Optimal population prediction of sandhill crane recruitment based on climate‐mediated habitat limitations

Abstract

Summary 1.Prediction is fundamental to scientific enquiry and application; however, ecologists tend to favour explanatory modelling. We discuss a predictive modelling framework to evaluate ecological hypotheses and to explore novel/unobserved environmental scenarios to assist conservation and management decision-makers. We apply this framework to develop an optimal predictive model for juvenile (<1 year old) sandhill crane Grus canadensis recruitment of the Rocky Mountain Population (RMP). We consider spatial climate predictors motivated by hypotheses of how drought across multiple time-scales and spring/summer weather affects recruitment. 2. Our predictive modelling framework focuses on developing a single model that includes all relevant predictor variables, regardless of collinearity. This model is then optimized for prediction by controlling model complexity using a data-driven approach that marginalizes or removes irrelevant predictors from the model. Specifically, we highlight two approaches of statistical regularization, Bayesian least absolute shrinkage and selection operator (LASSO) and ridge regression. 3. Our optimal predictive Bayesian LASSO and ridge regression models were similar and on average 37% superior in predictive accuracy to an explanatory modelling approach. Our predictive models confirmed a priori hypotheses that drought and cold summers negatively affect juvenile recruitment in the RMP. The effects of long-term drought can be alleviated by shortterm wet spring-summer months; however, the alleviation of long-term drought has a much greater positive effect on juvenile recruitment. The number of freezing days and snowpack during the summer months can also negatively affect recruitment, while spring snowpack has a positive effect. 4. Breeding habitat, mediated through climate, is a limiting factor on population growth of sandhill cranes in the RMP, which could become more limiting with a changing climate (i.e. increased drought). These effects are likely not unique to cranes. The alteration of hydrological patterns and water levels by drought may impact many migratory, wetland nesting birds in the Rocky Mountains and beyond. 5. Generalizable predictive models (trained by out-of-sample fit and based on ecological hypotheses) are needed by conservation and management decision-makers. Statistical regularization improves predictions and provides a general framework for fitting models with a large number of predictors, even those with collinearity, to simultaneously identify an optimal predictive model while conducting rigorous Bayesian model selection. Our framework is impor-*Corresponding author. E-mail: bgerber@colostate.edu †Retired.Published 2015. This article is a U.S. Government work and is in the public domain in the USA 2015, 84, 1299-1310 doi: 10.1111/1365-2656.12370 tant for understanding population dynamics under a changing climate and has direct applications for making harvest and habitat management decisions. Journal of Animal Ecology

Do we need demographic data to forecast plant population dynamics?

“…Therefore, we avoid selecting among models with all possible combinations of climate covariates, and instead use Bayesian ridge regression to regulate, or constrain, the posterior distributions of each climate covariate (Gerber et al . ; Hooten & Hobbs ). Ridge regression is a specific application of statistical regularization that seeks to optimize model generality by trading off bias and variance.…”

Extreme site fidelity as an optimal strategy in an unpredictable and homogeneous environment

“…LASSO regularizes model parameters, thereby accommodating numerical issues due to multicollinearity of covariates and providing variable selection by removing effects of covariates. The result is an optimal predictive model that is coherently interpretable in terms of important ecological effects (see Gerber, Kendall, Hooten, Dubovsky, & Drewien, ; Hooten & Hobbs, ). We used the same procedure to model nest success as a multinomial outcome to evaluate whether predation on nesting females or eggs drives nest failure and whether it varied spatially or temporally (0 = nest failed or was abandoned, 1 = nest failed due to the female or eggs being depredated, 2 = nest success).…”

Predicting species distributions: unifying model selection and scale optimization for multi‐scale occupancy models

“…The magnitude of shrinkage, and therefore the degree of regularization, can be optimized by identifying the prior distribution that results in optimal predictive scores calculated on out‐of‐sample data (e.g., Gerber et al. ). In this way, the sparsity of the model which includes all predictor variables is optimized directly for the purpose of out‐of‐sample prediction.…”