Climate change poses a significant threat to marine biodiversity, with marine ectotherms exhibiting particular sensitivity to warming due to their narrow thermal safety margins and full occupancy of their potential latitudinal ranges. Unlike terrestrial counterparts, marine ectotherms are more vulnerable to physiological stress from temperature fluctuations. This sensitivity, coupled with rapid climate change and fewer dispersal constraints, has led to range shifts towards newly suitable habitats, resulting in higher rates of local extirpation, colonization, and species richness changes in the oceans compared to land. While equatorward range edges experience abundance declines and local extirpations, higher latitudes witness abundance increases and species influx. The influence of temperature on both high and low-latitude range limits highlights the importance of considering both the loss of suitable habitats and the emergence of thermal opportunities in previously unsuitable regions. Numerous studies have projected significant future changes in marine biodiversity composition, primarily due to species losses in the tropics and invasions at higher latitudes. However, these studies often lack the temporal resolution needed to understand when and how fast these changes will occur. This research addresses this gap by modeling the future temporal dynamics of thermal exposure and opportunity for marine species assemblages globally, focusing on understanding the relative timing of these processes and their interplay in shaping biodiversity responses.

Existing research highlights the sensitivity of marine organisms to warming, especially marine ectotherms, which fully occupy their potential latitudinal ranges, making their range boundaries closely track temperature changes. Compared to terrestrial ectotherms, marine ectotherms show higher susceptibility to climate change-induced physiological stress due to narrower thermal safety margins. This sensitivity, alongside rapid climate change and fewer dispersal limitations, has fueled range shifts towards new suitable habitats. This has resulted in elevated rates of local extirpation, colonization, species richness alteration, and faster community composition turnover in marine ecosystems. Studies have shown that while equatorward range edges face abundance decline and local extirpations, poleward edges exhibit abundance increases and new species influxes. Prior research assessing future risks to marine biodiversity using climate change projections to alter the thermal seascape has forecasted substantial changes in community composition, primarily owing to increasing species loss in the tropics and species invasion at higher latitudes. However, these studies often concentrate on a limited number of future time points, usually towards the end of the century, lacking temporal detail to fully understand the timing and speed of changes.

This study utilized yearly sea surface temperature projections from nine climate models within the Coupled Model Intercomparison Project Phase 6 (CMIP6) framework, obtained through the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP3b). These models were chosen based on process representation, historical performance, data availability, structural independence, and climate sensitivity, encompassing both low and high climate sensitivity models. Species distribution data were sourced from AquaMaps, a comprehensive database employing ecological niche modeling. Species occurring exclusively below 200 m depth were excluded. The final dataset included 21,696 species, with five phyla (Chordata, Mollusca, Arthropoda, Cnidaria, and Echinodermata) contributing approximately 92% of the species. Realised thermal niche limits were estimated using geographical distributions and yearly climate data from the historical period (1850-2014). This accounted for temporal climate variability. The maximum (Tmax) and minimum (Tmin) annual mean sea surface temperatures experienced by each species across its range were calculated, excluding outliers. The mean Tmax and Tmin were used as upper and lower realised niche limits, respectively. To constrain the emergence of thermal opportunities, depth and dispersal limitations were considered, using data from the ETOPO1 Global Relief Model and AquaMaps. Two dispersal rates (10 km year⁻¹ and 50 km year⁻¹) were used to create buffer zones around species distributions, limiting the areas where opportunities could arise. Niche unfilling was addressed by excluding grid cells where temperature fell within a species’ thermal limits but were deemed unsuitable by AquaMaps. Thermal exposure was estimated using the biodiversity climate horizon framework, identifying species experiencing at least five consecutive years of temperatures exceeding their realised thermal niche limits. Thermal opportunity emergence was assessed similarly but focused on the number of years suitable grid cells experience temperatures within a species’ thermal niche limits. Several metrics, including magnitude, timing, and abruptness, were employed to characterize exposure and opportunity dynamics. The duration of opportunities was also analyzed, categorizing them as persistent (open until 2100) or transient (closing before 2100), further differentiating transient opportunities by cause (cold or warm exposure). Analyses were conducted using R version 4.3.3.

