainable land use with high-value cash crops, and by
supporting the smallholder farmers with an encouraging policy environment. What is,
however, needed most for a change is a political will of the government that recognises the
importance of agricultural and forestry development in poverty reduction.

136
REFERENCES

Abu Ghaida, D. & Klasen, S. 2004. The Economic and Human Development Costs of
Missing the Millennium Development Goal on Gender Equity. World Bank
Working Paper. Report No. 29710. Washington, D.C. The World Bank.
Agroforestree Database. 2006. ICRAF
http://www.worldagroforestrycentre.org/Sites/TreeDBS/aft.asp (12.3.2006).
Akyeampong, E., Hitimana, L., Torquebiau, E. and Munyemana, P.C. 1999.
Multistarata agroforestry with beans, bananas, and Grevillea robusta in the
highlands of Burundi. Expl. Agric. 35: 357-369. Cambridge University Press.
Alden Wily. L. 2003. Community-based land tenure management: Questions and
answers about Tanzanias new Village Land Act, 1999. IIED (International
Institute for Environment and Development. Issue paper no. 120. 59 p.
http://www.iied.org/pubs/pdf/full/9295IIED.pdf (26.11.2007).
Alden Wily, L. 2001. Forest Management and Democracy in East and Southern
Africa: Lessons from Tanzania. Gatekeeper Series no. 95. International
Institute for Environment and Development. Sustainable Agriculture and Rural
Livelihoods Programme. 20 p.
Aldrich, M. 2000. Tropical montane cloud forests. Pp. 390-397. In: Price, M.F. &
Butt, N. (eds.). Forests in Sustainable Mountain development: a State of
Knowledge. Report for 2000, Task Force on Forests in Sustainable Mountain
Development. CABI Publishing. UK & USA. 590 p.
Amani Butterfly Project. 2007. http://www.amanibutterflyproject.org/ (30.11.2007).
Amani Nature Reserve General Management Plan. 1998. Ministry of Natural
Resources and Tourism, Forestry and Beekeeping Division, Tanzania.
Amara, D.S. & Kamara, A.Y. 1998. Growth and yield of Gliricidia sepium (Jack.)
Walp. provenances on an acid sandy clay loam soil in Sierra Leone.
International Tree Crops Journal 9: 169-178.
Anderson, L.S & Sinclair, F.L. 1993. Ecological interactions in agroforestry systems.
Agroforestry Abstracts 6(2): 57-91.
Angelsen, A. & Kaimowitz, D. 2004. Is Agroforestry Likely to Reduce Deforestation?
Pp: 87-106. In: Schroth, G., da Fonseca, G.A.B., Harvey, C.A., Gascon, D.,

137
 Vasconcelos, H.L. and Izac, A.N. Agroforestry and Biodiversity Conservation
in Tropical Landscapes. 575 p. Island Press.
ASB (Alternatives to Slash and Burn research consortium). Undated. Alternatives to
slash and burn: A global initiative. Nairobi, ICRAF.
Ashton, M.S., Gunatilleke, C.V.S., Singhakumara, B.M.P. and Gunatilleke, I.A.U.N.
2001. Restoration pathways for rain forest in southwest Sri Lanka: a review of
concepts and models. Forest Ecology and Management 154: 409-430.
Atta-Krah, K., Kindt, R., Skilton, J.N. and Amaral, W. 2004. Managing biological and
genetic diversity in tropical agroforestry. Agroforestry System 61: 183-194.
Azeria, E.T., Sanmartin, I., s, S., Carlson, A. and Burgess, N. 2007. Biogeographic
patterns of the East African coastal forest vertebrate fauna. Biodiversity and
Conservation 16(4): 883-912.
Balmford, A., Moore, J.L., Brooks, T., Burgess, N., Hansen, L.A., Williams, P. and
Rahbek, C. 2001. Conservation Conflicts across Africa. Science 291(5513):
2616-2619.
Beets, W.C. 1990. Raising and Sustaining Productivity of Smallholder Farming
Systems in the Tropics. 738 p.
Binagi, E.R. 1992. Consolidation of environmental education for adults: critique of
FINNIDA-funded forestry projects in Tanzania. A case study of the East
Usambara Catchment Forest Project. Technical Paper 7. EUCFP. Forestry and
Beekeeping Division & Finnish Forest and Park service, Dar es Salaam &
Vantaa. 23 p.
Biodiversity Hotspots. 2007. Club of Amsterdam Journal, June, Issue 89.
http://www.clubofamsterdam.com/press.asp?contentid=702#top (27.7.2007).
Black Pepper Cultivation. 2007. Spices Board of India, Ministry of Commerce,
Government of India http://www.indianspices.com/html/ind_sp_farm_02.html.
(30.5.2007).
Blomlay, T. & Ramadhani, H. 2006. Going to scale with Participatory Forest
Management: Early lessons from Tanzania. International Forestry Review
8(1): 93-100.
Boesen, J. & Friis-Hansen, E. 2001. Soil Fertility Management in Semi-Arid
Agriculture in Tanzania: Farmers Perceptions and Management Practices.
CDR Working Papers. Centre for Development Research, Denmark. 31 p.
Brown, L.H. 1981. The conservation of forest islands in areas of high human density.
African Journal of Ecology. 19: 27-32.

138
Burgess, N.D., Balmford, A., Cordeiro, N.J., Fjels, J., Kuper, W., Rahbek, C.,
Sanderson, E.W., Scharlemann, J.P.W., Sommer, J.H. and Williams, P.H.
2007a. Correlations among species distributions, human density and human
infrastructure across the high biodiversity tropical mountains of Africa.
Biological Conservation 134(2): 164-177.
Burgess, N.D., Butynski, T.M., Cordeiro, N.J., Doggart, N.H., Fjelds, J., Howell,
K.M., Kilahama, F.B., Loader, S.P., Lovett, J.C., Mbilinyi, B., Menegon, M.,
Moyer, D.C., Nashanda, E., Perkin, A., Rovero, F., Stanley, W.T. and Stuart,
S.N. 2007b. The biological importance of the Eastern Arc Mountains of
Tanzania and Kenya. Biological Conservation 134(2): 209-231.
Burgess, N. & Kilahama, F. 2005. What are the main threats facing the Eastern Arc
Forests and how serious are they? The Arc Journal, Issue 19. Tanzania Forest
Conservation Group. Pp. 16-17.
Business Times, Tanzania. 2003. Tanzania: Recognize spice industry potential. 11
April, 2003. Das es Salaam, Tanzania.
CEPF (Critical Ecosystem Partnership Fund). 2005. Ecosystem profile: Eastern Arc
Mountains & Coastal Forests of Tanzania & Kenya. Conservation
International. Centre of Insect Physiology and Ecology.
www.cepf.net/ImageCache/cepf/content/pdfs/final_2eeasternarc_2eep_2epdf/v
2/final.easternarc.ep.pdf. (15.10.2005).
Chambers, R. 1997. Whose Reality Counts? Putting the first last. Intermediate
Technology Publications. The Bath Press, Bath. 297 p.
Current, D.E. & Scherr, S. J. 1995. Farmer costs and benefits from agroforestry and
farm forestry projects in Central America and the Caribbean: Implications for
policy. Agroforestry Systems 30: 87-103.
De Costa, W.A.J.M., Surenthran P. and Attanayake K.B. 2005. Tree-crop interactions
in hedgerow intercropping with different tree species and tea in Sri Lanka: 2.
Soil and plant nutrients. Agroforestry Forum 63: 211-218.
Dolan, T. 2006. Sustainable agriculture and rural development a response to the
realities of rural Africa. In: Garrity, D., Okono, A., Grayson, M. and Parrott, S.
(eds.). World Agroforestry into the Future. World Agroforestry Centre.
Nairobi. Pp. 129-134.
Dommergues, Y. R. 1995. Nutrient fixation by trees in relation to soil nitrogen
economy. Nutrient Cycling in Agroecosystems 42(1-3): 215-230.

