Catskill mountain water resources: Vulnerability, hydroclimatology, and climate-change sensitivity

We present an initial assessment of the potential impact of climate change on water supply in the Metropolitan East Coast (MEC) region of the U.S. National Assessment of the Potential Consequences of Climate Variability and Change. A version of the Thornthwaite water-balance model is applied to one of six basins in the Catskill Mountains that together provide water for approximately 10 million people in New York City and other municipalities. In addition to Thornthwaite's original soil moisture reservoir, the model includes the snow pack water reservoir of Willmott, Rowe, and Mintz (1985), a ground-water storage term, and several additional modifications. Following a review of the vulnerability of water supplies and historical hydroclimatology of this region, we estimate (1) the sensitivity of water supply to altered temperature and precipitation regimes and (2) the potential impacts of specific climate-change scenarios used by national and regional climate,change assessments. The sensitivity of runoff to temperature changes is approximately 6 percent per degree C; its sensitivity to precipitation changes is approximately 1.5-2 percent per percent change in precipitation, for annual mean values. Under all scenarios, rising temperatures will lead to significantly diminished water supplies unless precipitation increases dramatically. Due to disagreement between precipitation projections from different models and scenarios, projected changes in mean annual water supply range from approximately +10 percent to -30 percent by the 2080s. Under the driest scenario, water supplies under mean climatic conditions will be comparable to the worst extended drought period of the twentieth century in this region. Equally important are the likely effects on the annual cycle, which include an earlier peak runoff and a reduction of the snowpack by at least 50 percent. Considered in the context of likely increased demands, these changes may be significant.