Compute block covariance matrix

Calculates the covariance matrix for a block based on a variogram model, discretising the block into pixels and computing covariances between all pairs of points within the block.

Usage

blockCorr(lat, rsl, model, nx, ny)

Arguments

lat

Latitude in degrees.

rsl

Resolution vector with x and y resolution in degrees (e.g., c(xres, yres)).

model

A variogram model object from the gstat package.

nx

Number of pixels in the x-direction.

ny

Number of pixels in the y-direction.

Value

A numeric matrix representing the covariance between all pairs of points within the block.

Examples

 #Compute covariance matrix for a 5x5 pixel block:
  library(gstat)
  vgm_model <- vgm(psill = 1, model = "Sph", range = 100, nugget = 0.1)
  res <- c(0.02, 0.02)  # 0.02-degree resolution
  result <- blockCorr(lat = 40, rsl = res, model = vgm_model, nx = 5, ny = 5)
  dim(result)  # Should be 25 x 25
#> [1] 25 25