Spatial Statistical Model for the Optimal Combination of Direct and Indirect Precipitation Measurements

Abstract:

Precipitation is highly variable in space and time. It has traditionally been measured with rain-gauges at point locations and the records are overwhelmingly daily. Drastic reductions in gauge networks are being experienced worldwide and many are plagued by error. On the other hand different methods such as radar or satellites provide indirect and partly erroneous information on precipitation. Due to the high variability it is essential to use all possible available sources for the reliable estimation of precipitation at ungauged locations. A general spatial statistical model is presented which can be used:

• to infill precipitation in the form of probability distributions for locations with incomplete records
• to interpolate precipitation from infilled and incomplete records
• to reflect the influence of topography and other deterministic features on precipitation
• to consider indirect measurements in the form of covariates and/or constraints.

The method is using Gaussian copula based constrained simulation, and offers a very flexible combination of the different sources of information. As simulation methodology it can be directly used for uncertainty assessment. Examples from South-West Germany and from South Africa are used to demonstrate the methodology.