anomaly: anomaly

In caliver: Calibration and Verification of Gridded Model Outputs

Description

This function calculates the anomaly (number of standard deviations from the mean climatology) of a forecast layer.

Usage

anomaly(r, b)

Arguments

r	is the RasterLayer to compare to the climatology.
b	RasterBrick/Stack containing the historical observations or a proxy (typically a reanalysis) that is used to derive the climatological information.

Details

The objects r and b should be comparable: same resolution and extent. More information on anomaly is available here: https://bit.ly/2Qvekz4. To estimate fire climatology one can use hindcast or reanalysis data. Examples of the latter are available from Zenodo: https://zenodo.org/communities/wildfire.

Value

The function returns a RasterLayer with extent, resolution and land-sea mask matching those of r. Values are the number standard deviations from the historical mean values.

Examples

## Not run: 
  # Generate dummy RasterLayer
  r <- raster(nrows = 1, ncols = 1,
              xmn = 0, xmx = 360, ymn = -90, ymx = 90, vals = 0.3)
  names(r) <- as.Date("2018-01-01")
  # Generate dummy RasterBrick
  b <- raster::brick(lapply(1:(365 * 3),
                  function(i) raster::setValues(r, runif(raster::ncell(r)))))
  names(b) <- seq.Date(from = as.Date("1993-01-01"),
                       to = as.Date("1995-12-31"),
                       by = "day")
  # Compute anomaly
  x <- anomaly(r, b)

  # This plots nicely using rasterVis::levelplot(), see example on GWIS
  # (\url{https://gwis.jrc.ec.europa.eu}
  rasterVis::levelplot(x, col.regions = colorRamps::matlab.like(n = 11))

## End(Not run)

caliver documentation built on Feb. 19, 2021, 5:07 p.m.