Empirical Relationship between Large Dams and the Alteration in Extreme Precipitation

Abstract

This study looks at the empirical relationship between the presence of large dams and the potential alteration in extreme precipitation patterns in their vicinity. The global analysis indicates that extreme precipitation has altered considerably more than mean precipitation during the last century. We found this alteration to be more pronounced during the postdam period where the 99th percentile of precipitation experienced an average of 4% increase per year in magnitude. While the density of dams within a given radius did not correlate tangibly with the change in the percentile value, the frequency of rain (average number of rainy days per year) was found to have twice as much correlation during the postdam period than during the predam period. In general, dams in the regions of southern Africa, India, the Western United States, and Central Asia were found to have increased extreme precipitation more than other regions. It also appeared that large dams alter extreme precipitation patterns more in the arid/semiarid regions more than other places. The study confirms that the impact of large dams on extreme precipitation is clearly a function of surrounding mesoscale and land-use conditions and that more research is necessary to gain insights on the physical mechanisms of precipitation alteration by dams. What is needed hereafter to understand how a reservoir triggers changes in precipitation patterns and affects dam safety is a coupled land-atmosphere modeling approach. Due to the interactions of the atmospheric processes with surface water, understanding and predicting the effect that human-modified flood-frequency behavior has on sustainable dam design and reservoir operations cannot be achieved by stand-alone hydrologic-hydraulic models as has been historically pursued by the engineering profession.