Assessment of Satellite-based Precipitation Products in Hydrologic Modeling: A Case Study from the Oum Er Rbia Watershed, Morocco


Racha El Kadiri, Western Michigan University


Precipitation is the most important forcing parameter in hydrological modeling, yet it is largely unknown in the arid Middle East. We assessed the magnitude, probability of detection, and false alarm rates of various rainfall satellite products (e.g., TRMM) compared to in situ gauge data (~30 stations) across the Our Er Rbia Watershed in Morocco. The basin is one of the largest watersheds in the country with an area of 35,000 km2. Precipitation over the basin is relatively high with an average of 268mm/year according to TRMM (1998-2008). The existing gauges indicate that the average annual precipitation across the Tadla and Coastal Plains region is 260mm/year and 390mm/year across the Atlas Mountains.

Following the assessment of satellite products against in situ gauge data, we evaluated the effects (e.g., runoff and recharge amounts) the various rainfall products due to difference in spatial and temporal resolution had on rainfall-runoff models (SWAT). Specifically, we performed a four-fold exercise: (1) The first stage focused on the analysis of the rainfall products; (2) the second stage involved the construction of a rainfall-runoff model using gauge data; (3) the third stage entailed the calibration of the model against flow gauges and/or dams storage variability, and (4) model simulation using satellite-based rainfall products using the calibrated parameters from the initial simulation.

The preliminary results evaluate the potential for using publicly available remote sensing datasets in lieu of field gauges for data sparse and inaccessible regions. This will address one of the major difficulties facing hydrologists while constructing representative rainfall-runoff models in the absence of field data as it is the case of most of the North African watersheds.