Large Fisheries Declines Linked to Compound and Extreme Climate Events

Large Fisheries Declines Linked to Compound and Extreme Climate Events | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Large Fisheries Declines Linked to Compound and Extreme Climate Events William Cheung, Thomas Frölicher, Juliano Palacios-Abrantes, Isabella Morgante This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8109072/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Climate extremes are increasingly disrupting marine ecosystems and fisheries. However, evidence on extreme ocean temperature effects on fish stocks is mixed, and the combined impacts of heat and productivity extremes remain unclear. Using three decades of global data encompassing 6,659 time-series for 1,246 species across 254 regions, we conducted a risk-based analysis to quantify how local extreme high temperatures and low ocean productivity that these species were exposed to, alone and together, affect local fisheries catches. These events, particularly when compounded, sharply increase the likelihood of local extreme low catch events, especially in tropical and subtropical regions. Species critical to food security and conservation are disproportionately affected, with widespread risks in socio-economically vulnerable countries. Without adaptation, extreme events’ risks are projected to strongly intensify by the mid-21st century. Strengthening monitoring and climate-responsive fisheries management is urgently needed to build resilience. Main Text Earth and environmental sciences/Ecology/Climate-change ecology Earth and environmental sciences/Climate sciences/Ocean sciences/Marine biology Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction The increasing frequency and intensity of extreme ocean climate events 1 , 2 are causing localized declines in marine life, altering ecosystem structure, and disrupting fisheries 3 – 5 . These changes result in substantial socio-economic impacts such as fishery closures, revenue losses, and diminished livelihoods 6 – 8 . Marine heatwaves, or extreme high ocean temperatures (TempX), have emerged as prominent drivers of fisheries disruption, though empirical evidence remains geographically and taxonomically limited 7 , 9 . In contrast, the effects of biogeochemical extremes 10 , such as extreme low net primary production (NPPX), and their interactions with high TempX are far less understood, especially at spatial and temporal scales relevant for fisheries management 11 – 14 . Despite advances in physical oceanography and ecosystem modelling, empirical, global-scale tests of how such local compound extremes translate into local realized fisheries outcomes have been lacking. Advancing our understanding of how individual and compounded marine extremes affect fisheries is essential for assessing emerging risks to seafood security and for guiding strategies to enhance the resilience of ocean management under a rapidly changing climate 15 . Temperature and net primary production (NPP) are key drivers of the distribution, biomass and production of species important to fisheries 16 – 18 . However, a recent meta-analysis of biomass and temperature time-series found that marine heatwaves have muted and variable impacts on demersal fish populations 9 . In contrast, observed high TempX has driven rapid biomass declines and fisheries closures of commercially important species (e.g., 19, 20 ). Theoretical models further project that extreme temperature and its compounding effects with biogeochemical extremes reduce fish biomass 6 , 13 . These contrasting empirical and modelling results may reflect limitations in available datasets and the complex, region- and species- specific interactions among environmental variables, ecosystems and other human drivers 9 , 21 . This underscores a critical need to reconcile the evidence base and robustly assess the risks that climate-driven extremes experienced by marine species pose to fisheries at the local scale. This study quantifies the impacts of annual TempX and NPPX, and compounded TempX with NPPX events on local fish stocks and fisheries at the global scale. Here, environmental and fisheries extremes arise from variability superimposed on long-term mean trends (see Supplementary Fig. 1, Methods). We focus on annual extremes as these events often have the greatest impacts on fisheries 6 . By integrating species biogeography with fisheries catch, temperature and net primary production time-series data from 1993 to 2019 with statistical modeling (see Methods), we evaluate two main hypotheses: (1) high TempX and low NPPX are associated with extreme low fisheries catches across 254 EEZs and 1,246 exploited fishes and invertebrates, (2) the strength and direction of these associations vary across regions and functional groups, and (3) the likelihood of extreme low fisheries catch events will increase under future climate change. Our analysis was structured in three tiers (see Supplementary Fig. 2 and Method). First, we pooled all observations across species–EEZ groups, treating each species–EEZ–year combination as an independent sample to estimate a relationship between extreme environmental conditions exposed to the specific species-EEZ unit and extreme low fisheries catches at the EEZ level, which we then aggregated globally. Second, we conducted EEZ-specific analyses to assess how these relationships vary geographically. Third, we grouped species into functional categories and repeated the analysis to evaluate responses at the functional-group level. We adopt a risk-based framework applied to 6,659 fisheries and environmental time-series data to quantify how environmental extremes alter the likelihood of extreme low fisheries catches across species at the functional groups and Exclusive Economic Zone levels, as well as globally aggregated. Using a relative-probability approach analogous to epidemiological risk analysis (see Method), we estimate how much the odds of extreme low catch events, defined as the lowest decile of catch anomalies, increase during years of TempX and/or NPPX. This framework captures the probabilistic nature of extreme and compound climate risks, revealing not only whether extremes coincide with drops in fisheries production, but also how strongly they amplify the underlying risk compared to non-extreme conditions. It therefore bridges empirical observation and theory by reconciling the frequency of observed co-occurrences with the modeled amplification of low catch likelihood under extreme or compound stressors. Significant global effects of univariate and compounded extremes on catches Globally, across all EEZs, our analysis indicates that univariate extreme environmental conditions significantly increase the likelihood of extreme low fisheries catches (odds ratio > 1; p < 0.05, Fig. 1 ). Here, extreme events are defined as annual mean anomalies falling beyond the 10th (for NPP) or 90th (for temperature) percentile of the historical distribution (see Method). Odds ratio (OR) quantifies the strength of association between environmental extremes and the likelihood of an extreme low fisheries catch event, with OR > 1 indicating increased odds relative to baseline (non-extreme) conditions in the time-series. We estimated odds ratios using logistic regression models, where the occurrence of an extreme low catch (binary response) was modeled as a function of extreme environmental conditions. For example, OR of 1.5 indicates that the likelihood of occurrence of low catches is 50% higher during extreme conditions compared to baseline years. In contrast, OR below 1 indicates a reduced likelihood of low catches during extreme conditions. Two consecutive years with high TempX were identified as the most important univariate driver, increasing the odds of an extreme low catch by 40% (95% CI: odds ratio = 20–65%) compared to baseline conditions. Considering only extra-ordinary TempX (the most extreme conditions in the historical record) did not further elevate the odds of extreme low catches compared with ordinary high TempX (Fig. 1 ). In contrast, univariate NPPX events significantly affected catch anomalies only when the lowest NPP in a time-series were classified as extreme. Compound TempX with NPPX events amplified the risk of extreme low catches (Fig. 1 ). Several combinations of co-occurring or sequential TempX and NPPX events significantly increased the odds of extreme low catch events beyond those of univariate extremes. For instance, the occurrence of lagged TempX with concurrent extra-ordinary NPPX yielded odds ratios up to 1.8 (CI = 1.5–2.1), indicating nearly an 80% higher likelihood of extreme low catch relative to baseline conditions. The impacts of high TempX on catches can be attributed to the pervasive effects of temperature on marine species and ecosystems 22 , 23 (see Supplementary Table 1 and Supplementary Fig. 3). Extreme high temperature impairs physiological performance, growth, and reproduction in fish and invertebrates, particularly during early life stages, thereby reducing cohort strength and subsequent year-class abundance 19 , 22 – 24 . These impacts may appear with a time lag as affected cohorts enter the fishery 4 , 25 , and recovery from prolonged extremes can take multiple years 26 , 27 . Low NPPX can compound these effects by limiting food availability, disrupting trophic interactions, and reducing prey resilience 17 , 28 . Delayed catch responses may also reflect lags in fisheries and their management, such as the timing of fishery closures or effort reductions 29 . The synergistic stress from high TempX and low NPPX thus create a cascade of social-ecological disturbances that undermines the abundance and availability of fishery catches, with potential localized exceptions due to shifts in species composition or fishing behaviour. While our main analysis focuses on surface-layer extremes, additional analyses examining extreme conditions at different depths show that surface-layer extremes are strongly associated with demersal catch declines, likely due to tight oceanographic and ecological coupling between surface and bottom layers that facilitates the propagation of thermal and productivity stress through the water column (see Supplementary Fig. 3). Using bottom-layer temperature anomalies yields consistent patterns and does not alter the main conclusions of this analysis. We tested the effects of “reverse” extremes (i.e., low TempX and high NPPX), the results of which are presented in a subsequent section. Spatial patterns in the effects of extreme environmental conditions Our spatial analysis revealed that high TempX significantly increased the odds of extreme low catches across some, but not all, EEZs (Fig. 2 A). Substantive elevation of odds ratios (OR ≥ 2) was most apparent along the eastern boundary upwelling systems, particularly the Humboldt and California Currents, as well as in the tropical eastern Pacific, including Peru, Ecuador, and Central America. Significant hotspots are also prevalent across small island nations in the Pacific, such as the Solomon Islands and Northwestern Hawaiian Islands, and in parts of the Indian Ocean, including Seychelles and Maldives. Additionally, semi-enclosed seas such as the Mediterranean Sea, Red Sea, and Persian Gulf exhibited consistently high odds of extreme low catches during TempX events, highlighting regional sensitivity to thermal extremes. These regions are also vulnerable to long-term ocean warming due to the narrow thermal tolerance of local species and limited adaptive capacity 30 , 31 . Our results suggest that these vulnerabilities may extend to acute temperature extremes, not just long-term warming trends, reinforcing concerns that climate-driven thermal variability poses a disproportionate threat to these fisheries. Low NPPX showed distinct spatial patterns from temperature extremes (Fig. 2 B). Elevated odds of extreme low catches were significantly (p < 0.05) associated with low NPPX in part of the temperate and subarctic regions, including the Barents Sea, Arctic Canada, North Sea, and northwest African coast, as well as parts of the Caribbean. Low NPPX also significantly (p < 0.05) increased the likelihood of low catches in the western Arabian Sea and around Mauritius, and several small island states in the South Pacific. These patterns reflect the ecological and biogeochemical role of primary production in supporting fish biomass and trophic transfer, particularly in regions where productivity is highly variable annually due to fluctuations in monsoons, nutrient upwelling, or ice dynamics 17 , 18 . These findings suggest that low NPPX affects fisheries through mechanisms that may be distinct from thermal stress. Compounded high TempX and low NPPX amplified OR of extreme low catches beyond those associated with univariate extremes across many EEZs (Fig. 2 C). These EEZs are found in regions including the eastern tropical Pacific, South China Sea, and eastern Indian Ocean. In several areas, including the southeast coast of Australia, significant increases in low catch odds occurred only when both temperature and productivity extremes co-occurred, highlighting the nonlinear and compounding nature of multi-stressor impacts on fisheries. In contrast, many non-significant areas (p > 0.05) were concentrated in shelf seas at intermediate latitudes, including the northwest and