1 Spatial distribution of mangrove health index on three genera

2 dominated zones in Benoa Bay, Indonesia

3 I PUTU SUGIANA1,2, ANAK AGUNG EKA ANDIANI1, I GUSTI AYU ISTRI PRADNYANDARI DEWI1, I
4 WAYAN GEDE ASTAWA KARANG3,♥, ABD. RAHMAN AS-SYAKUR2,3, I WAYAN EKA DHARMAWAN4,♥♥
1
5 Bali Research Center. Jl. Gunung Talang VI-C No. 10A, Padangsambian, Denpasar 80117, Bali, Indonesia.
2
6 Environmental Research Center, Udayana University. Jl. PB. Sudirman, Denpasar, Bali 80234, Indonesia.
3
7 Department of Marine Science, Udayana University. Jl. Raya Kampus Unud, Jimbaran, Badung, Bali 80361. ♥email: [email protected]
4
8 Research Center for Oceanography, National Research and Innovation Agency, Jl. Pasir Putih I, No 1, North Jakarta 14430, Jakarta, Indonesia.
♥♥
9 email: [email protected]
10
11 Abstract. Mangrove distribution depends on environmental setting gradients forming variable forest structures from sea to landward.
12 The functionality of mangrove services for each zone is influenced by area, healthiness and forest structure. A spatial study of mangrove
13 stratification was conducted in Benoa Bay Bali which experienced high coastal development pressures during the last decade. The study
14 aimed to determine mangrove health index distribution (MHI) and IVI? along three genera-dominated zones. Random forest method
15 was applied to classify the areas of distributed genera along the Bay. Forest structure was assessed on with 54 quadratic plots
16 representing each zone condition. Finally, the spatial distribution of the MHI was estimated using a high accuracy model and
17 distinguished into three healthiness categories. The forest was mainly composed of Rhizophora and Sonneratia-dominated zones at
18 approximately 51% and 45% of the total mangrove area, respectively. A narrow Bruguiera dominated zone was found in the most
19 landward area, only covering 4% of the mangrove area. Overall accuracy and a kappa coefficient indicated high accuracy of forest
20 classification at 97% and 0.94 respectively. We found that 47.44% of mangrove areas could be classified into the highest healthiness
21 category indicating that the mangrove forest in Benoa Bay is in the excellent condition. The Rhizophora zone made a significant
22 contribution to the entire forest state since the excellent category coverage in this zone was approximately 73.80%. The mean value of
23 MHI in Rhizophora zone pixels was more than 67%. Bruguiera and Sonneratia zones had a majority of areas in the moderate
24 healthiness category.

25 Keywords: Community structure, mangrove forest, mangrove health index, zonation

26 INTRODUCTION

27 Mangrove species is a unique coastal ecosystem livinglive in tidal environments and is are highly adapted to a wide
28 range of salinity gradients and anoxic substrates (Srikanth et al. 2015; Naido, 2016). Its Their distribution depends on
29 various estuarine profiles, i.e., coastal geomorphological type (Darmadi et al. 2012), salinity gradient (Ball, 1998; Yang et
30 al. 2013) and nutrient content (Lovelock et al. 2009; Frederika et al. 2021). The salinity gradient formed through tidal and
31 freshwater mixing in estuarine areas has contributed to plant diversity due to an optimum salinity range preference among
32 mangrove species (Barik et al. 2017). The gradients of environmental conditions on mangrove areas are responsible for
33 forest structure from the sea to the landward area in a distribution perpendicular to the coastline (Sreelekshmi et al. 2018).
34 Forest stratification describes mangrove distribution patterns through environmental gradients where salt-tolerant species
35 would be found along the seaward area. On the other hand, the landward zone is mainly occupied by mangrove plants with
36 a low salinity tolerance (Sugiana et al. 2021).
37 The mangrove ecosystem has significant roles in coastal areas in terms of ecological, physical and socio-economic
38 perspectives. Ecologically, the ecosystem provides preferred habitat for living marine and terrestrial organisms both as
39 feeding and nursery grounds. Mangrove forests can reduce damage to coastal areas caused by strong waves, hurricanes and
40 tsunamis (Setiawan, 2013; Karimah, 2017; Adilah, et al. 2018). Coastal mangrove communities around small islands are
41 important to cope with sea-level rise effects and control saltwater intrusion along coastal areas (Gilman et al. 2006;
42 Wilson, 2017). Mangrove ecosystems bring socio-economic benefits to local people forprovide ecosystem goods such as
43 wood, food and medicinal compounds (Walters et al. 2008; Mojiol et al. 2016). On the other hand, the aesthetic value of
44 mangrove forests has potential for ecotourism purposes which support community livelihood (Friess, 2017; Singgalen,
45 2020). In terms of climate change, mangrove has been considered one of the most effective carbon sequestration areas for
46 reducing greenhouse gasses in the atmosphere and mitigating global warming (Heriyanto and Subiandono, 2016; Maher et
47 al. 2018).
48 Several studies demonstrated that forest area, healthiness and complexity of their zones influences the magnitude of
49 ecosystem function and services delivered to adjacent communities (Brander et al. 2012; Carugati et al. 2018; Zhang et al.
50 2022). Indonesia has the largest mangrove forests in the world sincewith 22.4% of the global mangrove area exists in this
51 archipelagic country (Giri et al. 2011). ) and It has been considered that the Indonesian mangrove forests contribute has a
52 significant contribution to regulating global climate and delivers high-value ecosystem services. However, the forest area

1
53 has decreased massively by 18.209 ha. yr-1in the two last decades, predominantly caused by land-use change for
54 aquacultures and oil palm cultivation (Richard and Friess, 2012; Kusmana, 2014; Arifanti et al. 2019; Tosiani et al. 2020;
55 Arifanti et al. 2021). Mangrove degradation in Indonesia has resulted in declines in forest area and degraded forest quality
56 related to standing size, distribution, and coverage (source?). A previous study developed theThe mangrove health index
57 (MHI) is used to assess Indonesian mangrove quality through three major parameters related to forest qualitythat are…?
58 canopy coverage, diameter and sapling density (Dharmawan et al. 2020b). The index was successfully modelled in spatial
59 scale and highly correlated to a combination of several mangrove indices based on remote sensing analysis, i.e., NBR,
60 SIPI, ARVI and GCI (Nurdiansah and Dharmawan, 2021). Since it has been newly developed, MHI functionality needs to
61 be studied tested comprehensively.
62 This study aimed to determine the spatial distribution of MHI along distinguished forest zones focused on Benoa Bay,
63 Bali. The wetland mangrove forest in the Bay experienced a massive threat from coastal development activities in the last
64 decade even though it is mostly included in a national forest management area; TAHURA Ngurah Rai Grand Forest Park
65 (TAHURA) Bali Ministry of Forestry and Environment. This study assessed the range of MHI values for each zone and
66 described a holistic interpretation of mangrove healthiness. Community structure for each zone was included to clarify
67 species composition and stand structure for each zone.

68 MATERIALS AND METHODS

69 Study description
70 This study area was located in the mangrove forest around in the semi-enclosed Benoa Bay, Bali (8o42’16.2”S-
71 8 47’48.1”S, 115o14’50.8”-115o10’28.1”). Mangrove forest in the Bay is regulated asin the Ngurah Rai Grand Forest Park
o

72 (TAHURA) by the Ministry of Forestry and Environment, consisting of a protection and utilization area of about 1.132,00
73 ha (Figure 1). The rest is a settlement, open land and waterbody, approximately 16.27 ha, 49.35 ha and 144.01 ha
74 consecutively (BPKH Wilayah VII Denpasar, 2013). Previous studies by Wiyanto and Faiqoh, (2015); Andiani et al.
75 (2021); Dewi et al. (2021); Sugiana et al. (2021) found that there wereidentified three mangrove genera which dominated
76 in this area such as : Sonneratia, Rhizophora and Bruguiera. Generally, mangrove in Benoa Bay had muddy to gravel soil
77 characteristics and was categorized as a semi-enclosed Bay. Rhizophora and Bruguiera dominated mangrove forests
78 mostly hadare associated with a muddy substrate, with low salinity levels < 25‰, and anoxic conditions. In contrast,
79 Sonneratia dominated mangrove forests hadis associated with muddy sand, with high salinity levels > 29 ‰, and
80 oxygenated conditions (Sugiana et al. 2021).

