Research

Looking Back to Plan the Future
by Rex Johnson, U.S. Fish and Wildlife Service, Michael Kjellsen, South Dakota State University, and Duane Pool, Ducks Unlimited, Inc.

Before Europeans settled the area, wetlands were a dominant feature of the glaciated tallgrass Prairie Pothole Region of Minnesota and Iowa. In the region's low relief landscapes, wetlands were so extensive that they prohibited large-scale farming. Even travel was impossible during some seasons. These landscapes were of unparalleled importance to wildlife, particularly continental waterbird populations. However, the lure of fertile soils and a strong work ethic converged in a mammoth undertaking—draining the tallgrass prairie and its wetlands. Today, in many areas of Minnesota and Iowa, fewer than 10% of original wetlands remain.

In October 2000, a Restorable Wetlands Working Group assembled to begin the process of mapping all of the restorable depressional wetlands in the glaciated tallgrass prairie. The first step was to determine which techniques, if any, could be used to reliably map drained wetlands in agricultural landscapes. A pilot project was launched to evaluate the accuracy of three potential delineation techniques: digital hydric soils databases (Soils), digital elevation models (DEM), and manual stereoscopic photointerpretation on high-altitude color infrared aerial photographs (PI).

The project covered nearly 4,000 square miles of different land forms and wetland characteristics in Jackson, Rice, Grant, and Polk Counties in Minnesota. After mapping was completed, some 1,500 drained wetlands were observed in the field to assess the accuracy of each technique. Results indicate that only photointerpretation provided reliable results (Table 1).

Delineation of historic wetlands using photointerpretation in the Okabena 7.5-minute quadrangle in east-central Jackson County vividly illustrates the potential of humans to alter the natural landscape (Fig. 1). The Okabena quadrangle historically encompassed more than 8,940 acres of depressional wetland or 27% of the total land area within the quadrangle. After nearly 100 years of agricultural drainage, approximately 1,280 acres of the historic depressional wetlands remain, representing an 86% reduction. When empirical models used to estimate duck pairs on individual wetlands are applied to the historic and current wetland habitat within the Okabena quadrangle, they estimate a 92% reduction in the habitat potential for common dabbling duck species.

The Okabena quadrangle's wetland density once exceeded that of most of the remaining U.S. Prairie Pothole Region. Without strong incentives for wetland conservation, the Okabena quadrangle foretells one possible future for much of the mixed-grass Prairie Pothole Region farther west.

Fortunately, wetlands once drained for agriculture may be restored to many of their historic functions. In highly fragmented ecosystems such as the tallgrass prairie portion of the North American Waterfowl Management Plan's Prairie Pothole Joint Venture, restoration of multiple wetland functions and values is most effective when highly focused at priority restoration landscapes. Data on the historic distribution of wetlands are an important part of developing strategic regional habitat restoration plans.

For more information, contact Michael Kjellsen, Wildlife and Fisheries Sciences, Box 2140B, South Dakota State University, Brookings, South Dakota 57007, 605-688-5894, michael_kjellsen@sdstate.edu.

Table 1. Error rates (%) by technique and restorable wetland size class. For each pair of error rates, commissional errors (features that were mapped in error) appear first, followed by omissional errors (restorable wetlands that were missed).

Restorable Wetlands Working Group

U.S. Fish and Wildlife Service
Ducks Unlimited, Inc.
Minnesota Department of Natural Resources
The Minnesota Board of Water and Soil Resources
U.S. Army Corps of Engineers
The Nature Conservancy
Natural Resources Conservation Service
Minnesota Pollution Control Agency

Restoring Native Tallgrass Prairie in Fescue Pastures
by Roger Applegate, Kansas Department of Wildlife and Parks

Tall fescue, a cool-season grass, was brought to the United States from Europe in the late 19th century. The University of Kentucky released K31 tall fescue for planting in the 1940s. This fescue variety caught on as an option for cattle forage before warm-season grasses begin growth in the spring and after warm-season grasses have become dormant in late fall.

There are an estimated 120 to 140 million acres of tall fescue in the United States, with most occurring in portions of the midwest and southeast. In southeastern Kansas, there are approximately 679,000 acres of tall fescue pastures. Nearly 27 percent of this acreage is in just three counties, where tall fescue had been seeded into existing tallgrass prairie pastures.

To provide adequate cattle forage, tall fescue requires annual inputs of nitrogen. The grass develops a dense sod that impedes growth of many native plants that normally would provide structural and functional diversity to the prairie. Toxins produced by a symbiotic fungal endophyte cause reproductive problems with voles, mice, and various livestock. Numbers of northern bobwhites and cottontails have declined in areas of extensive fescue, and the greater prairie-chicken is nearly gone from the fescue region of eastern Kansas.

