fassf

An R package with various improved and original functions for spatial analysis in R. These are mostly wrappers that improve or add functionality. Greatly improves speed and efficiency of various spatial operations.

installation: devtools::install_github("noahdasanaike/fassf/fassf")

Examples:

• obtain all respective values from a global 1km-level tiff file for 40,000 sf polygons in 8 minutes (w/o exceeding 1gb of memory usage)
• calculate distance to nearest of 3,200 sf polygons for each of 32,000 sf polygons in 3 minutes
• obtain average outcome values for 100 sf polygons across 18,000 intersecting sf polygons in <1 second

Current functionality:

• faster averages (or measure functions) from spatial intersections
  • intersection_fun(need_df, have_df, "var", fun = "mean") or intersection_mean(want_df, have_df, "var")
• obtain values from raster for polygons
  • need_raster_means_sf <- raster_polygon_values("values.tif", need_raster_means_sf, fun = "mean")
• faster match each x with a y polygon
  • need_labels_df$labels <- intersection_match(need_labels_df, labels_df, "labels")
• faster st distance
  • need_distance$distance <- faster_st_distance(need_distance_df, have_distance_sf)
• add latitude and longitude columns
  • need_coords_sf <- add_lat_lon(need_coords_sf)
• ruggedness calculations (median absolute deviation)
  • need_ruggedness_sf$ruggedness <- ruggedness_mad(need_ruggedness_sf)
• small area nightlight density estimates
  • need_nightlight_sf <- nightlight_estimates(1992:2021, need_nightlight_sf)
• generate voronoi polygons for coordinates bounded by a polygon
  • need_voronoi_sf$geometry <- better_voronoi_polygons(need_voronoi_sf, bounding_polygon_sf)
• geocode query with geocode.xyz and openstreetmap
  • addresses <- osm_query_sf(c("1737 Cambridge St, Cambridge, MA", "20 Bradston St, Boston, MA")) * note: latter function occasionally returns invalid (broken) geometry polygons, working to fix
• faster st filter
• query openstreetmap, get polygons (see below)
• drive time calculation between points with openstreetmap
• split polygon by line
• rotate polygon

Examples:

• Obtain all "highway:residential" values from OpenStreetMap, drop values with missing names, and save to "roads_usa.RDS" by census-designated place geometries, adding a column for the name of the latter:

osm_data_sf(usa_localities,
            shape_names = usa_localities$placens,
            quiet = FALSE,
            key = "highway",
            value = "residential",
            drop_na = "name",
            save = "road_names/roads_usa.RDS")