Northern Prairie Wildlife Research Center
Our study demonstrates that water-level fluctuations are greater in wetlands located in areas of intensive agricultural activity relative to those in more natural grassland settings. The extent to which these increased fluctuations affect the plant, invertebrate, and ultimately the vertebrate communities of these wetlands is unknown. However, impacts are likely substantial if the increased fluctuations in a wetland are sufficient to alter the concentrations and species of salts present. In addition, increased fluctuations were directly related to increased runoff from adjacent uplands. How sediments and agricultural chemicals transported to wetlands in runoff affect plant, invertebrate, and vertebrate communities is poorly understood. Future studies should concentrate on determining how agricultural alteration of the natural hydrology of prairie wetlands affects their water chemistry and biota.
The differences observed in water-level fluctuations among the 3 wetland classes are related to the unique hydrology of each wetland class in the semiarid Prairie Pothole Region. In addition to surface runoff, semipermanent wetlands receive a large proportion of their water as input from ground water. This input of ground water has a stabilizing effect on water levels of semipermanent wetlands and results in wetlands of this class typically retaining water throughout the year. By contrast, seasonal, and to an even greater extent, temporary wetlands are dependent upon surface runoff for the majority of their water input. In addition, water loss in semipermanent wetlands is largely from evapotranspiration; whereas water loss in temporary wetlands is primarily from recharge to ground-water tables. Water loss in seasonal wetlands may be attributable to ground-water recharge and evapotranspiration, depending upon the wetland's hydrologic function and position in the landscape (Winter and Rosenberry 1995). The greater influence of surface runoff in the maintenance of water levels in temporary and seasonal wetlands highlights the critical role of vegetation in mitigating impacts from precipitation events.
While it would be highly desirable to use continuous water-level monitoring systems to monitor water-level fluctuations, the high cost of these devices precludes their use except on a very limited scale. The low cost devices we designed and tested in this study performed well despite being exposed to a severe winter and a spring thaw. The only major problem we encountered with the devices installed at CLSA occurred in 1993, when we received record rainfall that flooded area wetlands to depths beyond the capacity of our devices. The reason our prototype water-level recorders failed was because the guide rods were too short. We recommend that the lengths of the devices exceed maximum anticipated water depths. Of the 36 recorders placed in our study wetlands, 1 malfunctioned due to corrosion of its guide rod and 3 were destroyed by cattle. To avoid corrosion problems, we recommend that guide rods be made with a thicker copper plating. The thinly plated welding rods used in this study often had scratches and other imperfections that facilitated corrosion. Also, we suggest that the PVC casings of the devices be constructed out of 10.2-cm I.D. pipe instead of the 7.6-cm I.D. pipe we used. This would allow use of a larger, more buoyant float whose mass would overcome any friction associated with minor corrosion and fouling of the guide rod. Finally, if water-level recorders are installed in wetlands where cattle may be present, efforts should be made to exclude the cattle from the area immediately surrounding the recorders.