Northern Prairie Wildlife Research Center
Just as the remains of aquatic invertebrates can be used to identify temporary and seasonal wetlands in the prairie pothole region (Euliss et al. 2001), they can also be used to refine delineations of these difficult-to-recognize wetland classes. Since aquatic invertebrates are strictly limited to areas of standing water and hydric soils and hydrophytes require only saturated soil conditions, it follows that invertebrate delineations will not overestimate wetland extent (i.e., they will fall at or below the elevation of standard delineations). Our delineations of the wetland edge using invertebrates were consistently at the same location as the standard delineations or closer to the wetland center for both seasonal and temporary wetlands in grassland; in grassland wetlands, we never found the remains of aquatic invertebrates outside of the wetland basin as delineated using standard indicators. Thus, while the presence of aquatic invertebrate remains is a reliable wetland indicator, the absence of remains does not always mean that the sample was collected outside of a wetland as defined by the standard delineation.
In wetland basins disturbed by agricultural tillage, the conservative relationship between the invertebrate delineations and the standard delineations broke down; many of our invertebrate delineations of seasonal and temporary wetlands in cropland were on the upland side of our standard delineations. A simple explanation is that invertebrate remains were transported higher in basins during the tillage process. However, additional factors may have contributed to the differences in delineations of grassland and cropland wetlands. Euliss and Mushet (1996) found that water-level fluctuations (the difference between maximum water level and minimum water level during a growing season) were greater in wetlands with tilled catchments than in wetlands with grassland catchments. They attributed part of this fluctuation to increased runoff from tilled catchments, resulting in deeper pool levels in cropland wetlands after major precipitation events. Such increases in pool depths would tend to deposit invertebrate remains at higher relative elevations in cropland wetlands relative to grassland wetlands. Similarly, van der Kamp et al. (1999) found that cultivated wetlands in the northern prairie region of Canada had lower pool levels after conversion to a brome/alfalfa mixture grassland than wetlands that remained in cultivation. They attributed the lower pool levels of wetlands in grassland areas to reduced runoff from spring snowmelt after conversion of the catchment back to grassland.
In addition to transportation during tillage and greater pool depth depositing invertebrate remains at higher elevations in cultivated wetlands, the inherent disturbance of the soils and plant communities resulting from tillage may have led to a migration of soil and plant indicators toward the wetland center and thus lowered the elevations of our standard delineations. In order to maximize production, agricultural producers routinely till and plant as much of a wetland's basin as water conditions permit. In temporary wetlands, this usually means the entire basin is cultivated. In seasonal wetlands, cultivation of the entire basin may not be possible in years with above-normal precipitation. However, even in wet years, outer margins of wetlands are usually tilled and planted, and the entire basin may be tilled for weed control if it dries later in the season. Kantrud and Newton (1996) found that over 77 percent of the area of all wet-meadow zones (the outer-most wetland zone) in agricultural watersheds had recently been cultivated. This cultivation resulted in the wet-meadow zone of wetlands in agricultural landscapes having lower species richness and a greater percentage of unvegetated bottom as compared to wetlands in grassland landscapes. The direct disturbance of the soils and increased percentage of unvegetated bottom resulting from tillage of the outer zone of prairie wetlands may yield standard delineations closer to the wetland center in cropland wetlands than in grassland wetlands, where the soil and plant indicators are undisturbed.
The maximum elevation around a wetland at which invertebrate remains will be deposited should be closely related to the elevation of pool levels (a relatively level plane). However, soil and plant indicators may extend further into the upland due to ground-water mounds or seeps. Thus, the invertebrate delineations should be less variable than among standard delineations based on soils or hydrophytes. With only one exception (a seasonal wetland in cropland), we found this to be the case. In both temporary and seasonal wetlands, and in both cropland and grassland habitats, the elevations of the wetland edge based on invertebrate remains had lower standard deviations within wetland basins than delineations using standard methods. In the single wetland where the deviation of the invertebrate delineated elevations was greater, the difference was not due to an exceptionally high standard deviation among the elevations of the six invertebrate delineations (5.9 cm) but instead was the result of an especially low standard deviation (1.2 cm) among the elevations of the six standard delineations. The close relationship between invertebrate delineations and maximum pool levels supports our argument that invertebrate delineations will be closer to the wetland center than standard delineations if hydric soils and hydrophytes extend further into the upland due to ground-water influences.
As expected, we were unable to identify the wetland edge in some (three out of five) tilled temporary wetlands using standard hydric soil and vegetation indicators. The standard delineations in these wetlands failed due to the destruction of soil and plant indicators by agricultural practices such as tillage. However, in two of these three wetlands, our invertebrate delineations also failed to identify the wetland edge. Euliss and Mushet (1999) found greatly reduced numbers of aquatic invertebrate resting eggs and shells in tilled temporary wetlands. In their study, cladoceran resting eggs were completely absent from the soils in 14 of the 19 tilled temporary wetlands sampled, whereas cladoceran resting eggs were absent from only one of the 19 temporary wetlands sampled in grassland. The absence of invertebrate remains in some of our tilled wetlands further supports our conclusion that, while the presence of aquatic invertebrate remains is a reliable wetland indicator, the absence of remains does not always imply that the location is not within a wetland. While beyond the scope of most routine field delineations, incubation of samples in aquaria (Euliss et al. 2001) may provide additional information useful in identifying wetland boundaries.
Wetland delineation is an attempt to locate an exact point along a continuum of hydrologic, substrate, and biotic characteristics distributed from a wetland's center into the upland. Not only is this point indefinite, but it is also continually moving as wetlands expand and contract in response to normal interannual wet/dry cycles of the prairie pothole region. Land-use impacts resulting in increased fluctuations of water levels and the destruction and disturbance of soil and plant indicators in agricultural wetlands exacerbates our ability to delineate wetlands in the prairie pothole region accurately. Aquatic invertebrate remains provide an additional wetland indicator that can be used to delineate wetlands when other methods fail or are not feasible, especially in agricultural landscapes.