81 Forest classification determination

82 Mangrove forest was classified into threeby the dominant genera; Rhizophora, Sonneratia, and Bruguiera based on
83 previous studies (Wiyanto and Faiqoh, 2015; Andiani et al. 2021; Dewi et al. 2021; Sugiana et al. 2021). The classification
84 analysis was applied using the Random Forest approach in the Google Earth Engine platform (Ghorbanian et al. 2021). A
85 cloud-free Sentinel 2A-SR image was extracted by aggregating collected images from September 2020 to April 2021 and
86 reducing them using the median value for each bands. Study The study area was bordered through the national mangrove
87 polygons specifically in Benoa Bay area which are managed by the Ministry of Forestry and Environment. Training and
88 validation data for each zone were defined through two approaches; by conducting a ground-truth survey on each pixel
89 1886 pixels as training data and 1218 pixels for validation) and interpretation from a high-resolution image on Google
90 Earth and Planet Lab. Producer, consumer, and overall accuracy were analyzed to detect classification performances as
91 well as a kappa coefficient. The number of training data was frequently added to perform the best classification result
92 which accurately separated mangrove zones in Benoa Bay. Finally, this study involved as many as 1886 pixels as training
93 data and 1218 pixels for validation.

94 Mangrove community structure measurement

95 Density, dominance and frequency for each species and stands level determined from measured DBH value were used
96 to calculate relative values and then the important value index (IVI). Fifty-four quadratic-10m x 10m-plots were randomly
97 distributed to represent mangrove structure in each genera-dominated zone, i.e. 25 plots in the Sonneratia zone, 22 plots in
98 the Rhizophora zone and seven plots of Bruguiera zone (Figure 1). MangroveS species identification was applied for each
99 measured stand based on Tomlinson (2016) and Kitamura et al. (1997). The DBH (diameter at breast height) of all stands
100 was measured on each plot and then classified into three (DBH≥5 cm) and sapling category (DBH<5 cm). Species
101 identification was applied for each measured stand based on Tomlinson (2016) and Kitamura et al. (1997). Density,
102 dominance and frequency for each species and stands level determined from measured DBH value and presence were
103 extracted for calculating relative values and then the important value index (IVI). Field data collection was run in the
104 MonMang 2.0 app to efficiently record and process the data. Mangrove community height for each zone was estimated
105 using a trigonometric approach by measuring several angles of the top-most forest canopy using a protractor at a 10-meter
106 distance. Hemispherical photography was applied to estimate mangrove canopy coverage by taking 9 photos vertically
107 upwards for each plots in each zone (Dharmawan, 2020). A smartphone with camera resolution of at least 3MP captured

2
108 squared output hemisphere photographs from scattered positions on each plot. Each picture was analyzed using ImageJ
109 software to count pixel numbers. Canopy coverage percentage was calculated by comparing the pixel number of
110 vegetation objects to the total pixels on each photograph.
111 Field data collection was entered? in the MonMang 2.0 app to efficiently record and process the data. The Shapiro-
112 Wilk normality test was applied to univariate data from each zone, i.e., sapling and tree density, canopy coverage
113 percentage, estimated tree height, and measured stem diameter. Analysis of variance (ANOVA) was generated for
114 normally distributed parameters followed by the Tukey HSD test using genera-dominated zone as factors. Statistical
115 analysis was proceeded by an open-source application, R-studio (Lubis, 2021).

116
117
118 Figure 1. Distribution of forest assessment plots representing each mangrove zone i.e. Sonneratia (blue dots), Rhizophora (red dots) and
119 Bruguiera (yellow dots).
120

121 Spatial distribution of MHI along forest zones

122 A cloud-free Sentinel-2A imagery was used to delineate forest classification in each classified zone coverage. The
123 spatial distribution of MHI was analyzed based on a high accuracy (R 2 = 0.831) model developed by Nurdiansah and
124 Dharmawan (2021), which combined four selected vegetation indices, i.e., NBR (Normalized Burn Ratio); GCI (Green
125 Chlorophyll Index); SIPI (Structure Insensitive Pigment Index) and ARVI (Atmospherically Resistant Vegetation Index)
126 and utilized several Sentinel bands Near Infrared (NIR); Shortwave Infrared - (SWIR), red green and blue bands.
127 MHI = 102.12*NBR – 4.64*GCI +178.15*SIPI + 159.53*ARVI - 252.39
128 where:
129 NBR = (NIR - SWIR) / (NIR + SWIR)
130 GCI = (NIR/green) - 1
131 SIPI = (NIR - blue) / (NIR - red)
132 ARVI = (NIR - 2.red + blue) / (NIR + 2.red + blue)
133
134 The model successfully assessed MHI spatial distribution through cloud-free satellite images. Min-max standardization
135 was applied to standardized MHI range values. Mangrove healthiness was classified into three basic categories based on
136 the MHI range, i.e., poor (0-33%), moderate (33%-66%) and excellent (66%-100%) (Dharmawan and Ulumuddin, 2021).
137 A descriptive analysis was conducted along pixels on each zone to determine several values: mean, range (minimum and
138 maximum), standard deviation, median, and first and third quartile.

139 RESULTS AND DISCUSSION

140 Mangrove classification

141 Mangrove forest in Benoa Bay was clearly distinguished into three genera-dominated zones, which were Sonneratia,
142 Rhizophora and Bruguiera (Figure 2). These genera had different distribution preferences along the Bay. Sonneratia-
143 dominated zones tended to grow in the seaward area, while the Bruguiera forest was mainly found near the land and
144 occupied a smaller area. This result was supported by a previous study that found R. apiculata, S. alba, and B.
145 gymnorrhiza were the three major dominant plants in Benoa Bay mangroves (Sundra, 2016). Martiningsih et al. (2015)

3
146 found only two main dominant species, R. apiculata and S. alba, in Benoa Bay Bali. They did not find Bruguiera
147 domination in their perpendicular transect lines. The Bruguiera dominated zone was distributed along a limited and narrow
148 area in the landward zone, which was difficult to cover using a limited line transect. Another study also revealed that R.
149 apiculata and S. alba were the cosmopolitan species in this area, while Bruguiera mixed with those species (Prinasti et al.
150 2020). The Sonneratia-dominated zone was mostly composed of monospecific stands of S. alba, which was considered a
151 native plant due to a larger tree diameter size of more than 15 cm (Dewi et al. 2021).
152 In this study, Rhizophora had a majority distribution in the middle to the landward area of mangrove forest both as a
153 growing rehabilitated mangrove and as native stands. Rhizophora species were mostly selected and planted 30 years ago
154 from 1992 to 1999 through The Development of Sustainable Mangrove Management Project established by a collaboration
155 between Japan International Cooperatian Agency (JICA) and Indonesia’s Ministry of Forestry to restore mangroves in
156 TAHURA forest in Benoa Bay from abandoned ponds (JICA, 1999). This rehabilitation program was successful and
157 provided valuable data for future rehabilitation activities in Indonesia.
158

159
160 Figure 2. Spatial distribution of mangrove genera-dominated zones in Benoa Bay, Bali
161
162
163 Forest classification results had a high overall accuracy and kappa coefficient at 97% and 0.94, respectively (Table 1).
164 Those values indicated that classification based on training data through Random Forest was highly accurate in
165 determining the genera-dominated zones' distribution in Benoa Bay. McHugh (2012) explained that a kappa statistic value
166 in the highest range (0.9 – 1.0) indicated a high correlation between training and validation data variation. A similar
167 technique has been applied by several studies to classify mangrove forests. Jhonnerie et al. (2015) implemented a random
168 forest technique with 81% and 0.76 respective overall accuracy and kappa statistics to define mangrove forests using
169 Landsat 5 TM and Alos Palsar Imageries. Another study declared a better accuracy at 95.89% with a kappa coefficient of
170 0.95 based on Worldview imagery (Jiang et al. 2021). Sentinel imagery had a better spatial resolution; hence classification
171 resulted in better accuracy. Based on Sentinel satellite imagery, mangrove ecosystem mapping in Qeshm Island, Iran, also
172 had high overall accuracy and kappa coefficient at 93.23% and 0.92, respectively (Ghorbanian et al. 2021). Sentinel was
173 fitted to classify mangroves among species levels using random forest classification (Behera et al. 2021). However, the
174 number and distribution of training data significantly affected classification accuracy and error (Millard and Richardson,
175 2015).
176
177
178 Table 1. Accuracy assessment for forest classification analysis based on field data validation for each mangrove zone
179
Accuracy Tests Zone
Sonneratia Rhizophora Bruguiera
Producer Accuracy 0.99 0.94 0.93
Consumer Accuracy 0.98 0.95 0.86
Overall Accuracy 97%

