Dynamic habitat suitability modelling reveals rapid poleward distribution shift in a mobile apex predator

Abstract

Many taxa are undergoing distribution shifts in response to anthropogenic climate change. However, detecting a climate signal in mobile species is difficult due to their wide-ranging, patchy distributions, often driven by natural climate variability. For example, difficulties associated with assessing pelagic fish distributions have rendered fisheries management ill-equipped to adapt to the challenges posed by climate change, leaving pelagic species and ecosystems vulnerable. Here, we demonstrate the value of citizen science data for modelling the dynamic habitat suitability of a mobile pelagic predator (black marlin, Istiompax indica) within the south-west Pacific Ocean. The extensive spatial and temporal coverage of our occurrence data set (n = 18 717), collected at high resolution (~1.85 km(2) ), enabled identification of suitable habitat at monthly time steps over a 16-year period (1998-2013). We identified considerable monthly, seasonal and interannual variability in the extent and distribution of suitable habitat, predominately driven by chlorophyll a and sea surface height. Interannual variability correlated with El Nino Southern Oscillation (ENSO) events, with suitable habitat extending up to ~300 km further south during La Nina events. Despite the strong influence of ENSO, our model revealed a rapid poleward shift in the geometric mean of black marlin habitat, occurring at 88.2 km decade(-1) . By incorporating multiple environmental factors at monthly time steps, we were able to demonstrate a rapid distribution shift in a mobile pelagic species. Our findings suggest that the rapid velocity of climate change in the south-west Pacific Ocean is likely affecting mobile pelagic species, indicating that they may be more vulnerable to climate change than previously thought.

Rapid shifts in distribution and high‐latitude persistence of oceanographic habitat revealed using citizen science data from a climate change hotspot

“…Given that spatial shifts in large numbers of marine species are expected with climate warming (Poloczanska et al., ; Sunday, Bates, & Dulvy, ), pragmatic approaches that utilize increasingly available marine‐based citizen science data sources (Bonney et al., ; Dickinson et al., ; Pecl, Barry, et al., ) to quantify species redistributions are required (Hill et al., ). Here, we demonstrate the utility of a marine habitat suitability model fitted using citizen science data for quantifying climate‐driven spatiotemporal shifts in oceanographic habitat, while accounting for the effects of natural intra‐ and interannual climate variability.…”

Spatial and Host-Related Variation in Prevalence and Population Density of Wheat Curl Mite (Aceria tosichella) Cryptic Genotypes in Agricultural Landscapes

“…Niche models that are routinely employed for predicting species distributions often focus solely on abiotic factors when evaluating the potential habitat suitability for a species of interest (e.g. [10, 11]). However, consideration of biotic interactions in modelling species distributions should improve our ability to predict distribution patterns [12].…”

Oceanographic habitat suitability is positively correlated with the body condition of a coastal‐pelagic fish

“…Predictions of habitat quality for locations where suitability values exceed the occurrence threshold could therefore be a novel method for investigating the physiology and ecology of other species that associate with oceanographic features. For example, in the development of a habitat suitability model for black marlin ( Istiompax indica ) in the Tasman Sea, Hill et al () identified a threshold value of 0.282 to partition between habitat that was suitable and unsuitable for the occurrence of this species. Spatial variability in habitat suitability values above such thresholds could serve as a proxy for physiology or ecological responses but further comparisons between field data and model predictions are required to test these relationships.…”

“…Twenty thousand pseudo‐absences were selected to (a) ensure that environmental variability occurring over the spatiotemporal extent encompassed by the occurrence dataset was adequately captured, (b) comply with Barbet‐Massin, Jiguet, Albert, and Thuiller () who recommend selecting a large number (i.e.> 10,000) of pseudo‐absences when using regression techniques to develop species distribution models, and (c) facilitate comparisons with habitat suitability models for other pelagic fishes from eastern Australia (see Champion, Hobday, Tracey, et al, ) that were also developed using approximately 20,000 pseudo‐absences (e.g. Brodie et al () and Hill et al () who used 20,000 and 23,242 pseudo‐absences, respectively). Explanatory oceanographic variables were matched to the resulting set of occurrence and pseudo‐absence data using the Spatial Dynamics Ocean Data Explorer (Hartog, Hobday, & Jumppanen, ).…”

“…Species distribution or habitat suitability models have become valuable tools for assessing spatiotemporal variation in environmental habitat quality (Brodie et al, ; Hazen et al, ). These models have been successfully applied to predict the seasonal distributions of coastal‐pelagic fishes (Brodie et al, ) and to identify shifts in their core (Dell, Wilcox, Matear, Chamberlain, & Hobday, ; Hill, Tobin, Reside, Pepperell, & Bridge, ) and range edge habitats under global change (Champion, Hobday, Tracey, & Pecl, ; Robinson, Hobday, Possingham, & Richardson, ). While predictions created using habitat suitability models are commonly dependent on the probability of species occurrence (Robinson et al, ), these predictions may also be used to provide novel insights into the physiological status of individuals across broad spatial extents.…”