A machine‐learning approach for extending classical wildlife resource selection analyses

Abstract

Resource selection functions (RSFs) are tremendously valuable for ecologists and resource managers because they quantify spatial patterns in resource utilization by wildlife, thereby facilitating identification of critical habitat areas and characterizing specific habitat features that are selected or avoided. RSFs discriminate between known‐use resource units (e.g., telemetry locations) and available (or randomly selected) resource units based on an array of environmental features, and in their standard form are performed using logistic regression. As generalized linear models, standard RSFs have some notable limitations, such as difficulties in accommodating nonlinear (e.g., humped or threshold) relationships and complex interactions. Increasingly, ecologists are using flexible machine‐learning methods (e.g., random forests, neural networks) to overcome these limitations. Herein, we investigate the seasonal resource selection patterns of mule deer (Odocoileus hemionus) by comparing a logistic regression framework with random forest (RF), a popular machine‐learning algorithm. Random forest (RF) models detected nonlinear relationships (e.g., optimal ranges for slope and elevation) and complex interactions which would have been very challenging to discover and characterize using standard model‐based approaches. Compared with standard RSF models, RF models exhibited improved predictive skill, provided novel insights about resource selection patterns of mule deer, and, when projected across a relevant geographic space, manifested notable differences in predicted habitat suitability. We recommend that wildlife researchers harness the strengths of machine‐learning tools like RF in addition to “classical” tools (e.g., mixed‐effects logistic regression) for evaluating resource selection, especially in cases where extensive telemetry data sets are available.

Summer weather conditions influence winter survival of honey bees (Apis mellifera) in the northeastern United States

“…Random Forests develop a large number of decision trees using a random sampling of variables, then average across all trees to produce an ensemble (forest) fit 39 , 46 . The RF technique is very efficient when working with datasets comprising a large number of predictors 47 , and when the relationships among variables are nonlinear or complex, because it is a flexible distribution-free method 48 . Given the complexity and nonlinearity of the dataset used in this study, RF was preferred to a linear regression method, and allowed the development of a reliable empirical model without prior knowledge of the relationship between the survival and the predictors 49 .…”

Resource selection of a montane endemic: Sex-specific differences in white-bellied voles (Microtus longicaudus leucophaeus)

“…We fit generalized linear mixed-effects logistic regression models with individual as a random effect and a binomial use vs. availability design, with the lme4 package in R for within home range scale selection. To reduce bias based on unequal known locations, we used a random intercept term assigned to each individual [ 56 , 59 ]. At the patch scale, to compare each vole location with its random location, we used conditional mixed-effects logistic regression models using the mclogit function in R with a binomial error structure and logit link function [ 2 , 11 , 56 ].…”

Multi-Level, Multi-Scale Modeling and Predictive Mapping for Jaguars in the Brazilian Pantanal.

“…Previous research has also applied ML methods like RF for movement studies for point selection functions for mule deer (16) and Florida panther (17). Zeller (2018) explored non-parametric and semiparametric statistical modeling approaches for use in resource selection functions to derive predictive distribution maps for large carnivore conservation (18).…”