Table 1 — Vegetation and other features of grazed and ungrazed 

 study plots, Summit Creek, ID, 1989 



Ungrazed Grazed 



Item 



Mean' 



SD 



Mean' 



SD 



P2 



Graminoid 













Biomass (g/m^) 



267.6 



254.2 



36.7 



30.8 



<0.01 



Canopy cover (%) 



61.6 



30.5 



51.9 



30.3 



.08 



Heigfit (m) 



.18 



.09 



.06 



.03 



<.01 



Forb 













Biomass (g/m^ 



24.9 



29.8 



11.9 



14.6 



<.01 



Canopy cover (%) 



12.0 



13.2 



11.5 



11.9 



.81 



Height (m) 



.07 



.05 



.03 



.02 



<.01 



Shrub 













Biomass (g/m^ 



71.1 



164.5 



75.0 



212.7 



.91 



Canopy cover (%) 



7.4 



14.5 



7.3 



13.9 



.97 



Height (m) 



.34 



.21 



.26 



.12 



.01 



Other 













Bare ground (%) 



20.3 



25.0 



23.3 



24.2 



.50 



Litter (%) 



10.4 



11.5 



9.1 



11.6 



.55 



Rock (%) 



.64 



1.52 



2.27 



5.89 



.04 



Lichen-moss (%) 



.30 



1.15 



.84 



2.27 



.10 



'/? = 60 except for forb and shrub height means for which n ranged from 50 

 to 58. 



^Probability associated with unpaired Hests. P of less than 0.10 was con- 

 sidered significant. 



and Brewer's blackbirds were territoria] only on the grazed 

 area. Red-winged blackbirds nested only on the ungrazed 

 plot. Wide-ranging raptorial birds, although commonly 

 seen, were not included in the analysis. Transient species 

 were also excluded. Other birds, observed as visitors to 

 the study site, are listed in the appendix. 



We found little difference between the grazed and un- 

 grazed plots in total breeding bird density (table 2). But 

 estimates of total bird biomass differed markedly on the 

 two plots. Biomass on the grazed plot (226 g/ha) was al- 

 most twice that (121 g/ha) on the ungrazed plot (table 2). 

 The difference in total biomass was almost entirely due to 

 the presence of large shorebirds (killdeer, willet, long- 

 billed curlew) that were breeders only on the grazed plot. 

 Species richness and our estimate of bird species diversity 

 (the reciprocal of Simpson's index) were larger on the 

 grazed plot, again as a result of the presence of the three 

 shorebirds that established breeding territories only on 

 the grazed plot. 



Curlew populations are declining in some areas of the 

 Western United States as habitat is lost to agriculture 

 and other land development (Ryser 1985). It is a short-to- 

 midgrass prairie nesting species (Pampush 1980), often 

 nesting in moist meadows near streams and lakes, as well 

 as dry upland habitats (Harrison 1979). Long-billed cur- 

 lews prefer breeding habitats containing short grass, bare 

 ground, shade, and abundant invertebrate prey (Pampush 

 1980). Livestock grazing tends to maintain the low vegeta- 

 tional profile apparently preferred by curlews as breeding 

 habitat. At Summit Creek, we observed them most often 

 near the stream in mesic herbaceous communities domi- 

 nated by grasses, sedges, and rushes. 



Killdeer and willets also nest in open habitats where 

 vegetation is sparse and low, usually within short flight 



distances to feeding areas (Palmer 1967). Both are 

 ground-inhabiting species, building their nests and forag- 

 ing there. We saw willets most often wading in the 

 stream or pecking and probing for insects along the shore- 

 line. Killdeer were usually seen either in flight or on the 

 ground within a few meters of the stream in the most 

 open habitats. Distraction displays and other territorial 

 behaviors exhibited by both the killdeer and willet were 

 noted only on the grazed plot. Taylor (1986) reported a 

 positive response by killdeer to grazing in southeastern 

 Oregon. In North Dakota, killdeer and willets were ob- 

 served in significantly greater densities in grazed habitats 

 (Renken and Dinsmore 1987). 



Savannah sparrows, numerically dominant on both 

 study plots, were found in greater numbers on the un- 

 grazed plot (table 2). Kantrud (1981) similarly found a 

 negative response by savannah sparrows to grazing in 

 North Dakota native grasslands. A preference of this 

 species for idle or lightly grazed areas was also noted by 

 Owens and Myres (1973). This sparrow fi-equents open 

 fields and meadows and is most commonly found in moist, 

 grassy habitats in Idaho (Burleigh 1972). It is usually 

 restricted to the vicinity of streams, ponds, lakes, and 

 irrigation systems — often where soils are alkaline (Ryser 

 1985). Savannah sparrows have an affinity for habitats 

 with a rank growth of vegetation and a dense ground 

 cover (Linsdale 1938), a condition existing in more abun- 

 dance in the ungrazed habitat on the Summit Creek site. 

 Most savannah sparrow territories on the study area were 

 located in the mat muhly/hummock community type al- 

 though other plant communities were often included 

 within territorial boundaries. 



Table 2 — Density (pairs/40 ha), diversity, and other attributes of 



breeding bird populations on grazed and ungrazed study 



plots. Summit Creek, ID, 1989 





Foraging 



Nesting 



Density 



Species 



guild' 



guild^ 



Ungrazed 



Grazed 



Killdeer 



GGI 



GRN 





4.4 



Willet 



SPI 



GRN 



+ 



3.1 



Long-billed curlew 



GFO 



GRN 



+ 



1.8 



Vesper sparrow 



GFO 



GRN 



8.4 



7.6 



Savannah sparrow 



GFO 



GRN 



39.1 



24.9 



Red-winged blackbird 



GFO 



CRN 



12.0 



+ 



Western meadowlark 



GGI 



GRN 



8.0 



6.2 



Brewer's blackbird 



GFO 



GBN 



+ 



17.3 





Total pairs/40 ha 



67.5 



65.3 





Total individuals/km^ 



338 



327 





Biomass^ (qlha) 



121 



226 





Species richness (n) 



4 



7 





Species diversity^ (1/Ip_^) 2.52 



4.07 



'After DeGraaf and others (1985). GGI = ground gleaning insectivore, 

 SPI = shoreline probing insectivore, GFO = ground foraging omnivore. 



^After Harrison (1979). GRN = ground nesier, CRN = cattail, rush, sedge, 

 reed, grass, and bush nester, GBN = ground and bush nester. 



^+ indicates bird observed infrequently (less than ttiree registrations). 



*Spedes weights from Dunning (1984). 



'After Hill (1973). Here, p-\s the proportional abundance of the nspecles 

 in a sample. 



3 