139
Doody, K.Z., Howell, K.M. and Fanning, E. (eds.) 2001. Amani Nature Reserve: A
biodiversity survey. Technical paper 52. Frontier Tanzania. East Usambara
Conservation Area Management Programme (EUCAMP). Ministry of Natural
Resources and Tourism, Tanzania, Forestry and Beekeeping Division;
Department of International Development Co-operation, Finland,
Metshallitus Consulting and Frontier-Tanzania, University of Dar es Salaam,
Society for Environmental Exploration. Tanga. 145 p.
Dubois, O. 2003. Forest-Based Poverty Reduction: A Brief Review of Facts, Figures,
Challenges and Possible Ways Forward. Pp. 65-83. In: Oksanen, Pajari,
Tuomasjukka (eds.) Forests in Poverty Reduction Strategies: Capturing the
Potential. EFI Proceedings No. 47. European Forest Institute. 206 p.
East Usambara Mountains. 1988. Forests and Forestry. Amani Forest Inventory and
Management Plan Project. Forest Division of Tanzania. Joint Venture
Finnmap-Silvestria. FINNIDA.
Eastern Arc Mountains Information Source. 2001. What is it all about in the
conservation of natural forests in the East Usambaras?
http://www.easternarc.org/eucamp/whatitisabout.html (15.6.2003).
Edwards, N., Tokar, M. Maxwell, J. 1997. Agribusiness Development in Sub-Saharan
Africa: Optimal Strategies and Structures. Technical Paper No. 83. SD
Publication Series. Office of Sustainable Development. Bureau for Africa.
133 p.
Ellman, A. 1996. Handing over the stick? Participatory village forest planning and
management in the East Usambara Mountains, Tanzania. East Usambara
Catchment Forest Project Technical Paper No. 18. Forestry and Beekeeping
Division & Finnish Forest and Park service, Dar es Salaam & Vantaa. 88 p.
Ehardt, C.L., Jones, T.P. and Butynski, T.M. 2005. Protective Status, Ecology and
Strategies for Improving Conservation of Cercocebus sanjei in the Udzungwa
Mountains, Tanzania. International Journal of Primatology 26(3): 557-583.
Escobal, J.A. 2001. The benefits of roads in rural Peru: A transaction costs approach.
Grupo de Anlisis para el Desarrollo (GRADE). Lima.
EUCAMP. 2001. Mid-Term Review of the East Usambara Conservation Area
Management Programme, Phase III, Tanzania. Ministry for Foreign Affairs of
Finland. 67 p.
EUCAMP. 2002. Administrative report 38. Annual Report 2001/2002. MNRT,
Forestry and Beekeeping Division, Tanzania, Ministry of Foreign Affairs,

140
 Indufor/Meshallitus Group, Finland & EU, Delegation of the European
Commission in the United Republic of Tanzania.
EUTCO (East Usambara Tea Company). 1994. Feasibility study for an outgrower tea
and fuelwood project. East Usambara Tea Company, Tanga.
Evans, J. 1982. Plantation Forestry in the Tropics. Clarendon Press, Oxford, 472 p.
Ewel, J.J. 1999. Natural systems as models for the design of sustainable systems of
land use. Agroforestry Systems 45: 1-21.
FAO. 2006. Better forestry, less poverty. A practitioners guide. FAO Forestry Paper
149. Rome. 66 p.
FAO. 2004. Forestry country reports.
http://www.fao.org/forestry/site/countryinfo/en/tza (30.5.2007).
FAO. 2003. Forests and poverty alleviation. Pp. 61-73. State of the Worlds Forests.
151 p.
FAO. 1995. Non-wood forest products for rural income and sustainable forestry.
NWFPs No. 7. Rome. 138 p.
FAO. 2000. Trees Outside Forests. Concept Paper. Forest Conservation Working
Paper Series. Food and Agriculture Organization of the United Nations, Rome.
Fernandes, E.C.M. 2000. Case study: Chagga agroforests on Mount Kilimanjaro,
Tanzania. Pp. 284-287. In: Price, M.F. & Butt, N. (eds.). Forests in Sustainable
Mountain Development: a State of Knowledge. Report for 2000, Task Force
on Forests in Sustainable Mountain Development. CABI Publishing, 2000. UK
& USA. 590 p.
Fernandes, E.C.M., OKtingati, A. and Maghembe J. 1989. The Chagga
homegardens: a multi-storeyed agroforestry cropping system on Mount
Kilimanjaro (Northern Tanzania). Pp. 309-332. In: Nair, P.K.R. (ed.).
Agroforestry Systems in the Tropics. Vol. 31. Kluwer with ICRAF. 664 p.
Financial Times. 2005. Tanzania set to produce first organic cotton this year.
17.8.2005.
Finnish Saintpaulia Society, 1997. http://www.saintpaulia.fi/ (24.11.2007).
Fjelds, J & Lovett, J.C. 1997. Biodiversity and environmental stability. Biodiversity
and Conservation 6: 315-323.
Fowler, S. & Nyambo, B. 1996. Survey of invasive weeds affecting biodiversity in the
forests of the East Usambara Mountains, Tanzania 1-6 October 1995.
Technical Paper 20. East Usambara Catchment Forest Project. Forestry and

141
 Beekeeping Division & Finnish Forest and Park service, Dar es Salaam &
Vantaa. 20 p.
Freling, D. (ed.) 1998. Paths for Change. Experiences in Participation and
Democratisation in Lindi and Mtwara Regions, Tanzania. Rural Integrated
Project Support (RIPS) Programme, Phase II. Ministry for Foreign Affairs of
Finland, Department for International Development Cooperation. Oy Finnagro
Ab. 128 p.
Garrity, D.P. 1993. Agroforestry and the Sustained Productivity of Asias Humid
Uplands. In: Mottema, J.W.T. and Stolz, D.R. (eds.) Upland Agriculture in
Asia. Proceedings of a Workshop Held in Bogor, Indonesia April 6-8, 1993.
CGPRT Centre. No 30: 225-233.
Garrity, D.P. 2004. Agroforestry and the achievement of the Millennium Development
Goals. Agroforestry Systems 61: 5-17.
Gholz, H.L. (ed.) 1987. Agroforestry: Realities, Possibilities and Potentials. Kluwer
Academic. 227 p.
Glover, E.K. 2005. Tropical dryland rehabilitation Case study on participatory forest
management in Gedaref, Sudan. Doctoral thesis. University of Helsinki, Dept.
of Forest Ecology, Viikki Tropical Resources Institute (VITRI). 183 p.
Gomez, A.A. & Gomez, K.A. 1983. Multiple Cropping in the Humid Tropics of Asia.
IDRC. 248 p.
Guardian. 2006. Europe, Canada Commend Tanzanian cotton. 10.01.2006.
Gunathilake, H.A.J. & Wasanthe, H.G. 2004. Gliricidia for production of Green
Manure and Green Energy in Coconut Plantations. In: Zoysa A.K.N and Ziyad
Mohamed M.T. (eds.) Proceeding of the first symposium on Plantation Crop
Research Current Trends and Future Challenges. The Tea Research Institute
of Sri Lanka. Pp. 43-52.
Hamilton A.C. & Bengsted-Smith, R. (eds.) 1989. Forest Conservation in the East
Usambara Mountains, Tanzania. IUCN, Gland and Cambridge. 392 p.
Hamilton, A.C. 1989. Safeguarding the Resources of the East Usambaras. In:
Hamilton A.C. & Bengsted-Smith, R. (eds.) Forest Conservation in the East
Usambara Mountains, Tanzania. IUCN, Gland and Cambridge. pp. 57-71.
Hamunen H. 1998. Early successional dynamics of woody vegetation on abandoned
agricultural land in the East Usambaras, Tanzania. M.Sc. thesis, University of
Joensuu, Faculty of Forestry, Finland.
Harwood, C.E. 1989. Grevillea robusta: an annotated bibliography. ICRAF. 123 p.

142
Hazell, P., Poulton, C., Wiggins, S. and Dorward, A. 2007. The Future of Small Farms
for Poverty Reduction and Growth. 2020 Discussion Paper 42. International
Food Policy Research Institute. Washington D.C. 38 p.
Hellin, J. 2006. Better Land Husbandry: From Soil Conservation to Holistic Land
Management. Science Publishers, Enfield and Plymouth. 325 p.
Hoekstra, D.A. 1987. Economics of Agroforestry. Agroforestry systems 5(3):
293-300.
Hokkanen, T.J. 2002. Adapting UNESCO MAB-principles to East Usambara Area,
Tanzania. East Usambara Conservation Area Management Programme.
Technical Paper No. 62. EUCAMP, Tanga, Tanzania; Metshallitus, Vantaa;
Forestry and Beekeeping Division, Dar es Salaam. 60 p.
Howard, P.C. 1996. Baseline biological surveys in selected East Usambara forest
reserves and forests, 1995-96. Project evaluation report. East Usambara
Catchment Forest Project, Technical Paper 21. Forest and Beekeeping
Division & Finnish Forests and Park Service, Dar es Salaam & Vantaa.
Huang, W., Luukkanen, O., Johansson, S., Kaarakka, V., Risnen, S. and Vihemki,
H. 2002. Agroforestry for biodiversity conservation of nature reserves:
functional group identification and analysis. Agroforestry Systems 55:65-72.
Huang, W. 1998. Productive Coexistence and Gain in Agroforestry Systems. Ph.D.
thesis, University of Helsinki, Dept. of Forest Ecology, Viikki Tropical
Resources Institute (VITRI). Acta Forestalia Fennica 260. The Finnish Society
of Forest Science & the Finnish Forest Research Institute. 72 p.
Huxley, P.A. (ed.) 1983. Plant Research and Agroforestry. ICRAF. 617 p.
Ihalainen, L. 2006. Improved rubber agroforestry system RASI in West Kalimantan,
Indonesia Biodiversity and farmers perceptions. M.Sc. thesis, University of
Helsinki, Dept. of Forest Ecology, Viikki Tropical Resources Institute
(VITRI). 117 p.
Ikerra, S.I., Semu, E. and Mrema, J. 2006. Combining Tithonia diversifolia and
minjingu phosphate rock for improvement of P availability and maize grain
yields on a chromic acrisol in Morogoro, Tanzania. Nutrient Cycling in
Agroecosystems 76(2-3): 249-260.
Investing in Mountains. 1997. Innovative Mechanisms and Promising Examples for
Financing conservation and Sustainable Development. The Mountain Institute
and FAO. 48 p.

