Code, example data, and manuscript supplemental materials which accompany Swetnam, T.L.; Yool, S.R.; Roy, S.; Falk, D.A. On the Use of Standardized Multi-Temporal Indices for Monitoring Disturbance and Ecosystem Moisture Stress across Multiple Earth Observation Systems in the Google Earth Engine. Remote Sens. 2021, 13, 1448. https://doi.org/10.3390/rs13081448
Lead Author: Tyson Lee Swetnam
Co-Authors: Donald A. Falk , Samapriya Roy , & Stephen Yool.
The repository is organized (in the attempt) to enable reproducible research as part of the FAIR data principles.
You can (re)run these analyses using your own computer, on commercial cloud, or a data science workbench CyVerse Discovery Environment.
See our Wiki for additional details on the background for this work.
Clone the repo to your local or VM:
git clone https://github.com/tyson-swetnam/emsi
gh repo clone tyson-swetnam/emsi
Tabular time series data presented in the manuscript.
Raster data are from NASA data archive services (e.g., landsat, modis) and ESA data services. Data are also hosted as Google Earth Engine Collections. These raster data can be extracted to your local computer using our example /python/*.py
scripting and /rmd/*.Rmd
notebooks.
Previously extracted data are stored as .csv
and geotiff .tif
files on CyVerse Data Store, and R Markdown notebooks are provided in this repo which wil extract and render these data as the tables and figures used in the main text.
Additional derivative imagery layers from Google Earth Engine (GEE) are hosted on CyVerse DataCommons.
Original commercial imagery are available from Planet.com.
We are hosting RStudio Server containers on CyVerse for reproducing the Rmd figures.
Dockerfiles w/ automated container builds hosted on
Container can be run on CyVerse Discovery Environment
Other CyVerse VICE images: https://github.com/cyverse-vice
Notebooks (Python3) for data analyses and extraction with GEE & Planet Labs enabled Jupyter Notebooks.
The JavaScript code used within Earth Engine Code UI for the intial time series point extractions, cloud-free NDVI calculations, and EMSI calculations within GEE. The EMSI utils code repo could be found here once added it should appear in your reader repositories. This contains code to generate a multiband EMSI product and to then generate a video from the multiband imagery.
You can also clone the repo scripts for Earth Engine directly by running:
git clone https://earthengine.googlesource.com/users/tswetnam/emsi
Calculate and export EMSI time series videos from GEE:
git clone https://earthengine.googlesource.com/users/samapriya/emsi-utils
R Markdown files with statistical data analyses for figures in manuscript.
These Rmd are also hosted as HTML on the CyVerse Data Store
US-Mexico Border grasslands example
Python .py
files for extracting image collections to Google Drive.