In recent years, farmers and wildlife and range managers have become disenchanted with tall fescue and have started looking for ways to convert fescue back to native tallgrass prairie. The conventional methods have been to treat pastures with herbicides followed by planting with grass/forb seed mixtures or to plow pastures and plant crops before reseeding. Both methods are labor intensive and costly.

A promising alternative for converting overseeded fescue pastures is to withhold nitrogen fertilization, terminate cattle grazing, and burn pastures in the spring following the cessation of grazing. Withholding nitrogen and spring burning set back fescue growth, while cessation of grazing protects the warm-season grasses that are present but suppressed by grazing and competition with fescue.

Three years of trials using this protocol in southeastern Kansas show that species richness and abundance of small mammals increase after fescue conversion. After the first year of grazing cessation, tall fescue increased, but by the third year, it declined. The canopy cover of native prairie plants increased over the 3 years.

The present goal is to convert additional fescue pastures and monitor those that have been converted to determine the long-term efficacy of this technique. If converted pastures continue to be maintained after cattle grazing has been reintroduced, the protocol described here will provide a less expensive alternative than the traditional methods of transforming tall fescue to tallgrass prairie. With time it is hoped that enough fescue will be converted to provide the basis for re-establishing greater prairie-chickens in large areas of eastern Kansas.

For more information, contact Roger Applegate, Research and Survey Office, Kansas Department of Wildlife and Parks, P.O. Box 1525, Emporia, Kansas 66801-1525, (620) 342-0658, rogera@wp.state.ks.us.

The Importance of Beavers to Duck Populations in Wyoming
by Mark McKinstry and Paul Caffrey, Wyoming Cooperative Fish and Wildlife Unit and Stanley Anderson, Spatial Data and Visualization Center

Beaver activities broadly influence many plants and other animals. They are especially important in the western United States where riparian and wetland habitats comprise less than 2% of the landscape yet support 80% of wildlife species.

The objectives of our research were to identify streams in Wyoming where beavers are currently present, have been extirpated, have been used to manage riparian habitat, and are suitable for creating wetlands and improving riparian habitat. We surveyed 125 public managers knowledgeable on beaver occupancy for streams within their districts. Each manager was sent 1:100,000-scale Bureau of Land Management topographic maps of their district, with instructions to highlight streams that met the research objectives. Managers were asked to limit reporting to only those streams for which they had direct and reliable knowledge. Seventy-two managers responded, providing information for 18% of Wyoming's 95,720 stream miles.

We transferred information into ARC INFO and ArcView and a digital line graph, 1:100,000-scale coverage for Wyoming. Where a manager's highlighting did not extend entirely to the end of a reach where a node was located, we extended the classification to the next node (1% of stream highlights). ArcView generated total stream lengths by stream order and the four categories of beaver occupancy.

Waterfowl surveys were conducted in 1994 from May 15 through June 15 between 0700 and 1000 hours to census active breeding birds. We selected eight paired 1-kilometer reaches on first- through third-order streams throughout the State. The only difference between paired sites was the presence or absence of beavers and wetland habitat. Waterfowl were surveyed by walking 1-kilometer stream reaches and recording species and sex for each bird observed within the riparian zone. We also surveyed 10 ponds created by introduced beavers. At each site, we walked pond shorelines and recorded all wildlife seen during two visits between June 1 to July 15.

A total of 60 waterfowl, representing seven species were counted during the paired stream surveys: 19 green-winged teal, 18 mallards, 12 blue-winged teal, 5 cinnamon teal, 3 wood ducks, 2 gadwalls, and 1 American wigeon. All, save one blue-winged teal, were counted on stream reaches with beaver ponds. Four broods (two green-winged teal, one mallard, and one blue-winged teal) also were counted. Beaver-pond stream reaches averaged 7.5 ducks/kilometer of stream. The 10 ponds, each less than 2 years old, had three mallard hens and one blue-winged teal hen; all but one mallard hen were tending active nests.

Extrapolating only to the first- through third-order streams for which managers thought beavers could be used to improve habitat, we estimate that beaver-extirpated streams may have provided habitat for 19,000 ducks. Assuming that the percent of streams where beavers have been removed (23%) is representative of the State, and extrapolating only to first- through third-order streams, the reduction in beaver habitat may have affected over 240,000 waterfowl statewide.

We realize that critical studies have not been done directly linking waterfowl population levels with beaver abundance; however, we believe that beavers are fulfilling an important role in supporting Wyoming's waterfowl and other wetland-dependent species.

For more information, contact Mark McKinstry, Research Scientist, Wyoming Cooperative Fish and Wildlife Research Unit, P.O. Box 3166, Laramie, Wyoming 82071, (307) 766-5491, markmck@uwyo.edu.