From fine-scale foraging to broad-scale migration, animal movement is shaped by the distribution of resources. There is mounting evidence, however, that learning and memory also guide movement. Although migratory mammals commonly track resource waves, how resource tracking and memory guide long-distance migration has not been reconciled. We examined these hypotheses using movement data from four populations of migratory mule deer (n=91). Spatial memory had an extraordinary influence on migration, affecting movement 2–28 times more strongly than tracking spring green-up or autumn snow depth. Importantly, with only an ability to track resources, simulated deer were unable to recreate empirical migratory routes. In contrast, simulated deer with memory of empirical routes, used those routes and obtained higher foraging benefits. For migratory terrestrial mammals, spatial memory provides knowledge of where seasonal ranges and migratory routes exist, while resource tracking determines when to beneficially move within those areas.
spring_data
Spring_data and Fall_data represent databases of used (denoted by column ‘case’ equal to 1) and available steps during spring and fall migration for mule deer in the Red Desert, San Juan Basin, Wyoming Range North, and Wyoming Range South of Wyoming, Idaho, New Mexico, and Colorado (USA) from approximately 2011-2018. The dataset is organized based on a step selection function framework, and can be analyzed using conditional logistic regression. These data were derived from relocations of GPS collared females. Program R was used to create the data file in July of 2019. Column headings exactly match the variables described in the associated publication in Ecology Letters. Please contact Jerod Merkle (jmerkle@uwyo.edu) at University of Wyoming for questions. If not available, please contact any of the authors listed in the associated publication in Ecology Letters.
autumn_data
Spring_data and Fall_data represent databases of used (denoted by column ‘case’ equal to 1) and available steps during spring and fall migration for mule deer in the Red Desert, San Juan Basin, Wyoming Range North, and Wyoming Range South of Wyoming, Idaho, New Mexico, and Colorado (USA) from approximately 2011-2018. The dataset is organized based on a step selection function framework, and can be analyzed using conditional logistic regression. These data were derived from relocations of GPS collared females. Program R was used to create the data file in July of 2019. Column headings exactly match the variables described in the associated publication in Ecology Letters. Please contact Jerod Merkle (jmerkle@uwyo.edu) at University of Wyoming for questions. If not available, please contact any of the authors listed in the associated publication in Ecology Letters.