How the range of each response changes between two projections

Counts the cells a response is lost from, kept in and gained, between a map for the present and a map for another period or scenario, as BIOMOD_RangeSize() does. A cell is in the range where the response is predicted present.

Usage

range_change(now, later, threshold = NULL)

Arguments

now, later: The two maps: a terra::SpatRaster with a layer per response, such as project() with type = "binary" returns, or a matrix of cells by responses. Layers are matched by position and must be as many.
threshold: NULL where the maps are already 0 and 1. Otherwise one cut per response, or one for all, at or above which a cell is present; the cuts decision_threshold() learns for a fit are what to give.

Value

A timesift_range_change: a list with the table, one row per response, and the map, a raster or a matrix of the codes above, of the shape of the inputs.

Details

With L the cells lost, K the cells kept, G the cells gained and A the cells absent in both: the current range is L + K, the later one K + G, percent_loss is 100 L / (L + K), percent_gain is 100 G / (L + K) and change is percent_gain - percent_loss, the change in range size as a share of the current one. A cell that is NA in either map is left out of every count.

The map codes a cell as biomod2 does: -2 lost, -1 kept, 0 absent in both and 1 gained.

Examples

now <- cbind(sp1 = c(1, 1, 1, 0, 0, NA), sp2 = c(0, 0, 1, 1, 1, 1))
later <- cbind(sp1 = c(1, 0, 0, 1, 1, 1), sp2 = c(0, 0, 1, 1, 0, 0))
range_change(now, later)$table