143
Isham, J. 2002. The effect of Social Capital on Fertilizer Adoption: Evidence from
Rural Tanzania. The Journal of African Economies 11: 39-60.
lversen, S.T. 199l. The Usambara Mountains, N.E. Tanzania: history, vegetation and
conservation. Uppsala.
Jacob, V.J. & Alles, W.S. 1989. The Kandyan gardens of Sri Lanka: pp 181- 195. In:
Nair, P.K.R (ed.). Agroforestry Systems in the Tropics. Forestry Science. Vol.
31. Kluwer with ICRAF. 664 p.
Jama, B., Kwesiga, F. and Niang, A. 2006. Agroforestry innovations for soil fertility
management in sub-Saharan Africa: Prospects and Challenges. In: Garrity, D.,
Okono, A., Grayson, M. and Parrott, S. (eds.). World Agroforestry into the
Future. World Agroforestry Centre. Nairobi. Pp. 53-50.
Jama, B., Swinkels, R., and Buresh, R.J. 1997. Agronomic and Economic Evaluation
of Organic and Inorganic Sources of Phosphorus in Western Kenya.
Agronomy Journal 89: 597-604.
Johansson, S.G. Cunneyworth, P. Doggart, N. and Botterweg, R. 1998. Biodiversity
surveys in the East Usambara Mountains: preliminary findings and
management implications. Journal of East African Natural History 87 (1-2):
139-158.
Johansson, S. & Sandy, R. 1996. Protected Areas and Public Lands. Land Use in the
East Usambara Mountains. Technical Paper 28. East Usambara Catchment
Forest Project. Forestry and Beekeeping Division & Finnish Forest and Park
service, Dar es Salaam & Vantaa. 37 p.
Kaimowitz, D. 2003. Not by Bread Alone. Forest and Rural Livelihoods in Sub-
Saharan Africa pp. 45-63. In: Oksanen, Pajari, Tuomasjukka (eds.) Forests in
Poverty Reduction Strategies: Capturing the Potential. EFI Proceedings No.
47, 2003. European Forest Institute. 206 p.
Kajembe, G.C. & Mwaseba, D.M. 1994. The extension and communication
programme for the East Usambara Catchment Forest Project. Technical Paper
11. Forestry and Beekeeping Division & Finnish Forest and Park service, Dar
es Salaam & Vantaa. 70 p.
Kang, B.T. 1993. Soil tillage in Africa: needs and challenges. FAO Soils Bulletin 69.
Kang, B.T. & Mulongoy, K. 1987. Gliricidia sepium as a source of green manure in
an alley cropping system. In: Withington, D., Glover, N. and Brewbaker, J.L.
(eds.), Gliricidia sepium (Jacq.) Walp.: Management and Improvement,

144
 Proceedings of a workshop at CATIE, Turialba, Costa Rica. NFTA Special
Publication 87-01, pp. 44-49.
Kaoneka, A.R.S., Ngaga, Y.M. and Monela, G.C. 2000. Case Study: Forests in the
Usambara Mountains: historical perspective and future prospects. Pp. 97-102.
In Price, M. & Butt, N. 2000. Forests in Sustainable Mountain Development:
A State of Knowledge Report for 2000. IUFRO Research Series 5. 2.
Mountain Communities and Their Forests. CABI publishing. UK & USA.
590 p.
Kaoneka, A.R.S. & Solberg, B. 1994. Forestry related land use in the West Usambara
Mountains, Tanzania. Agriculture, Ecosystems and Environment 49: 207-215.
Katigula, M.I.L. 1999. Dependence of local communities on protected forests as
sources of non-timber forest products and its role in forest management. A
case study of the East Usambara Protected Natural Forests, Tanzania. M.Sc.
Thesis in socio-economic information for natural resources management.
International Institute for Aerospace Survey and Earth Sciences. Enschede.
116 p.
Katila, P. 2008. Devolution of forest-related rights: Comparative analyses of six
developing countries. Doctoral thesis. University of Helsinki, Dept. of Forest
Ecology, Viikki Tropical Resources Institute (VITRI).162 p.
Kerkhof, P. 1990. Agroforestry in Africa: a survey of project experience. 1.
Developing countries. Forestry. The Panos Institute. 213 p.
Kessy, J.F. 1998. Conservation and utilization of natural resources in the East
Usambara Forest Reserves: Conventional views and local perspectives.
Tropical resource management papers. Ph.D. thesis, Wageningen Agricultural
University. 168 p.
Keswani, C.L. & Ndunguru, B.J. (eds.) 1982. Intercropping. Proceedings of the
Second Symposium on Intercropping in Semi-Arid Areas, held at Morogoro,
Tanzania, 4-7 August, 1980. IDRC (Monographs) (IDRC), no. 186.
International Development Research Centre. Ottawa. 168 p.
Kiriinia, C. 1999. The Best way to manage grevillea. Agroforesty Today (June). Pp.
34-35.
Kobb, D. 1998. Forest royalties in Tanga Region: Paper versus reality. East Usambara
Catchment Forest Project. Technical Paper 48. Ministry of Natural Resources
and Tourism, Tanzania, Forestry and Beekeeping Division and Department of

145
 International Development Cooperation, Finland, Finnish Forest and Park
Service. Tanga. 65 p.
Koning, N. 2002. Should Africa Protect its Farmers to Revitalise its Economy?
Gatekeeper Series no. 105. IIED. 16 p.
Koponen, J. 1988. People and production in late precolonial Tanzania: History and
Structures. Monographs of the Finnish Society for Development Studies No 2.
Transactions of the Finnish Anthropological Society No 23. Studia Historica
28. Finnish Society for Development Studies; Finnish Anthropological
Society; Finnish Historical Society; and Scandinavian Institute of African
Studies. 434 p.
Korikanthimanth, V.S. 2001. Cardamom (small). In: Peter, K.V. (ed.). Handbook of
herbs and spices. Vol. 1. Woodhead. Pp. 123-133.
Korikanthimath, V.S., Mulge, R., Hedge, R., Hiremath, G.G. and Hosmani M.M.
1998. Crop Combination and yield pattern in coffee mix cropped with
cardamom. Journal of Plantation Crops 26(1): 41-49
Kumar, B. M., Kumar, V.S. and Mathew, T. 1995. Floristic attributes of small
cardamom (Elettaria cardamomum (L.) Maton) growing in the Western Ghats
of peninsular India. Agroforestry Systems 31: 275-289.
Kumar, B.M. & Nair, P.K.R. 2004. The enigma of tropical homegardens. Agroforestry
Systems 61:135-152.
Land and People. 2003. www.worldagroforestry.org/. (22.09.2005)
Lawless, J. 1995. The Illustrated Encyclopedia of Essential Oils. Element Books. Pp.
56-64.
Leakey, R., Tchoundjeu, Z., Schreckenberg, K., Simons, T., Shackleton, S., Mander,
M., Wynberg, R., Shackleton, C. and Sullivan, C. 2006. Trees and markets for
agroforestry tree products: Targeting poverty reduction and enhanced
livelihoods. In: Garrity, D., Okono, A., Grayson, M. and Parrott, S. (eds.).
World Agroforestry into the Future. World Agroforestry Centre. Nairobi. Pp.
11-22.
Liyanage, M de S., Tejwani, K.G. and Nair, P.K.R., 1989. Intercropping under
coconuts in Sri Lanka. pp.165-179. In: Nair P.K.R. (ed.). Agroforestry
Systems in the Tropics. Forestry Science. Vol. 31. Kluwer with ICRAF. 664 p.
Lovett, J.C. & Wasser, S.K. 1993. Biogeography & ecology of the rain forests of
Eastern Africa. Cambridge University Press. 351 p.

