92nd American Meteorological Society Annual Meeting (January 22-26, 2012)

Thursday, 26 January 2012
Modeling the Major Pakistan Floods of 2010: A Coupled Atmospheric-Hydrologic Forecasting Scheme for the Indus Valley
Hall E (New Orleans Convention Center )
Kristofer Shrestha, Georgia Institute of Technology, Atlanta, GA; and V. E. Toma and P. J. Webster

We investigate the predictability of 1-15-day river discharge and the potential for flooding over the Indus Valley. Using the European Centre for Medium Range Weather Forecasts (ECMWF) 15-day Ensemble Prediction System (EPS), earlier studies showed that the heavy rainfall events of July-August 2010 that affected the regions of northern Pakistan were predictable in both magnitude and timing with a high probability 6-8 days in advance. The present study focuses on a coupled atmospheric-hydrologic forecasting scheme that includes the statistically rendered ensemble meteorological parameters forecasts from the ECMWF model as input variables for the Water Budget Record from Variable Infiltration Capacity (VIC: 2) macroscale hydrologic model. We test the sensitivity of the coupled probabilistic model by attempting to forecast the major floods in Pakistan during July-August 2010.

Earlier studies aimed at forecasting Bangladesh flooding used a simpler hydrological model. However, given the more complex terrain, the VIC routing model incorporates subgrid variability in the soil moisture storage capacity, drainage from the lower soil moisture zone as a nonlinear recession, and the inclusion of topography allowing for orographic precipitation/temperature lapse rates resulting in more realistic hydrology in mountainous regions. Runoff transport is assumed to be linear, causal, stable, and time invariant with a non-negative unit hydrograph. Integrations over the course of the 2010 monsoon period show forecasts of plausible flooding first in the North Pakistan and later flood plain inundation in southern Pakistan. The results show that Pakistan would benefit from the implementation of an operational hydrological forecasting scheme and warning system, with high correlations between the observed river discharge and the forecast discharge up to 6-8 days in advance. The warning and communication system could be based on the Bangladesh system that became operational in 2007.

Supplementary URL: