Northern Prairie Wildlife Research Center
U.S. Fish and Wildlife Service, Northern Prairie Wildlife Research Center, Rural Route 1, Box 96C, Jamestown, ND 58401-9736
The amount of water held in individual wetland basins depends not only on local climate patterns but also on ground-water flow regime, soil type, vegetation, and surrounding land use. Most wetland basins in the northern prairies alternate between being wet (i.e., holding water) and dry. To assess the potential effect of climate change on the number of prairie wetlands holding water in a given year, one must first determine how much of the variability in number of wet basins is accounted for by climatic variables.
In this paper, I discuss the results of a modelling effort to assess the effect of climate on the number of wet basins in the prairie pothole region from 1972 through 1987. I used multiple linear regression to examine the relationship between climate variables and percentage of wet ponds. Independent variables included monthly, seasonal, and annual temperature and precipitation, and monthly snowfall in Canada. The effects of one-year lags were examined for the seasonal and annual data. Two indices were calculated, the Thornthwaite Moisture Index (1) and a Conserved Soil Moisture Index (2). All calculations were based on a year defined to be June-to-May.
I divided the study area into 3 parts, Canadian parkland, Canadian grassland, and Dakota grassland, and developed models separately for each area. For parkland, the best model explained 65% of the variation in percent wet basins:
where pond1 is the number of ponds counted the previous May, ttlpc is total precipitation for the pond-year, sprtma is mean maximum spring temperature, and falltma is mean maximum fall temperature. The best model for Canadian grassland differed from the parkland model, but still explained 63% of the variation in percent wet basins:
where fallpc1 is total precipitation during the fall previous to the pond-year and daprt is the difference between mean maximum and mean minimum temperatures in April. The Dakota model included the same variables as the Canadian grassland model, but magnitudes of the coefficients were different.
When temperatures were increased by 3°C in subsequent model runs, the Canadian grassland and Dakota models predicted only small depressions in the percentage of wet basins. The parkland model, however, indicated substantial decreases in the availability of wetlands with increased temperatures. One important consequence of this geographical difference in wetland response to temperature is the decreased probability of finding better conditions as waterfowl migrate farther north. During nearly half of the period 1972-1987, wetland conditions were better in the parkland than in the Dakotas. Under the scenario of increased temperature, wetland conditions in the parkland are predicted to be better than in the Dakotas in only 1 of the 15 years simulated.