Converting Point Estimates of Daily Rainfall onto a Rectangular Grid

Rainfall measurements are commonly observed at point locations and these values are required inter alia by Hydrologists and Engineers to be converted to a raster format. The most common method that is used is the inverse distance weighting interpolation technique. This technique uses the location and magnitude of the rainfall to determine estimates of rainfall at unmeasured locations.

South Africa occupies an areal extent of approximately one million square kilometres. Daily rainfall is measured, in real-time using telemetry, at approximately 170 locations in South Africa and at a further 1,900 locations daily that are then assimilated monthly into a national daily rainfall database by the South African Weather Bureau.

Areal estimates of daily rainfall are of paramount importance especially in water-poor countries such as South Africa. Researchers often require that an estimate of rainfall at ungauged points are determined in order for them to assess the spatial distribution of the rainfall amounts.

The techniques, using the Arc/Info GIS, will be discussed in this paper to assist researchers in converting point estimates of daily rainfall to a raster surface that can be used to describe the daily rainfall in an areal context.