Characterizing the impacts of water resources infrastructure, humans, and hydrologic nonstationarity on changes in flood risk across the Himalaya region

Wednesday, 17 December 2014: 4:45 PM
Desiree D Tullos, Oregon State University, Corvallis, OR, United States
As flood control infrastructure reaches its design life, and climate change, population growth, and urban migration increase flood risk, the historical paradigm of store-then-release floodwaters behind rigid infrastructure is of decreasing physical and socioeconomic value. Instead, a new paradigm of sustainable flood management is emerging, which can be framed in the context of three elements that can contribute to and/or mitigate flood risk: 1) water resources infrastructure, 2) policies and socioeconomics, and 3) changing climates and land use. In this presentation, I present the results of analysis on the role of these three elements in contributing to flood risk of the Sutlej River (India) and the Koshi River (Nepal) basins for six historical flood events. The Himalaya region was selected based on the a) increasing intensity of monsoonal rains, b) increasing prevalence of glacial lake outburst floods, c) water resources management that achieves short-term development goals but lacks long-term sustainability, and d) other socio-economic, environmental, and geopolitical factors.

I develop and apply a flood risk management framework that is based on metrics for characterizing the losses associated with the three elements contributing to major floods in the Himalaya region. Derived from a variety of data sources, results highlight how, across different hydrogeologic settings and various flood magnitudes, the largest influences on high flood losses are associated with inflexible water resources infrastructure and inappropriate development and flood management policies. Particularly for the most destructive events, which are generally associated with landslides and other natural hazards in this region, the effectiveness of some types of traditional and inflexible flood management infrastructure, including large dams and levees, is limited.

As opposed to the probability of a particular flood event, findings illustrate the importance of the damages side of the flood risk equation, which is often the most controllable but disregarded element of flood risk management. In addition, results lead to a hypothesized matrix of appropriate flood management strategies for the types of flood events that occur in the hydrogeology and cultural settings of high mountain areas and the lowlands to which they drain.