|Estell, Richard - Rick|
Submitted to: Wildland Shrub Symposium Proceedings
Publication Type: Abstract only
Publication Acceptance Date: 6/10/2008
Publication Date: N/A
Citation: Interpretive Summary:
Technical Abstract: Plots on rangeland infested with one-seed juniper were exposed to high (small patches; 10m2/AU/day) or low (large patches; 60 m2/AU/day) density stocking (vs control plots without grazing) of goats and goats plus sheep (2 replicates/treatment) during a summer targeted grazing experiment. Frequency of appearance of cattle and mule deer on these plots were determined the following winter, spring, summer, and fall using automated cameras with infrared motion sensors. Individual counts at 2 minute intervals were used to predict selection indices (SI; i.e., frequency of animals in plots compared to controls) and model probability of deer presence in patches as a function of cattle counts. Both deer (SI: 0.00) and cattle (SI: 0.40; P < 0.001) avoided large patches which had been grazed by goats + sheep. These patches received heaviest utilization of herbaceous vegetation (73.5%) the previous summer. Deer selected small patches that had been grazed by goats + sheep (SI: 2.24; P = 0.001); whereas cattle selected large patches (SI: 1.50; P = 0.058) and avoided small patches grazed by goats alone (SI: 0.59; P = 0.052). Cattle exhibited greatest preference for grazed patches in summer (SI: 3.18; P = 0.001), while deer avoided grazed patches in summer (SI: 0.21; P = 0.034) and spring (SI: 0.13; P = 0.016). The probability of mule deer presence on a given patch (y) was reduced by cattle (x) presence (y= e-0.128x; R2=0.40; P < 0.001). Targeted grazing programs with small ruminants could be used to create contrasting patches of different sizes to improve habitat for multiple rangeland ungulates.