146
Lulanda, L.L.L. 1998. Meeting the needs of the people through species domestication:
a basis for effective conservation of the Eastern Arc Mountain forest
biodiversity. Journal of East African Natural History 87 (1-2): 243-252.
Lundgren, B. 1985. Global deforestation, its causes and suggested remedies.
Agroforestry Systems 3: 91-95.
Lundgren, B. & Lundgren, L. 1982. Socio-economic effects and constraints in forest
management: Tanzania. In: Hallsworth, E.G. (ed.) Socio-economic Effects and
Constraints in Tropical Forest Management. John Wiley & Sons Ltd., New
York. pp. 43-52.
Lundgren, B & Nair, P.K.R. 1985. Agroforestry for soil conservation. In: El-Swaify,
S.A., Modenhauer, W.C. and Lo, A. (eds.). Soil Erosion and Conservation.
Ankeny, Iowa. Soil Conservation Society of North America. pp. 703-717.
Lundgren, B. & Raintree, J.B. 1983. Sustained Agroforestry. Pp. 1-26. In: B. Nestel,
(Ed.) Agricultural research for development: potentials and challenges in Asia.
The Hague, ISNAR.
Luukkanen, O., Katila, P., Elsiddig, E., Glover, E., Sharawi, H. and Elfadl, M. 2006.
Partnership between public and private actors in forest-sector development.
Options for dryland Africa based on experiences from Sudan with case studies
on Laos, Nepal, Vietnam, Kenya, Mozambique and Tanzania. Study
commission by the Ministry for Foreign Affairs of Finland. University of
Helsinki Tropical Forestry Reports 31. 119 p.
MacDicken, K.G. 1994. Selection and Management of Nitrogen-fixing Trees.
Winrock International, Morrilton, Arkansas and FAO, Bangkok.
Mandal, K.G., Misra, A.K., Hati, K.M., Bandyopadhyay, K.K., Ghosh, P.K. and
Mohanty M. 2004. Rice residue- management options and effects on soil
properties and crop productivity. Food, Agriculture & Environment 2 (1):
224-231.
Manuh, T. 1998. Women in Africa's development. Overcoming obstacles, pushing for
progress. Africa Recovery Briefing Paper. Number 11, April 1998. United
Nations Publication.
http://www.un.org/ecosocdev/geninfo/afrec/bpaper/maineng.htm (22.6.2007)
Masayanyika, S.W.S. 1995. Linking conservation with livelihood security. The case
of East Usambara Mountains, Amani Division, Tanga region, Tanzania. M.Sc.
thesis. Dept. of Land Resources and Urban Sciences. International Institute for
Aerospace Survey and Earth Sciences. Enschede. 109 p.

147
Masilingi, W.M.K. 1992. Consultancy Report of the Legal Establishment of the
Amani Nature Reserve. Technical Paper No. 6. East Usambara Catchment
Forest Project. Forestry and Beekeeping Division & Finnish Forest and Park
service, Dar es Salaam & Vantaa. 27 p.
Mastrantonio, J.L. & Francis, J.K. 2004. Tropical Forest Conservation. Pp 190-199.
In: Elevitch, C.R. (ed.). The Overstory Book, Cultivating Connections with
Trees, 2nd Edition. Permanent Agriculture Resources. 548 p.
Mbuya, L.P., Msanga, H.P., Ruffo, C.K., Birnie, A. and Tengns, B. 1994. Useful
trees and shrubs for Tanzania. Published by SIDAs Regional Soil
Conservation Unit, RSCU. 542 p.
McNeely, J. 2004. Forest Biodiversity. In The Overstory Book: Cultivating
Connection with Trees. 2nd Edition. Elevitch, G. (ed). Permanent Agriculture
Resources. USA. 526 p.
Meghji, Z. 2003. Participatory Forest Management in Tanzania. Paper submitted in
the XII World Forestry Congress, 2003. Quebec City, Canada
http://www.fao.org/DOCREP/ARTICLE/WFC/XII/0813-C1.HTM
(25.6.2007).
Melgar, W. 2003. Estado de la diversidad biolgica de los rboles y bosques de
Guatemala. Taller Regional sobre los Recursos Genticos Forestales de
Centroamrica, Cuba y Mxico. CATIE, Turrialba, Costa Rica, 24 al 29 de
noviembre 2002. Servicio de Desarrollo de Recursos Forestales Documento de
Trabajo FGR/53S. FAO, Rome. 65 p.
Mery, G., Alfaro, R., Kanninen, M. and Lobovikov, M. (eds.) 2005. Forests in the
Global Balance Changing Paradigms. IUFRO World Series Volume 17.
Helsinki. 318 p.
Milledge, S.A.H., Gelvas, A.K. and Ahrends, A. 2007. Forestry, Governance and
National Development: Lessons learned from logging boom in Southern
Tanzania. An Overview. TRAFFIC East/Southern Africa/Tanzania/
Development Partners Group/Ministry of Natural Resources of Tourism, Dar
es Salaam. 16 p.
Millenium Ecosystem Assessment. 2005. Ecosystems and Human Well-Being:
Biodiversity Synthesis. World Resources Institute, Washington D.C. 86 p.
Mishra, H.R. 2002. Mountains of the developing world: pockets of poverty or
pinnacles for prosperity. Unasylva - No. 208 2002. International year of
Mountains. FAO.

148
Montagnini, F., Cusack, D., Petit, B. and Kanninen, M. 2003. Environmental Services
of Native Tree Plantations and Agroforestry Systems in Central America.
Forest Ecology and Management 175 (1-3): 195-204.
Mose, L.O., Nzabi, W., Onyango, M., Gor, A.B., Moruri, S.A.N., Makworo, S.O.,
Okoko, E.N.K. and Kwach, D.J.K. 2000. Inorganic and organic fertiliser use:
potential for adoption at smallholder level in South West Kenya. Proceedings
of the 2nd Scientific Conference of the Soil Management and Legume Research
Network Projects. June 2000, Mombasa. Pp: 392-401.
http://www.kari.org/Legume_Project/Legume2Conf_2000/ (26.6.2007).
Mountains of the World. 2002. Why policies and instruments for mountain
development? Key issues, principles and strategies. Recognition of mountains
as important and specific areas of development. Mountains of the World.
Mountain Forests and Sustainable Development. Mountain Agenda. Prepared
for The World Summit on Sustainable Development. SDC. CDE. pp. 4-7.
Msikula S.N. 2003. The economics of improved agroforestry systems for income,
food security and biodiversity conservation in the East Usambara Mountains,
Tanzania. M.Sc. thesis, Management of Natural Resources for Sustainable
Agriculture. Sokoine University of Agriculture, Morogoro.119 p.
Muchiri, M.N. 2001. Yield and management of maize - Grevillea robusta agroforesty
system in Kenya. Ph.D. thesis, University of Joensuu, Faculty of Forestry.
42 p.
Mustalahti, I. H. 2007. Handling the stick: practices and impacts of participation in
forest management. Case study analyses of Finnish forestry assistance in
Tanzania, Mozambique, Laos and Vietnam. Ph.D. thesis, Danish Centre for
Forest, Landscape and Planning. Faculty of Life Sciences. University of
Copenhagen.
Myers, N.; Mittermeier, R.A.; Mittermeier, C.G; Da Fonseca, G.A.B.; and Kents, J.
2000. Biodiversity hotspots for conservation priorities. Nature 403: 853-858.
Myers, N. 1990. The biodiversity challenge: expanded hot-spots analysis.
Environmentalist 10: 243-256.
Nair, P.K.R. & Kutty, M.G. 2004. Cardamom (Elettaria cardamom) in Kerala, India.
In: Kusters, K. & Belcher, B. (eds.) Forest products, livelihoods and
conservation: case studies of non-timber forest product systems. Vol. 1 - Asia.
Bogor. CIFOR. Pp 133-150.

