USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Grays Lake Ecosystem

Evaluation of Management Practices in Wet Meadow Habitats
at Grays Lake National Wildlife Refuge, Idaho, 1997-2000

Dr. Jane Austin, Team Leader
U.S. Geological Survey
Northern Prairie Wildlife Research Center
Jamestown, ND 58401

Austin, J. E., W. H. Pyle, J. R. Keough, and D. H. Johnson.  2002.  
     Evaluation of management practices in wet meadow habitats at Grays Lake 
     National Wildlife Refuge, Idaho, 1997-2000.  Final report to U.S. Fish 
     and Wildlife Service-Region 1.  U.S. Geological Survey, Northern Prairie 
     Wildlife Research Center, Jamestown, ND.

We assessed the relative values of 4 management practices (idle, late season grazing, fall prescribed burning, and rotation of idle and summer grazing) to biotic resources of the grassland-wetland meadow ecosystem at Grays Lake during 1997-2000. Three replicates of each treatment were randomly assigned to 12 experimental units that bordered the deep emergent marsh. Biotic factors examined included the breeding bird community and abundance, nesting activity and nest success, small mammal abundance, plant community, and annual plant biomass production. Fall burns achieved treatment objectives, removing most residual vegetation across a range of cover types. Objectives for grazing treatments were mostly attained; however, vegetation use levels were insufficient for consistent attainment of treatment objectives.

Savannah sparrow, American coot, Canada goose, sandhill crane, mallard, and yellow-headed blackbird were the most common bird species present. Densities of 2 bird species (savannah sparrow and red-winged blackbird) were related to year effect only. The effect of unit on densities of redhead, lesser scaup, ruddy duck, sora, long-billed curlew, and common snipe likely reflects habitat differences among units. Densities of 6 species (eared grebe, canvasback, American coot, American avocet, willet, and common yellowthroat) were related to both year and unit effects. Treatment affected densities of 6 of the 29 species examined (mallard, northern shoveler, cinnamon teal, blue-winged teal, American crow, and yellow-headed blackbird); we found no common trend in response to treatments among those species. Overall, idled habitat did not stand out to be a valuable treatment, whereas grazing tended to have positive responses for a number of species. Burning was more likely to result in reduced bird densities than other treatments. We also describe the distribution of species observations among 8 different habitat types.

Of the 23 nesting species sampled in the experimental units, the most common were American coot, sandhill crane, Canada goose, American avocet, mallard, and cinnamon teal. Daily survival rates (DSRs) of dabbling duck nests (all species pooled) were negatively affected by fall grazing. We detected no effects of treatments on DSRs of Canada geese or sandhill crane nests. DSRs for sandhill crane nests were higher in 1998 than in 1999 or 2000 and were slightly higher than that in 1997. DSRs for coot nests were affected by both year and treatment; within-treatment differences among years were extensive. In 1998, when all units were idled, DSRs for coot nests were higher in units assigned to idle treatment than those assigned to fall-grazed or rotation treatment. DSRs for coot nests did not differ among treatment blocks in 1997 (all units idled) or 1999 (first year after treatments). We speculate that compaction of residual vegetation by snowpack reduced any differences between idle and treated units and thus lessened the value of idle habitat for most nesting birds. Nesting densities and nest success rates of Canada geese, dabbling ducks, and sandhill cranes were lower those that reported from Steel's (1952) study in 1949-1950, but differences in habitats and areas searched relative to our study make comparisons difficult. Nest success rates of sandhill cranes also were lower than those reported by Drewien (1973). Declines in nest success probably are related to changes in predator community.

We captured 5 species of small mammals (meadow vole, montane vole, deer mouse, vagrant shrew, and ermine). Populations of meadow and montane voles irrupted in 1998 then crashed in spring 1999; the most marked changes were in montane vole numbers. Capture rates of ermine and observation rates for striped skunks and raptors suggested a numerical response (higher recruitment) by these predators to higher prey abundance and possibly distributional shifts (movement into areas of more abundant microtines). We did not detect differences in capture rates among treatments for any species, but captures rates were relatively low in post-treatment years. We found little evidence to indicate that litter or vegetation affected capture rates, but sample sizes were limited.

Predator observations during field activities detected the presence of 20 predator species. The most commonly observed predators were buteos, corvids, northern harrier, eagles, falcons, striped skunk, and red fox. Seasonal variation in raptor species reflects migrational movements through the area in spring and fall. Mapping of mammalian predator observations indicated seasonal and annual variations in their activity patterns in the basin and on the experimental units. We noted high levels of fox and skunk activity in the southeast portion of the basin, an area also heavily used by nesting waterbirds.

The lakeshore habitats at Grays Lake support an abundance of breeding birds, particularly savannah sparrows, coots, Canada geese, mallards, sandhill cranes, avocets, and yellow-headed blackbirds. The lakeshore habitats also are important for foraging by marsh-nesting species such as white-faced ibis and Franklin's gulls. Strategies that increase habitat quality for foraging and nesting/brood-rearing habitat for these species would yield the greatest benefit to the regional bird community and populations. Conversely, abundance and production of dabbling ducks as a group were relatively low in the lakeshore area, suggesting that the lakeshore habitat probably does not provide a significant contribution to the regional population. The materials included here provide the necessary framework for continuing monitoring of key biotic resources in the lakeshore habitat of Grays Lake National Wildlife Refuge. This 4-year study should be viewed as the first step in a long-term study to fully address effects of management practices in a montane wetland environment.

