eSDM: A tool for creating and exploring ensembles of predictions from species distribution and abundance models

Species distribution modeling (SDM) in dynamic marine environments has enhanced our ecological understanding and ability to assess potential impacts to species of conservation concern at finer spatial scales than traditional methods. However, different data sets or analytical approaches often yield different modeled results, creating uncertainty and challenges in the decision-making process. For example, there are currently multiple SDMs for blue whales off the U.S. West Coast, and assessing spatial distribution shifts using these models is challenging because they predict absolute density, relative density, or probability of occurrence at varying spatial resolutions. One solution is ‘ensemble averaging’, where the output of multiple models is combined using a weighted or unweighted average. Such ensemble models are often more robust than individual models.

We present eSDM, an R package with a built-in graphical user interface. eSDM allows users to overlay SDM outputs (predictions) onto a single base geometry, create ensembles of these predictions, estimate ensemble uncertainty, and calculate performance metrics or create maps of original predictions and ensembles. Users can create ensembles of SDM predictions made at different spatial scales, using different data sources, and with different numerical scales to better evaluate spatial uncertainties and make informed conservation and management decisions.