149
Nair, P.K.R. 1984. Soil Productivity Aspects of Agroforestry. Science & Practice of
Agroforestry 1. ICRAF, Nairobi. 85 p.
National Soil Service. 1989. Towards Sustainable land use in the East Usambara
Mountains. TARO Agricultural Research Institute, Mlingano, Tanga,
Tanzania. Site Evaluation Report S12. 18 p.
Newmark, W. D. 2002. Conserving Biodiversity in East African Forests. A Study of
the Eastern Arc Mountains. Ecological Studies 155. Springer-Verlag. Berlin
Heidelberg. 197 p.
Newmark, W.D. 1998. Forest area, fragmentation and loss in the Eastern Arc
Mountains: Implications for the conservation of biological diversity. Journal of
East African Natural History 87 (1-2): 29-36.
Newmark, W.D. 1992. Recommendations for Wildlife Corridors and the Extension
and Management of Forest Reserves in the East Usambara Mountains,
Tanzania. East Usambara Catchment Forest Project. Technical Paper No. 4.
Forestry and Beekeeping Division & Finnish Forest and Park Service. Vantaa,
Helsinki.
NFTA 1989. Gliricidia production and use. Nitrogen Fixing Tree Association. 44 p.
NFTA 1987. Gliricidia sepium (Jacq.) Walp.: Management and Improvement.
Proceedings of a Workshop held at CATIE, Turrialba, Costa Rica, June 1987.
Nitrogen Fixing Tree Association Special Publication 87-101. 255 p.
Nkonya, E., Pender, J., Jagger, P., Sserunkuuma, D., Kaizzi, C. and Ssali, H. 2004.
Strategies for sustainable land management and poverty reduction in Uganda.
Research Report 133. International Food Policy Research Institute.
Washington D.C. 136 p.
NRI (Natural Resources Institute) 2005. Biomass-Energy Toolbox. Growth and
development of short-rotation coppice crops.12 p.
www.nri.org/biomass/conference-papers/gliricidia_sepium-species -notes.pdf.
(21.11. 2006).
Nyagawa, S. & Kahamela, A. 2005. Awareness of the Arc. The Arc Journal, Issue 19.
Tanzania Forest Conservation Group. Pp.10.
Okigbo, B. N. 1990. Home Gardens in Tropical Africa. Pp. 21-40. In: Landauer, K &
Brazil, M (eds.). Tropical Home Gardens. UNU. Japan. 257 p.
Oksanen, T., Pajari, B., and Tuomasjukka, T. (eds.) 2003. Forests in Poverty
Reduction Strategies: Capturing the Potential. EFI Proceedings No. 47.
European Forest Institute. 206 p.

150
Owen, M. 1992. East Usambaras Conservation and Development Project. Forest
Product Survey Report. Bellerie Foundation. Nairobi.
Pandey, D.N. 2007. Multifunctional agroforestry systems in India. Current Science 92
(4): 455-463.
Pannell, D.J. 1999. Social and economic challenges in the development of complex
farming systems. Agroforestry Systems 45: 393-409.
Patil, B.P. 1989. Cut down fertilizer nitrogen need of rice by Gliricidia green manure.
Indian Farming 29: 34-35.
Peyre, A., Guidal, A., Wiersum, K.F. and Bongers, F.J.J.M. 2006. Dynamics of
homegarden structure and function in Kerala, India. Agroforestry Systems 66
(2): 101-115.
Pfliegner, K. & Burgess, N. 2005. What are the Eastern Arc Mountain forests worth to
Tanzania? The Arc Journal, Issue 19. Tanzania Forest Conservation Group.
Pp. 11-12.
Place, F., Adato, M., Hebinck, P. and Omosa, M. 2005. The Impact of Agroforestry-
Based Soil Fertility Replenishment Practices on the Poor in Western Kenya.
Research Report 142. International Food Policy Research Institute.
Washington DC.166 p.
Place, F. & Prudencio, Y-C. 2006. Policies for improved land management in
smallholder agriculture: The role for research in agroforestry and natural
resource management. In: Garrity, D., Okono, A., Grayson, M. and Parrott, S.
(eds.). World Agroforestry into the Future. World Agroforestry Centre.
Nairobi. Pp. 71-78.
Plan, 2007. Because I am a Girl. The State of the Worlds Girls 2007. Plan UK and
Plan International. London. 190 p.
Pretty, J., Thompson, J. and Hinchcliffe, F. 1996. Sustainable Agriculture: Impacts on
Food Production and Food Security. Gatekeeper Series. No. 60. IIED.
International Institute for Environment and Development.
Price, M. & Kohler, T. 2000. Why forests and mountains? Key issues in mountain
forestry. Multifunctionality - a global concept for forest management.
Mountains of the World. Mountain Forests and Sustainable Development.
Mountain Agenda. Prepared for the Commission on Sustainable Development
and its 2000 Spring Session. SDC, CDE. Pp. 4-7.
Purseglove, J.W., Brown, E.G., Green, C.L. and Robbins, S.R.J. 1981. Spices.
Longman Group Limited, New York. 719 p.

151
Quisumbing, A.R. 2003. Household Decisions, Gender, and Development: A
Synthesis of Recent Research. International Food Policy Research Institute,
Washington D.C.
Rao, M.R., Palada, M.C. and Becker, B. N. 2004. Medicinal and aromatic plants in
agroforestry systems. Agroforestry Systems 61: 107-122.
Ravindran, P.N. 2000. Black Pepper: Piper nigrum. CRC Press. 553 p.
Ravindran, P. N. & Madhusoodanan, K.J. (eds.) 2002. Cardamom: The Genus
Elettaria. Medicinal and Aromatic Plants - Industrial Profiles. London. Taylor
& Francis. 374 p.
Rodgers, W.A., Nabanyumya, R., Mupada E. and Persha, L. 2002. Community
conservation of closed forest biodiversity in East Africa: can it work?
Unasylva. No. 209. Forest Biological Diversity. Vol. 53. 2002/2.
http://www.fao.org/documents (25.6.2007).
Rodgers, W.A. 1993. The conservation of the forest resources of eastern Africa: past
influences, present practices and future needs. In: J.C. Lovett & S.K. Wasser
(eds.), Biogeography & ecology of the rainforests of eastern Africa, pp. 283-
330. Cambridge University Press.
Rosemeyer, M.E. 2001. Improving Agroecosystem Sustainability Using Organic
(Plant-Based) Mulch. Pp. 67-91. In: Gliessman, S.R.(ed.) Agroecosystem
Sustainability: Developing Practical Strategies. CRC Press.210 p.
Ruiz-Perez, M., Belcher, B., Achdiawan, R., Alexiades, M., Aubertin, C., Caballero,
J., Campbell, B., Clement, C., Cunningham, T., Fantini, A., de Foresta, H.,
Carcia Fernandez, C., Gautam, K.H., Hersch Martinez, P., de Jong, W.,
Kusters, K., Kutty, M.G., Lopez, C., Fu, M., Martinez Alfaro, M.A., Nair,
T.K.R., Ndoye, O., Ocampo, R., Rai, N. Ricker, M., Schreckenberg, K.,
Shackleton, S., Shanley, P., Sunderland, T. and Youn, Y. 2004. Report:
Markets drive the specialization strategies of forest peoples. Ecology and
Society. 9(2): Article 4. 23 p.
Risnen, S. 2001. Biodiversity on tropical agroforesty systems. On farm studies in
the East Usambaras, Tanzania. M.Sc. thesis, University of Helsinki, Dept. of
Forest Ecology, Viikki Tropical Resources Institute (VITRI). 111 pp.
Sah, S. 1996. Use of farmers knowledge to forecast areas of cardamom cultivation.
An application of a participatory land suitability analysis in East Usambaras,
Tanzania. ITC. 161 p.

152
Sanchez, P.A. & Benites, J.R. 1987. Low-input cropping for acid soils of the humid
tropics. Science 238: 1521-1527.
Sanchez, P.A., Buresh, R.J. and Leakey, R.R.B. 1997. Trees, soils, and food security.
Biological Transactions of the Royal Society 352: 949-961. London, Great
Britain.
Sanchez, P.A. 1987. Soil productivity and sustainability in agroforestry systems. Pp.
205-223. In: Steppler, H.A. & Nair, P.K.R. (eds.) Agroforestry: a decade of
development. ICRAF, Nairobi.
Sayer, J.A., Harcourt, C.S. and Collins, N.M. 1992. The Conservation Atlas of
Tropical Forests. Africa. MacMillan. London.
Sayer, J. 1991. Rain forest buffer zones: Guidelines for protected area managers.
IUCN, Cambridge.
Scherr, S. 1999. Soil Degradation: a threat to developing-country food security by
2020? Food, Agriculture and the Environment Discussion Paper 27.
International Food Policy Research Institute. Washington D.C. 63 p.
Schroeder, P. 1993. Agroforestry systems: integrated land use to store and conserve
carbon. Climate Res. 3: 53-60.
Sengoe, K.C. 1994. The assessment of impact of Cardamom production on the land
and forest resources in the Eastern Usambara Mountains, Muheza District,
Tanga Region, Tanzania. M.Sc. in Rural Resource Management. University of
Wales, Bangor. 113 p.
Sharma, H.R. 1998. Micro-level Evidence of Impacts on Agrobiodiversity Caused by
the Change from Subsistence to Cash Crop Farming in the Hills. In: Partap, T
and Sthapit, B (eds.) Managing Agrobiodiversity Farmers Changing
Perspectives and Institutional responses in the Hindu Kush-Himalyan Region.
ICIMOD (International Centre for Integrated Mountain Development)/IPGRI
(International Plant Genetic Resources Institute), Kathmandu, 1998, 439 p.
Sharma, R, Sharma, E. and Purohit, A.N. 1997. Cardamom, mandarin and nitrogen-
fixing trees in agroforestry systems in Indias Himalayan region. II Soil
nutrient dynamics. Agroforestry Systems 35: 255-268.
Shelutete, M.1996. Focus on Women. Report of a Consultancy on Conservation and
Women in East Usambara. Technical Paper 27. EUCAMP, Tanga. Forestry
and Beekeeping Division & Finnish Forest and Park service, Dar es Salaam &
Vantaa. 38 p.

