Seminar : A Dynamic Risk Management Framework for Water Resources Science

Posted on: January 9, 2017

Speaker: Arun Ravindranath

Date: January 20th, 2017

Time: 2 pm

Location: Steinman Hall, Grove School of Engineering, City College

Water is arguably the most important resource on earth. Much of a country’s GDP, and hence the prosperity of its people, is reliant on its water resources and water systems. Water plays an irreplaceable role on this planet in general. As with any resource, it is limited, and is vulnerable to the forces of change that bear their weight on water systems, which, broadly speaking, are non-stationary, dynamically shifting risks and vulner- abilities that can be attributed to variabilities in climate and societal conditions. Systematic long-term varia- tions in climate characteristically display quasi-periodic behavior on the interannual, decadal and multi- decadal scales. Such shifts in climate regimes manifest as long periods of wet or dry hydroclimatic condi- tions in various regions across the globe. Water systems are impacted in crucial ways as a result of such vari- abilities in climate, and this impacts everything that is served by these water systems. The decisions and poli- cies of human societies also impact water systems in significant ways. It is therefore essential that any mean- ingful approach to water systems management and water resources allocation take these risks into account. The information resulting from such an approach can be used to better inform water policy, risk manage- ment, infrastructure investments and water allocation schemes. We are concerned in particular with the sys- tematic variations in streamflow resulting from natural and anthropogenic influences. We assert that climate variability and impacts on natural resources are crucial factors affecting streamflow globally and that such multi-time scale variability and change increases risk to water sustainability and water systems. Dam opera- tions, a necessary means for collecting the water brought by streamflow without jeopardizing downstream ecosystems, are governed by rules and policies that are contingent on the highly variable behavior of stream- flow. Hence, we need to understand how to develop a holistic dynamic risk management system that will inform future infrastructure investments, intra- and interstate water allocation schemes, water policies and improve water and environmental sustainability in spite of changing climate and societal conditions. The first step in such an undertaking is to examine the variations in streamflow resulting from climate and socie- ty. Low-frequency modes of variability in climate change over long time scales, and streamflow records are often comparatively short. Streamflow reconstructions and hindcasts are therefore necessary in extending streamflow records, and we present a network model with a hierarchical Bayesian regression framework em- bedded within it as a novel reconstruction method. The reconstructions are informed by regional tree ring chronologies as well as feeder streamflow sites. Our results for selected sites in the Missouri River Basin show good reliability in streamflow reconstructions and adjusted R-squared values as high as 0.97. A thor- ough analysis of water policy in the Delaware River Basin has also been done, with an understanding that we will eventually aim to connect the modeling and quantitative aspect of the work with our understanding of the policy aspect of the work. This is the general idea driving the dynamic risk management perspective dis- cussed in this abstract.

Download Abstract/Flyer here