north Atlantic and the south Atlantic (grey shading, Fig. 2 ). In these regions, biogeographic shifts have already altered catch composition, with warm‐adapted species potentially more resilient to acute extremes 32 . In some EEZs, lower quality or resolution of catch statistics may also reduce the ability to detect robust relationships. These factors may help explain the weaker apparent influence of high TempX, low NPPX, or their co‐occurrence in these shelf sea regions. In addition, a limited number of EEZs exhibited effects in the opposite direction (odds ratio < 1). However, the spatial extent of these effects was limited compared to the dominant trend of increased vulnerability under extreme climate stressors, and in most cases, it is because of the association with significant negative effects of TempX (e.g., the Azores) or NPPX (e.g., Central Indonesia). Functional group-specific sensitivity to extremes The sensitivity of fisheries to climate extremes varied by functional groups, reflecting variation in responses due to differences in ecological traits and biogeography (Fig. 3 ). Across all extreme event types, sharks and rays, small pelagics, and commercial invertebrates were among the most sensitive groups, with the highest median odds ratios of extreme low catches (Fig. 3 A). These groups also experienced the most frequent and severe observed declines, with median catch drops exceeding 40% during extreme events (Fig. 3 B). Functional groups with broader distributions or higher mobility, such as benthopelagic and large pelagic fishes, exhibited more variable responses. In addition, reef-associated fishes showed heightened and widespread risk from NPPX and compounded TempX with NPPX, potentially due to their strong dependence on biogenic habitats, with the indirect effects stemming from coral bleaching, habitat degradation, and disrupted trophic interactions in addition to the direct temperature impacts from environmental extremes. These patterns of sensitivity to extremes and their potential implications for countries vulnerable to climate change were exemplified in distinct case studies (Fig. 3 , see Supplementary Fig. 4 and Supplementary Table 1 for illustrative examples for each case studies). Small pelagic fisheries in Peru, dominated by anchoveta ( Engraulis ringens ), showed extreme sensitivity to prolonged high temperature anomalies, with odds of extreme low catches increasing more than ninefold and median declines around 60% (Fig. 3 C) that is consistent with the close association of anchoveta biomass and its fisheries to El Niño Southern Oscillation 33 . Our analysis indicates that in the Solomon Islands, medium and large pelagic fisheries experienced up to a 41-fold increase in the odds of low catches under NPP extremes (Fig. 3 D), consistent with observed declines in NPP and the eastward displacement of tuna, such as skipjack tuna ( Katsuwonus pelamis ), during El Niño events 34 . Reef-associated fisheries in the Maldives were highly sensitive to lagged temperature extremes, with odds ratios exceeding 4 and typical catch losses of ~ 25%; in line with the very high vulnerability of tropical coral reef ecosystems in Indian Ocean, and globally, to heatwaves and other climatic extremes 35 , 36 (Fig. 3 E). Moreover, in the Galápagos Islands, shark and ray fisheries, such as those for blue shark ( Prionace glauca ), were particularly sensitive to singular high-temperature extremes, which disrupted thermocline structure and shifted vertical species distributions, affecting their catchability by fisheries (Fig. 3 G). Marine heatwaves have led to sharp declines in shark abundance in the waters around the Galápagos Islands, particularly in shallow habitats, due to altered thermal structure and prey availability ( 39 ). In temperate systems, commercial invertebrate fisheries in the United Kingdom were particularly vulnerable to compounded TempX and NPPX events (Fig. 3 F). Although compound extremes are rarely isolated as the sole drivers of historical invertebrate declines in UK waters, emerging evidence suggests that the combined effects of ocean warming, acidification, and productivity changes impair growth, recruitment, and survival of shellfish species 37 . Recent marine heatwaves have already caused major disruptions to UK shellfish fisheries, highlighting the growing threat that climate-driven extremes pose to invertebrate fisheries in temperate regions 38 . Overall, the functional-group contrasts illustrate the diverse mechanisms linking TempX, NPPX and low catches. The high sensitivity of sharks, rays, and small pelagics to co-occurring TempX–NPPX events may arise from habitat compression and altered vertical structure during stratification events, which constrain foraging and migration corridors. For demersal and reef-associated species, strong coupling between surface and bottom layers means that surface productivity shocks can propagate to benthic prey and habitat conditions. Beyond these physical drivers, other environmental, socio-economic, and management factors, such as changes in market demand, fishing effort, policy enforcement, and ecosystem interactions, also contributed to extreme low catches in many cases, often acting in combination with climatic extremes in complex and region-specific ways (Table S1 ). Our analysis of the “reverse” extremes, i.e., cold-year (low temperature) and high productivity (high NPP) events, revealed that these conditions do not produce mirror-image high-catch anomalies to those low catches being driven by high TempX and low NPPX (see Supplementary Fig. 5). Low TempX and high NPPX, whether singular or compound, yielded inconsistent and limited increases in extreme catch odds. Such asymmetric responses have been documented economically for the impacts of El Niño Southern Oscillation, with losses incurred during El Niño are rarely offset by gains in La Niña periods 39 . Biological lags in stock recovery and socio-economic delays further suppress rebounds 29 . In addition, the effects of high TempX and low NPPX extremes on low versus high extreme catch outcomes amongst EEZ-functional groups were only weakly positively correlated (Pearson’s r = 0.047, p < 0.001). While some fish stocks might have benefited from extreme conditions, these gains tend to occur in different species or regions than those that were negatively affected. These findings underscore the importance of diversification of fishing activities as a strategy for adapting to the growing risks posed by climatic extremes 40 . Projecting the future risks of low catches under TempX and NPPX We applied temperature and NPP projections from three Earth system models from the Coupled Model Intercomparison Project Phase 6 to our statistical model of extreme catch ORs to predict occurrences of future stock declines under the ‘strong mitigation, low emissions’ (SSP1-RCP2.6) and ‘no mitigation, high emissions’ (SSP5-RCP8.5) scenarios (see Method), the results show that the share of fish stocks experiencing extreme low catches in vulnerable EEZs is projected to increase from about 11% in the early 1990s to almost 16% by the mid-21st century, with similar trajectories under both low- and high-emission scenarios (Fig. 4 ). The similarity of trajectories between SSP1-2.6 and SSP5-8.5 reflects climate-ocean-system inertia and the dominance of near-term warming already committed under both scenarios, rather than implying that mitigation is unimportant. This suggests that uncertainties associated with emissions pathways in projecting the mid-century impacts of temperature (TempX) and productivity (NPPX) extremes on catches are relatively low. Climate-driven disruptions to fisheries are already unfolding and are expected to intensify in the next few decades regardless of mitigation, underscoring the urgency of adaptation. Reconciling conflicting evidence on temperature extremes and fish stock responses Reconciling conflicting evidence on temperature extremes and fish stock responses Our analysis helps resolve recent inconsistencies in the literature regarding the impact of marine heatwaves on fish stock biomass. Specifically, we examined demersal fish species in EEZs overlapping with those assessed by 9 and confirmed that high TempX alone exert limited and variable influence on the odds of extreme low catches of this subset of stocks between the 1990s and 2010s (see Supplementary Fig. 6). In contrast, low NPPX, particularly when lagged or combined across multiple years, significantly elevated the odds of low catches (OR > 1, p < 0.05). Notably, the strongest signal emerged when TempX and NPPX occurred in sequence, with lagged compounding effects nearly doubling the likelihood of low catch anomalies. This clarifies that productivity disruptions, not examined by Fredston et al. 9 , rather than heatwaves alone, may be associated with stock and fishery declines in some systems. Moreover, the geographic focus of Fredston et al 9 on the Northeast Pacific and North Atlantic regions, where TempX links to catch drops are comparatively weak (Fig. 2 ), further explains why their findings may not generalize globally or across other functional groups. In contrast, our findings align with global modeling studies that have identified extreme high temperatures and compound climate events as key drivers of fish biomass declines 13 , 26 , highlighting that persistent or compounded temperature and NPP extremes are particularly destabilizing for fisheries. The congruence between modeled probabilities and real-world disruptions reinforces the robustness of our statistical approach and underscores the urgent need to integrate extreme event risk into fisheries management and climate adaptation strategies. Considerations of key uncertainties Definitions of extremes and anomalies are based on historical baselines that may not reflect species-specific thresholds or ecological tipping points 41 . Our statistical models capture associations rather than mechanistic causality, and may not fully account for complex biological responses such as recruitment lags, range shifts, or altered food-web interactions, or linkages with the socio-economic and governance dimensions of fisheries 42 (Table S1 ). Regional variability in catch reporting, differences in monitoring capacity and management frameworks or actions introduce further uncertainty. Moreover, projections assume that past relationships between environmental extremes and catch anomalies remain stable under future conditions, an assumption challenged by potential species adaptation and evolving fisheries and management responses 43 , 44 . Importantly, co-occurring stressors such as overfishing, habitat degradation, or deoxygenation were not explicitly included, which may underestimate cumulative risks. These uncertainties underscore the importance of integrative, process-based models and improved monitoring to improve our knowledge to support adaptive fisheries governance. Conclusions Our findings underscore the urgent need for climate-adaptive fisheries management that addresses not only long-term environmental trends but also the growing risks of acute disruptions from marine extremes. Several sources of uncertainty warrant consideration (see Method for details), but these uncertainties generally suggest that our estimates of climate-driven fishery disruption are conservative. Strategies such as adaptive harvest control rules, early warning systems, diversification of target species, and strengthened monitoring and forecasting capacity will be essential for enhancing resilience. These needs are especially critical for at-risk fisheries in socio-economically vulnerable countries, where resources, scientific infrastructure, and human capacity are often limited. Targeted investments and international support will be vital to ensure these regions can respond effectively to the compounding pressures of physical and biogeochemical extremes. Methods Fisheries time-series data We obtained catch data from the Sea Around Us reconstruction database ( www.seaaroundus.org ), which estimates annual marine fisheries catches (in tonnes) by EEZ and species from 1991 to 2019. We follow the EEZ classification by the Sea Around Us, noting that it subdivides the EEZs of 198 coastal states into 280 regions (e.g., Canada’s EEZ is divided in Arctic, Atlantic and Pacific), including islands territories (updated 1 July 2015, available from https://www.seaaroundus.org ). In total, we analyzed 6,569 time-series comprising 178,256 catch records. Unlike landing-based datasets that attribute catches to the country of landing, the Sea Around Us data reconstructs where the catch was taken, assigning it to the spatial origin of fishing activity. This distinction is critical for our study, which aims to assess the spatial overlap between environmental extremes (e.g., temperature or NPP anomalies) and the regions where fish were actually harvested (see Supplementary Fig. 2). The reconstructed database uses a wide variety of sources to estimate total removals, including industrial, artisanal, subsistence, recreational catch, and discards, many of which are absent from official national reports 45 . It incorporates missing components of catch through the use of published and grey literature, fisheries reports, and expert knowledge, thereby producing a more comprehensive time series by species and country. While catch