downscale: Perform EOT-Based Spatial Downscaling

In environmentalinformatics-marburg/ESD:

Description

Perform EOT-based spatial downscaling based on user-defined predictor and response domains.

Usage

downscale(pred, resp, neot = 1L, newdata, nprd = 1L, var = NULL,
  filename = "", format = "GTiff", overwrite = TRUE, ...)

Arguments

pred, resp	Raster* objects used as predictor and response domains during EOT analysis.
neot	integer. Number of EOT modes to calculate.
newdata	Raster* object passed to predict.
nprd	integer. Number of EOT modes used for prediction.
var	numeric. Minimum amount of variance to be explained by the EOT modes. If specified, 'nprd' is adjusted automatically.
filename, format, overwrite	Arguments passed to predict and its call to writeRaster.
...	Additional arguments passed to eot.

Value

A Raster* object.

Author(s)

Florian Detsch

References

Appelhans T, Detsch F, Nauss T (2015) remote: Empirical Orthogonal Teleconnections in R. Journal of Statistical Software 65(10), 1-19, doi:10.18637/jss.v065.i10. Available online: https://www.jstatsoft.org/article/view/v065i10.

Examples

## raw data
plot(albGIMMS[[49:52]])

## downscaled data
prd <- downscale(pred = albGIMMS[[1:48]], resp = albMODIS[[1:48]], neot = 10L,
                 newdata = albGIMMS[[49:96]], var = 0.85)
plot(prd[[1:4]])                  

environmentalinformatics-marburg/ESD documentation built on May 16, 2019, 7:49 a.m.