New Publication by Scott Steinschneider Garners "Editor's Choice" Selection in Journal of Water Resources Planning and Management

A new study led by NE CASC Affiliated Investigator Scott Steinschneider, a faculty member at Cornell University, has been recognized as an Editor’s Choice selection in the Journal of Water Resources Planning and Management. The publication, "Influence of Subseasonal-to-Annual Water Supply Forecasts on Many-Objective Water System Robustness under Long-Term Change,” explores a range of issues connected to the use of subseasonal-to-seasonal hydrologic forecasts to enhance the performance of water systems in the context of climate change. 

Water systems can improve the services they provide to people and ecosystems by using subseasonal-to-seasonal (1-12 month) hydrologic forecasts to guide operations. However, little is known about whether the use of subseasonal-to-seasonal forecasts can help water systems more flexibly adapt to longer term (decadal) climate change; whether such a strategy could maintain adequate levels of service for multiple stakeholder groups with conflicting interests; and whether the answers to these questions depend on the lead time and skill of the forecasts themselves.

These are the questions explored in Steinschneider’s work through a case study on the outflow management plan of the Lake Ontario - St. Lawrence River system. The article uses optimization techniques to find operating policies (i.e., dam release rules) that are tailored for the historical record of hydrology and forecasts at different lead times and levels of skill. These policies are designed to balance tradeoffs between six different stakeholder groups, including upstream and downstream flooding, hydropower, commercial navigation, recreational boating, and ecosystem health. Steinschneider’s team then tested whether these policies, originally trained on the historic record, perform well under a large set of hydrologic scenarios that reflect plausible future climate conditions. 

Results show that policies which use longer-lead and nosier forecasts are better able to navigate a wide range of future hydrologic conditions. This counterintuitive outcome highlights the potential to overfit operating policies to historical information and perfect foresight, so that the policy does not work well when faced with previously unseen hydrologic conditions. The study also finds that for some stakeholder interests, trends in simple hydrologic indicators can be a useful way to determine when changes in system management should be triggered to adapt to changing climate conditions. However, this is not the case for all stakeholder interests in the system, suggesting that for some interest groups, it might be hard to change system management to preserve water services as climate change unfolds.