Anthropogenic climate change is driving a rapid reorganisation of global biodiversity, particularly in the ocean, with profound implications for both ecosystems and humans. This climate-driven redistribution of marine biota is altering the species observed by ocean users, like divers, fishers and beachcombers, in their local areas. Observations made by marine citizen scientists can provide early indications of emerging range shifts, making these valuable for understanding the drivers of early stages of marine species redistributions. Identifying factors contributing to the early stages of marine range shifts provides important evidence for assessing the vulnerability of marine species distributions to climate change.
Our study was an international collaboration led by Heather Hunt (University of New Brunswick) and Jorge Molinos (Hokkaido University) that utilised citizen science observations of species outside their historical distributions logged with the Australian Range Extension Database and Mapping Project (Redmap; https://www.redmap.org.au/). Redmap is a citizen science project, created by senior author Gretta Pecl around a decade ago, that requests submission of photographs and associated data for marine species ‘unusual’ for the given location (i.e. out-of-range observations) in Australia to a website or smartphone application (e.g. Fig. 1).
Fig. 1. Spearfisher Derrick Cruz, pictured with a coral trout (Plectropomus leopardus) caught off eastern Australia’s temperate coastline. This is an example of an observation poleward of the species historical distribution limit that was logged with Redmap.
Photographic observations submitted to Redmap are subjected to scientist-verification via semi-automated crowd-sourcing of a network of approximately 80 taxonomic experts from 26 institutions throughout Australia (see Pecl et al 2019 for more details at https://www.frontiersin.org/articles/10.3389/fmars.2019.00349/full). We utilised Redmap observations to calculate annual maximum out-of-range distances as both the latitudinal distance (i.e. direct latitudinal extension distance) and distance along the coastline (i.e. coastal extension distance) relative to the historical range boundaries as of 2012 for 61 species of marine fishes, reptiles and invertebrates (Fig. 2). These formed our response variables for assessing drivers of the early stages of marine species redistributions.
Fig. 2. Schematic showing the differences between the two metrics used for estimating maximum annual out-of-range extension distances: (a) latitudinal distances (DL) calculated as differences in latitude between an observation and the historical southern latitude for the corresponding species, and (b) along-the-coast distances (DC) calculated as least-cost-path distances constrained to the coastline between the historical southern limit and the observation.
Our comprehensive assessment included the effects of long-term warming, short-term climatic extremes (i.e. heatwaves and cold spells), the strength of ocean currents and their directional agreement with climate warming as well as species traits, on maximum out-of-range distances. The combined effects of these factors had the capability to explain 48% and 57% of the total variance in latitudinal and coastal extension distances, respectively.
For coastal extension distances, an interaction between climate velocity (which is a measure of the speed and direction of migrating isotherms), current velocity, and the directional agreement between currents and thermal gradients had the highest relative importance. This finding highlights that complex interactive effects among physical factors contribute to the early stages of marine range shifts. For both latitudinal and coastal extension distances, we found a positive relationship between extension distances and climate velocity, with large range extensions occurring in locations where high climate velocity displaced thermal niches quickly (Fig. 3) These results build upon our existing understanding that out-of-range observations of marine species are linked to longer-term climatic processes1.
Species traits were also important, but to a lesser extent. For example, the position of species in the water column made meaningful contributions to the distance species had shifted along the coast, with greater distances associated with pelagic than demersal or benthic species. Carnivores and omnivores were also associated with greater latitudinal extension distances than herbivores.
Fig. 3. The velocity of climate change, its directional agreement with ocean surface currents, and the locations of out-of-range observations. The velocity of climate change and its directional agreement with average mean annual surface currents. Inset panels show the out-of-range observations from (a) western Australia, (b) eastern Australia and (c) Tasmania (n = 127) coloured by their corresponding maximum annual coastal extension distances relative to the historical southern distribution limit of each species.
By providing the most complete understanding to date of the physical and biological variables that contribute to early stages of species range extensions in the ocean, our work highlights the complex interactive effects between multiple physical process that contribute to the early stages of marine species redistributions. Species traits (particularly the position of species in the water column and their trophic category) also slightly increased the amount of variation explained by our models. Taken together, these findings support the use of correlative and trait-based assessments of marine species vulnerability to climate-driven range shifts. This work is also a demonstration of the immense value of citizen science data for informing our understanding of the factors driving the redistribution of life in the ocean.
1 Fogarty HE, Burrows MT, Pecl GT, Robinson LM, Poloczanska ES (2017) Are fish outside their usual ranges early indicators of climate‐driven range shifts? Global Change Biology 23: 2047–2057.