Scale of inference: on the sensitivity of habitat models for wide‐ranging marine predators to the resolution of environmental data

Abstract

Understanding and predicting the responses of wide‐ranging marine predators such as cetaceans, seabirds, sharks, turtles, pinnipeds and large migratory fish to dynamic oceanographic conditions requires habitat‐based models that can sufficiently capture their environmental preferences. Marine ecosystems are inherently dynamic, and animal–environment interactions are known to occur over multiple, nested spatial and temporal scales. The spatial resolution and temporal averaging of environmental data layers are therefore key considerations in modelling the environmental determinants of habitat selection. The utility of environmental data contemporaneous to animal presence or movement (e.g. daily, weekly), versus synoptic products (monthly, seasonal, climatological) is currently debated, as are the trade‐offs between near real‐time, high resolution and composite (i.e. synoptic, cloud‐free) data fields. Using movement simulations with built‐in environmental preferences in combination with both modelled and remotely‐sensed (ROMS, MODIS‐Aqua) sea surface temperature (SST) fields, we explore the effects of spatial and temporal resolution (3–111 km, daily–climatological) in predictive habitat models. Results indicate that models fitted using seasonal or climatological data fields can introduce bias in presence‐availability designs based upon animal movement datasets, particularly in highly dynamic oceanographic domains. These effects were pronounced where models were constructed using seasonal or climatological fields of coarse (> 0.25 degree) spatial resolution. However, cloud obstruction can lead to significant information loss in remotely‐sensed data fields. We found that model accuracy decreased substantially above 70% data loss. In cloudy regions, weekly or monthly environmental data fields may therefore be preferable. These findings have important implications for marine resource management, particularly in identifying key habitats for populations of conservation concern, and in forecasting climate‐mediated ecosystem changes.

Outstanding Challenges in the Transferability of Ecological Models

“…For example, poorly resolved predictors might not capture important aspects of a species' ecology, and relate only indirectly to observed patterns of occurrence and biogeography [31,32]. Where possible, the scale(s) over which the processes of interest operate should therefore drive predictor choice, with sensitivity tests advisable [31]. As habitat availability, and thus perceived preference, also often link to scale [33], models will be sensitive to the extent of the study region, especially for fragmented habitats and steep environmental gradients [8].…”

“…Importantly, data resolution influences model fit, prediction, and by extension, transferability. For example, poorly resolved predictors might not capture important aspects of a species' ecology, and relate only indirectly to observed patterns of occurrence and biogeography [31,32]. Where possible, the scale(s) over which the processes of interest operate should therefore drive predictor choice, with sensitivity tests advisable [31].…”

Projecting Marine Mammal Distribution in a Changing Climate

“…Several characteristics of global climate and earth system models have shaped their application in projecting marine mammal distribution and abundance under climate change (Stock et al, 2011). The coarse resolution of global climate models (often 1-2 • , or 100-200 km ocean grids in past assessments of the Intergovernmental Panel on Climate Change, IPCC) can generate significant coastal biases (Scales et al, 2017). Large intermodel differences in regional climate change projections limit confidence in scenarios for the future states of drivers of marine resource distribution and abundance (Hawkins and Sutton, 2009;Frölicher et al, 2016).…”

Fishermen Follow Fine-Scale Physical Ocean Features for Finance

“…These coarse graining experiments have been used to examine the relationship between pelagic predators and relatively large areas of the ocean, characterized by different biophysical properties (Scales et al, 2017). Coarse graining experiments like these move the focus away from the behavior of individual fishermen and fishing vessels, and instead ask questions about where groups of vessels (fleets) operate.…”