Austin, J. E., J. R. Keough, and W. H. Pyle.  2003.  Effects of habitat 
     management practices on plant cover types of uplands and wetlands at Grays 
     Lake National Wildlife Refuge, Idaho, 1997-2000.   U.S. Geological Survey, 
     Northern Prairie Wildlife Research Center, Jamestown, ND.  Final report to 
     U.S. Fish and Wildlife Service May 2003.

Abstract:  The U.S. Fish and Wildlife Service needed scientific information to address concerns about the compatibility of wetland and upland management practices on Grays Lake National Wildlife Refuge, Idaho; a long-term monitoring framework was also needed. We conducted a study to address these needs during 1997-2000. We describe plant cover types found in 12 management units of the refuge and evaluate the response of plant composition and biomass production of plant cover types to 4 management practices: continuous idle, fall prescribed burning, fall grazing, and rotation of idle and summer grazing. Three replicates of each management treatment were randomly assigned to 12 experimental units that bordered the interior marsh. Plant data were collected from mid-June to late July each year on 700 plots systematically located within the units. We used cluster analysis on the 1998 cover data to derive 8 plant cover types: marsh, spikerush slough, Baltic rush, moist meadow, alkali, moist upland, dry upland, and unclassified. We compared changes in plant composition, biomass production, and species richness between 1998, when all units had been idled for 3 years, and 1999 (1 year post-treatment) and 2000 (2 years post-treatment).

We found few substantial shifts in the composition of plant cover types for any treatment. Changes in idle were as great as the active fall-burn, fall-graze, and rotation treatments. Compositional changes tended to be greatest in marsh, spikerush, and Baltic rush, reflecting decreased cover of key individual species in these habitats (Baltic rush, mat muhly, and common spikerush) following treatment. In marsh, increases in the cover of open water and aquatic species in 1999 and 2000 probably were related to variation in precipitation and water levels among years. Compositional changes in spikerush suggested rhizomes of spikerush and other marsh species were damaged or destroyed by burning or fall grazing. In Baltic rush, cover of marsh sedges increased after 2 years in all plots; negative effects of fire or trampling by grazing appeared to be temporary (first year only), and perennial species were able to colonize the wet soils in the second year. Moist upland plots in fall-graze units had greater cover of woody species after 1 and 2 years of treatment.

After 1 year of treatment, biomass production increased in spikerush slough plots of idle units and in moist upland, dry upland, and unclassified plots in rotation units. Conversely, production declined in marsh and spikerush slough plots in fall-burn units, marsh and Baltic rush plots in fall-graze units, and marsh and spikerush slough plots in rotation units. All significant differences in biomass production between 1998 and 2000, after 2 years of treatment, were negative: moist meadow plots in idle units; marsh, spikerush slough, Baltic rush, moist meadow, moist upland, and unclassified plots in fall-burn units; and marsh, spikerush slough, and Baltic rush plots in rotation units. Production in plots in fall-graze units in 2000 did not differ from 1998 for any cover type. In general, treatments involving cattle grazing seemed to have more persistent, negative impacts on biomass production of wetter cover types such as marsh, spikerush slough, and Baltic rush but had mixed impacts on drier cover types. Effects of burning on production tended to be shorter-lived.

Species richness (number of species) tended to be >1 species/ha in units with substantial areas of upland habitat, whereas units with <1 species/ha (009, 19A, 03B, and 18B) had relatively little upland habitat. We found no evidence that species richness declined following 1 or 2 years of treatment. Overall, patterns of change suggest that factors associated with year overwhelmed any potential treatment effect on species richness.

We examined distribution of plant cover types relative to elevation to assess the influence of hydrology and topography on plant communities. Marsh and spikerush slough were largely limited to elevations below the median spring high water level (6386.4 ft). Baltic rush occurred largely at elevations within 0.5 ft of the median spring high water level. Moist meadow, moist upland, and dry upland occupied the widest range of elevations. The variability in these cover types likely is related to factors influencing microhabitat conditions such as soil texture and subsurface irrigation (affecting soil moisture). Moist upland and dry upland were primarily limited to elevations above the median spring water levels.

In conclusion, we found few substantial impacts of treatments on cover types of the experimental lakeshore units. Shifts in composition of plant cover types may be attributed to both climatic factors and moderate treatment effects. Negative effects of active treatments on biomass production were most apparent in wetter habitats and in grazing treatments. Impacts of fall burning seemed short lived. Our results provide only preliminary information on the effects of the 4 habitat management treatments. We recommend that treatment types and pattern be standardized on the same units as in this study for another 4-6 years and data collection be continued at least intermittently to facilitate long-term evaluation of habitat responses to treatments.

Next Research Page -- Breeding Ecology of Sandhill Cranes at Grays Lake National Wildlife Refuge, Idaho
Back to Research
Return to Grays Lake Ecosystem Contents

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Webmaster
Page Last Modified: Friday, 01-Feb-2013 18:01:47 EST
Sioux Falls, SD [sdww55]