TR(BR)-126 : Development of disaggregation techniques

Abstract

In several analysis we do come across a situation when a relationship between the basin or region average value of precipitation and individual station values of precipitation located within/ outside the basin is needed.

Present report deals with various statistical techniques based on frequency analysis approach, correlation concept and disaggregation techniques which are developed to establish a statistical correlation between historical precipitation value of a station and it's corresponding average basin value. In all, ten methods have been developed and tried and results compared. Monthly precipitation data of Sher and Damanganga basins is used in the analysis. Disaggregated values when averaged, match well with the original average basin values. Outcomes of different methodologies are compared based upon different error criteria and results show that methods based upon frequency analysis approach and disaggregation technique perform well.

These two techniques are used to interpolate GCM output (monthly precipitation values) i.e. average value corresponding to 2xCO2 enriched atmosphere, which are generally available at grid points only, to local scale i.e. station values.

The methodology, as discussed in the report may prove a handy tool to deal with spatial disaggregation of precipitation and may be extended for other hydrological parameters as well.