npphen: an R-package for non-parametric reconstruction of vegetation phenology and anomaly detection using remote sensing

For ecologists, the challenge at using remote sensing tools is to convert spectral data into ecologically relevant information like abundance, productivity or traits distribution. Among these features, plant phenology is one of the most used variables in any study applying remote sensing to plant ecology and it has formally considered as one of the Essential Biodiversity Variables. Currently, satellite imagery make possible cost-efficient monitoring of land surface phenology (LSP), but methods applicable to different ecosystems are not available. Here, we introduce the npphen R-package developed for remote sensing LSP reconstruction and anomaly detection using non-parametric techniques. The package implements basic and high-level functions for manipulating vector and raster data to obtain high resolution spatial and temporal LSP reconstructions. Advantages of npphen are: its flexibility to describe any LSP pattern (suitable for any ecosystem), it handles time series or raster stacks with and without gaps, and it provides confidence interval for the expected LSP at yearly basis, useful to judge anomaly magnitudes. We present two study cases to show how npphen can successfully reconstruct and map LSP and anomalies for contrasting ecosystems.