reconstructions offer critical advantages for spatially explicit analyses, they are subject to caveats 45 , 46 . The methods rely on assumptions and expert input in data-poor situations, which may introduce uncertainty or bias. The reconstruction approach and quality of underlying data also vary across taxa and regions, and current global fisheries statistics tend to exhibit a Global North bias, with better resolution and species identification in those regions than in the Global South. Historical sea temperature and net primary production data We used historical observation-based ocean temperature and model-based net primary production (NPP) datasets from the Copernicus Marine Environment Monitoring Service (CMEMS) to quantify environmental conditions associated with extreme fisheries catch anomalies. Sea water temperature was derived from the Global Ocean Ensemble Reanalysis product 47 , which provides monthly fields of global ocean temperature from 1993 to 2022 at 0.25° × 0.25° spatial resolution across 50 vertical levels. To match the resolution of other datasets, we bilinearly regridded the temperature data to a 0.5° × 0.5° global grid. Our main analysis used annual mean surface temperature (0–100 m). Moreover, we tested the effects of using depth strata that were more specific to species’ depth distribution: (i) surface layer (0–100 m), (ii) epipelagic layer (0–200 m), (iii) mesopelagic (200–1000 m) and (iv) bottom layer (deepest grid cell in the data product). In addition, we tested the effects of using average temperature across the depth strata. Each species was linked to the most relevant thermal layer based on its maximum depth from FishBase and SeaLifeBase: pelagic species with a depth below 200 m were linked to epipelagic temperature, demersal species to bottom-layer temperature, and upper water column species (maximum depth above 100 m) to surface layer (0-100 m). Net primary production (NPP) data were obtained from the biogeochemical hindcast for global ocean produced at Mercator-Ocean 48 , which provides global estimates of biogeochemical variables from 1993 onward using the PISCES model coupled with ocean reanalysis data. NPP is provided monthly at 0.25° × 0.25° spatial resolution. The robustness of the Mercator-Ocean hindcast dataset is supported by quantitative validation metrics detailed in the CMEMS Quality Information Document 49 , which reports seasonal and global agreement for primary productivity and related biogeochemical variables via bias, RMSD, and Estimated Accuracy Numbers (EANs) against satellite and climatology reference datasets. We selected this dataset because of its consistent temporal coverage, physical-biogeochemical coupling, and relatively high spatial resolution, which together offer robust estimates of NPP anomalies across EEZs and ocean basins. We applied the same bilinear interpolation and regridding approach to convert this dataset to a 0.5° × 0.5° grid, and computed depth integrated annual total NPP by summing daily values for each year and grid cell. The processed annual temperature and NPP fields were used to characterize environmental exposure for species across EEZs, identify climate extremes, and assess their association with extreme low and high catch anomalies in global fisheries. Projected future sea water temperature and net primary production under climate change We used future projections of sea surface temperature (SST) and net primary production (NPP) under climate change to assess future trends in the probability of extreme low catch events. Output from three Earth System Models (ESMs) were used for the analysis: IPSL-CM6A-LR, GFDL-ESM4, and MPI-ESM1-2-HR. Model outputs were taken from emissions scenarios SSP5-8.5 and SSP1-2.6 with simulations performed using the Coupled Model Intercomparison Project 6 (CMIP6) framework. The raw ESM data were provided on a 1° × 1° regular global grid at a monthly temporal resolution. To harmonize spatial resolution across datasets, we regridded the monthly outputs to a 0.5° × 0.5° regular grid using bilinear interpolation. For near-coastal grid cells where bilinear interpolation introduced missing values due to partial ocean coverage, we applied a distance-weighted extrapolation method. This method employed the haversine formula to calculate distances between grid cell centroids, and applied inverse-distance weighting to estimate values from surrounding valid ocean cells. Following spatial interpolation, the monthly data was aggregated into annual means for SST and sums for NPP (SST and NPP) for each grid cell. Kyoto Analyzing the effects of extreme temperature and net primary production on catches Prior to identifying extreme events, all environmental time series were statistically detrended to account for long-term climate trends that may confound interannual variability. We employed generalized additive models (GAMs) with cubic splines to fit temporal trends. The significance of the temporal trend was evaluated at α = 0.05. If significant, the residuals from the GAM were retained as detrended values; otherwise, mean-centered anomalies were used. This approach allowed us to isolate short-term deviations from long-term trends, thereby improving the robustness of our extreme event definitions. The detrended values were then used to compute anomalies for each variable. To quantify the effects of extreme temperature and net primary production conditions on fisheries catches, we applied generalized linear modeling (GLM) across individual species–EEZ combinations. We defined extreme events as annual anomalies in ocean temperature (Temp) and net primary production (NPP) that elevated beyond the 90th (high) or fell below the 10th (low) percentiles of their respective historical anomalies distributions of the detrended time-series. We also identified extra-ordinary extremes using the maximum and minimum values in the time-series for Temp and NPP, respectively. Climate anomalies were matched to each species based on sea surface temperature exposure and the EEZ(s) in which they occurred. Similarly, reversed extremes are defined as the 10th or 90th percentiles of the historical anomalies distributions of the detrended ST and NPP, respectively. For each species–EEZ time series, we modeled the probability of experiencing an extreme low catch year. This was defined as the bottom decile of the catch anomaly distribution, as a binary response variable using logistic regression. Predictor variables included indicators of extreme high temperature, low NPP, and their 1-year lagged values, along with their addition and interaction terms to capture compound event effects. To evaluate competing hypotheses about the importance of individual and compound stressors, we constructed and compared all combinations of predictor sets using GLMs and ranked them based on Akaike Information Criterion (AIC). Odds ratios (ORs) were derived from the coefficients of generalized linear models (GLMs) with a binomial error distribution and logit link. The GLMs estimated the relationship between the occurrence of extreme low catch events (binary response: 1 = event, 0 = no event) and predictors representing temperature extremes (TempX), primary production extremes (NPPX), and their combinations. For each predictor, the model coefficient (β) from the logit scale was exponentiated (OR = e^β) to express the change in odds of an extreme low catch event associated with the presence of the extreme condition, relative to baseline (non-extreme) conditions. We applied this modelling framework to all EEZ-species time-series units globally, and to individual EEZ and functional group separately. Model-averaged effect estimates were obtained using AIC weights, from which we derived weighted odds ratios (ORs) and associated 95% confidence intervals for each environmental extreme scenario. We considered effects statistically significant if the confidence interval did not include OR = 1. We presented model results globally and across EEZs and functional groups to assess spatial and taxonomic patterns in climate sensitivity. To identify key combinations of extremes, we developed a scenario matrix of binary combinations of the four predictor variables and computed model-averaged log-odds differences from a “baseline” (no extreme) condition. As a sensitivity analysis, we additionally fitted generalized estimating equations (GEEs) with an AR(1) correlation structure to account for temporal autocorrelation within species–EEZ panels using the geepack package in R (v. 4.4.0). This analysis examines the effects of accounting for temporal autocorrelation within species–EEZ panels and repeated measures through time on the estimated ORs of CatchX in relation to TempX and NPPX. The GEE-based results were consistent with those of the GLMs, showing no significant differences in the estimated odds ratios or their relative rankings across predictor combinations, confirming the robustness of our findings to temporal dependence in the data (see Supplementary Fig. 7). To assess future trends in the probability of extreme low catch events, we applied the fitted GLMs to Earth system model (ESM) projections of SST and NPP under two Shared Socioeconomic Pathways (SSP1-2.6 and SSP5-8.5). We calculated annual scenario-specific probabilities of low catch events, aggregated these at the EEZ and functional group levels, and derived time-series estimates of the proportion of species with extreme low catch events under each climate scenario. We then calculated the global average proportion amongst the EEZs weighted by the number of species assessed therein. All analyses were implemented in R (version 4.3.2). Declarations Acknowledgments: This research was enabled in part by support provided by the Digital Research Alliance of Canada (alliancecan.ca). We are thankful for the Sea Around Us and the Corpunicus data service that make fisheries and oceanographic data available in the public domain. Such data are critical for our research. We acknowledge funding support from: Natural Sciences and Engineering Research Council of Canada Discovery Grant (WLC); Social Sciences and Humanity Research Council of Canada Partnership Grant (WLC). Author contributions: Conceptualization: WLC; Methodology: WLC, TLF, JPA, IM; Investigation: WLC, IM; Visualization: WLC, TLF, JPA, IM ; Funding acquisition and project administration: WLC; WLC wrote the initial draft and all authors contribute to the writing, review and editing of the original draft. Competing interests: Authors declare that they have no competing interests Materials & Correspondence: William W.L. Cheung, [email protected] Data availability: All datasets and code underlying the results are made available via github repository [https://github.com/coruubc/CompoundExtremes.git]. There will be no restrictions on data availability or access when the manuscript is published. References Frölicher TL, Fischer EM (2018) Gruber, N. Marine heatwaves under global warming. Nature 560:360–364 Oliver ECJ et al (2018) Longer and more frequent marine heatwaves over the past century. Nat Commun 9:1324 IPCC. Climate Change (2023) : Synthesis Report. A Report of the Intergovernmental Panel on Climate Change. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC Geneva Switz. 24 (2023) Smale DA et al (2019) Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nat Clim Change 9:306–312 Smith KE et al (2023) Biological Impacts of Marine Heatwaves. Annu Rev Mar Sci 15:119–145 Cheung WW et al (2021) Marine high temperature extremes amplify the impacts of climate change on fish and fisheries. Sci Adv 7:eabh0895 Smith KE et al (2021) Socioeconomic impacts of marine heatwaves: Global issues and opportunities. Science 374:eabj3593 Villaseñor-Derbez JC, Arafeh-Dalmau N, Micheli F (2024) Past and future impacts of marine heatwaves on small-scale fisheries in Baja California, Mexico. Commun Earth Environ 5:623 Fredston AL et al (2023) Marine heatwaves are not a dominant driver of change in demersal fishes. Nature 1–6 Gruber N, Boyd PW, Frölicher TL, Vogt M (2021) Biogeochemical extremes and compound events in the ocean. Nature 600:395–407 Le Grix N, Zscheischler J, Laufkötter C, Rousseaux CS, Frölicher TL (2021) Compound high-temperature and low-chlorophyll extremes in the ocean over the satellite period. Biogeosciences 18:2119–2137 Le Grix N, Zscheischler J, Rodgers KB, Yamaguchi R, Frölicher TL (2022) Hotspots and drivers of compound marine heatwaves and low net primary production extremes. Biogeosciences 19:5807–5835 Le Grix N, Cheung WWL, Reygondeau G, Zscheischler J, Frölicher TL (2023) Extreme and compound ocean events are key drivers of projected low pelagic fish biomass. Glob Change Biol. https://doi.org/10.1111/gcb.16968 Wyatt AM, Resplandy L, Marchetti A (2022) Ecosystem impacts of marine heat waves in the northeast Pacific. Biogeosciences 19:5689–5705 Capotondi A, Alexander MA, Bond NA, Curchitser EN, Scott JD (2012) Enhanced upper ocean stratification with climate change in the CMIP3 models. 