POTENTIAL EFFECTS OF GLOBAL WARMING ON WATERFOWL POPULATIONS BREEDINGIN THE NORTHERN GREAT-PLAINS

The Prairie Pothole Region (PPR) of the Northern Great Plains is the m ost important breeding area for waterfowl in North America. Historical ly, the size of breeding duck populations in the PPR has been highly c orrelated with spring wetland conditions. We show that one indicator o f climate conditions, the Palmer Drought Severity Index (PDSI), is str ongly correlated with annual counts (from 1955 to 1996) of both May po nds (R-2 = 0.72, p < 0.0001) and breeding duck populations (R-2 = 0.69 , p < 0.0001) in the Northcentral U.S., suggesting the utility of PDSI as an index for climatic factors important to wetlands and ducks. We then use this relationship to project future pond and duck numbers bas ed on PDSI values generated from sensitivity analyses and two general circulation model (GCM) scenarios. We investigate the sensitivity of P DSI to axed changes in temperature of 0 degrees C, +1.5 degrees C, +2. 5 degrees C, and +4.0 degrees C in combination with fixed changes in p recipitation of -10%, +0%, +7%, and +15%, changes spanning the range o f typically-projected values for this region from human-induced climat ic change. Most (11 of 12) increased temperature scenarios tested resu lt in increased drought (due to greater evapotranspiration under wanne r temperatures) and declining numbers of both wetlands and ducks. Assu ming a doubling of CO2 by 2060, both the equilibrium and transient GCM scenarios we use suggest a major increase in drought conditions. Unde r these scenarios, Northcentral U.S. breeding duck populations would f luctuate around means of 2.1 or 2.7 million ducks based on the two GCM s, respectively, instead of the present long-term mean of 5.0 million. May pond numbers would fluctuate around means of 0.6 or 0.8 million p onds instead of the present mean of 1.3 million. The results suggest t hat the ecologically and economically important PPR could be significa ntly damaged by climate changes typically projected. We make several r ecommendations for policy and research to help mitigate potential effe cts.