Pixelwise Correlation-Based Landscape Classification (PiCo)(LAR-19015-1)

PiCo was written in attempt to automate and regionalize the Climate Landscape Response (CLaRe) metrics developed by Wallace et al 2016. The CLaRe metric system was created to map invasive buffelgrass in the southwestern United States. This grass both propagates and benefits from increased wildfire events, and is a threat to the local ecosystems within the Sonoran Desert. Buffelgrass responds to precipitation quicker than native vegetation. This behavior is what CLaRe captures. Correlation values derived from regression analysis between Normalized Difference Vegetation Index (NDVI) and precipitation values are used to separate pixels invaded by buffelgrass from those that are not. PiCo, written in R, performs a pixelwise regression analysis to produce rasters whose correlation values can be evaluated to target buffelgrass.