117:1–23 Cheung WWL, Close C, Lam V, Watson R, Pauly D (2008) Application of macroecological theory to predict effects of climate change on global fisheries potential. Mar Ecol Prog Ser 365:187–197 Stock CA et al (2017) Reconciling fisheries catch and ocean productivity. Proc. Natl. Acad. Sci. 114, E1441—-E1449 Heneghan RF et al (2021) Disentangling diverse responses to climate change among global marine ecosystem models. Prog Oceanogr 198:102659 Szuwalski CS, Aydin K, Fedewa EJ, Garber-Yonts B, Litzow MA (2023) The collapse of eastern Bering Sea snow crab. Science 382:306–310 Mills KE et al (2013) Fisheries management in a changing climate: lessons from the 2012 ocean heat wave in the Northwest Atlantic. Oceanography 26:191–195 Free CM et al (2019) Impacts of historical warming on marine fisheries production. Science 363:979–983 Pörtner HO, Peck MA (2010) Climate change effects on fishes and fisheries: towards a cause-and-effect understanding. J Fish Biol 77:1745–1779 Alter K et al (2024) Hidden impacts of ocean warming and acidification on biological responses of marine animals revealed through meta-analysis. Nat Commun 15:2885 Alfonso S, Gesto M, Sadoul B (2021) Temperature increase and its effects on fish stress physiology in the context of global warming. J Fish Biol 98:1496–1508 Pershing AJ et al (2015) Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery. Science 350:809–812 de Guibourd V, Gascuel D, Reygondeau G, Cheung WW (2024) L. Large potential impacts of marine heatwaves on ecosystem functioning. Glob Change Biol 30:e17437 Suryan RM et al (2021) Ecosystem response persists after a prolonged marine heatwave. Sci Rep 11:6235 du Pontavice H, Gascuel D, Reygondeau G, Stock C, Cheung WW (2021) Climate-induced decrease in biomass flow in marine food webs may severely affect predators and ecosystem production. Glob Change Biol 27:2608–2622 Free CM et al (2023) Impact of the 2014–2016 marine heatwave on US and Canada West Coast fisheries: Surprises and lessons from key case studies. Fish Fish 24:652–674 Bindoff NL et al (2019) Changing Ocean, Marine Ecosystems, and Dependent Communities Jones MC, Cheung WWL (2015) Multi-model ensemble projections of climate change effects on global marine biodiversity. ICES J Mar Sci 72:741–752 Watson RA et al (2013) Global marine yield halved as fishing intensity redoubles. Fish Fish 14:493–503 Gutiérrez D, Akester M, Naranjo L (2016) Productivity and sustainable management of the Humboldt Current large marine ecosystem under climate change. Environ Dev 17:126–144 Bell JD et al (2013) Mixed responses of tropical Pacific fisheries and aquaculture to climate change. Nat Clim Change 3:591–599 Bessell-Browne P, Epstein HE, Hall N, Buerger P, Berry K (2021) Severe heat stress resulted in high coral mortality on Maldivian Reefs following the 2015–2016 El Niño Event. in Oceans vol. 2 233–245MDPI Brown CJ, Mellin C, Edgar GJ, Campbell MD, Stuart-Smith RD (2021) Direct and indirect effects of heatwaves on a coral reef fishery. Glob Change Biol 27:1214–1225 Townhill B, Artioli Y, Pinnegar J, Birchenough S (2022) Exposure of commercially exploited shellfish to changing pH levels: how to scale-up experimental evidence to regional impacts. ICES J Mar Sci 79:2362–2372 Harvey F (2024) Catastrophic marine heatwaves are killing sealife and causing mass disruption to UK fisheries. The Guardian https://www.theguardian.com/environment/2024/nov/23/catastrophic-marine-heatwaves-are-killing-sealife-and-causing-mass-disruption-to-uk-fisheries Osgood GJ, White ER, Baum JK (2021) Effects of climate-change-driven gradual and acute temperature changes on shark and ray species. J Anim Ecol 90:2547–2559 Cline TJ, Schindler DE, Hilborn R (2017) Fisheries portfolio diversification and turnover buffer Alaskan fishing communities from abrupt resource and market changes. Nat Commun 8:14042 Amaya DJ et al (2023) Marine heatwaves need clear definitions so coastal communities can adapt. Nature 616:29–32 Hertz T et al (2024) Eliciting the plurality of causal reasoning in social-ecological systems research. Ecol Soc 29 Reusch TBH (2014) Climate change in the oceans: Evolutionary versus phenotypically plastic responses of marine animals and plants. Evol Appl 7:104–122 Moore JW, Schindler DE (2022) Getting ahead of climate change for ecological adaptation and resilience. Science 376:1421–1426 Zeller D et al (2016) Still catching attention: Sea Around Us reconstructed global catch data, their spatial expression and public accessibility. Mar Policy 70:145–152 Pauly D, Zeller D (2016) Catch reconstructions reveal that global marine fisheries catches are higher than reported and declining. Nat Commun 7:10244 Mercator Ocean International (2023) Global Ocean Ensemble Physics Reanalysis (GLOBAL_MULTIYEAR_PHY_ENS_001_031). EU Copernicus Marine Service https://doi.org/10.48670/moi-00024 Mercator Océan International (2024) Global Ocean Biogeochemistry Hindcast (GLOBAL_MULTIYEAR_BGC_001_029). Copernicus Marine Service https://doi.org/10.48670/moi-00019 Perruche C, Szczypta C, Paul J, Drévillon M (2024) Quality Information Document for GLOBAL_MULTIYEAR_BGC_001_029 . https://doi.org/10.48670/moi-00019 doi:10.48670/moi-00019 Additional Declarations There is NO Competing Interest. Supplementary Files CheungetalSupplementary261025formatted.docx Supplementary Information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8109072","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":546527133,"identity":"74dc0251-727d-4966-969e-cfee775438a0","order_by":0,"name":"William Cheung","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6UlEQVRIiWNgGAWjYHACNjDJD+dLEKtFsoFkLQYHiNVicID52YOPbXfsNh9vfvjg5446BoPbzQ8YftTg08Jmbjiz7VnytjPHjA17zxxmMLhzzICx5xhuLWYHeNikedsOJ5vdyGGT4G07wGBwI4eBGepa3Fr+ArUYz8hh//m3rQ6q5R8BLYxth+0MJHLYmHnbmCFaGNtwa7E/zGYm2XPucIIE0C/Ssm2HeSSBfjnY24dbi2R78zOJH2WH7fnbmx9+fNtWJ8d3Gxh0P77h1gJ0NhgkNkD5PCDiAB4NCAcSo2gUjIJRMApGKAAAW09QQmJ/rkoAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-9998-0384","institution":"University of British Columbia","correspondingAuthor":true,"prefix":"","firstName":"William","middleName":"","lastName":"Cheung","suffix":""},{"id":546527134,"identity":"aa91cd81-c155-452d-9153-faf069132b76","order_by":1,"name":"Thomas Frölicher","email":"","orcid":"https://orcid.org/0000-0003-2348-7854","institution":"University of Bern","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Frölicher","suffix":""},{"id":546527135,"identity":"23ebbce5-98bf-4aa0-a9c6-d8a49458ed11","order_by":2,"name":"Juliano Palacios-Abrantes","email":"","orcid":"https://orcid.org/0000-0001-8969-5416","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Juliano","middleName":"","lastName":"Palacios-Abrantes","suffix":""},{"id":546527136,"identity":"85c6c280-1a7d-4758-9fad-75f2cd4ddd88","order_by":3,"name":"Isabella Morgante","email":"","orcid":"https://orcid.org/0009-0006-2208-3451","institution":"The University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Isabella","middleName":"","lastName":"Morgante","suffix":""}],"badges":[],"createdAt":"2025-11-13 21:40:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8109072/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8109072/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96236680,"identity":"b586354d-199d-4da5-97d8-737d85e09aa1","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6374951,"visible":true,"origin":"","legend":"","description":"","filename":"CheungetalfullmsfinalformattedNatComm.docx","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/131c9672a1463513c887188a.docx"},{"id":96236675,"identity":"a14f77da-547b-4a7b-b2cd-ac27ed08bd45","added_by":"auto","created_at":"2025-11-19 06:14:04","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5539,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS2590332.json","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/0b933672f15fd2177ac5f5b0.json"},{"id":96236685,"identity":"1c21251a-0c98-476f-ad3f-6bb391698b05","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7198840,"visible":true,"origin":"","legend":"","description":"","filename":"CheungetalSupplementary261025formatted.docx","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/15cf3a73c9f3a720e1fd404f.docx"},{"id":96253316,"identity":"1141d002-ed00-42a8-bd75-82682236e440","added_by":"auto","created_at":"2025-11-19 07:42:18","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109648,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS25903320enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/a7990ae936bcc65f00a2618b.xml"},{"id":96253217,"identity":"38cd2bf1-55b8-40e4-b2f4-6614fc493f09","added_by":"auto","created_at":"2025-11-19 07:42:09","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":551019,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/a191c3b7a068ab27574ad592.png"},{"id":96236688,"identity":"64ca558a-c9cd-49ab-8d89-4cb73fbbf1a3","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1233430,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/f2c5fa842b9be90df74330f4.png"},{"id":96253374,"identity":"7e4c1823-3fd8-4827-96c1-f71d9eaa1d11","added_by":"auto","created_at":"2025-11-19 07:42:22","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":186119,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/92b080599725e76c980e42c3.jpeg"},{"id":96236683,"identity":"8a44aae1-70e3-4386-bd95-748c93ea8135","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":418463,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/6c272041a897ae85fe2be8f3.png"},{"id":96236689,"identity":"23dbfbaf-b254-4591-9c73-8f70de40d2ea","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":81043,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/1323a4255c4e14de7b93b732.png"},{"id":96252954,"identity":"6d72565f-4662-4eaa-9e9e-d8e22a071c9c","added_by":"auto","created_at":"2025-11-19 07:41:42","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":177732,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/b9ad7e14ed69d3726e43fe61.png"},{"id":96236686,"identity":"67ad45a9-2d53-44fa-911f-ad6249148b76","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":42462,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/2644fb0193dffba9d0eb3478.png"},{"id":96252010,"identity":"2d744025-7bb9-40dc-9b21-1d4350e9beb2","added_by":"auto","created_at":"2025-11-19 07:40:19","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":115069,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/a0a053b7535f98b878e35b95.png"},{"id":96236691,"identity":"26d23307-9da6-4233-a326-6ef3c4fea2d6","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":106679,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS25903320structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/a6c4299ca63513abcca368b3.xml"},{"id":96236692,"identity":"25d6a671-57a7-4e2c-9e7f-80d6e03db066","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":116636,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/9d409efb25c5e8e4c3433a97.html"},{"id":96236673,"identity":"9a3659b1-e68f-47da-85ea-eaf4fc77baf1","added_by":"auto","created_at":"2025-11-19 06:14:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":622610,"visible":true,"origin":"","legend":"\u003cp\u003eEstimated Odds Ratios (OR) for the occurrence of extreme low fish catch events in relation to extreme high temperature (T), extreme low net primary production (N), and their 1-year lagged effects (T_lag and N_lag) across all exclusive economic zones (EEZs) and exploited fishes and invertebrates over the period 2002 to 2019. Variables marked with an asterisk (†) and in darker red and green colour represent extra-ordinary extreme high temperature or low NPP events while the lighter red and green colours represent ordinary extreme events (10 percentile outliers) occurring within the time series. Grey dots are statistically not significant (p\u0026gt;0.05) while the blue dot indicates the estimated OR is significantly below 1 (p\u0026lt;0.05). (+) refers to additive effects while (X) is interaction terms capturing the compound effects of T, N and their lags in the generalized linear modelling (see Method). Error bars indicate 95 percent confidence intervals.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/da7367e877b8e4a862c58314.png"},{"id":96236676,"identity":"52a1e32c-2bf0-4f0b-9d7d-e3e7c6cb7af0","added_by":"auto","created_at":"2025-11-19 06:14:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1168261,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal maps of the estimated effects of environmental extremes on the odds ratio of extreme low‐catch events relative to normal conditions between 2002 to 2019. (A) Extreme high temperature effects; (B) Extreme low NPP effects; (C) Combined effect of co‐occurring high temperature and low NPP extremes. Analysis included events that occurred as 90 percentile and 10 percentile of the high temperature and low NPP anomalies, respectively. Shaded areas represent non-significant effects (p\u0026gt;0.05). EEZ area without colour indicates that estimated odds ratio is not available, primarily due to insufficient compound TempX and NPPX co-occurrences in the time period for robust statistical analysis.