Plot how much data each time step actually has

Satellite ocean colour is missing wherever cloud, ice, or low sun angle blocked the view, and those gaps are not spread evenly — they cluster in particular seasons and latitudes. A series can look complete in a table and be mostly empty in winter.

Usage

plot_coverage(env_dat, vars = NULL, main = "Data coverage by time step", ...)

Arguments

env_dat

an sf POINT object from accessEnvDat()

vars

which variables to show; NULL uses all covariate columns

main

plot title

...

passed to graphics::plot()

Value

a data frame of the plotted coverage, invisibly, with one row per time step per variable

Details

This plots the fraction of cells carrying a value in each time step, which is the thing to look at before trusting a monthly mean or deciding whether fill_satellite_gaps() is worth the seam it introduces.

See also

fill_satellite_gaps(), upscale_time(), whose min_coverage argument acts on the same quantity

Examples

if (FALSE) { # \dontrun{
chl <- accessEnvDat(vars = "CHL", years = 2010, months = 1:12, bounding_box = bb)

plot_coverage(chl)
# The winter months are the ones to be careful with.
} # }