A Multi-Scale Approach to Assessment of Climate Change Impacts on Hydrologic Response of Watershed Systems: 1. Quantifying Regional Trends and their Uncertainty

The current ways of assessing the impacts of climate change on watershed systems are inadequate: they are based on an ad hoc selection of climate models; they focus on metrics at very coarse scales (hundreds-to-thousands of square km) detached from the reality of human activities and ecosystem services at the local (often the stream reach or floodplain) scales; and they do not yield any assessment of uncertainty associated with watershed modeling and projections into the future. This research aims to bridge the coarse-scale projections and the multi-scale, space-time connectivity of watershed systems, propagating information on climate signals through the entire watershed system: from headwater areas to stream channels, and to the details of flow characteristics. A comprehensive program of modeling and field observations is developed. The project will focus on the state of Michigan, where a number of observed metrics already demonstrate trends consistent with a warming climate, including shorter winters, higher mean annual temperatures, and higher frequency of heavy precipitation events. A number of case studies is developed along the state-wide gradient of climate and hydrological regimes. The premise used is that watersheds of 1st- to- 2nd orders can be used as "warning systems" (sentinels) of climate change for the hydrological regimes of drainage basins of higher orders. Multi-model ensembles of climate change projections from the World Climate Research Programme's Coupled Model Intercomparison Project are downscaled for the locations of these basins using an "in-house" stochastic downscaling methodology. The inferred probabilistic information is used to assess climate change impacts on watershed systems for early (2010-2039), mid- (2040-2069), and late century (2070-2099) periods. Changes in first- and second-order moment properties of essential characteristics of hydrological regimes are investigated for the sentinel basins using a multi-scale, physically-based framework of modeling watershed surface/subsurface processes and flow hydrodynamics, integrated through a capability of Nested Dynamics Modeling.