Hydrologic Evaluation of the TMPA-3B42V7 Precipitation Data Set over an Agricultural Watershed Using the SWAT Model

Real-time availability of satellite-based precipitation products gives the hydrologic prediction community an opportunity to enhance hydrologic prediction for a watershed or river basin. In the present study, the latest tropical rainfall-measuring mission (TRMM) multisatellite precipitation analysis (TMPA) research product, 3B42V7, was assessed against a gauge-based Indian Meteorological Department (IMD) gauge-based gridded data set using statistical and contingency table methods for an agricultural watershed in India. A comparative analysis of the TRMM and IMD data sets was carried out on daily, monthly, seasonal, and yearly bases for 16years (1998-2013). The analysis revealed that the TRMM data set performed reasonably well but showed significant biases. Although it underestimated watershed-averaged daily precipitation (bias=-28.65%, correlation coefficient=0.42), its results improved for monthly precipitation (bias=-26.17%, correlation coefficient=0.86). Moreover, its rainfall detection capability was determined to be better during the monsoon season than during the nonmonsoon season. In all timescales, however, it usually underestimated heavy rains. The utility of the TRMM precipitation data set in hydrologic modeling was evaluated via the semidistributed soil and water assessment tool (SWAT) hydrologic model. Using the IMD gauged data-calibrated SWAT model, TRMM data set-based simulation showed limited hydrologic prediction on a daily scale, whereas its prediction capability was fairly good on a monthly scale. The SWAT model exhibited remarkable improvement in both daily and monthly simulation when recalibrated with TRMM precipitation data. However, when driven by IMD data, the model always performed better than its TRMM-driven counterpart. The analysis indicated that the use of the TRMM precipitation data set can be a compensating approach after suitable bias correction and that it has the potential for hydrologic prediction in data-sparse regions.