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/4fb8a2d53664466d348f7738.png"},{"id":96236674,"identity":"889e90cc-17ed-4836-9c91-9019e6bbf7e2","added_by":"auto","created_at":"2025-11-19 06:14:04","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":165069,"visible":true,"origin":"","legend":"\u003cp\u003eFisheries-specific sensitivities to extreme high temperature and low net primary production events. (A) Odds ratios (ORs) for extreme low catches associated with temperature extremes (TempX), NPP extremes (NPPX), and compounded TempX–NPPX events across seven functional groups. Bars show median ORs across species–Exclusive Economic Zone (EEZ) combinations. Colors indicate the proportion of EEZs where significant effects were detected (A) and the total number of EEZs where the species groups are exposed to extreme conditions (B). (B) Corresponding observed median percentage declines in catch during extreme low catch years, highlighting the severity of biomass reductions. (C–G) Case study examples show ORs and 95% confidence intervals for specific fisheries with significant responses: small pelagic fishes in Peru (C), medium and large pelagic fishes in the Solomon Islands (D), reef-associated fishes in the Maldives (E), sharks and rays in the Galápagos Islands (G), and commercial invertebrates in the United Kingdom (F). Symbols denote the type of extreme events that had significant effects: T (high temperature), T† (extra-ordinary high temperature), N (low NPP), and N† (extra-ordinary low NPP), with lags and compound (+ for additive and x for interactive) events indicated where relevant. For the boxplots in A and B, the vertical boundaries of each box represent the 25th percentile, median, and 75th percentile values; the horizontal lines show the lower and upper ranges, and the red dots mark identified outliers.\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/a87e8be38e897332d627a96d.jpeg"},{"id":96236678,"identity":"334a08b1-c9df-47eb-b87d-869903458522","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":378107,"visible":true,"origin":"","legend":"\u003cp\u003eProjected increases in extreme low catch events under climate change. Time series of the percentage of stocks (EEZ–species combinations) experiencing extreme low catches—defined as observed catches falling below the lower bounds expected from non-extreme years—under two Shared Socioeconomic Pathways (SSP1-2.6 in blue, SSP5-8.5 in orange). Solid lines represent the mean projections across three Earth System Models (ESMs) based on fitted statistical relationships. Shaded ribbons denote the combined uncertainty from the statistical model (95% confidence intervals) and inter-model spread among the ESMs. Grey line and area represent historical periods of the ESMs.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/86cfe7614ce3828e4b3a01cc.png"},{"id":96257248,"identity":"347d9efe-d45d-4b52-b5e1-10f97dd98436","added_by":"auto","created_at":"2025-11-19 07:51:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2882066,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/5c436b4a-dee4-4451-89b7-887c5af57375.pdf"},{"id":96236677,"identity":"a454f598-02d0-4027-b7cf-5bb18cfa81a3","added_by":"auto","created_at":"2025-11-19 06:14:05","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":7198840,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"CheungetalSupplementary261025formatted.docx","url":"https://assets-eu.researchsquare.com/files/rs-8109072/v1/18f82e30a1776cadf125e9d1.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Large Fisheries Declines Linked to Compound and Extreme Climate Events","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe increasing frequency and intensity of extreme ocean climate events\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e are causing localized declines in marine life, altering ecosystem structure, and disrupting fisheries \u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. These changes result in substantial socio-economic impacts such as fishery closures, revenue losses, and diminished livelihoods\u003csup\u003e\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Marine heatwaves, or extreme high ocean temperatures (TempX), have emerged as prominent drivers of fisheries disruption, though empirical evidence remains geographically and taxonomically limited\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. In contrast, the effects of biogeochemical extremes\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, such as extreme low net primary production (NPPX), and their interactions with high TempX are far less understood, especially at spatial and temporal scales relevant for fisheries management\u003csup\u003e\u003cspan additionalcitationids=\"CR12 CR13\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Despite advances in physical oceanography and ecosystem modelling, empirical, global-scale tests of how such local compound extremes translate into local realized fisheries outcomes have been lacking. Advancing our understanding of how individual and compounded marine extremes affect fisheries is essential for assessing emerging risks to seafood security and for guiding strategies to enhance the resilience of ocean management under a rapidly changing climate\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eTemperature and net primary production (NPP) are key drivers of the distribution, biomass and production of species important to fisheries\u003csup\u003e\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. However, a recent meta-analysis of biomass and temperature time-series found that marine heatwaves have muted and variable impacts on demersal fish populations \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. In contrast, observed high TempX has driven rapid biomass declines and fisheries closures of commercially important species (e.g.,\u003csup\u003e19, 20\u003c/sup\u003e). Theoretical models further project that extreme temperature and its compounding effects with biogeochemical extremes reduce fish biomass \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. These contrasting empirical and modelling results may reflect limitations in available datasets and the complex, region- and species- specific interactions among environmental variables, ecosystems and other human drivers \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. This underscores a critical need to reconcile the evidence base and robustly assess the risks that climate-driven extremes experienced by marine species pose to fisheries at the local scale.\u003c/p\u003e\u003cp\u003eThis study quantifies the impacts of annual TempX and NPPX, and compounded TempX with NPPX events on local fish stocks and fisheries at the global scale. Here, environmental and fisheries extremes arise from variability superimposed on long-term mean trends (see Supplementary Fig.\u0026nbsp;1, Methods). We focus on annual extremes as these events often have the greatest impacts on fisheries \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. By integrating species biogeography with fisheries catch, temperature and net primary production time-series data from 1993 to 2019 with statistical modeling (see Methods), we evaluate two main hypotheses: (1) high TempX and low NPPX are associated with extreme low fisheries catches across 254 EEZs and 1,246 exploited fishes and invertebrates, (2) the strength and direction of these associations vary across regions and functional groups, and (3) the likelihood of extreme low fisheries catch events will increase under future climate change.\u003c/p\u003e\u003cp\u003eOur analysis was structured in three tiers (see Supplementary Fig.\u0026nbsp;2 and Method). First, we pooled all observations across species\u0026ndash;EEZ groups, treating each species\u0026ndash;EEZ\u0026ndash;year combination as an independent sample to estimate a relationship between extreme environmental conditions exposed to the specific species-EEZ unit and extreme low fisheries catches at the EEZ level, which we then aggregated globally. Second, we conducted EEZ-specific analyses to assess how these relationships vary geographically. Third, we grouped species into functional categories and repeated the analysis to evaluate responses at the functional-group level. We adopt a risk-based framework applied to 6,659 fisheries and environmental time-series data to quantify how environmental extremes alter the likelihood of extreme low fisheries catches across species at the functional groups and Exclusive Economic Zone levels, as well as globally aggregated. Using a relative-probability approach analogous to epidemiological risk analysis (see Method), we estimate how much the odds of extreme low catch events, defined as the lowest decile of catch anomalies, increase during years of TempX and/or NPPX. This framework captures the probabilistic nature of extreme and compound climate risks, revealing not only whether extremes coincide with drops in fisheries production, but also how strongly they amplify the underlying risk compared to non-extreme conditions. It therefore bridges empirical observation and theory by reconciling the frequency of observed co-occurrences with the modeled amplification of low catch likelihood under extreme or compound stressors.\u003c/p\u003e"},{"header":"Significant global effects of univariate and compounded extremes on catches","content":"\u003cp\u003eGlobally, across all EEZs, our analysis indicates that univariate extreme environmental conditions significantly increase the likelihood of extreme low fisheries catches (odds ratio\u0026thinsp;\u0026gt;\u0026thinsp;1; p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Here, extreme events are defined as annual mean anomalies falling beyond the 10th (for NPP) or 90th (for temperature) percentile of the historical distribution (see Method). Odds ratio (OR) quantifies the strength of association between environmental extremes and the likelihood of an extreme low fisheries catch event, with OR\u0026thinsp;\u0026gt;\u0026thinsp;1 indicating increased odds relative to baseline (non-extreme) conditions in the time-series. We estimated odds ratios using logistic regression models, where the occurrence of an extreme low catch (binary response) was modeled as a function of extreme environmental conditions. For example, OR of 1.5 indicates that the likelihood of occurrence of low catches is 50% higher during extreme conditions compared to baseline years. In contrast, OR below 1 indicates a reduced likelihood of low catches during extreme conditions. Two consecutive years with high TempX were identified as the most important univariate driver, increasing the odds of an extreme low catch by 40% (95% CI: odds ratio\u0026thinsp;=\u0026thinsp;20\u0026ndash;65%) compared to baseline conditions. Considering only extra-ordinary TempX (the most extreme conditions in the historical record) did not further elevate the odds of extreme low catches compared with ordinary high TempX (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In contrast, univariate NPPX events significantly affected catch anomalies only when the lowest NPP in a time-series were classified as extreme.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eCompound TempX with NPPX events amplified the risk of extreme low catches (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Several combinations of co-occurring or sequential TempX and NPPX events significantly increased the odds of extreme low catch events beyond those of univariate extremes. For instance, the occurrence of lagged TempX with concurrent extra-ordinary NPPX yielded odds ratios up to 1.8 (CI\u0026thinsp;=\u0026thinsp;1.5\u0026ndash;2.1), indicating nearly an 80% higher likelihood of extreme low catch relative to baseline conditions.\u003c/p\u003e\u003cp\u003eThe impacts of high TempX on catches can be attributed to the pervasive effects of temperature on marine species and ecosystems \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e (see Supplementary Table\u0026nbsp;1 and Supplementary Fig.\u0026nbsp;3). Extreme high temperature impairs physiological performance, growth, and reproduction in fish and invertebrates, particularly during early life stages, thereby reducing cohort strength and subsequent year-class abundance \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan additionalcitationids=\"CR23\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. These impacts may appear with a time lag as affected cohorts enter the fishery \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e, and recovery from prolonged extremes can take multiple years \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Low NPPX can compound these effects by limiting food availability, disrupting trophic interactions, and reducing prey resilience \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Delayed catch responses may also reflect lags in fisheries and their management, such as the timing of fishery closures or effort reductions \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. The synergistic stress from high TempX and low NPPX thus create a cascade of social-ecological disturbances that undermines the abundance and availability of fishery catches, with potential localized exceptions due to shifts in species composition or fishing behaviour.\u003c/p\u003e\u003cp\u003eWhile our main analysis focuses on surface-layer extremes, additional analyses examining extreme conditions at different depths show that surface-layer extremes are strongly associated with demersal catch declines, likely due to tight oceanographic and ecological coupling between surface and bottom layers that facilitates the propagation of thermal and productivity stress through the water column (see Supplementary Fig.\u0026nbsp;3). Using bottom-layer temperature anomalies yields consistent patterns and does not alter the main conclusions of this analysis. We tested the effects of \u0026ldquo;reverse\u0026rdquo; extremes (i.e., low TempX and high NPPX), the results of which are presented in a subsequent section.