A Cycle Short Course Lehi G University

Download as pdf or txt
Download as pdf or txt
You are on page 1of 43

Acycle: Time-series analysis software

for paleoclimate research and education

Mingsong Li
Linda Hinnov, Lee Kump
Penn Sate
Sept. 6, 2019
OUTLINE
1. Introduction

2. Getting Started

3. Acycle Graphic User Interface (GUI)

4. Case study: Carnian cyclostratigraphy


Time series analysis
• a statistical technique that deals with time series data.
• accounts for the fact that data points taken over time may
have an internal structure (such as trend or periodicity)
that should be accounted for.

• Economic Forecasting • STEM (Science, technology,


• Sales Forecasting engineering, and
mathematics)
• Population Forecasting • …
• Stock Market Analysis
• Yield Projections
Climate Time Series

Monthly average CO2 at Mauna Loa Observatory, Hawaii

https://www.esrl.noaa.gov/gmd/ccgg/trends/

Global benthic marine δ18O stack (global ice volume and temperature)

(Lisiecki & Raymo, 2005) (Hinnov, 2013)


Download this paper
Logo designed by Hewei Duan
Total: 1782 unique visitors
(Sept 2018 – Sept 2019)
http://mingsongli.com/acycle
Data source: https://analytics.google.com
What they say

“It is truly an amazing contribution to the geosciences


community … It opens up much needed access to these
powerful tools for a wide audience in the sedimentary geology
and paleoclimate community. … A really marvellous job."
—— Dr. J. Fred Read (Emeritus Professor, Virginia Tech)

"His Acycle software will become the standard tool for


time-scale applications by all international workers."

—— Dr. James G. Ogg (Professor, Purdue University)

“Not only is this software powerful and effective, it is also simple


to use and therefore benefits researchers and at all levels within
the paleoclimatology community, from novices to experts.”

—— Dr. Paul E. Olsen (Professor, Columbia University)


2. Getting Started
* Stand-alone versions of Acycle only needs Runtime, not MatLab

* MatLab Runtime is not MATLAB!

* MatLab Runtime is free!

Dropbox | OneDrive | BaiduCloud | GitHub

Read more
•Wiki: https://github.com/mingsongli/acycle/wiki
•or: https://github.com/mingsongli/acycle/blob/master/doc/AC_Users_Guide.pdf
•or: Acycle "Help" menu - "Manual",
•or: /doc/AC_Users_Guide.pdf
2. Getting Started

Acycle for… MatLab Stand-alone **


Platform Mac Windows
Requirement MatLab Runtime 2019a (Free)
MatLab
download Runtime for Mac Runtime for Win

Version Installer Green Installer Green

Size 63.5 MB 97 MB 91 MB 122 MB 121 MB


How to install Acycle2.0_Users_Guide.pdf (Chapter 3)

** Need to install the MatLab Runtime 2019a (!!!)


Fast installation:

1. Install MatLab Runtime 2019a (free)


You don’t need to install MatLab (heavy and expensive)!
After the installation, you don’t need to run runtime yourself
If you have installed MatLab 2019a, this step can be skipped

2. Download Acycle2.0 GREEN version


No installation is needed.

3. Follow the next slide to start up Acycle 2.0.

** Need to install the MatLab Runtime 2019a (!!!)


Startup
double-click

1. Go to the installation folder

2. Right click “Acycle” file, choose “Show Package Content”

3. Drag the applauncher file to the dock


Acycle Graphic User Interface (GUI)

Don’t close this window!

Press CTRL + C keys to kill Acycle


Data Requirement

comma-, table- or space-delimited text (.txt)


comma-separated values files (.csv)

No header is permitted.

Most data files should contain two columns of series.

depth
or time value
Plot Pro
Insolation

Based on the MatLab code inso.m by Jonathan Levine (2001), UC Berkeley.


modified by Peter Huybers (Harvard)
modified by Mingsong Li (Penn State, 2018)
Astronomical Solution
Signal/Noise Generator
Math Time series

Case studies:
1. Insolation
2. Laskar 2004 astronomical solution
3. Carnian cyclostratigraphy (~235 Ma, Triassic)
Carnian cyclostratigraphy (~235 Ma, Triassic)

Proxy: gamma-ray with higher intensities


indicating higher average clay contents

Target: learn typical data process steps

Wayao

Zhang et al., 2015 Palaeo-3


Flowchart of the cyclostratigraphic
analysis in Acycle software (1)
Flowchart of the cyclostratigraphic
analysis in Acycle software (2)
Step 1. Load Data

The gamma ray data entitled


“Example-WayaoCarnianGR0.txt”
will be loaded and displayed in
the Acycle main window.
Step 2. Data Preparation (optional)

Users can sort data in ascending


order. Two or more values for the
same time (or depth) may be
averaged with the "Unique"
function.
Step 3. Interpolation

Stratigraphic depth or time series are typically irregularly spaced due to


uncertain timescales or difficulty in data collection. This necessitates
interpolation to generate uniformly spaced time (or depth) series.