4
 Kappa Coefficient 0.94
180
181
182 According to area calculation analysis for each zone, Rhizophora forest had the most extensive mangrove area with
183 approximately 603.56 ha or about 51% of the forest area in Benoa Bay (Figure 3). Similarly, the Sonneratia forest
184 occupied an extensive area mainly along the seaward zone, with about 45% of the total mangrove forest. Only 41.65 ha or
185 about 4% of mangrove forest was defined as a Bruguiera-dominated forest.
186 In this study, Rhizophora had a majority distribution in the middle to the landward area of mangrove forest both as a
187 growing rehabilitated mangrove and as native stands. Rhizophora species were mostly selected and planted 30 years ago
188 from 1992 to 1999 through The Development of Sustainable Mangrove Management Project established by a collaboration
189 between Japan International Cooperatian Agency (JICA) and Indonesia’s Ministry of Forestry to restore mangroves in
190 TAHURA forest in Benoa Bay from abandoned ponds (JICA, 1999).There previously have been only limited studies to
191 assess mangrove area coverage for each zone in the Bay. The contribution of past rehabilitation programs was significant
192 in the forest zones area in Benoa Bay. Approximately 200 ha of abandoned ponds was rehabilitated using Rhizophora
193 seedlings (JICA, 1999). A temporal analysis found that the mangrove area declined to 96.21 ha during a five-year
194 observation from 2015 to 2020 in Benoa Bay (Nurhaliza et al. 2021).
195 The distribution of each mangrove genera was influenced by environmental parameters gradient, and the area changes
196 for each zone were driven mainly by coastal development activities. Sonneratia forest area has experienced a gradual
197 decline due to high sedimentation rates, which were triggered by highway and harbour construction through reclamation
198 activities. Andika et al. (2015) revealed that sedimentation area increased significantly by 1966.14 ha during the initial
199 stage of highway construction in 2015. Consequently, pneumatophores of Sonneratia were buried by sediment deposition;
200 hence oxygen uptake was inhibited. This case drove a dieback phenomenon in the mangrove forest (Nardin et al. 2021).
201

202
203
204 Figure 3. Area of each genera-dominated zone (S=Sonneratia; R=Rhizophora and B = Bruguiera) in Benoa Bay’s mangroves
205

206 Mangrove community structure

207 The species composition varied among mangrove zones in Benoa Bay. Sonneratia-dominated zones had the lowest
208 species diversity compared to other zones, with only four and two species on tree and sapling levels, respectively (Table
209 2). A previous study inIn other Sonneratia-dominated forest in Southeast Sulawesi, three to had more diversity; seven
210 species were identified (Rahman et al. 2014). Other comparison studies in a similar mangrove zone in in Southeast
211 Sulawesi Riau, Center Sulawesi and West Java provinces only found three species (Rahman et al. 2014; Ramdani et al.
212 2015; Schaduw, 2020; Siregar et al. 2022). The seaward zone of mangrove forest on a Papuan small island was only
213 composed of two species which were dominated by S. alba, and another species was R. stylosa (Nurdiansah and
214 Dharmawan, 2021). Sonneratia. alba had a majority distribution in Sonneratia-dominated this zone (IVI: 266.35%),
215 followed by three other Rhizophora species, i.e., R. apiculata, R. mucronata and R. stylosa, at tree level and only R.
216 apiculata at sapling level. It was reflected in the IVI of S. alba in this zone at 266.35% (Figure 4). In oceanic mangroves or
217 mangrove zones which closed to the sea, S. alba had fully dominated with a maximum IVI value (Dharmawan and
218 Pramudji, 2019).
219
220
221 Table 2. Mangrove species composition on each stand category alongin three genera-dominated zones in Benoa Bay, Bali
Zone
Mangrove species Sonneratia Rhizophora Bruguiera
Tree Sapling Tree Sapling Tree Sapling
S. alba + + + + + -
R. apiculata + + + + + +

5
 R. mucronata + - + + + +
R. stylosa + - + + - -
L. racemosa - - + + - -
B. gymnorrhiza - - + + + +
X. granatum - - + + - -
A. marina - - - - + +
Number of species 4 2 7 7 5 4
222 Note: (+): present; (-): absent
223
224 In contrast, theThe Rhizophora-dominated zone was the most diverse mangrove in this study area. There were with
225 seven species found both at tree and sapling levels (Table 2). Mangrove species R. apiculata, R. mucronata and R. stylosa
226 were the main component of this zone based on their IVI value xxx%, xxx% and xxx% respectively, followed by S. alba,
227 L. racemosa, X. granatum and B. gymnorrhiza (Table 1). Other research conducted on Indonesian mangroves in Gorontalo
228 (Kasim et al. 2019), North Maluku (Serosero et al. 2020), East Kalimantan (Edwin et al. 2021), and West Lombok
229 (Sukuryadi et al. 2021) has found a greater number of species inventory which were atrecorded 9, 10, 15 and 12 species,
230 respectively. However, species composition in Benoa Bay’s Rhizophora zone had a higher species number than Mare
231 Island (Akbar et al. 2016), Jakarta Bay (Sari et al. 2019), Situbondo, East Java (Hariyanto et al. 2019), Mantehage-Paniki
232 Islands, North Sulawesi (Opa et al. 2019) and Cirebon (Purwanto et al. 2022). Several mangrove species were co-
233 dominantly identified with Rhizophora, such as A. alba, B. hainesii, B. parviflora, E. agallocha, H. littoralis, L. littorea, S.
234 hydrophyllacea, S. ovata and X. moluccensis, as found in Riau (Prianto et al. 2006), and A. officinalis, B. cylindrica, B.
235 sexangula, S. caseolaris in North Sumatra, Indonesia (Harefa et al. 2022). The importance value index of the Rhizophora
236 genus in this zone was 217.23%, while the cumulative IVI of other species was only 82.77% (Figure 4). Additional
237 research conducted on Indonesian mangroves in North Gorontalo (Usman et al. 2013); North Sumatera (Hotden et al.
238 2014); West Bangka (Rosalina and Rombe, 2021) and Benoa Bay (Dewi et al. 2021) found that the Rhizophora genus has
239 a higher value compared to other species. The IVI implies that the Rhizophora genus had a significant role in the research
240 location.
241 A narrow area of theThe Bruguiera-dominated zone consisted of five mangrove species in the tree stand category and
242 four species on sapling levels (Table 2). B. gymnorrhiza was dominated in this zone at IVI 160.13%, followed by S. alba,
243 which was only found at the tree level, while R. apiculata, R. mucronata and A. marina were observed in both stand levels
244 (Table 2, Figure 4). The number of mangrove species found in this zone was more diverse than Sorong, Papua (Yanti et al.
245 2021), Okinawa Island, Japan (Kamruzzaman et al. 2017), and Kosrae Island, Federated States of Micronesia (Krauss and
246 Allen, 2003) which only found three mangrove species. B. gymnorrhiza was found to be monospecific on a small Papuan
247 island (Dharmawan and Pramudji, 2019).
248

249
250
251 Figure 4. Importance value index (IVI) of each mangrove species along zones (S=Sonneratia; R=Rhizophora and B = Bruguiera) in
252 Benoa Bay, Bali
253
254
255 Overall, there were eight species of mangrove found in the Benoa Bay along all zones. Previous studies in the Bay
256 have reported other species such as Aegiceras floridum, Avicennia officinalis, A. rumphiana, B. cylindrica, B. sexangula,
257 Ceriops tagal and S. caseolaris (Wiyanto and Faiqoh, 2015; Prinasti et al. 2020). The distribution of mangrove species
258 along forest zones is mainly determined by variation in environmental characteristics. Mangrove substrates in Benoa Bay
259 are mainly dominated by mud in the land zone to the middle area and sandy soils in the sea zone. Prinasti et al. (2020) and
260 Imamsyah et al. (2021) reported that the soil types in Benoa Bay ranged from fine sand to gravel with dominant coarse
261 sand. Sugiana et al. (2021) found that the mangrove landward and middle zone in Benoa Bay had a predominant muddy
262 substrate and significantly lower salinity levels than seaward sites. Those studies have confirmed the mangrove species