153
Shimizu, J.Y., Spence, L.A., Martins, E.G. and Araujo, A.J. 2002. Genetic and
phenotypic variations in early growth performances of grevillea trees for use in
agroforestry systems. International Forestry Review 4(2): 128-132.
Simons, A.J. & Leakey, R.R.B. 2004. Tree domestication in tropical agroforestry.
Agroforestry Systems 61: 167-181.
Sitaubi, L.A. 1990. Foliar Nitrogen levels of some dry zone hardwood species in
Malawi. Pp.305-310. In: Prinsley R.T (ed). Agroforestry for sustainable
production. Economic Implications. Commonwealth Science Council. London.
Sivaraman, K., Kandiannan, K,. Peter, K. V. and Thankamani, C. K. 1999. Agronomy
of black pepper (Piper nigrum L.) a review. Journal of Spices and Aromatic
Crops 8 (1): 1-18.
Smith, L.C., Ramakrishnan, U., Ndaiye, A., Haddad, L. and Martorelle, R. 2003. The
Importance of Women's Status for Child Nutrition in Developing Countries.
Research Report. 131. International Food Policy Research Institute,
Washington D.C.
Social Capital Gateway. 2006. http://www.socialcapitalgateway.org/ (20.3.2006).
Soini, E. 2005. Changing livelihoods on the slopes of Mt. Kilimanjaro, Tanzania:
Challenges and opportunities in the Chagga homegarden system. Agroforestry
Systems 64: 157-167.
Steppler, H.A. & Nair, P.K.R. 1987. Agroforestry a decade of development. ICRAF.
335 p.
Stocking, M. & Perkin, S. 1992. Conservation-with-development: and application of
the concept in the Usambara Mountains, Tanzania. Transactions - Institute of
British Geographers. Oxford. 17: 337-349.
Sundet, G. 2005. The 1999 Land Act and Village Land Act. A technical analysis of
the practical implications of the Acts. Oxfam. 21 p.
http://www.oxfam.org.uk/resources/learning/landrights/downloads/1999_land_
act_and_village_land_act.rtf (26.11.2007).
Swai, E.S., Minja, F.N. and Zylsra, L. 1992. Dairy Development Programme in
Tanga, Tanzania. In: Kategile, J. A. & Mubi, S. (eds.). 1992. Future of
livestock industries in East and southern Africa. Proceedings of a workshop
held at Kadoma Ranch Hotel, Zimbabwe, 20-23 July 1992. ILCA
(International Livestock Centre for Africa), Addis Abeba. 227 p.
Swallow, B., Boffa, J-M. and Scherr, S.J. 2006a. The potential for agroforestry to
contribute to the conservation and enhancement of landscape biodiversity. In:

154
 Garrity, D., Okono, A., Grayson, M. and Parrott, S. (eds.). World Agroforestry
into the Future. World Agroforestry Centre. Nairobi. Pp. 95-102.
Swallow, B., Russell, D. and Fay, C. 2006b. Agroforestry and environmental
governance. In: Garrity, D., Okono, A., Grayson, M. and Parrott, S. (eds.).
World Agroforestry into the Future. World Agroforestry Centre. Nairobi. Pp.
85-94.
Tejwani, K.G. 1994. Agroforestry in India. Oxford & IBH Publishing Co. Ptv. Ltd.
Pp. 61-65.
TFCG (Tanzania Forest Conservation Group). 2007a. The Usambara Mountains:
Derema Corridor facilitating the payment of compensation, The Arc Journal
Issue 20. 27 p.
TFCG. (Tanzania Forest Conservation Group). 2007b. Community Forest
Conservation Network Unit. http://www.tfcg.org/docs/cfcn.htm (30.11.2007).
Thankamani, C. K., Kandiannan, K., Mathew, P.A. and Srinivasan, V. 2007.
Propagate Black Pepper through serpentine method. Spice India 20:17-20.
The Bugwood Network. 2001. The University of Georgia College of Agricultural and
Environmental Sciences and Warnell School of Forest Resources,
(http://www.easternarc.org/html/map.html (8.8.2001).
The World Bank Country Study. 2000. Agriculture in Tanzania Since 1986. Follower
or Leader of Growth? Government of the United Republic of Tanzania, The
World Bank, International Food Policy Research Institute. Washington D.C.
167 p.
The World Bank. 2005 'Social Capital', PovertyNet 15.9.2006
http://www.worldbank.org/poverty/scapital/whatsc.htm.
The World Bank. 1995. Towards Gender Equality: The role of Public Policy.
Education and Social Policy Department. 43 p.
Tomich, T.P., Kilby, P. and Johnston, B.F. 1995. Transforming Agrarian Economies.
Cornell University Press, Ithaca, New York.
Transparency International. 2006. http://www.infoplease.com/ipa/A0781359.html
(22.11.2007).
Uliwa, P. & D. Fischer. 2004. Assessment of Tanzanias Producer Organizations.
Experience and environment. USAID Tanzania, Economic Growth office.56 p.
Umapathy, G. & Kumar, A. 2000. The occurrence of arboreal mammals in the rain
forest fragments in the Anamalai Hills, South India. Biological Conservation
92: 311-319.

155
UN, Agenda 21. 1992. Chapter 13 - Managing Fragile Ecosystems: Sustainable
Mountain Development. Rio de Janeiro, on June 14, 1992.
UNCTAD. 2004. Trading opportunities for organic food products from developing
countries. Strengthening research and policy-making capacity on trade and
environment in developing countries. United Nations Conference on Trade and
Development. New York and Geneva. 162 p.
UNDP. 1995. Agroecology: Creating the Synergism for a Sustainable Agriculture.
UNDP Guidebook Series.87 p.
UNICEF. 2006. School Fees for Africa: Coming to grips with an elusive promise,
UNICEF Press Centre. Nairobi, 4 April 2006.
http://www.unicef.org/media/media_33184.html (26.7.2007).
URT (United Republic of Tanzania). 1997a. Agricultural and Livestock Policy.
Ministry of Agriculture and Cooperatives. United Republic of Tanzania. Dar
es Salaam. 161 p. http://www.tanzania.go.tz/ (26.11.2007).
URT (United Republic of Tanzania). 1997b. Cooperation Development Policy.
Ministry of Agriculture and Cooperatives. United Republic of Tanzania. Dar
es Salaam. 21 p. http://www.tanzania.go.tz/ (26.11.2007).
URT (United Republic of Tanzania). 2006. Derema Forest Corridor: East Usambara
Mountains. Resettlement Action Plan for farm plots displaced for biodiversity
conservation in the Derema Forest Corridor. Prepared for consideration of
compensation funding by the World Bank. Ministry of Natural Resources and
Tourism, Forestry and Beekeeping Division & Tanzania Forest Conservation
and Management Project (TFCMP). 56 p.
URT (United Republic of Tanzania). 2002a. First progress report on Tanzania without
Poverty. A plain language guide to: The United Republic of Tanzanias
Poverty Reduction Strategy Paper, Progress Report 2000/01 August 2001,
printed in Jan 2002. 48 p. http://www.hakikazi.org/ (2.3.2006).
URT (United Republic of Tanzania). 1998. National Forest Policy. Ministry of
Natural Resources and Tourism, The United Republic of Tanzania. Dar es
Salaam.
URT (United Republic of Tanzania). 1992. National Policy on Women in
Development in Tanzania. Ministry of Community Development, Women
Affairs and Children. United Republic of Tanzania. Dar es Salaam. 27 p.
http://www.tanzania.go.tz/ (26.11.2007).

