Validation and Intercomparison of Satellite Rainfall Products over Guiana Shield

Abstract:

Four satellite-based rainfall estimate algorithms TRMM-TMPA 3B42 (Tropical Rainfall Measuring Mission Multi Satellite Precipitation analysis) V7 (Version 7) and RT (Real Time), PERSIANN (Precipitation Estimation from Remote Sensing Information using Artificial Neural Network) and CMORPH (Climate Prediction Center MORPHing technique) are evaluated at daily and monthly time scales, and a spatial resolution of 0.25° longitude/latitude. The reference data come from to rain gauges network over French Guiana and North Brazil. The comparison between the products is obtained with quantitative and qualitative statistical analysis. Results are discussed in terms of the accuracy of the products, the effect of climate variability, and differences between products. The validation and intercomparison of these products is done for the whole as well as different parts of the area. Validation results are reasonably satisfactory for daily and monthly rainfall with a correct match in spatial distribution in rainfall between regions. There is a better correlation for monthly precipitation. On average the bias vary from -0.55 mm to -2.05 mm for daily and from -16.44 mm to -62.04 mm for monthly, the root mean square error vary from 15.15 mm to 16.06 mm for daily and from 159.62 mm to 172.65 mm for monthly. All products underestimate the strong rainfall regimes and overestimate low rainfall regimes. Structures in the occurrence of rainfall are better represented than structures in rainfall amounts. TRMM-V7 and PERSIANN show better performance, CMORPH and TRMM-RT lower, although the differences between products are very low.