6
263 distribution pattern in the Bay. S. alba tends to grow in sandy, rocky or coral rubble substrates usually located in higher
264 salinity areas (Noor et al. 2006; Dharmawan and Pramudji, 2020; Nurdiansah and Dharmawan, 2021). This species also
265 tolerates high salinity levels (Pillai and Harilal, 2016), and even sandy to gravel mud soil (Bessie et al. 2013; Lewerissa et
266 al. 2018). S. alba has been considered a pioneer species that forms the main layer of the seaward side of the forest where
267 there is sandy substrate (Goltenboth et al. 2006; Jenoh et al. 2016; Mughofar et al. 2018). Forests dominated by this
268 species contained less diversity (Rahman et al. 2014; Supriadi et al. 2015; Wiyanto and Faiqoh, 2015; Andiani et al. 2021;
269 Dewi et al. 2021; Sugiana et al. 2021). Rhizophora and Bruguiera were frequently found in muddy substrates and lower
270 salinity areas, even though Bruguiera had a preference for less submerged substrates (Primantara et al. 2019; Khairunnisa
271 et al. 2020).
272
273 In terms of stand distribution, the most significant tree stand density was found in the Rhizophora-dominated zone at
274 an average of 4750 ± 2867 stands.ha -1, while Sonneratia forest had the lowest tree density at 2332 ±786 stands.ha -1 (Table
275 3). The number of trees stands in the Sonneratia zone had a significant difference (ANOVA: p < 0.05) compared to
276 Rhizophora and Bruguiera dominated zones. among zones. Similar conditions were found in East Kalimantan (Ardiansyah
277 et al. 2012) and Middle Maluku (Nanulaitta et al. 2019), where S. alba dominated zone had lower tree and sapling density
278 than R. apiculata, also R. mucronata in Benoa Bay (Andiani et al. 2021; Dewi et al. 2021; Sugiana et al. 2021) and B.
279 gymnorrhiza in Biak Regency, Papua (Rambu et al. 2019). Tree density was highly correlated to the stand diameter
280 average. The high density of mangroves causes competition in getting nutrients and solar radiation, which disrupts the
281 lateral growth of mangroves (Syukri et al. 2018). This study found that the Sonneratia-dominated zone had the largest
282 average trunk diameter at 14.12±3.15 cm, while the Rhizophora and Bruguiera zones were found at 9.94±2.39 cm and
283 7.86±1.91 cm, respectively. Dharmawan et al. (2020b) explained that less-disturbed forests dominated by S. alba tended to
284 have a larger trunk size due to the sparser distribution in certain areas. Compared to other studies, Sonneratia stands
285 diameters in Benoa Bay seems to be lower than Belitung Island with 15.46±1.92 cm of diameter average (Suyarso et al.
286 2018) and Ternate with 16.63±1.73 cm of mangrove diameter average (Arbi et al. 2018). Compared to similar research in
287 West Halmahera, North Maluku, which found similar species to Benoa Bay, tree density in Sonneratia, Rhizophora and
288 Bruguiera mangroves were lower than Benoa Bay at 1200 stands/ha, 3600 stands/ha and 2000 stands/ha respectively (Gabi
289 et al. 2021). A dense mangrove forest tends to have a small stands diameter. The result was supported by Dharmawan et al.
290 (2020a), who conducted research in Biak-Numfor Regency and Siringoringo et al. (2017) in North Nias Regency; while S.
291 alba dominated the mangroves forest, it mostly had low density with a large diameter.
292
293
294 Table 3. Mean value of tree and sapling density (stands.ha-1), diameter (cm), height (m) and canopy coverage (%) on each mangrove
295 zones in Benoa Bay, Bali
296
Zone
Parameter
Sonneratia Rhizophora Bruguiera
-1
Tree Density (stands.ha ) 2332 ±786ᵃ 4750 ± 2867ᵇ 5157 ± 1615ᶜ
-1
Sapling Density (stands.ha ) 356 ± 324ᵃ 700 ± 1120ᵃ 1514 ± 884ᵇ
Diameter (cm) 14.12 ± 3.15ᵃ 9.94 ± 2.39ᵇ 7.86 ± 1.91ᶜ
Height (m) 9.14 ± 2.29ᵃ 10.57 ± 2.39ᵃ 10.55 ± 3.00ᵃ
% Canopy Cover 53.00 ± 12.72ᵃ 75.33 ± 6.23ᵇ 78.08 ± 6.51ᵇ
297 Notes: Letters a, b, and c represented the results of the one-way ANOVA followed by the Tukey HSD test which different letters given
298 had indicated the significant difference (p < 0.05) result of each parameter, among mangrove zones
299
300
301 In contrast with tree density, sapling stands in the Sonneratia mangrove zone were found at a lower density than in
302 other zones at 356 ± 324 stands.ha-1. However, the average had no significant difference to Rhizophora zones (ANOVA: p
303 > 0.05). The highest sapling density was found in the Bruguiera zone with 1514 ± 883 stands.ha -1, significantly different
304 from other zones (ANOVA: p < 0.05). In pristine or non-rehabilitated mangrove forest mangroves, sapling density in
305 Rhizophora was relatively limited since tree stands dominated space (Dharmawan and Widyastuti, 2017) and nutrient
306 competition (Koch, 1997). In contrast, degraded mangrove forests have been found to have many saplings; for instance,
307 from 2010 to 2017, the mangrove forest in East Java experienced a massive decrease caused by illegal logging of 39,756.2
308 ha (Rudianto and Bengen, 2020). A previous study on Belitung Island found a variable comparison of sapling density in
309 each genera-dominated zone (Akhrianti et al. 2019). The forest had no saplings in Sonneratia forest, while Bruguiera had
310 a limited number at only 320 stands/ha, and a frequent harvested Rhizophora forest was found to have sapling density
311 density of up to 3600 stands/ha compared to Benoa Bay. The low density of mangrove sapling in the Sonneratia zone was
312 caused by the allelopathic secretion ability of S. alba, which inhibits lower stands growth (Xin et al. 2013; Lin et al. 2015;
313 Zhang et al. 2018).

7
314 Mangrove canopy coverage in Benoa Bay ranged from 52.99%±12.72% to 78.08% ± 6.51%. Sonneratia-dominated
315 forests had the lowest canopy coverage, which was significantly different (p < 0.05) from both Rhizophora and Bruguiera
316 zones (Table 3).. The average value of canopy cover from the Sonneratia mangrove zone was 52.99%±12.72%, while
317 Rhizophora and Bruguiera zones were more than 75%. The species S. alba, which dominated in the Sonneratia zone,
318 tended to have a lower coverage due to sparse leaf density and lower stand distribution. At the same time, Rhizophora and
319 Bruguiera were characterized by a canonical shape of canopy with individual dense leaves (Dharmawan, 2020). A
320 previous study also found that the Sonneratia forest had a lower canopy coverage both in Benoa Bay (Andiani et al. 2021;
321 Dewi et al. 2021; Sugiana et al. 2021) and also other locations, i.e., at Tidore Island (Nurdiansah and Dharmawan, 2018)
322 and Middleburg-Miossu Island, West Papua (Nurdiansah and Dharmawan, 2021).
323 Forest height average among sites in Benoa Bay was not significantly different (ANOVA: p > 0.05). The sizeheight of
324 the mangrove stands in the Sonneratia, Rhizophora and Bruguiera zones were 9.14±2.29 m, 10.57±2.39 m and 10.55±3.00
325 m in height, respectively. Another study in North Maluku found mangrove forests reached 29.43±3.38 m in height (Arbi et
326 al. 2018). Mangroves in Padaido and Middleburg-Miossu Iisland also had the higher diameter growth up to 16.64±1.74 m
327 (Dharmawan et al. 2019); 15.49±0.64 m (Nurdiansah and Dharmawan, 2021); 12.29±0.60 m (Siringoringo et al. 2019);
328 and 18.25±0.76 m (Suyarso et al. 2018) consecutively. Those relationships are mainly due to nutrient, light, water, and
329 CO2 uptake competition (Mustika et al. 2014).

330 Spatial distribution of MHI

331 The spatial distribution of forest healthiness was dynamic, depending on various threats. Anthropogenic activities
332 around the mangrove areas do affect their growth and distribution. Dharmawan (2021) reported that reduced
333 anthropogenic activities around mangroves allowed them to grow and spread their seeds optimally.
334
335 Mangrove forest in Benoa Bay was categorized as having a moderate health state with an estimated mean MHI value
336 of 60.91±10.06% and ranging from 0.16% – 99.98% (Table 4). This study had similar results on MHI values to Liki and
337 Middleburg-Miossu islands, estimated at 60.03% and 60.70% consecutively respectively (Nurdiansah and Dharmawan,
338 2021). Those sites also had a similar range of density, diameter and canopy coverage as major components for MHI
339 calculation compared to Benoa Bay. There have been limited studies on spatial MHI distribution since the model was only
340 recently developed in 2021. On the other hand, the model clearly estimated spatial mangrove healthiness in other sites in
341 Indonesia, which can be accessed through the MonMang app. The specific area of the Rhizophora-dominated zone was
342 found with the highest MHI average at 67.15±9.08%. Its value ranged from 0.40% to 96.91%, while the Sonneratia zone
343 had the lowest MHI, 60.34±8.93% ranging from 0.16% to 99.98%. Mangrove growth was supported by environmental
344 conditions and less anthropogenic activities around it. Low anthropogenic pressure provides opportunities for mangrove
345 seedlings and saplings to grow optimally. Moreover, the seedlings spread more effectively when there is less
346 anthropogenic threat. The success of mangrove growth increases the MHI (Dharmawan, 2021).
347
348
349 Table 4. Descriptive statistical parameters of MHI value on each genera-dominated zone in Benoa Bay mangrove
350
Descriptive Statistical Forest Zone
Parameters Sonneratia Rhizophora Bruguiera All zones
Mean ± Standard deviation
60.34±8.93 67.15±9.08 65.13±4.87 60.91±10.06
(%)
Min 0.16 0.40 38.95 0.16
Max 99.98 96.91 76.58 99.98
Median (Q2) 62.,75 70.25 65.,61 63.24
Q1 57.,76 66.25 62.,12 58.76
Q3 65.,75 72.25 68.,86 66.24