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eSpatial patterns in the effects of extreme environmental conditions\u003c/h2\u003e\u003cp\u003eOur spatial analysis revealed that high TempX significantly increased the odds of extreme low catches across some, but not all, EEZs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Substantive elevation of odds ratios (OR\u0026thinsp;\u0026ge;\u0026thinsp;2) was most apparent along the eastern boundary upwelling systems, particularly the Humboldt and California Currents, as well as in the tropical eastern Pacific, including Peru, Ecuador, and Central America. Significant hotspots are also prevalent across small island nations in the Pacific, such as the Solomon Islands and Northwestern Hawaiian Islands, and in parts of the Indian Ocean, including Seychelles and Maldives. Additionally, semi-enclosed seas such as the Mediterranean Sea, Red Sea, and Persian Gulf exhibited consistently high odds of extreme low catches during TempX events, highlighting regional sensitivity to thermal extremes. These regions are also vulnerable to long-term ocean warming due to the narrow thermal tolerance of local species and limited adaptive capacity \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Our results suggest that these vulnerabilities may extend to acute temperature extremes, not just long-term warming trends, reinforcing concerns that climate-driven thermal variability poses a disproportionate threat to these fisheries.\u003c/p\u003e\u003cp\u003eLow NPPX showed distinct spatial patterns from temperature extremes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Elevated odds of extreme low catches were significantly (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) associated with low NPPX in part of the temperate and subarctic regions, including the Barents Sea, Arctic Canada, North Sea, and northwest African coast, as well as parts of the Caribbean. Low NPPX also significantly (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) increased the likelihood of low catches in the western Arabian Sea and around Mauritius, and several small island states in the South Pacific. These patterns reflect the ecological and biogeochemical role of primary production in supporting fish biomass and trophic transfer, particularly in regions where productivity is highly variable annually due to fluctuations in monsoons, nutrient upwelling, or ice dynamics \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. These findings suggest that low NPPX affects fisheries through mechanisms that may be distinct from thermal stress.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eCompounded high TempX and low NPPX amplified OR of extreme low catches beyond those associated with univariate extremes across many EEZs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). These EEZs are found in regions including the eastern tropical Pacific, South China Sea, and eastern Indian Ocean. In several areas, including the southeast coast of Australia, significant increases in low catch odds occurred only when both temperature and productivity extremes co-occurred, highlighting the nonlinear and compounding nature of multi-stressor impacts on fisheries.\u003c/p\u003e\u003cp\u003eIn contrast, many non-significant areas (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05) were concentrated in shelf seas at intermediate latitudes, including the northwest and north Atlantic and the south Atlantic (grey shading, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In these regions, biogeographic shifts have already altered catch composition, with warm‐adapted species potentially more resilient to acute extremes \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. In some EEZs, lower quality or resolution of catch statistics may also reduce the ability to detect robust relationships. These factors may help explain the weaker apparent influence of high TempX, low NPPX, or their co‐occurrence in these shelf sea regions. In addition, a limited number of EEZs exhibited effects in the opposite direction (odds ratio\u0026thinsp;\u0026lt;\u0026thinsp;1). However, the spatial extent of these effects was limited compared to the dominant trend of increased vulnerability under extreme climate stressors, and in most cases, it is because of the association with significant negative effects of TempX (e.g., the Azores) or NPPX (e.g., Central Indonesia).\u003c/p\u003e\u003c/div\u003e"},{"header":"Functional group-specific sensitivity to extremes","content":"\u003cp\u003eThe sensitivity of fisheries to climate extremes varied by functional groups, reflecting variation in responses due to differences in ecological traits and biogeography (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Across all extreme event types, sharks and rays, small pelagics, and commercial invertebrates were among the most sensitive groups, with the highest median odds ratios of extreme low catches (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). These groups also experienced the most frequent and severe observed declines, with median catch drops exceeding 40% during extreme events (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Functional groups with broader distributions or higher mobility, such as benthopelagic and large pelagic fishes, exhibited more variable responses. In addition, reef-associated fishes showed heightened and widespread risk from NPPX and compounded TempX with NPPX, potentially due to their strong dependence on biogenic habitats, with the indirect effects stemming from coral bleaching, habitat degradation, and disrupted trophic interactions in addition to the direct temperature impacts from environmental extremes.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThese patterns of sensitivity to extremes and their potential implications for countries vulnerable to climate change were exemplified in distinct case studies (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, see Supplementary Fig.\u0026nbsp;4 and Supplementary Table\u0026nbsp;1 for illustrative examples for each case studies). Small pelagic fisheries in Peru, dominated by anchoveta (\u003cem\u003eEngraulis ringens\u003c/em\u003e), showed extreme sensitivity to prolonged high temperature anomalies, with odds of extreme low catches increasing more than ninefold and median declines around 60% (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC) that is consistent with the close association of anchoveta biomass and its fisheries to El Ni\u0026ntilde;o Southern Oscillation \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Our analysis indicates that in the Solomon Islands, medium and large pelagic fisheries experienced up to a 41-fold increase in the odds of low catches under NPP extremes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD), consistent with observed declines in NPP and the eastward displacement of tuna, such as skipjack tuna (\u003cem\u003eKatsuwonus pelamis\u003c/em\u003e), during El Ni\u0026ntilde;o events \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Reef-associated fisheries in the Maldives were highly sensitive to lagged temperature extremes, with odds ratios exceeding 4 and typical catch losses of ~\u0026thinsp;25%; in line with the very high vulnerability of tropical coral reef ecosystems in Indian Ocean, and globally, to heatwaves and other climatic extremes \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). Moreover, in the Gal\u0026aacute;pagos Islands, shark and ray fisheries, such as those for blue shark (\u003cem\u003ePrionace glauca\u003c/em\u003e), were particularly sensitive to singular high-temperature extremes, which disrupted thermocline structure and shifted vertical species distributions, affecting their catchability by fisheries (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG). Marine heatwaves have led to sharp declines in shark abundance in the waters around the Gal\u0026aacute;pagos Islands, particularly in shallow habitats, due to altered thermal structure and prey availability (\u003cem\u003e39\u003c/em\u003e). In temperate systems, commercial invertebrate fisheries in the United Kingdom were particularly vulnerable to compounded TempX and NPPX events (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). Although compound extremes are rarely isolated as the sole drivers of historical invertebrate declines in UK waters, emerging evidence suggests that the combined effects of ocean warming, acidification, and productivity changes impair growth, recruitment, and survival of shellfish species \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Recent marine heatwaves have already caused major disruptions to UK shellfish fisheries, highlighting the growing threat that climate-driven extremes pose to invertebrate fisheries in temperate regions \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eOverall, the functional-group contrasts illustrate the diverse mechanisms linking TempX, NPPX and low catches. The high sensitivity of sharks, rays, and small pelagics to co-occurring TempX\u0026ndash;NPPX events may arise from \u003cem\u003ehabitat compression\u003c/em\u003e and altered vertical structure during stratification events, which constrain foraging and migration corridors. For demersal and reef-associated species, strong coupling between surface and bottom layers means that surface productivity shocks can propagate to benthic prey and habitat conditions. Beyond these physical drivers, other environmental, socio-economic, and management factors, such as changes in market demand, fishing effort, policy enforcement, and ecosystem interactions, also contributed to extreme low catches in many cases, often acting in combination with climatic extremes in complex and region-specific ways (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eOur analysis of the \u0026ldquo;reverse\u0026rdquo; extremes, i.e., cold-year (low temperature) and high productivity (high NPP) events, revealed that these conditions do not produce mirror-image high-catch anomalies to those low catches being driven by high TempX and low NPPX (see Supplementary Fig.\u0026nbsp;5). Low TempX and high NPPX, whether singular or compound, yielded inconsistent and limited increases in extreme catch odds. Such asymmetric responses have been documented economically for the impacts of El Ni\u0026ntilde;o Southern Oscillation, with losses incurred during El Ni\u0026ntilde;o are rarely offset by gains in La Ni\u0026ntilde;a periods \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Biological lags in stock recovery and socio-economic delays further suppress rebounds \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. In addition, the effects of high TempX and low NPPX extremes on low versus high extreme catch outcomes amongst EEZ-functional groups were only weakly positively correlated (Pearson\u0026rsquo;s r\u0026thinsp;=\u0026thinsp;0.047, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). While some fish stocks might have benefited from extreme conditions, these gains tend to occur in different species or regions than those that were negatively affected. These findings underscore the importance of diversification of fishing activities as a strategy for adapting to the growing risks posed by climatic extremes \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Projecting the future risks of low catches under TempX and NPPX","content":"\u003cp\u003eWe applied temperature and NPP projections from three Earth system models from the Coupled Model Intercomparison Project Phase 6 to our statistical model of extreme catch ORs to predict occurrences of future stock declines under the \u0026lsquo;strong mitigation, low emissions\u0026rsquo; (SSP1-RCP2.6) and \u0026lsquo;no mitigation, high emissions\u0026rsquo; (SSP5-RCP8.5) scenarios (see Method), the results show that the share of fish stocks experiencing extreme low catches in vulnerable EEZs is projected to increase from about 11% in the early 1990s to almost 16% by the mid-21st century, with similar trajectories under both low- and high-emission scenarios (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The similarity of trajectories between SSP1-2.6 and SSP5-8.5 reflects climate-ocean-system inertia and the dominance of near-term warming already committed under both scenarios, rather than implying that mitigation is unimportant. This suggests that uncertainties associated with emissions pathways in projecting the mid-century impacts of temperature (TempX) and productivity (NPPX) extremes on catches are relatively low. Climate-driven disruptions to fisheries are already unfolding and are expected to intensify in the next few decades regardless of mitigation, underscoring the urgency of adaptation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Reconciling conflicting evidence on temperature extremes and fish stock responses","content":"\u003cdiv class=\"Heading\"\u003eReconciling conflicting evidence on temperature extremes and fish stock responses\u003c/div\u003e\u003cp\u003eOur analysis helps resolve recent inconsistencies in the literature regarding the impact of marine heatwaves on fish stock biomass. Specifically, we examined demersal fish species in EEZs overlapping with those assessed by \u003csup\u003e9\u003c/sup\u003e and confirmed that high TempX alone exert limited and variable influence on the odds of extreme low catches of this subset of stocks between the 1990s and 2010s (see Supplementary Fig.