Distance to a contour of a covariate

Distance from every point to the nearest place where a covariate crosses a given value. Position relative to a contour is often what matters rather than the value itself — plankton distributions track the shelf break, and "20 km inshore of the 100 m isobath" locates that far better than "depth = 85 m" does.

Usage

distance_to_contour(env_dat, var, levels, per = c("km", "m"), prefix = NULL)

Arguments

env_dat

an sf POINT object with one row per location and time step, as datamatch's access functions return

var

covariate whose contours are wanted

levels

one or more values to contour

per

distance unit for the result: "km" (default) or "m"

prefix

stem for the new column names; defaults to var

Value

env_dat with one distance column per level. A level lying entirely outside the covariate's range in a time step gives NA for that step.

Details

A cell is treated as on the contour if level falls between its value and that of any neighbour, so the contour is found even though no cell sits exactly on it.

Examples

if (FALSE) { # \dontrun{
# Distance to the shelf break and two other isobaths
env <- distance_to_contour(env, "DEPTH", levels = c(50, 100, 200))
# -> DEPTH_dist_50, DEPTH_dist_100, DEPTH_dist_200

# Works on any field: distance to the 10 degree isotherm
env <- distance_to_contour(env, "SST", levels = 10)
} # }