8
351

352
353 Figure 5. Distribution of MHI for each mangrove zone
354
355
356
357 According to the MHI spatial distribution area, most Benoa Bay mangrove were was categorized as being in excellent
358 condition over an area of approximately 564.96 ha (47.74%) (Figure 5 and 6). Meanwhile, moderate and poor categories
359 covered about 40.76% and 11.51% of the total mangrove area in the Bay, respectively. The excellent mangrove category
360 was mainly found in the Rhizophora-dominated zone; aroofund 445 ha, whereas the moderate and poor categories were
361 mainly found in the Sonneratia-dominated zone.
362 Field documentation of MHI categories on each zone was provide in Figure 7. The spatial distribution of forest
363 healthiness was dynamic, depending on various threats. Anthropogenic activities around the mangrove areas do affect their
364 growth and distribution. Dharmawan (2021) reported that reduced anthropogenic activities around mangroves allowed
365 them to grow and spread their seeds optimally.

366
367 Figure 6. Area of each MHI category (left) and their proportion (right) on each genera-dominated zone in Benoa Bay, Bali.
368

9
369
370
371 Figure 7. Field observation for each category of MHI along mangrove zones (a, b, and c were a respective poor, moderate, and
372 excellent condition of Sonneratia-dominated forest; d, e, and f represented poor, moderate, and excellent category of Rhizophora
373 dominated zone, respectively; while g and h were moderate and excellent Bruguiera forest

374
375 Figure 8. Mangrove dieback phenomenon in Benoa Bay, Bali due to massive sedimentation on the seaward area
376
377
378 A high proportion of MHI in the poor category was contributed by the Sonneratia zone, followed by Rhizophora and
379 Bruguiera zones. The Dieback dieback event showed by the drierchanged of mangrove twigs that become drier and
380 reduces the leaves' absence in mangrove stands and seedlings, around them which primarily triggered by sedimentation
381 progress in Benoa Bay, was the main cause of the high contribution of poor category area of MHI in the Sonneratia zone
382 (Figure 8). (Figure 8). The dieback effect is primarily triggered by sedimentation progress in Benoa Bay. Monitoring the
383 Sentinel-2A satellite image in January 2019 showed that the mangrove area in Benoa Bay that experienced dieback was
384 8.95 ha, where it continued to expand from 2017 to 2019 (Prasetio, 2019). The dieback event that occurred in Benoa Bay
385 mainly affected the Sonneratia, Rhizophora and Avicennia species since these species are mostly found seaward, and this
386 zone was more influenced by sedimentation (MongaBay, 2019).

10
387 ACKNOWLEDGEMENTS

388 The authors would like to thank Dr. Bruce Campbell for his constructive criticism of the manuscript. In addition, we
389 would like to thank COREMAP-CTI Project implemented by National Research and Innovation Agency which was
390 funded by World Bank through Capacity Building Research Programme 2021, who supported partial funding during field
391 activities and the manuscript writing assistance. We are also very grateful to other parties who have assisted and
392 contributed to this research. The authors declare no conflicts of interest affecting the publication of this paper.

393 REFERENCES

394 Adilah D, Akbar AA, Jumiati. 2018. Peran ekosistem mangrove bagi mitigasi pemanasan global. J Civ Eng 18(2): 1-8. DOI:10.26418/jtsft.v18i2.31233
395 Akbar N, Baksir A, Tahir I, Arafat D. 2016. Community structure of mangrove in Mare Island, Tidore City, Maluka Utara Province. J of Aquat Coast and
396 Fish Sci 5(3): 133-142. DOI:10.13170/depik.5.3.5578
397 Akhrianti I, Nurtjahya E, Franto, Syari IA. 2019. Condition of Mangrove Community at North Mendanau Coastal and Batu Dinding Island in Belitung
398 Regency. J Aquat Resour 13 (1): 12-26. DOI:10.33019/akuatik.v12i1.856
399 Andiani AAE, Karang IWGA, Putra ING, Dharmawan IWE. 2021. Relationship among mangrove stand structure parameters in estimating the
400 community scale of aboveground carbon stock. J of Mar Sci. and Technol 13(3): 485-498. DOI: 10.29244/jitkt. v13i3.36363
401 Andika IBMB, Kusmana C, Nurjaya IW. 2017. The impact of the construction of Bali Mandala Toll Road to the mangrove ecosystem at Benoa
402 Bay Bali. J Nat Resour and Environ Manag 9(3): 641-657. DOI: 10.29244/jpsl.9.3.641-657
403 Arbi UY, Utama RS, Wahidin JSN, Budiyanto A, Makatipu PC, Wibowo K, Harahap A, Rahmawati S, Irawan A, Rizqi MP, Kotta R, Huwae R, Djabar
404 M, Nurdiansah D, Kusnadi A, Akbar N, Patty SI, Sumarno. 2018. Monitoring kondisi terumbu karang dan ekosistem terkait Perairan Ternate dan
405 sekitarnya. COREMAP-CTI, Pusat Penelitian Oseanografi LIPI, Jakarta.
406 Ardiansyah WI, Pribadi R, Nirwani S. 2012 Struktur dan komposisi vegetasi mangrove di Kawasan Pesisir Pulau Sebatik, Kabupaten Nunukan,
407 Kalimantan Timur. J of Mar Res 1(2): 203–215. DOI: 10.14710/jmr.v1i2.2039
408 Arifanti VB, Kauffman JB, Hadriyanto D, Murdiyarso D, Diana R. 2019. Carbon dynamics and land use carbon footprints in mangrove-converted
