Smooth components of wavelet decompositions at different scale levels. The upscaling is performed by the upscale function for variable aridity belonging to carlinadata data set. The data represent a square region. (Any region is extended to the next or next but one square of 2nx2n grid cells and is padded with predefined values, default is mean value, by the function provided. Thus the data recorded is available in a form that enables wavelet analyses.) Level=0 displays the raw, full-resolution predictor values, which are then "aggregated" by wavelets to ever coarser resolutions. Values increase from black to white. This function can be applied to any variable of interest, e.g. predictor, response or residuals.

  Part of: Carl G, Levin S, Kühn I (2018) spind: an R Package to Account for Spatial Autocorrelation in the Analysis of Lattice Data. Biodiversity Data Journal 6: e20760.