“Example-WayaoCarnianGR0-rsp0.33.txt”.
Step 4. Detrending
Detrending is a critical step for power spectral analysis to ensure that data
variability oscillates about a zero mean, and to avoid power leakage from very
low-frequency components into higher frequencies of the spectrum.

Read More: Detrending


Step 5. Power spectral analysis

Power spectral analysis evaluates the distribution of time series variance


(power) as a function of frequency. The primary use of power spectral analysis
is for the recognition of periodic or quasi-periodic components in a data series

Read More: Spectral Analysis


Step 6. Evolutionary power spectral analysis

Read More
Step 7. Correlation coefficient

To estimate the optimal sedimentation rate.

Read More: COCO/eCOCO

Similar tool: TimeOpt / eTimeOpt


You will have the following figure and a log file saving all settings.

It tells the most likely sedimentation rate is ~9 cm/kyr,


with a null hypothesis (no orbital forcing) significance level of <1%
All seven orbital parameters are used in the estimation.
eCOCO analysis to track variable sedimentation rate

Similar tool: eTimeOpt


eCOCO analysis to track variable sedimentation rate
Step 8. Filtering
Filters are also essential tools to aid in the isolation of specific frequency
components in the paleoclimate data series.

Read More
Step 9. Age model and tuning
transform original data (usually in the depth domain) to tuned data
(usually in the time domain) when an age model file is available
Select “Example-WayaoCarnianGR0-rsp0.33-80-LOWESS-gaus-0.028+-0.006.txt”

a, Build age model

Example-WayaoCarnianGR0-rsp0.33-80-LOWESS-gaus-0.028+-0.006-agemod-405-max.txt

b, Age Scale
Tuned data:
“Example-WayaoCarnianGR0-TD-Example-WayaoCarnianGR0-rsp0.33-80-
LOWESS-gaus-0.028+-0.006-agemod-405-max.txt”
Mini robot
Other useful tools
Sedimentary noise model
Lag-1 autocorrelation coefficient (ρ1)
dynamic noise after orbital tuning (DYNOT) model

Li, et al., 2018 Ancient ocean Credit: Hewei Duan


Nature Communications
Smoothing
‘Image’ Tool
1. Select an image file
2. Select “Show Image”
3. (Optional) Transfer an RGB
image to Grayscale image:
Select “RGB to Grayscale”
‘Image’ Tool

4: Image Profile
5: Press ALT key, select 2 cursors
6. Press “Enter” in the terminal window
Step 6
Plot Digitizer

Example-PlotDigitizer.jpg
Acknowledgement

Astronomical solution: Jacques Laskar,


Richard Zeebe
Changepoint: Eric Ruggieri
Insolation: Jonathan Levine, Peter Huybers
Padding: Nicolas R. Thibault
Spectral Moments: Matthias Sinnesael
TimeOpt/eTimeOpt: Steve Meyers
Linda Hinnov Lee Kump
Wavelet: Christopher Torrence, Gilbert Compo

Feng Cheng, Anne-Christine Da Silva, Hewei Duan,


Yanan Fang, Marco Franceschi, Daniel R. Franco,
Xiaoni Hu, Dorothée Husson, Arsenio Muñoz Jiménez,
Dongyang Liu, James Ogg, Paul Olsen, J. Fred Read,
Chen Shen, Chuanyue Wang, Mathieu Martinez, Meng
Wang, Xu Yao, Qiyan Zhang, Yang (Wendy) Zhang,
Christian Zeeden
Make it a better software for your own research.
ANY feedback is highly appreciated!!!

Email: [email protected]; [email protected]


https://github.com/mingsongli/acycle/issues

Please cite:

Li, M., Hinnov, L., Kump, L. (2019) Acycle: Time-series analysis software for
paleoclimate research and education. Computers & Geosciences 127, 12-22.

and other relevant publications.

Thank you!

You might also like