\u0026nbsp;6). In contrast, low NPPX, particularly when lagged or combined across multiple years, significantly elevated the odds of low catches (OR\u0026thinsp;\u0026gt;\u0026thinsp;1, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Notably, the strongest signal emerged when TempX and NPPX occurred in sequence, with lagged compounding effects nearly doubling the likelihood of low catch anomalies. This clarifies that productivity disruptions, not examined by Fredston et al. \u003csup\u003e9\u003c/sup\u003e, rather than heatwaves alone, may be associated with stock and fishery declines in some systems. Moreover, the geographic focus of Fredston et al \u003csup\u003e9\u003c/sup\u003e on the Northeast Pacific and North Atlantic regions, where TempX links to catch drops are comparatively weak (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), further explains why their findings may not generalize globally or across other functional groups. In contrast, our findings align with global modeling studies that have identified extreme high temperatures and compound climate events as key drivers of fish biomass declines \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e, highlighting that persistent or compounded temperature and NPP extremes are particularly destabilizing for fisheries. The congruence between modeled probabilities and real-world disruptions reinforces the robustness of our statistical approach and underscores the urgent need to integrate extreme event risk into fisheries management and climate adaptation strategies.\u003c/p\u003e"},{"header":"Considerations of key uncertainties","content":"\u003cp\u003eDefinitions of extremes and anomalies are based on historical baselines that may not reflect species-specific thresholds or ecological tipping points \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Our statistical models capture associations rather than mechanistic causality, and may not fully account for complex biological responses such as recruitment lags, range shifts, or altered food-web interactions, or linkages with the socio-economic and governance dimensions of fisheries \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Regional variability in catch reporting, differences in monitoring capacity and management frameworks or actions introduce further uncertainty. Moreover, projections assume that past relationships between environmental extremes and catch anomalies remain stable under future conditions, an assumption challenged by potential species adaptation and evolving fisheries and management responses \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. Importantly, co-occurring stressors such as overfishing, habitat degradation, or deoxygenation were not explicitly included, which may underestimate cumulative risks. These uncertainties underscore the importance of integrative, process-based models and improved monitoring to improve our knowledge to support adaptive fisheries governance.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eOur findings underscore the urgent need for climate-adaptive fisheries management that addresses not only long-term environmental trends but also the growing risks of acute disruptions from marine extremes. Several sources of uncertainty warrant consideration (see Method for details), but these uncertainties generally suggest that our estimates of climate-driven fishery disruption are conservative. Strategies such as adaptive harvest control rules, early warning systems, diversification of target species, and strengthened monitoring and forecasting capacity will be essential for enhancing resilience. These needs are especially critical for at-risk fisheries in socio-economically vulnerable countries, where resources, scientific infrastructure, and human capacity are often limited. Targeted investments and international support will be vital to ensure these regions can respond effectively to the compounding pressures of physical and biogeochemical extremes.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003eFisheries time-series data\u003c/h2\u003e\u003cp\u003eWe obtained catch data from the Sea Around Us reconstruction database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003ca href=\"http://www.seaaroundus.org\" target=\"_blank\"\u003ewww.seaaroundus.org\u003c/a\u003e\u003c/span\u003e\u003cspan address=\"http://www.seaaroundus.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), which estimates annual marine fisheries catches (in tonnes) by EEZ and species from 1991 to 2019. We follow the EEZ classification by the Sea Around Us, noting that it subdivides the EEZs of 198 coastal states into 280 regions (e.g., Canada\u0026rsquo;s EEZ is divided in Arctic, Atlantic and Pacific), including islands territories (updated 1 July 2015, available from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.seaaroundus.org\u003c/span\u003e\u003cspan address=\"https://www.seaaroundus.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). In total, we analyzed 6,569 time-series comprising 178,256 catch records. Unlike landing-based datasets that attribute catches to the country of landing, the Sea Around Us data reconstructs where the catch was taken, assigning it to the spatial origin of fishing activity. This distinction is critical for our study, which aims to assess the spatial overlap between environmental extremes (e.g., temperature or NPP anomalies) and the regions where fish were actually harvested (see Supplementary Fig.\u0026nbsp;2).\u003c/p\u003e\u003cp\u003eThe reconstructed database uses a wide variety of sources to estimate total removals, including industrial, artisanal, subsistence, recreational catch, and discards, many of which are absent from official national reports \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. It incorporates missing components of catch through the use of published and grey literature, fisheries reports, and expert knowledge, thereby producing a more comprehensive time series by species and country.\u003c/p\u003e\u003cp\u003eWhile catch reconstructions offer critical advantages for spatially explicit analyses, they are subject to caveats \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. The methods rely on assumptions and expert input in data-poor situations, which may introduce uncertainty or bias. The reconstruction approach and quality of underlying data also vary across taxa and regions, and current global fisheries statistics tend to exhibit a Global North bias, with better resolution and species identification in those regions than in the Global South.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eHistorical sea temperature and net primary production data\u003c/h2\u003e\u003cp\u003eWe used historical observation-based ocean temperature and model-based net primary production (NPP) datasets from the Copernicus Marine Environment Monitoring Service (CMEMS) to quantify environmental conditions associated with extreme fisheries catch anomalies.\u003c/p\u003e\u003cp\u003eSea water temperature was derived from the Global Ocean Ensemble Reanalysis product \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e, which provides monthly fields of global ocean temperature from 1993 to 2022 at 0.25\u0026deg; \u0026times; 0.25\u0026deg; spatial resolution across 50 vertical levels. To match the resolution of other datasets, we bilinearly regridded the temperature data to a 0.5\u0026deg; \u0026times; 0.5\u0026deg; global grid. Our main analysis used annual mean surface temperature (0\u0026ndash;100 m). Moreover, we tested the effects of using depth strata that were more specific to species\u0026rsquo; depth distribution: (i) surface layer (0\u0026ndash;100 m), (ii) epipelagic layer (0\u0026ndash;200 m), (iii) mesopelagic (200\u0026ndash;1000 m) and (iv) bottom layer (deepest grid cell in the data product). In addition, we tested the effects of using average temperature across the depth strata. Each species was linked to the most relevant thermal layer based on its maximum depth from FishBase and SeaLifeBase: pelagic species with a depth below 200 m were linked to epipelagic temperature, demersal species to bottom-layer temperature, and upper water column species (maximum depth above 100 m) to surface layer (0-100 m).\u003c/p\u003e\u003cp\u003eNet primary production (NPP) data were obtained from the biogeochemical hindcast for global ocean produced at Mercator-Ocean \u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e, which provides global estimates of biogeochemical variables from 1993 onward using the PISCES model coupled with ocean reanalysis data. NPP is provided monthly at 0.25\u0026deg; \u0026times; 0.25\u0026deg; spatial resolution. The robustness of the Mercator-Ocean hindcast dataset is supported by quantitative validation metrics detailed in the CMEMS Quality Information Document \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e, which reports seasonal and global agreement for primary productivity and related biogeochemical variables via bias, RMSD, and Estimated Accuracy Numbers (EANs) against satellite and climatology reference datasets. We selected this dataset because of its consistent temporal coverage, physical-biogeochemical coupling, and relatively high spatial resolution, which together offer robust estimates of NPP anomalies across EEZs and ocean basins. We applied the same bilinear interpolation and regridding approach to convert this dataset to a 0.5\u0026deg; \u0026times; 0.5\u0026deg; grid, and computed depth integrated annual total NPP by summing daily values for each year and grid cell.\u003c/p\u003e\u003cp\u003eThe processed annual temperature and NPP fields were used to characterize environmental exposure for species across EEZs, identify climate extremes, and assess their association with extreme low and high catch anomalies in global fisheries.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eProjected future sea water temperature and net primary production under climate change\u003c/h2\u003e\u003cp\u003eWe used future projections of sea surface temperature (SST) and net primary production (NPP) under climate change to assess future trends in the probability of extreme low catch events. Output from three Earth System Models (ESMs) were used for the analysis: IPSL-CM6A-LR, GFDL-ESM4, and MPI-ESM1-2-HR. Model outputs were taken from emissions scenarios SSP5-8.5 and SSP1-2.6 with simulations performed using the Coupled Model Intercomparison Project 6 (CMIP6) framework.\u003c/p\u003e\u003cp\u003eThe raw ESM data were provided on a 1\u0026deg; \u0026times; 1\u0026deg; regular global grid at a monthly temporal resolution. To harmonize spatial resolution across datasets, we regridded the monthly outputs to a 0.5\u0026deg; \u0026times; 0.5\u0026deg; regular grid using bilinear interpolation. For near-coastal grid cells where bilinear interpolation introduced missing values due to partial ocean coverage, we applied a distance-weighted extrapolation method. This method employed the haversine formula to calculate distances between grid cell centroids, and applied inverse-distance weighting to estimate values from surrounding valid ocean cells. Following spatial interpolation, the monthly data was aggregated into annual means for SST and sums for NPP (SST and NPP) for each grid cell.\u003c/p\u003e\u003cp\u003eKyoto\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eAnalyzing the effects of extreme temperature and net primary production on catches\u003c/h2\u003e\u003cp\u003ePrior to identifying extreme events, all environmental time series were statistically detrended to account for long-term climate trends that may confound interannual variability. We employed generalized additive models (GAMs) with cubic splines to fit temporal trends. The significance of the temporal trend was evaluated at α\u0026thinsp;=\u0026thinsp;0.05. If significant, the residuals from the GAM were retained as detrended values; otherwise, mean-centered anomalies were used. This approach allowed us to isolate short-term deviations from long-term trends, thereby improving the robustness of our extreme event definitions. The detrended values were then used to compute anomalies for each variable.\u003c/p\u003e\u003cp\u003eTo quantify the effects of extreme temperature and net primary production conditions on fisheries catches, we applied generalized linear modeling (GLM) across individual species\u0026ndash;EEZ combinations. We defined extreme events as annual anomalies in ocean temperature (Temp) and net primary production (NPP) that elevated beyond the 90th (high) or fell below the 10th (low) percentiles of their respective historical anomalies distributions of the detrended time-series. We also identified extra-ordinary extremes using the maximum and minimum values in the time-series for Temp and NPP, respectively. Climate anomalies were matched to each species based on sea surface temperature exposure and the EEZ(s) in which they occurred. Similarly, reversed extremes are defined as the 10th or 90th percentiles of the historical anomalies distributions of the detrended ST and NPP, respectively.