jrod55/enviropolate

EnviroPolate

This repo contains a series of scripts for interpolating data from a sparse grid onto a dense grid. Included are scripts for fitting several different 2d interpolation models (nearest neighbor, inverse distance weighting, GAM 2d-splines), and performing cross-validation to choose the best model to proceed with for interpolation. An example is provided using the following environmental variables: PPFD (µmol m-2 s-1), Air temperature (°C), Air humidity (%), Air VPD (kPa). The example data consists of these variables, collected every 15 minutes for one day in the PhenoArch phenotyping platform. The sparse grid consists of data collected from 10 sensors distributed in the platform, and the dense grid is a physical metric coordinate system to interpolate the environmental variables down to the individual pot-level spatial resolution.

Example

These scripts should be run in sequential order:

• 1_env_interpolation_model_comparison.R
• 2_env_interpolation.R
• 3_env_interpolation_plots.R