sytbru/CV-clustered-data: First release

R scripts related to the manuscript "Dealing with clustered samples for assessing map accuracy by cross-validation" in Ecological Informatics.

DATA 

The input data are supposed to be in a directory "data". The data can be downloaded from Zenodo: DOI:10.5281/zenodo.6513429

agb.tif = above ground biomass (AGB) map

AGBstack.tif  = covariates used for predicting AGB

aggArea.tif = coarse grid used for simulation in the model-based methods

ocs.tif = soil organic carbon stock (OCS) map

OCSstack.tif = covariates used for predicting OCS

strata.xxx = geo-strata used (shp) for generating the clustered samples

TOTmask.tif = mask of the area covered by the covariates

RUNING THE SCRIPTS 

First, the samples need to be prepared by running the scripts sample_*.R

Next, the cross-validation scripts named CV_*.R can be run.

Start by running "CV_random.R", as the other CV_*.R scripts depend on the results it produces.

The script "CV_model_based.R" should be run before running "CV_heteroscedastic.R".

The script figs.R can be used for reproducing several of the figures shown in the manuscript. Here it is assumed that the full set of results has been generated (see WARNING below).

WARNING 

Note that running the (single core) scripts with the full sample size and number of replications as used in the paper requires a very long time to complete. Set n_samp, n_CV and nsim to numbers << 100 to check the approach without reproducing all the results. The code can easily be adapted to run on multiple cores.