409 aquaculture: The case of the Mahakam Delta, Indonesia. For Ecol and Manag 432: 17-29. DOI: 10.1016/j.foreco.2018.08.047
410 Arifanti VB, Novita N, Tosiani A. 2021. Mangrove deforestation and CO2 emissions in Indonesia. IOP Conf Ser Earth Environ Sci. October.
411 DOI:10.1088/1755-1315/874/1/012006
412 Ball M. 1998. Mangrove species richness in relation to salinity and waterlogging: a case study along the Adelaide River floodplain, northern Australia.
413 Glob Ecol & Biogeogr Lett 7(1): 73-82. DOI: 10.2307/2997699
414 Barik J, Mukhopadhyay A, Ghosh T, Mukhopadhyay SK, Chowdhury SM, Hazra S. 2018. Mangrove species distribution and water salinity: an indicator
415 species approach to Sundarban. J of Coast Conserv 22(2): 361-368. DOI:10.1007/s11852-017-0584-7
416 Behera MD, Barnwal S, Paramanik S, Das P, Bhattyacharya BK, Jagadish B, Behera SK. 2021. Species-level classification and mapping of a mangrove
417 forest using random forest—utilisation of AVIRIS-NG and sentinel data. Remote Sens 13(11): 1-17. DOI: 10.3390/rs13112027
418 Bessie DM, Schaduw JN, Reppie E, and Lasut MT. 2013. Community structure of mangrove at Marine Tourism Park of Kupang Bay, East Nusa
419 Tenggara. Aquat Sci & Manag (1): 3-9. DOI: 10.35800/jasm.0.0.2013.2270
420 Brander LM, Wagtendonk AJ, Hussain SS, McVittie A, Verburg PH, de Groot RS, van der Ploeg S. 2012. Ecosystem service values for mangroves in
421 Southeast Asia: A meta-analysis and value transfer application. Ecosyst Serv 1(1): 62-69. DOI:10.1016/j.ecoser.2012.06.003
422 Carugati L, Gatto B, Rastelli E, Lo Martire M, Coral C, Greco S, Danovaro R. 2018. Impact of mangrove forests degradation on biodiversity and
423 ecosystem functioning. Sci Rep 8(1): 1-11. DOI: 10.1038/s41598-018-31683-0
424 Darmadi, Lewaru MW, Khan AMA. 2012. Struktur komunitas vegetasi mangrove berdasarkan karakteristik substrat di Muara Harmin Desa Cangkring
425 Kecamatan Tan Cantigi Kabupaten Indramayu. J of Fish and Mar Sci 3(3): 347-358.
426 Dewi IGAIP, Faiqoh E, As-syakur AR, Dharmawan IWE. 2021. Natural regeneration of mangrove seedlings in Benoa Bay, Bali. J of Mar Sci and
427 Technol 13(3): 395-410. DOI: 10.29244/jitkt.v13i3.36364
428 Dharmawan IWE, Hadi TA, Arbi UY, Makatipu PC, Rahmawati S. 2020a. Monitoring kesehatan terumbu karang dan ekosistem terkait di Kabupaten
429 Biak-Numfor tahun 2020. COREMAP-CTI LIPI, Jakarta.
430 Dharmawan IWE, Pramudji. 2019. Mangrove community structure in Papuan Small Islands, case study in Biak Regency. IOP Conf Ser Earth Environ
431 Sci Volume 550 International Conference Mangroves and Its Related Ecosystems 2019. Purwokerto, Indonesia, 21 – 23 August 2019. [Indonesia].
432 DOI:10.1088/1755-1315/550/1/012002
433 Dharmawan IWE, Suyarso, Ulumuddin YI, Prayudha B, Pramudji. 2020b. Manual for mangrove community structure monitoring and research in
434 Indonesia. NAS Media Pustaka, Makassar.
435 Dharmawan IWE, Widyastuti A. 2017. Pristine mangrove community in Wondama Gulf, West Papua, Indonesia. Mar Res in Indones 42(2): 73-82. DOI:
436 10.14203/mri.v42i2.175
437 Dharmawan IWE. 2020. Hemispherical Photography: Analisis Tutupan Kanopi Komunitas Mangrove. Nas Media Pustaka, Makassar.
438 Dharmawan IWE. 2021. Mangrove health index distribution on the restored post-tsunami mangrove area in Biak Island, Indonesia. IOP Conf Ser Earth
439 Environ Sci. Makassar, Indonesia, 5-6 June 2021. [Indonesia]. DOI: 10.1088/1755-1315/860/1/012007
440 Dharmawan IWE, Utama RS, Giyanto LP, Aji PC, Makatipu, Irawan A. 2019. Monitoring kondisi kesehatan terumbu karang dan ekosistem pesisir
441 terkait di TWP. Padaido, Biak-Numfor. COREMAP-CTI, LIPI, Jakarta.
442 Dharmawan IWE, Ulumuddin YI. 2021. Mangrove Community Structure Data Analysis, A Guidebook for Mangrove Health Index (MHI) Training. Nas
443 Media Pustaka, Makassar.
444 Edwin M, Sulistyorini IS, Poedjirahajoe E, Faida LRW, Purwanto RH. 2021. Structure and dominance of species in mangrove forest on Kutai National
445 Park, East Kalimantan, Indonesia. J of Trop For Manag 27(1): 59-59. DOI: 10.7226/jtfm.27.1.59
446 Frederika YC, Ihsan YN, Riyantini I. 2021. Nutrient profile and mangrove vegetation composition in the coastal waters of Indramayu. J of Mar Sci 7(1):
447 42-51. DOI: 10.20956/jiks.v7i1.12879
448 Friess DA. 2017. Ecotourism as a tool for mangrove conservation. Sumatra J of Disaster Geogr and Geogr Educ (SJDGGE) 1(1): 24-35.
449 Gabi FG, Sondak CFA, Kumampung DRH, Darwisito S, Ompi M, Rembet UNWJ. 2021. Gamlamo mangrove community structure, Jailolo Subdistrict,
450 West Halmahera, North Maluku Province. J of Coast and Trop Sea 9(3): 34-43. DOI: 10.35800/jplt.9.3.2021.36489
451 Ghorbanian A, Zaghian S, Asiyabi RM, Amani M, Mohammadzadeh A, Jamali S. 2021. Mangrove ecosystem mapping using Sentinel-1 and Sentinel-2
452 satellite images and random forest algorithm in Google Earth Engine. Remote Sens 13(13): 2-18. DOI: 10.3390/rs13132565
453 Gilman EL, Ellison J, Jungblut V, Van Lavieren H, Wilson L, Areki F, Yuknavage K. 2006. Adapting to Pacific Island mangrove responses to sea level
454 rise and climate change. Clim Res 32(3): 161-176. DOI:10.3354/cr032161

