Evaluation of gridded multi-satellite precipitation (TRMM -TMPA) estimation performance in the Upper Indus Basin (UIB)

Data with acceptable gridded resolution of daily climatic variables are critical for hydrological and water resource modeling. The gauge-based interpolation methods usually used by hydrologic models in general do not cover the spatial heterogeneity of the variability of climatic variables in a catchment. The errors in the interpolated data fields have the potential to significantly bias hydrological model calibrations and water balances calculations. Two major issues, especially in these type of hydrometeorological studies, are the possible sparsity of the data sampling points (gauge stations), and the discontinuities in the data series or the quality of the temporal records. Resolving these issues usually comprise the first part of most climatological, environmental and hydrological studies. Nowadays an increasing number of gridded climatic products such as satellite or radar based gridded data are readily available which, though they have high spatial resolution, may show considerable errors and biases. Nevertheless, these gridded data sets can be used in conjunction with gauge based data to get better rainfall estimates. The present study aims to evaluate the capability of the “Tropical Rainfall Measurement Mission” (TRMM) “Multi-satellite Precipitation Analysis” (TMPA) in estimating appropriate precipitation rates in the Upper Indus basin (UIB) for subsequent use in hydrological watershed modeling in the wake of climate change and the dependency of the estimate performance on the time scale. To that avail statistical analyses and comparison of the TMPAproducts with gauge measurements in the UIB are carried out on monthly, seasonal and annual sums. The results show large biases in the TMPA(TRMM) precipitation estimates for the UIB, as well as high false alarm and miss ratios. The correlation of the TMPAestimates with ground-based gauge data increases considerably and almost in a linear fashion with increasing temporal aggregation, i.e. time scale. There is no sign of any uniform trends of underor overestimation of the gauge data by the TMPAestimates. In spite of these discrepancies between TMPAestimates and gauge data, the use of such TRMM-satellite products in hydrological watershed modeling, endeavored presently by the authors, may be a valuable alternative in datascarce regions like the UIB, but must be taken with a grain of salt. © 2015 The Authors. Published by Elsevier B.V. Peer-review under responsibility of organizing committee of ICWRCOE 2015. 2 Khan, Koch and Chinchilla / Aquatic Procedia 00 (2015) 000–000