Predicting Greater Prairie-Chicken Nest Success from Vegetation and Landscape Characteristics

Abstract

To aid management of prairie habitat for nesting greater-prairie chickens (Tympanuchus cupido pinnatus), we tested whether vegetation and landscape variables could be used to predict prairie-chicken nest success. We monitored 60 nests during the 1990-92 breeding seasons in southwestern Missouri. Nest success ranged from 28 to 40% over 3 years ((x) over bar = 35%). We identified 2 2-variable models (logistic regression) incorporating litter (horizontal, residual) and woody cover (P < 0.001) or forb and grass cover (P < 0.001) at nests as the best predictors of nest success. Litter cover at the nest was the Lest single predictor of nest success (P = 0.001). Models incorporating litter cover and distance of nests to edge or tree also predicted nest success (both Ps = 0.004). However, distance of nests to edge or tree alone or in combination did not predict nest success tall Ps > 0.5), Nest sites with litter cover >25% had a failure rate twice that of nests with 5% was present at nests (P = 0.01), when forb cover was less than or equal to 5% (P = 0.009), or when grass cover was less than or equal to 25% (P = 0.02). We suggest that managers can use litter accumulation of >25% as a cue to initiate management action such as burning, grazing, or haying.