Sensitivity Analyses of a Population Projection Model of Ring-Necked Pheasants

Abstract A large literature exists on population dynamics of ring-necked pheasant (Phasianus colchicus) in North America, but there has not been an attempt to formulate a matrix model nor a sensitivity analysis of the relationships between vital rates and population finite growth rate (Λ) that can be used to guide management. We summarized demographic data available from a 5-year field study in Iowa, USA, collected in Kossuth County (low composition of perennial habitat) and Palo Alto County (high composition of perennial habitat) into a 2-stage (young and adult) matrix projection model. We estimated Λ₁ (the dominant eigenvalue of the deterministic matrix), the stable age distribution (ω), relative reproductive value vector (n?), other demographic parameters, and Λ_iid, a bootstrap estimate of growth that includes interannual variation in life history parameters. We analyzed the relative importance of vital rates on population growth rate using sensitivity and elasticity of both matrix elements and lower-level parameters such as winter survival and nest success. We first characterized general life history using averaged data from both areas and all years that yielded Λ₁ = 1.086, and a stable stage distribution of. Minimum success of the initial nesting attempt (H₁) that would maintain Λ ≥ 1 under average conditions was estimated to be 42%. Changes in Λ₁ were most sensitive to survival of chicks during brood rearing (S_B), followed in importance by survival during the subsequent winter (S_W), followed by H₁. We followed the general analyses with analyses that helped us to focus on the differences in the landscapes of northwest Iowa. Λ_iid was ≥1 in only 9% of simulations of data from Kossuth whereas estimated Λ_iid was ≥1 in 88% of simulations from Palo Alto. Our analyses of the relative importance and interactions between S_B, S_W, and total recruitment (M, including H₁ and renesting), if combined with data more specific to a local area, would guide management aimed at affecting multiple life history stages whose relative importance will vary across the landscape.