11
455 Giri C, Ochieng E, Tieszen LL, Zhu Z, Singh A, Loveland T, Duke N. 2011. Status and distribution of mangrove forests of the world using earth
456 observation satellite data. Glob Ecol and Biogeogr 20(1): 154-159. DOI: 10.1111/j.1466-8238.2010.00584.x
457 Göltenboth F, Timotius KH, Milan PP, Margraf J. 2006. Ecology of Insular Southeast Asia – The Indonesian Archipelago. Elsevier, Amsterdam.
458 Harefa MS, Nasution Z, Mulya MB, Maksum A. 2022. Mangrove species diversity and carbon stock in silvofishery ponds in Deli Serdang District, North
459 Sumatra, Indonesia. J of Biol Divers 22(2): 655-662. DOI: 10.13057/biodiv/d230206
460 Hariyanto S, Fahmi AK, Soedarti T, Suwarni EE. 2019. Vegetation and community structure of mangrove in Bama Resort Baluran National Park
461 Situbondo East Java. J of Biol & Biol Educ (Biosaintifika) 11(1): 132-138. DOI: 10.15294/biosaintifika.v11i1.19111
462 Heriyanto NM, Subiandono E. 2016. Role of mangrove biomass in carbon sink, in Kubu Raya, West Kalimantan. J of Policy Anal 13(1): 1–12. DOI:
463 10.20886/jakk.2016.13.1.1-12
464 Simamora HP, Khairijon, Isda MN. 2014. Analisis vegetasi mangrove di ekosistem mangrove Desa Tapian Nauli 1 Kecamatan Tapian Nauli Kabupaten
465 Tapanuli Tengah Provinsi Sumatera Utara. Stud Online J in Math and Nat Sci 1(2): 1-10. DOI: 10.17969/jimfp.v6i4.18137
466 Imamsyah A, Bengen DG, Ismet MS. 2020. Struktur Vegetasi Mangrove Berdasarkan Kualitas Lingkungan Biofisik di Taman Hutan Raya Ngurah Rai
467 Bali. J of Environ Sci (Ecotrophic) 14(1): 88-99. DOI: 10.24843/EJES.2020.v14.i01.p08
468 Jenoh EM, Robert EMR, Lehmann I, Kioko E, Bosire JO, Ngisiange N, Guebas FD, Koedam N.2016. Wide ranging insect infestation of the pioneer
469 mangrove Sonneratia alba by two insect species along the Kenyan Coast. PLoS One 11(5): 1-15. DOI: 10.1371/journal.pone.0154849
470 Jhonnerie R, Siregar VP, Nababan B, Prasetyo LB, Wouthuyzen S. 2015. Random forest classification for mangrove land cover mapping using Landsat 5
471 TM and Alos Palsar imageries. Procedia Environ Sci 24(2015): 215-221. DOI: 10.1016/j.proenv.2015.03.028
472 Jiang Y, Zhang L, Yan M, Qi J, Fu T, Fan S, Chen B. 2021. High-resolution mangrove forests classification with machine learning using worldview and
473 uav hyperspectral data. Remote Sens 13(8): 1-21. DOI:10.3390/rs13081529
474 JICA. 1999. Model Pengelolaan Hutan Mangrove Lestari. Kerjasama JICA dengan Departemen Kehutanan, Jakarta.
475 Kamruzzaman M, Osawa A, Deshar R, Sharma S, Mouctar K. 2017. Species composition, biomass, and net primary productivity of mangrove forest in
476 Okukubi River, Okinawa Island, Japan. Reg Stud in Mar Sci 12(2017): 19-27. DOI:10.1016/j.rsma.2017.03.004
477 Karawang Post. 2021. Rehabilitasi mangrove Tahura Bali akan segera menjadi percontohan di Indonesia. PT Kolaborasi Mediapreneur Nusantara,
478 Bandung. https://karawangpost.pikiran-rakyat.com/nasional/pr-1422761031/rehabilitasi-mangrove-tahura-bali-akan-segera-menjadi-percontohan-di-
479 indonesia
480 Karimah. 2017. Peran ekosistem hutan mangrove sebagai habitat untuk organisme laut. J of Trop Biol 17(2): 51-58. DOI: 10.29303/jbt.v17i2.497
481 Kasim F, Kadim MK, Nursinar S, Karim Z, Lamalango A. 2019. Comparison of true mangrove stands in Dudepo and Ponelo islands, north Gorontalo
482 district, Indonesia. J of Biol Divers (Biodiversitas) 20(1): 259-266. DOI:10.13057/biodiv/d200142
483 Khairunnisa, C., E. Thamrin, & H. Prayogo. 2020. Species diversity of mangrove vegetation in Dusun Besar Village Pulau Maya District, Kayong Utara
484 Regency. J of Sustain For 8(2): 325-336. DOI: 10.26418/jhl.v8i2.40074
485 Kitamura S, Anwar C, Amalyos C, Baba S. 1997. Buku Panduan Mangrove di Indonesia: Bali & Lombok. ISME.
486 Koch MS. 1997. Rhizophora mangle L. seedling development into the sapling stage across resource and stress gradients in Subtropical Florida 1.
487 Biotropica 29(4): 427-439. DOI: 10.1111/j.1744-7429.1997.tb00037.x
488 Krauss KW, Allen JA, Cahoon DR. 2003. Differential rates of vertical accretion and elevation change among aerial root types in Micronesian mangrove
489 forests. Estuar Coast and Shelf Sci 56(2): 251-259. DOI: 10.1016/S0272-7714(02)00184-1
490 Kusmana, C. 2014. Distribution and current status of mangrove forests in Indonesia. In Mangrove ecosystems of Asia (pp. 37-60). Springer, New York,
491 NY. DOI:10.1007/978-1-4614-8582-7_3
492 Lewerissa YA, Sangaji M, Latumahina M B. 2018. Mangrove management based on type of the substrate at Ihamahu Water Saparua Island. J of Aquat
493 Resour Manag 14(1): 1-9.
494 Li N, Chen P, Qin C. 2015. Density storage and distribution of carbon in mangrove ecosystem in Guangdong’s coastal areas. Asian Agric Res 7(2): 62 –
495 73. DOI: 10.22004/ag.econ.202107
496 Lovelock CE, Ball MC, Martin KC, C Feller I. 2009. Nutrient enrichment increases mortality of mangroves. PloS One 4(5): e5600.
497 DOI:10.1371/journal.pone.0005600
498 Lubis Z. 2021. Statistika terapan untuk ilmu-ilmu sosial dan ekonomi. ANDI, Yogyakarta.
499 Maher DT, Call M, Santos IR, Sanders CJ. 2018. Beyond burial: lateral exchange is a significant atmospheric carbon sink in mangrove forests. Biol Lett
500 14(7): 1-4. DOI: 10.1098/rsbl.2018.0200
501 Martiningsih NGAGE, Suryana IM, Sutiadipraja N. 2015. Analisa vegetasi hutan mangrove di Taman Hutan Raya (Tahura) Bali. J of Agric based on
502 Ecosyst Balance 5(09): 90187.
503 McHugh ML. 2012. Interrater reliability: the kappa statistic. Biochemia Medica 22(3): 276-282. DOI:10.11613/BM.2012.031
504 Millard K, Richardson M. 2015. On the importance of training data sample selection in random forest image classification: A case study in peatland
505 ecosystem mapping. Remote Sens 7(7): 8489-8515. DOI: 10.3390/rs70708489
506 Mojiol AR, Kodoh J, Wahab R, Majuki M. 2016. Contribution of non-wood forest product to the local community living near mangrove forest in Kudat,
507 Sabah. J of Trop Resour and Sustain Sci 4(1): 38-41.
508 MongaBay. 2019. Degradasi Mangrove Indonesia: Fenomena Dieback Pada Kawasan Teluk Benoa Bali. MongaBay-Indonesia.
509 https://www.mongaBay.co.id/2019/02/11/degradasi-mangrove-indonesia-fenomena-dieback-pada-kawasan-teluk-benoa-bali/.
510 Mughofar A, Masykuri M, Setyono P. 2018. Zonation and composition of mangrove forest vegetation at Coast Cengkrong, Cengkrong Village, District
511 of Karanggandu, Trenggalek East Jawa Provinsi. J of Nat Resour and Environ Manag 8(1): 77-85. DOI: 10.29244/jpsl.8.1.77-85
512 Mustika DI, Rusdiana O, Sukendro A. 2014. The development of Rhizophora apiculata at mangrove nursery of Muara Teluk Naga Village, Tangerang
513 District, Banten. J of Bonorowo Wetl 4(2): 108-116. DOI: 10.13057/bonorowo/w040204
514 Naidoo G. 2016. The mangroves of South Africa: An ecophysiological review. South Afr J of Bot 107: 101-113. DOI: 10.1016/j.sajb.2016.04.014
515 Nanulaitta EM, Tulalessy A, Wakano D. 2019. Analysis of mangrove’s rapidity as one of the ecowicate indicators in The Alariano Beach Waters of
516 Amahai Sub-District, Maluku Central District. The Small Isl For J 3(2): 217–226. DOI: 10.30598/jhppk.2019.3.2.217
517 Nardin W, Vona I, Fagherazzi S. 2021. Sediment deposition affects mangrove forests in the Mekong delta, Vietnam. Cont Shelf Res 213: 1-29. DOI:
518 10.1016/j.csr.2020.104319
519 Noor YR, Khazali M, Suryadiputra INN. 2006. Panduan Pengenalan Mangrove di Indonesia. PHKA/WI-IP, Bogor.
520 Nurdiansah D, Dharmawan IWE. 2018. Mangrove community in Coastal Area of Tidore Islands. Oceanol and Limnol in Indones, 3(1): 1-9. DOI:
521 10.14203/oldi.2018.v3i1.63
522 Nurdiansah D, Dharmawan IWE. 2021. Community structure and healthiness of mangrove in Middleburg-Miossu Island, West Papua. J of Mar Sci and
523 Technol 13(1): 81-96. DOI:10.29244/jitkt.v13i1.34484
524 Nurhaliza AP, Damayanti A, Dimyati M. 2021. Monitoring area and health changes of mangrove forest using multitemporal landsat imagery in Taman
525 Hutan Raya Ngurah Rai, Bali Province. IOP Conf Ser Earth Environ Sci February. DOI: 10.1088/1755-1315/673/1/012050
526 Opa ET, Kusen JD, Kepel RC, Jusuf A, Lumingas LJ. 2019. Community structure of mangrove in Mantehage Island and Paniki Island, North Sulawesi,
527 Indonesia. Aquac Aquar Conserv & Legis 12(4): 1388-1403.
528 Pillai NG, Harilal CC. 2016. Surveillance of the tolerance limit of Sonnearia alba Sm. to certain hydrogeochemical parameters from heterogenous natural
529 habitats of Kerala, South India. Int Res J of Biol Sci 5(12): 28-37.