The study revealed distinct spatial and temporal patterns in thermal exposure and opportunity. Exposure was concentrated in the tropics, while opportunities arose mostly at higher latitudes. Regions projected to experience the highest magnitude change (opportunity or exposure >10% of species richness) generally showed either high exposure or high opportunity, but rarely both. Lowering emissions significantly reduced exposure, but the effect on opportunities was less dramatic, with the magnitude of opportunities halved while exposure reduction was around 100-fold when contrasting high and low emission scenarios. Across all emission scenarios, opportunities started emerging immediately and followed a similar trajectory of gradual increase until approximately 2040. After this, opportunities continued to emerge but at a slower rate with higher greenhouse gas concentrations leading to more opportunities. In contrast, exposure generally began later and was more abrupt. Until approximately 2060, exposure patterns were similar across all scenarios, but after 2060, it accelerated more significantly under higher emission scenarios, considerably exceeding the number of opportunities by the end of the century under a high emission scenario (SSP5-8.5). The earlier emergence of opportunities compared to exposure is attributable to the close match between marine ectotherm latitudinal range limits and thermal limits. Opportunities and exposure showed distinct abruptness patterns. Exposure became more abrupt with higher greenhouse gas emissions, while opportunity abruptness showed an opposite trend. The tropics generally experienced more abrupt changes. The timing of both events was later under higher emission scenarios. Most thermal opportunities (79-97%, depending on the scenario) were projected to persist beyond 2100. Higher dispersal rates increased the number of opportunities but did not significantly alter the proportion of persistent opportunities. Transient opportunities were primarily caused by cold exposure under lower emission scenarios and warm exposure under high emission scenarios. Warm-exposed transient opportunities tended to be longer under SSP5-8.5, while cold-exposed ones were mostly concentrated in temperate and polar regions and lasted longer under lower emission scenarios. The study found that across all scenarios, opportunities consistently exceeded exposure in magnitude until around mid-century. The analyses demonstrated that increasing dispersal rates predictably increased the magnitude of opportunity but didn’t substantially alter the overall trends in timing, abruptness, and duration.

The findings highlight the significant role of thermal opportunities in driving marine biodiversity change, particularly in the near and mid-term. The study’s observation that opportunities generally exceed exposure in magnitude, except under high-emission scenarios, aligns with empirical evidence showing faster poleward range shifts compared to equatorward shifts in marine species. This signifies that poleward range shifts and range expansions are likely to be more prevalent, especially under lower warming scenarios. Colonization of new locations may increase resilience and decrease extinction risk for many species. However, the influx of new species into recipient assemblages can introduce risks through competition and disruption of established trophic interactions. The study’s finding that most opportunities will arise in areas with low exposure suggests that low exposure does not imply low risk from climate change, as opportunities could lead to significant ecological impacts. While the emergence of an opportunity doesn’t guarantee biodiversity change, dispersal ability greatly affects the likelihood of successful colonization. The study’s permissive approach to identifying thermal opportunities within a dispersal-constrained buffer zone allowed the identification of the temporal dynamics of opportunities potentially reachable through rare long-distance dispersal events or persistent opportunities over several decades. The study acknowledges potential limitations, such as the exclusion of factors that may hinder colonization, such as biotic interactions and other environmental variables (e.g., oxygen, pH). Further, the focus on sea surface temperature means the findings may not fully capture the complexity of the three-dimensional thermal habitat shifts. Despite these limitations, the study provides a valuable framework for predicting future thermal seascape changes and guiding monitoring efforts.

This study provides crucial insights into the thermal constraints likely to drive future marine biodiversity changes. The emergence of thermal opportunities is an ongoing process, exceeding exposure until mid-century regardless of emission scenarios. These opportunities are projected to persist until the end of the century, increasing chances of successful colonization. Under high emission scenarios, exposure will likely overtake opportunity as a major driver of change towards the end of the century. The projections serve as an early warning system, informing targeted monitoring efforts and informing conservation and ecosystem management plans for potential reorganisation of marine assemblages.

The study’s focus on sea surface temperature and exclusion of species below 200 m depth might underestimate the complexity of three-dimensional thermal habitat changes. The use of a fixed five-year period to define exposure and opportunity events may not capture species responses under short periods of exposure. The analyses do not incorporate factors like local adaptation, biotic interactions (competition, predation), and other abiotic factors, which could influence colonization success and range shifts. While the study accounts for dispersal limitations, the actual colonization success is dependent on various ecological and evolutionary factors, beyond the scope of this study. The model may overestimate opportunity magnitudes and underestimate exposure magnitudes due to assumptions regarding species dispersal, thermal niche breadth, and the absence of other environmental stressors.