USGS Home
Contact USGS
Search USGS

Agriculture is the predominant factor affecting the landscape in the PPR (Kantrud et al., 1989). Wetland quality and function are degraded primarily by agricultural activities associated with annual crops, including drainage, cultivation of wetland basins, siltation due to soil erosion from adjacent cropland, and agricultural chemicals (Kantrud et al., 1989). Agricultural activities affecting upland habitats also affect duck populations (Bethke and Nudds, 1995) and production (Johnson et al., 1987) at local and landscape scales. Most dabbling ducks nest in grasses, forbs, or shrubs, and the success of their nesting effort is affected by the availability and type of cover (Johnson et al., 1987; Klett et al., 1988). Perennial grassland habitat has been lost and what remains is fragmented in areas where cropland is abundant. Greenwood et al. (1995) found that duck nest success in the Canadian PPR was negatively correlated with the amount of cropland present.

In 1989, the U.S. Environmental Protection Agency (EPA) initiated the Environmental Monitoring and Assessment Program (EMAP) to provide quantitative assessments of the status and long-term trends in the condition of ecological resources on a regional and national scale (Peterson, 1994). The ecological significance of the PPR and the stress imposed on this ecosystem by conversion of grassland to cropland was the reason the EPA selected the PPR for development and evaluation of ecological indicators to monitor wetland condition (Cowardin and Peterson, 1997). The first stage of EMAP (Prairie Pothole Pilot I Study) evaluated the performance of selected landscape- and field-level variables as indicators of environmental quality of wetlands and as tools to discriminate between wetlands in highly disturbed (crop agricultural) and less disturbed (grassland) landscapes (Peterson et al., 1997). One component of the Pilot I study evaluated estimated numbers of breeding ducks as an indicator of landscape conditions (i.e. wetland basin conditions and upland habitat availability; Cowardin and Sklebar, 1997). In that study, habitat data derived from aerial video and counts of ducks were used as input to models to predict the annual combined breeding population of five species of dabbling ducks: mallards, blue-winged teal (Anas discors), gadwalls (A. strepera), northern shovelers (A. clypeata), and northern pintails (A. acuta). The definition of landscape condition was dependent on the ratio of cropland area to total upland area; poor-condition areas had high ratios of cropland to upland area (median value = 91%) and almost no grassland cover. This definition, established for the Pilot I study, was based on the perception that wetlands in a complex containing predominantly cropland would probably be in a more degraded state than those containing predominantly grassland (Cowardin, 1997:11-12).

The Pilot I study showed that the estimated combined breeding population of the five duck species was an indicator that could separate two extremes in landscape condition (high vs. low ratio of cropland to total area of upland; Cowardin and Sklebar, 1997). More ducks were calculated on grassland-dominated study areas than on cropland-dominated study areas (P = 0.0038, 1992 and P < 0.001, 1993). The differences between the two extreme landscape conditions were apparent despite the fact that Cowardin and Sklebar (1997) observed no association between duck pairs and number of wetland basins as would be expected. Numbers of breeding pairs therefore showed promise as an indicator of landscape condition.

Our study is the next stage of EMAP studies, Prairie Pothole Pilot II. We used a larger data set to build upon the results of the Pilot I studies and to test hypothesized relations between indicators of landscape and wetland condition and waterfowl abundance (an indicator for bird communities). Our objective was to evaluate the use of duck numbers as indicators of landscape condition, where landscape condition is defined as the ratio of cropland area to total upland area.