156
URT (United Republic of Tanzania). 2002b. Prevention of Corruption Act. Chapter
329. United Republic of Tanzania. Dar es Salaam. http://www.tanzania.go.tz/
(26.11.2007).
URT (United Republic of Tanzania). 2004. Poverty Reduction Strategy: The Third
Progress Report 2002/03. The United Republic of Tanzania. Dar es Salaam.
71 p.
URT (United Republic of Tanzania). 1999a. The Land Act 1999. No. 4. United
Republic of Tanzania. Dar es Salaam.
URT (United Republic of Tanzania). 1999b. The Village Land Act 1999. No. 5.
United Republic of Tanzania. Dar es Salaam.
URT (United Republic of Tanzania). 2002c. The New Forest Act 2002. No. 14.
United Republic of Tanzania. Dar es Salaam.
URT (United Republic of Tanzania). 2003. Trade Policy. Trade Policy for a
Competitive Economy and Export-led Growth. Ministry of Industry and Trade.
United Republic of Tanzania. Dar es Salaam. 91 p. http://www.tanzania.go.tz/
(26.11.2007).
Urquijo J. 2004. Seguridad Alimentaria y Desarrollo Sostenible en Zonas Marginadas
de Guatemala: El papel de los bosques en la lucha contra la pobreza y la
inseguridad alimentaria. M.Sc. dissertation. Madrid, Espaa: Universidad
Politcnica de Madrid, Escuela Tcnica Superior de Ingenieros Agrnomos.
Veltheim, T. & Kijazi, M. 2002. Lessons learned on participatory forest management.
East Usambara Conservation Area Management Programme Technical Paper
61. FDB, Department of International Cooperation, Finland and Metshallitus
Consulting Oy. Tanga. 25 p.
Vihemki, H. 2001. Black pepper (Piper nigrum) cultivation in the East Usambaras,
Tanzania: Associated tree species and their effects on agroforestry system
management. M.Sc. thesis, University of Helsinki, Dept. of Forest Ecology,
Viikki Tropical Resources Institute (VITRI). 108 p.
Vihemki, H. 2005. Politics of Participatory Forest Conservation: Cases from the East
Usambara Mountains, Tanzania. The Journal of Transdisciplinary
Environmental Studies 4(2), Article 6. 16 p. http://www.journaltes.dk/
vol_4_no_2/NO6_HE_h.PDF (15.10.2006).
Wadsworth, F.H. 2004. Forests and Water. Pp 177- 178. In: Elevitch, C.R. (ed.). The
Overstory Book, Cultivating Connections with Trees, 2nd Edition. 548 p.

157
White, P. & Mustalahti, I. 2005. Finnish Forestry Assistance: Success Story or
Failure? Analyses of case studies from Sub-Saharan Africa and their possible
impacts on Poverty Reduction. Silva Carelica 48. University of Joensuu,
Faculty of Forestry. 105 p.
Wickama, J.M. & Mowo, J.G. 2001. Using local resources to improve soil fertility in
Tanzania. Managing Africas Soils No. 21. IIED.
Wilkinson, K & Elevitch, C.R. 2004. Nontimber Forest Products Enterprises. Pp. 265-
268. In: Elevitch, C.R. (ed.) The Overstory Book. Cultivating Connections
with Trees. Permanent Agriculture Resources. 548 p.
Woodcock, K.A. 1995. Indigenous knowledge and forest use: Two case studies from
the East Usambaras, Tanzania. Technical paper 22. East Usambara Catchment
Forest Project. Forestry and Beekeeping Division & Finnish Forest and Park
service, Dar es Salaam & Vantaa. 55 p.
Wurster, K. & Burgess, N. 2005. Fire. Is the fire problem getting better or worse in the
Eastern Arc Mountains? The Arc Journal, Issue 19. Tanzania Forest
Conservation Group. Pp. 18-19.
Yasu, H. 1999. The diffusion process of planting Grevillea robusta among rural
households in North-Central Tanzania. African Study Monographs 20(3):
119-145.
Young, A. 1989. Agroforestry for Soil Conservation. ICRAF. CABI. 276 p.
Young, A. 2004. Effects of Trees on Soils. Pp. 179-182. In: Elevitch, C.R. (ed.) the
Overstory Book, Cultivating Connections with Trees, 2nd Edition. 548 p.
Young, A. 1987. The potential of agroforestry for soil conservation. Part 11:
maintenance of fertility. ICRAF Working Paper 43. ICRAF, Nairobi. 135 p.
Zimmer, C. A. 2007. Biological Hot Spot in Africa, With New Species Still to
Discover. March 6, 2007. Science/The New York Times.

158
ANNEXES

Annex 1. Soil profiles

Profile number HP1 Mapping unit: HI.1 Agro-ecological zone: E2

Region: TANGA
District: Muheza
Map sheet no. map sheet Mnyuzi 130/1 and Hemagoma 110/3
Coordinates: S 05o 06.405, E 038o 36,801
Location: Emau hill, (about 3 Km from west of Amani Nature Reserve)
Elevation: 900 masl
Landform: Hilly; sloping. Slope 2-5% convex, > 100m, upper slope. Surface characteristics:
Erosion: none. Deposition: none surface stoniness: none. Flooding: none. Perched: none.
Natural drainage class: well drained. Natural vegetation: mostly agro forestry trees (Gliricidia
sepium, Grevillea robusta, Elettaria cardamomum, Piper nigrum). Land use: Agro forestry.
Forest clearing and burning is the major human influence in this place.

Described by G.J. Urassa and Kajiru on 5/8/2006.

Soils are very deep, well drained dark reddish brown sand clay loam on topsoils with
yellowish red and dark red sand clay in the sub soil.

Ap 0-10 cm: dark reddish brown (5YR3/3) dry, dark reddish brown (5YR3/3) moist, sand
clay loam; slightly hard dry, friable moist, slightly sticky and slightly plastic wet; moderate
very fine, fine and medium sub angular blocky; few very fine, common fine, many medium
pores; common very fine, medium fine roots; clear smooth boundary to

AB 10-14/15 cm: dark reddish brown (5YR3/4) dry, dark reddish brown (5YR/3/4) moist,
sand clay loam; slightly hard dry, friable moist, slightly sticky and slightly plastic wet; very
fine faint clay cutans, moderate very fine, fine and medium sub angular blocky; very few fine,
few fine, common fine and many medium pores; common very fine, medium fine roots; clear
smooth boundary to

Bt1 14/15 -60cm: yellowish reddish (5YR4/6) dry, dark reddish brown (5YR 4/6) moist,
sand clay slightly hard dry, friable moist, very sticky and very plastic wet; very fine faint
clay cutans, moderate fine and medium sub angular blocky; few very fine, common fine and
many medium pores; very few, medium and coarse roots; clear smooth boundary to

Bt2 60-170+ cm: dark red (2.5YR3/6) dry, dark red (2.5YR 3/6) moist, sand clay slightly
hard dry, friable moist, and very sticky wet and very plastic wet; very fine faint clay cutans,
moderate fine and medium sub angular blocky; few fine, common very fine and many fine
pores; very few, medium roots; clear smooth boundary

159
SOIL CLASSIFICATION: WRB (FAO 1988): Haplic Acrisols (Analytical data profile HP1)

Horizon Ap AB Bt1 Bt2

Depth (cm) 0-10 10-14/15 14/15-60 60-170+
Clay % 26 32 44 46
Silt % 16 18 10 8
Sand % 58 50 46 46
Texture SCL SCL SC SC
pH H2O (1:2.5) 5.2 4.9 4.8 4.8
pH KCL (1:2.5) 4.6 4.4 4.3 4.5
EC mS/cm 0.16 0.14 0.04 0.03
Organic C % 3.50 3.46 0.85 0.81
Total N % 0.33 0.25 0.08 0.07
C/N 12 13 10 12
Av. P, mg/kg 7.81 5.68 4.67 4.17
CEC, Cmols (+)/kg 19.06 15.84 9.13 9.22
Ex Ca, Cmols (+)/kg 4.72 3.08 0.28 0.32
Ex. Mg, Cmols (+)/kg 0.60 0.56 0.12 0.12
Ex. K, Cmols (+)/kg 0.12 0.10 0.04 0.02
Ex. Na, Cmols (+)/kg 0.09 0.10 0.07 0.15
B.S. % 50 49 45 50

Profile number MP1 Mapping unit: HI.2 Agro-ecological zone: E2

Described by G. J. Urassa and Kajiru on 5/8/2006.

Soils are very deep, well drained dark reddish brown sand clay loam on topsoils with
yellowish red and dark red clay in the sub soil.

