README.md

In RETURN-project/GEE.aux: What the Package Does (One Line, Title Case)

Calculate recovery metrics from sampled time series (GEE)

This repository contains JavaScript, python and R code and follows this workflow: - Javascript: Generate an image with pixels of interest (/inst/Javascript/SampleArea) - Javascript: Sample point locations using a stratified random sampling from the pixels of interest (/inst/JavaScript/SamplePoints) - Python: Extract time series and auxiliary information from the sampled locations to a csv file (/inst/python/dllLandsatFire.py) - R: Prepare data and derive recovery metrics (/vignettes/Recovery_indicators_fire.Rmd): - Generate a dataframe with data extracted from the GEE - Convert the irregular time series to regular ones - Fit piecewise regression (segmentation of the time series) - Calculate recovery metrics - Export shapefile with locations for which the recovery indicators do not equal NA - The shapefile with recovery indicators should be uploaded as asset to the GEE - Python: Download the covariates (/inst/python/dllCovar.py) - R: Plot the recovery indicators - R: Model the recovery indicator as function of covariates

The JavaScript code should be pasted in a Google Earth Engine (GEE) code editor (registration is needed to get access the first time). After clicking on 'run' button, a new 'run' button should appear for the specific task under the tasks tab. The task will start after clicking this button.

The python code can be run after installing the Earth Engine Python API. More information about the GEE can be found here and information about the python installation of the GEE can be found here.

RETURN-project/GEE.aux documentation built on Dec. 18, 2021, 8:46 a.m.