Loads the predictor importance (PI) data from the pi-pd.db sqlite database. PI estimates are provided for each stixel over which a model was run and are identified by a unique stixel ID in addition to the coordinates of the stixel centroid. PI estimates are for the occurrence model only.

load_pis(path, ext, return_sf = FALSE)



character; directory that the Status and Trends data for a given species was downloaded to. This path is returned by ebirdst_download() or get_species_path().


ebirdst_extent object; the spatiotemporal extent to filter the data to. The spatial component of the extent object must be provided in unprojected, latitude-longitude coordinates.


logical; whether to return an sf object of spatial points rather then the default data frame.


Data frame, or sf object if return_sf = TRUE, containing PI estimates for each stixel for both the occurrence and relative abundance models. The data are provided in a 'wide' format, with each row corresponding to the PI estimates for a give stixel for the occurrence count model, and the relative importance of each predictor in columns. Stixels are identified by a unique stixel_id, the centroid of the stixel in space and time is specified by the lat, lon, and date column, which expresses the day of year as a value from 0-1.


if (FALSE) {
# download example data
path <- ebirdst_download("example_data", tifs_only = FALSE)
# or get the path if you already have the data downloaded
path <- get_species_path("example_data")

# load predictor importance
pis <- load_pis(path)

# plot the top 15 predictor importances
# define a spatiotemporal extent to plot data from
bb_vec <- c(xmin = -86.6, xmax = -82.2, ymin = 41.5, ymax = 43.5)
e <- ebirdst_extent(bb_vec, t = c("05-01", "05-31"))
plot_pis(pis, ext = e, n_top_pred = 15, by_cover_class = TRUE)