Ap 0-9/10 cm: dark reddish brown (5YR3/3) dry, dark reddish brown (5YR/3/3) moist,
sand clay loam; slightly hard dry, friable moist, and slightly sticky wet and slightly plastic
wet; moderate very fine, fine and medium sub angular blocky; few fine, common fine, many
medium pores; common very fine, medium fine roots; clear smooth boundary to

AB 9/10-35 cm: yellowish reddish (5YR4/6) dry, reddish brown (5YR/4/4) moist, sand
clay; slightly hard dry, friable moist, and sticky and plastic wet; very fine faint clay cutans,
moderate fine and medium sub angular blocky; few fine, common fine and many medium
pores; common very fine, medium fine roots; clear smooth boundary to

Bt 35 -150+cm: Dark red (2.5YR3/6) dry, dark reddish brown (5YR 4/6) moist, sand clay
slightly hard dry, friable moist, and sticky wet and plastic wet; very fine faint clay cutans,
very fine and fine sub angular blocky; few fine, common fine and many medium pores; very
few, medium and coarse roots; clear smooth boundary

160
SOIL CLASSIFICATION: WRB (FAO 1988): Haplic Lixisols (Analytical data profile MP1)

Horizon Ap AB Bt1
Depth (cm) 0-9/10 9/10-35 35-150+
Clay % 30 42 44
Silt % 20 12 14
Sand % 50 46 42
Texture SCL SC C
pH H2O (1:2.5) 4.8 4.8 5.8
pH KCL (1:2.5) 4.1 4.2 4.5
EC mS/cm 0.17 0.07 0.16
Organic C % 3.65 2.53 1.37
Total N % 0.30 0.18 0.10
C/N 12 14 14
Av. P, mg/kg. 5.82 4.92 3.92
CEC, Cmols (+)/kg 14.7 9.98 9.48
Ex Ca, Cmols (+)/kg 2.12 0.68 0.20
Ex. Mg, Cmols (+)/kg 0.60 0.16 0.08
Ex. K, Cmols (+)/kg 0.11 0.06 0.03
Ex. Na, Cmols (+)/kg 0.12 0.13 0.65
B.S. % 44 52 65

Profile number MP2 Mapping unit: HI.3 Agro-ecological zone: E2

Described by G.J. Urassa and Kajiru on 5/8/2006.

Soils are shallow to moderately deep, well drained dark reddish brown sand clay loam on
topsoils with yellowish red and dark red and red sand clay in the sub soil.

Ap 0-10/11 cm: reddish brown (5YR4/4) dry, reddish brown (5YR4/3) moist, sand clay
loam slightly hard dry, friable moist, slightly sticky and slightly plastic wet; moderate very
fine, fine and medium sub angular blocky; few fine, common fine, many medium pores;
common very fine, medium fine, medium and coarse roots; clear smooth boundary to

At1 10/11-40/60 cm: Yellowish red (5YR4/6) dry, yellowish red (5YR4/6) moist, sand
clay to clay with gravel/stones; slightly hard dry, friable moist, and slightly sticky and plastic
wet; very fine faint clay cutans, moderate very fine, fine and medium sub angular blocky; few
fine, common fine and many medium pores; common very fine, many fine and medium roots;
clear smooth boundary to

Bt2 40/60-85cm: dark red (2.5YR3/6) dry, dark red (2.5YR 3/6) moist, sand clay to clay
with gravel/stones; slightly hard dry, friable moist, very sticky and very plastic wet; very fine
faint clay cutans, moderate very fine and fine sub angular blocky; few very fine, common fine
and many medium pores; very few, medium and coarse roots; clear smooth boundary to

B 85-150+ cm: red (2.5YR4/8) dry, red (2.5YR 4/6) moist, sand clay loam with
gravel/stones slightly hard dry, friable moist; sticky and plastic wet; moderate very fine and
fine sub angular blocky; very few very fine, and fine pores; very few, very fine roots; clear
smooth boundary

161
SOIL CLASSIFICATION: WRB (FAO 1988): Haplic Lixisols, lithic phase
(Analytical data profile MP2)

Horizon Ap AB Bt1 B
Depth (cm) 0- 10/11-40/60 40/60-85 85-150+
10/11
Clay % 32 40 46 32
Silt % 20 16 10 10
Sand % 48 44 44 56
Texture SCL SC-C SC-C SCL
(gravel/stones gravel/stones)
pH H2O (1:2.5) 4.9 5.1 5.3 5.5
pH KCL (1:2.5) 3.9 4.0 4.2 4.7
EC mS/cm 0.28 0.19 0.15 0.10
Organic C % 3.37 2.75 1.53 0.80
Total N % 0.35 0.25 0.12 0.05
C/N 15 15 15 17
Av. P, mg/kg. 6.27 4.52 4.40 3.20
CEC, Cmols (+)/kg 8.16 8.09 8.06 8.04
Ex Ca, Cmols (+)/kg 4.02 1.85 1.85 0.93
Ex. Mg, Cmols (+)/kg 0.08 0.20 0.04 0.08
Ex. K, Cmols (+)/kg 0.10 0.07 0.04 0.01
Ex. Na, Cmols (+)/kg 0.40 0.07 0.25 0.37
B.S. % 48 55 54 58

162
Annex II. Questionnaire for cardamom farmers in the East Usambara Mountains

(October-2005)

1. Name of the farmer: ___________________________ Gender: male / female

2. Name of the village: ______________
3. Distance of village from the border of Amani Nature Reserve? ________
4. How did you obtain your land?
a) Purchased
b) rented
c) inherited
d) share cropped
e) granted free
f) state forest lands
g) other, what? ___________________________________________

5. What is the size of your land in hectares or acres? __________

6a. Do you cultivate spices? Yes / No

6b. If you answered Yes to Question 6a, how big is the area under cultivation? ______

7a. Do you cultivate cardamom? Yes / No

7b. If you answered Yes to Question 7a, how big is the area under cultivation? _______

8a. Do you cultivate black pepper? Yes / No

8b. If you answered Yes to Question 8a, how many black pepper plants do you have?
__________

9a. Where do you cultivate cardamom?

a) in natural forest,
b) in home garden,
c) in natural forest and in home garden?

9b. If you cultivate cardamom in the home garden, which species do you intercrop it with?
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

9c. Do you recommend the intercropping to others? Yes / No

9d. If you answered Yes to Question 9c, give some reasons to support your answer?
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

9e. If you answered Yes to Question 9c, can you describe the nature of intercropping or
combination? (What is the density? Do you intercrop the cardamom between the tree rows,
etc)?
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

163
10a. Which is more profitable way to cultivate cardamom? a) in natural forest b) in home
garden?
10b. In your opinion, which is more sustainable way to cultivate cardamom:
a) in natural forest or
b) in home garden? Give reasons
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

11a. Which species are supporting black pepper?

___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

11b. Do you recommend the intercropping cardamom / black pepper with other crops and
trees to other farmers?
Yes / No

11c. If you answered Yes to Question 11b, give some reasons to support your
answer?_____________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

12. What is the annual production of a) cardamom? ____________ kg fresh / dry

b) black pepper? ____________ kg fresh / dry

13a. How significant is the cardamom income to your family?

a) necessary,
b) very important,
c) good extra income, but not necessary,
d) not so important

13b. How significant is black pepper income of the total income?

a) necessary,
b) very important,
c) good extra income, but not necessary,
d) not so important

14. How many years have you cultivated a) cardamom? ___________

b) black pepper? __________

15a. Why did you start to cultivate cardamom?

a) Traditional reason (e.g. father has cultivated also),
b) just wanted to try it,
c) because it is profitable,
d) other reason, what? _____________________________________________________

15b. Why did you start to cultivate black pepper?

a) Traditional reason (e.g. father has cultivated also),
b) just wanted to try it,
c) because it is profitable,
d) other reason, what? _____________________________________________________

164
16a. Do you use any
a) organic fertilizer (e.g. green manure, animal manure, compost, farm yard manure),
c) chemical fertilizer,
d) pesticides,
e) other, what? ___________________________________________________________

16b. If you answered No to Question 16a, can you give reasons in support of your answer?
a) The soil is good with nutrients
b) lack of money,
c) lack of knowledge,
d) lack of organic manure,
e) lack of manpower,
f) other reason, what?
______________________________________________________

16c. If you answered Yes to Question 16a, can you specify how often per year do you apply
it (a-d from 16a)? ______
16d. If you answered Yes to Question 16a, can you state your sources?
______________________________________

17. What was the price/kg of dried cardamom last year? _________ Sold where? _________
18. What was the price/kg of dried black pepper last year?_________ Sold where? ________

19a. Are you satisfied with the markets? Yes / No

19b. If you answered No to Question 19a, can you give reasons in support your answer?
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

20. If you think about the past years, has the cardamom production in your cultivation
a) increased
b) declined
c) maintained at the same level?

21. Have you noticed any harmful effects in the environment in the Usambaras? No /Yes

22. If you answered Yes to Question 21, what kind of effects have you noticed?
a) faster erosion,
b) decline in yields
c) changes in climate
d) slower rate of regeneration
e) other, what __________________________________________________________

23. Could cardamom cultivation be one of the reasons for the degradation of the environment?
Yes (choose a-e from the previous question) ____ / No

165
24. How have the conservation activities affected your cardamom cultivation?
a) not affected at all,
b) reduced the cultivation area,
c) forced to change the cultivation site
d) had to continue cultivation in the site without permission
e) Other effects, what?
_____________________________________________________________________
_____________________________________________________________________

25. What would you suggest for improving the spice farming and business, especially in the
case of cardamom in the East Usambara/Tanzania?
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

166