igorkf/soil-carbon-stocks

How can we get an accurate estimate of carbon stocks across a landscape with minimal ground truth data?

soil-carbon-stocks

Reproducing

Firstly, clone or download the repository. Now run all the following scripts, in order:

1. Sample points for each county:

src/01-sample-points.R

2. Download SSURGO variables (bulk density, SOC, etc.):

src/02-download-ssurgo.R

3. Download top daily 9 km resolution Volumetric Water Content (VWC) raster files from Crop-CASMA:

src/03-download-vwc.R

4. Preprocess VWC data:

src/04-preprocess-vwc.R

5. Download daily precipitation data from NASA POWER:

src/05-download-precipitation.R

6. Download evapotranspiration data from GEE:

src/06-download-evapotranspiration.R

Disclaimer: This part needs a lot of additional dependencies (python, rgee, etc.). Follow the instructions contained on the script to setup everything you need.

7. Merge all data:

src/07-merge-data.R

8. Run a linear mixed model:

src/08-model.R

Tested on:

• Windows 10
• Ubuntu 22