Prediction of mean adult survival rates of southern African birds from demographic and ecological covariates

摘要: Estimates of annual survival rates birds are valuable in a wide range studies population ecology and conservation. These include modelling to assess the impacts climatic change or anthropogenic mortality for many species which no reliable direct estimates available. We evaluate performance regression models predicting adult from values demographic ecological covariates available textbooks databases. estimated 67 using dead recoveries ringed southern Africa fitted five covariates: mean clutch size, body mass, age at first breeding, diet migratory tendency. Models including these explanatory variables performed well this set species, both when phylogenetic relatedness was taken into account generalized least squares (51% variation logit explained) it not (48%). Two independent validation tests also indicated good predictive power, as by high correlations observed with expected leave-one-out cross test data (35% explained), mark–recapture 38 African were predicted Clutch size mass most influential covariates, without inclusion effects, model only two (39 48% explained). Our models, version likely perform rate