Gridded precipitation mapping in mountainous terrain combining GRASS and R

The article deals with the generation of gridded regional-scale precipitation maps for the mountainous Tyrol region (Eastern Alps), covering an area of 39,600 km(2). The authors consider mean annual and seasonal precipitation for the period 1961-1990. The input data were collected from 533 rain gauges. To strengthen the sparse high-elevation data network, the precipitation at the equilibrium line altitude (ELA) of glaciers was calculated using an empirical temperature-precipitation relationship. For the interpolation from points to grid cells the authors developed and applied the open source algorithm gradgrid4, combining GRASS GIS and R functions. The computation of mean annual precipitation for each cell includes (1) calculating local averages of precipitation and elevation, (2) calculating the local vertical precipitation gradient, and (3) combining the results of (1) and (2) to generate a precipitation map at a cell size of 250 m. Mean seasonal precipitation was computed from the percentage of mean annual precipitation for each season. The modelled mean annual precipitation for 86.1% of the evaluated data points ranges within +/- 20% of the observed values. They authors conclude that gradgrid4 is highly suitable for regional-scale gridded precipitation mapping in mountain areas, given a sufficiently dense network of high-elevation data points.