12
530 Prianto E, Jhonnerie R, Firdaus R, Hidayat T, Miswadi M. 2006. Biodiversity and Ecological Structure of Mangrove Mature in Coastal Area of Dumai
531 District-Riau Province. J of Biol Divers (Biodiversitas) 7(4):327-332. DOI: 10.13057/biodiv/d070406
532 Primantara IKE, Darmadi AAK, Ginantra IK. 2019. Growth of several species of mangrove seedlings as seeds ready for planting in, Karhutla Centre of
533 Java Bali, and Nusa Tenggara. Simbiosis 7(1): 6-10. DOI: 10.24843/JSIMBIOSIS.2019.v07.i01.p02
534 Prinasti NKD, Dharma IGBS, Suteja Y. 2020. Struktur komunitas vegetasi mangrove berdasarkan karakteristik substrat di Taman Hutan Raya Ngurah
535 Rai, Bali. J of Mar and Aquat Sci 6(1): 90-99. DOI: 10.24843/jmas.2020.v06.i01.p11
536 Purwanto RH, Mulyana B, Satria RA, Yasin EHE, Putra ISR, Putra AD. 2022. Spatial distribution of mangrove vegetation species, salinity, and mud
537 thickness in mangrove forest in Pangarengan, Cirebon, Indonesia. J of Biol Divers (Biodiversitas) 23(3): 1383-1391. DOI: 10.13057/biodiv/d230324
538 Rahman, Yanuarita D, Nurdin N. 2014. Mangrove community structure in District Muna. Torani J of Fish and Mar Sci 24(2): 29-36. DOI:
539 10.35911/torani.v24i2.226
540 Rambu LP, Runtuboi F, Loinenak FA. 2019. Mangrove diversity and distribution based on substrate type in Coastal Coast of Syoribo Village East
541 Numfor District Biak Numfor District Papua Province. Indopacific J of Aquat Resour 3(1): 31-44. DOI: 10.46252/jsai-fpik-
542 unipa.2019.Vol.3.No.1.64
543 Ramdani F, Rahman S, Giri C. 2018. Principal polar spectral indices for mapping mangroves forest in South East Asia: study case Indonesia. Int J of
544 Digit Earth 12(10): 1103-1117. DOI: 10.1080/17538947.2018.1454516
545 Richards DR, Friess DA. 2016. Rates and drivers of mangrove deforestation inSoutheast Asia, 2000–2012. PNAS 113(2):344-349. DOI:
546 10.1073/pnas.1510272113
547 Rosalina D, Rombe KH. 2021. Structure and composition of mangrove species in West Bangka Regency. J of Airaha 10(1): 99-108. DOI:
548 10.15578/ja.v10i01.219
549 Rudianto R, Bengen DG. 2020. An analysis of mangrove forest damage due to illegal logging and its effect on carbon stock and absorption in East Java,
550 Indonesia. Res Sq 1-17. DOI: 0.21203/rs.3.rs-64373/v1
551 Sari NWP, Abrar M, Nurhasim RMS, Putra RD, Sinaga M, Jumsurizal, Lestari IF, Pangestyansah I, Rasidin A, Siantaru RSOR, Perisha B. 2019.
552 Monitoring kesehatan terumbu karang dan ekosistem terkait di kawasan konservasi Perairan Nasional, Taman Wisata Perairan, Kepulauan Anambas
553 dan laut sekitarnya. COREMAP-CTI, Pusat Penelitian Oseanografi LIPI, Jakarta.
554 Schaduw JNW. 2020. Mangrove community structure Morowali District, Central Sulawesi Province, case in Maratape Village, Lafeu Village, and
555 Labota Village. Platax Sci J 8(2): 264-273. DOI: 10.35800/jip.8.2.2020.31538
556 Serosero R, Riani E, Butet NA. 2020. Reproduction of coconut crabs (Birgus latro) in Daeo District, Morotai Island, North Maluku, Indonesia. Aquac
557 Aquar Conserv & Legis 13(4): 2002-2013.
558 Setiawan H. 2013. Ecological Status of Mangrove Forest at Various Thickness Levels. Wallacea J of For Res 2(2): 104 – 120.
559 DOI:10.18330/jwallacea.2013.vol2iss2pp104-120
560 Singgalen YA. 2020. Mangrove forest utilization for sustainable livelihood through community-based ecotourism in Kao Village of North Halmahera
561 District. J of Trop For Manag 26(2): 155-168. DOI: 10.7226/jtfm.26.2.155
562 Siregar AMH, Efriyeldi E, Nasution S. 2022. The structure of mangrove community in Sebauk Village, Bengkalis District, Bengkalis Regency, Riau
563 Province. J of Coast and Ocean Sci 3(1): 60-66. DOI: 10.31258/jocos.3.1.60-66
564 Siringoringo RM, Abrar M, Sari NWP, Hukom FD, Putra RD, Cappenberg H, Pramudji, Dharmawan IWE, Ernawati, Triyono, Sianturi OR. 2019.
565 Monitoring Kesehatan Terumbu Karang dan Ekosistem Terkait di Kabupaten Merauke, Papua. COREMAP-CTI LIPI, Jakarta.
566 Siringoringo RM, Suharsono, Sari NWP, Arafat Y, Arbi UY, Azkab H, Dharmawan IWE, Sianturi OR, Anggraeni K. 2017. Monitoring Kesehatan
567 Terumbu Karang dan Ekosistem Terkait Lainnya di Kabupaten Nias Utara. COREMAP CTI LIPI 2017, Jakarta.
568 Sreelekshmi S, Preethy CM, Varghese R, Joseph P, Asha CV, Nandan SB, Radhakrishnan CK. 2018. Diversity, stand structure, and zonation pattern of
569 mangroves in southwest coast of India. J of Asia-Pacific Biodivers11(4): 573-582. DOI: 10.1007/s11273-021-09791-y
570 Srikanth S, Lum SKY, Chen Z. 2016. Mangrove root: adaptations and ecological importance. Trees 30(2): 451-465. DOI: 10.1007/s00468-015-1233-0
571 Sugiana IP, Faiqoh E, Indrawan GS, Dharmawan IWE. 2021. Methane concentration on three mangrove zones in Ngurah Rai Forest Park, Bali. J of
572 Environ Sci 19(2): 422-431. DOI: 10.14710/jil.19.2.422-431
573 Sukuryadi, Harahab N, Primyastanto M, Hadi AP. 2021. Short Communication: Structure and composition of mangrove vegetation in Lembar Bay area,
574 West Lombok District, Indonesia. J of Biol Divers (Biodiversitas) 22(12): 5585-5592. DOI: 10.13057/biodiv/d221243s
575 Sundra IK. 2016. Pengelolaan hutan mangrove kawasan Taman Hutan Raya (TAHURA) Ngurah Rai Propinsi Bali. [Thesis]. Udayana University, Bali.
576 [Indonesia]
577 Supriadi, Romadhon A, Farid A. 2015. Struktur komunitas mangrove di Desa Martajasah Kabupaten Bangkalan. Indones J of Mar Sci and Technol 8(1):
578 44-51. 10.21107/jk.v8i1.812s
579 Suyarso, Suharsono, Supriyadi IH, Cappenberg H, Hukom FD, Souhoka J, Dharmawan IWE, Wilandari DA, Manuputty AE, Nagib I, Budiyanto A,
580 Picasouw J, Rasyidin A, Salatalohi A. 2018. Monitoring kondisi terumbu karang dan ekosistem terkait di perairan Belitung. COREMAP-CTI, Pusat
581 Penelitian Oseanografi LIPI, Jakarta.
582 Syukri M, Mashoreng S, Werorilangi S, Isyrini R, Rastina R, Faizal A, Tahir A, Gosalam A. 2018. Carbon stock analysis of mangrove in Bebanga
583 Mamuju Regency West Sulawesi. Prosiding Simposium Nasional Kelautan dan Perikanan V. 5 Mei 2018. [Indonesia].
584 Tomlinson, P.B. 2016. The botany of mangroves. Cambridge University Press. New York.
585 Tosiani, A. Purwanto, J., Djuariah, R., Usman, A.B. 2020. Akurasi Data Penutupan Lahan Nasional Tahun 1990–2016 Akurasi Data Penutupan Lahan
586 Nasional Tahun 1990–2016. Direktorat Inventarisasi dan Pemantauan Sumber Daya Hutan, Ditjen Planologi Kehutanan dan Tata Lingkungan,
587 Kementerian Lingkungan Hidup dan Kehutanan, Jakarta.
588 Usman L, Syamsuddin, Hamzah SN. 2013. Analisis vegetasi mangrove di Pulau Dudepo Kecamatan Anggrek Kabupaten Gorontalo Utara. Fish and Mar
589 Sci J 1(1): 11-17. DOI: 10.37905/.v1i1.1211
590 Walters BB, Rönnbäck P, Kovacs JM, Crona B, Hussain SA, Badola R, Dahdouh-Guebas F. 2008. Ethnobiology, socio-economics and management of
591 mangrove forests: A review. Aquat Bot 89(2): 220-236. DOI: 10.1016/j.aquabot.2008.02.009
592 Wilson R. 2017. Impacts of climate change on mangrove ecosystems in the coastal and marine environments of Caribbean Small Island Developing
593 States (SIDS). Caribbean Climate Change Report Card: Science Review 2017: 60-82.
594 Wiyanto DB, Faiqoh E. 2015. Analisis vegetasi dan struktur komunitas mangrove di Teluk Benoa. J of Mar and Aquat Sci 1(1): 1- 7. DOI:
595 10.24843/jmas.2015.v1.i01.1-7
596 Xin K, Zhou Q, Arndt SK, Yang X. 2013. Invasive capacity of the mangrove Sonneratia apetala in Hainan Island, China. J of Trop For Sci 25(1): 70–78.
597 Yang SC, Shih SS, Hwang GW, Adams JB, Lee HY, Chen CP. 2013. The salinity gradient influences on the inundation tolerance thresholds of mangrove
598 forests. Ecol Eng 51: 59-65. Yang SC, Shih SS, Hwang GW, Adams JB, Lee HY, Chen CP. 2013. The salinity gradient influences on the inundation
599 tolerance thresholds of mangrove forests. Ecological engineering 51: 59-65. DOI: 10.1016/j.ecoleng.2012.12.049
600 Yanti DI, Paruntu CP, Kepei RC, Mandagi SV, Tabalessy RR, Masengi MC. 2021. Suitability index and carrying capacity of mangrove tourism on Jeflio
601 Island, Indonesia. Aquac Aquar Conserv & Legis 14(5), 3145-3156.
602 Zhang Y, Liang FP, Li YYW, Zhang JW, Zhang SJ, Bai H, Liu Q, Zhong CYR, Li L. 2018. Allelopathic effects of leachates from two alien mangrove
603 species, Sonneratia apetala and Laguncularia racemosa on seed germination, seedling growth and antioxidative activity of a native mangrove species
604 Sonneratia caseolaris. Allelopath J 44(1): 119-130. DOI:10.26651/allelo.j/2018-44-1-1158

13
605 Zhang Y, Su T, Ma Y, Wang Y, Wang W, Zha N, Shao M. 2022. Forest ecosystem service functions and their associations with landscape patterns in
606 Renqiu City. Plos One 17(4): 1-29. DOI:10.1371/JOURNAL.PONE.0265015.

14