ABSTRACT Within the R ecosystem, packages like terra and sf are the go‐to solutions for most geospatial analyses, yet can struggle with large rasters and vectors. The Geographic Resources Analysis Support System , or GRASS , offers solutions that are often more efficient for large data. However, using GRASS through R requires users to become familiar with GRASS ‐specific syntax and data constructs. The fasterRaster package for R seamlessly connects to GRASS and enables analysis of large data sets. Modeled after the functions in terra , fasterRaster possesses over 200 methods for processing rasters and spatial vectors. fasterRaster also contains a growing number of specialty functions for hydrological, remote sensing, and topographical analysis. For small spatial objects, terra and sf will nearly always be faster, but for larger objects, fasterRaster can be several times faster, and for very large objects, it can succeed where other solutions fail. A pkgdown website contains documentation and a tutorial: https://adamlilith.github.io/fasterRaster/index.html .
Adam B. Smith (Thu,) studied this question.