\u003c/p\u003e\u003cp\u003eFor each species\u0026ndash;EEZ time series, we modeled the probability of experiencing an extreme low catch year. This was defined as the bottom decile of the catch anomaly distribution, as a binary response variable using logistic regression. Predictor variables included indicators of extreme high temperature, low NPP, and their 1-year lagged values, along with their addition and interaction terms to capture compound event effects. To evaluate competing hypotheses about the importance of individual and compound stressors, we constructed and compared all combinations of predictor sets using GLMs and ranked them based on Akaike Information Criterion (AIC).\u003c/p\u003e\u003cp\u003eOdds ratios (ORs) were derived from the coefficients of generalized linear models (GLMs) with a binomial error distribution and logit link. The GLMs estimated the relationship between the occurrence of extreme low catch events (binary response: 1\u0026thinsp;=\u0026thinsp;event, 0\u0026thinsp;=\u0026thinsp;no event) and predictors representing temperature extremes (TempX), primary production extremes (NPPX), and their combinations. For each predictor, the model coefficient (β) from the logit scale was exponentiated (OR\u0026thinsp;=\u0026thinsp;e^β) to express the change in odds of an extreme low catch event associated with the presence of the extreme condition, relative to baseline (non-extreme) conditions. We applied this modelling framework to all EEZ-species time-series units globally, and to individual EEZ and functional group separately.\u003c/p\u003e\u003cp\u003eModel-averaged effect estimates were obtained using AIC weights, from which we derived weighted odds ratios (ORs) and associated 95% confidence intervals for each environmental extreme scenario. We considered effects statistically significant if the confidence interval did not include OR\u0026thinsp;=\u0026thinsp;1. We presented model results globally and across EEZs and functional groups to assess spatial and taxonomic patterns in climate sensitivity. To identify key combinations of extremes, we developed a scenario matrix of binary combinations of the four predictor variables and computed model-averaged log-odds differences from a \u0026ldquo;baseline\u0026rdquo; (no extreme) condition.\u003c/p\u003e\u003cp\u003eAs a sensitivity analysis, we additionally fitted generalized estimating equations (GEEs) with an AR(1) correlation structure to account for temporal autocorrelation within species\u0026ndash;EEZ panels using the geepack package in R (v. 4.4.0). This analysis examines the effects of accounting for temporal autocorrelation within species\u0026ndash;EEZ panels and repeated measures through time on the estimated ORs of CatchX in relation to TempX and NPPX. The GEE-based results were consistent with those of the GLMs, showing no significant differences in the estimated odds ratios or their relative rankings across predictor combinations, confirming the robustness of our findings to temporal dependence in the data (see Supplementary Fig.\u0026nbsp;7).\u003c/p\u003e\u003cp\u003eTo assess future trends in the probability of extreme low catch events, we applied the fitted GLMs to Earth system model (ESM) projections of SST and NPP under two Shared Socioeconomic Pathways (SSP1-2.6 and SSP5-8.5). We calculated annual scenario-specific probabilities of low catch events, aggregated these at the EEZ and functional group levels, and derived time-series estimates of the proportion of species with extreme low catch events under each climate scenario. We then calculated the global average proportion amongst the EEZs weighted by the number of species assessed therein. All analyses were implemented in R (version 4.3.2).\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments: \u003c/strong\u003eThis research was enabled in part by support provided by the Digital Research Alliance of Canada (alliancecan.ca). We are thankful for the Sea Around Us and the Corpunicus data service that make fisheries and oceanographic data available in the public domain. Such data are critical for our research. We acknowledge funding support from: Natural Sciences and Engineering Research Council of Canada Discovery Grant (WLC); Social Sciences and Humanity Research Council of Canada Partnership Grant (WLC).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions: \u003c/strong\u003eConceptualization: WLC; Methodology: WLC, TLF, JPA, IM; Investigation: WLC, IM; Visualization: WLC, TLF, JPA, IM\u003cstrong\u003e; \u003c/strong\u003eFunding acquisition and project administration: WLC; WLC wrote the initial draft and all authors contribute to the writing, review and editing of the original draft.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests: \u003c/strong\u003eAuthors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMaterials \u0026amp; Correspondence: \u003c/strong\u003eWilliam W.L. Cheung, [email protected]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability: \u003c/strong\u003eAll datasets and code underlying the results are made available via github repository [https://github.com/coruubc/CompoundExtremes.git]. There will be no restrictions on data availability or access when the manuscript is published.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eFr\u0026ouml;licher TL, Fischer EM (2018) Gruber, N. Marine heatwaves under global warming. Nature 560:360\u0026ndash;364\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOliver ECJ et al (2018) Longer and more frequent marine heatwaves over the past century. Nat Commun 9:1324\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIPCC. Climate Change (2023) : Synthesis Report. A Report of the Intergovernmental Panel on Climate Change. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. \u003cem\u003eIPCC Geneva Switz.\u003c/em\u003e 24 (2023)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSmale DA et al (2019) Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nat Clim Change 9:306\u0026ndash;312\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSmith KE et al (2023) Biological Impacts of Marine Heatwaves. Annu Rev Mar Sci 15:119\u0026ndash;145\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheung WW et al (2021) Marine high temperature extremes amplify the impacts of climate change on fish and fisheries. Sci Adv 7:eabh0895\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSmith KE et al (2021) Socioeconomic impacts of marine heatwaves: Global issues and opportunities. Science 374:eabj3593\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVillase\u0026ntilde;or-Derbez JC, Arafeh-Dalmau N, Micheli F (2024) Past and future impacts of marine heatwaves on small-scale fisheries in Baja California, Mexico. Commun Earth Environ 5:623\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFredston AL et al (2023) Marine heatwaves are not a dominant driver of change in demersal fishes. Nature 1\u0026ndash;6\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGruber N, Boyd PW, Fr\u0026ouml;licher TL, Vogt M (2021) Biogeochemical extremes and compound events in the ocean. Nature 600:395\u0026ndash;407\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLe Grix N, Zscheischler J, Laufk\u0026ouml;tter C, Rousseaux CS, Fr\u0026ouml;licher TL (2021) Compound high-temperature and low-chlorophyll extremes in the ocean over the satellite period. Biogeosciences 18:2119\u0026ndash;2137\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLe Grix N, Zscheischler J, Rodgers KB, Yamaguchi R, Fr\u0026ouml;licher TL (2022) Hotspots and drivers of compound marine heatwaves and low net primary production extremes. Biogeosciences 19:5807\u0026ndash;5835\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLe Grix N, Cheung WWL, Reygondeau G, Zscheischler J, Fr\u0026ouml;licher TL (2023) Extreme and compound ocean events are key drivers of projected low pelagic fish biomass. Glob Change Biol. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.16968\u003c/span\u003e\u003cspan address=\"10.1111/gcb.16968\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWyatt AM, Resplandy L, Marchetti A (2022) Ecosystem impacts of marine heat waves in the northeast Pacific. Biogeosciences 19:5689\u0026ndash;5705\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCapotondi A, Alexander MA, Bond NA, Curchitser EN, Scott JD (2012) Enhanced upper ocean stratification with climate change in the CMIP3 models. 117:1\u0026ndash;23\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheung WWL, Close C, Lam V, Watson R, Pauly D (2008) Application of macroecological theory to predict effects of climate change on global fisheries potential. Mar Ecol Prog Ser 365:187\u0026ndash;197\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStock CA et al (2017) Reconciling fisheries catch and ocean productivity. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 114, E1441\u0026mdash;-E1449\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHeneghan RF et al (2021) Disentangling diverse responses to climate change among global marine ecosystem models. Prog Oceanogr 198:102659\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSzuwalski CS, Aydin K, Fedewa EJ, Garber-Yonts B, Litzow MA (2023) The collapse of eastern Bering Sea snow crab. Science 382:306\u0026ndash;310\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMills KE et al (2013) Fisheries management in a changing climate: lessons from the 2012 ocean heat wave in the Northwest Atlantic. Oceanography 26:191\u0026ndash;195\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFree CM et al (2019) Impacts of historical warming on marine fisheries production. Science 363:979\u0026ndash;983\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eP\u0026ouml;rtner HO, Peck MA (2010) Climate change effects on fishes and fisheries: towards a cause-and-effect understanding. J Fish Biol 77:1745\u0026ndash;1779\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlter K et al (2024) Hidden impacts of ocean warming and acidification on biological responses of marine animals revealed through meta-analysis. Nat Commun 15:2885\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlfonso S, Gesto M, Sadoul B (2021) Temperature increase and its effects on fish stress physiology in the context of global warming. J Fish Biol 98:1496\u0026ndash;1508\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePershing AJ et al (2015) Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery. Science 350:809\u0026ndash;812\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ede Guibourd V, Gascuel D, Reygondeau G, Cheung WW (2024) L. Large potential impacts of marine heatwaves on ecosystem functioning. Glob Change Biol 30:e17437\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSuryan RM et al (2021) Ecosystem response persists after a prolonged marine heatwave. Sci Rep 11:6235\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003edu Pontavice H, Gascuel D, Reygondeau G, Stock C, Cheung WW (2021) Climate-induced decrease in biomass flow in marine food webs may severely affect predators and ecosystem production. Glob Change Biol 27:2608\u0026ndash;2622\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFree CM et al (2023) Impact of the 2014\u0026ndash;2016 marine heatwave on US and Canada West Coast fisheries: Surprises and lessons from key case studies. Fish Fish 24:652\u0026ndash;674\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBindoff NL et al (2019) Changing Ocean, Marine Ecosystems, and Dependent Communities\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJones MC, Cheung WWL (2015) Multi-model ensemble projections of climate change effects on global marine biodiversity. ICES J Mar Sci 72:741\u0026ndash;752\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWatson RA et al (2013) Global marine yield halved as fishing intensity redoubles. Fish Fish 14:493\u0026ndash;503\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuti\u0026eacute;rrez D, Akester M, Naranjo L (2016) Productivity and sustainable management of the Humboldt Current large marine ecosystem under climate change. Environ Dev 17:126\u0026ndash;144\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBell JD et al (2013) Mixed responses of tropical Pacific fisheries and aquaculture to climate change. Nat Clim Change 3:591\u0026ndash;599\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBessell-Browne P, Epstein HE, Hall N, Buerger P, Berry K (2021) Severe heat stress resulted in high coral mortality on Maldivian Reefs following the 2015\u0026ndash;2016 El Ni\u0026ntilde;o Event. in \u003cem\u003eOceans\u003c/em\u003e vol. 2 233\u0026ndash;245MDPI\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrown CJ, Mellin C, Edgar GJ, Campbell MD, Stuart-Smith RD (2021) Direct and indirect effects of heatwaves on a coral reef fishery. Glob Change Biol 27:1214\u0026ndash;1225\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTownhill B, Artioli Y, Pinnegar J, Birchenough S (2022) Exposure of commercially exploited shellfish to changing pH levels: how to scale-up experimental evidence to regional impacts. ICES J Mar Sci 79:2362\u0026ndash;2372\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHarvey F (2024) Catastrophic marine heatwaves are killing sealife and causing mass disruption to UK fisheries. \u003cem\u003eThe Guardian\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.theguardian.com/environment/2024/nov/23/catastrophic-marine-heatwaves-are-killing-sealife-and-causing-mass-disruption-to-uk-fisheries\u003c/span\u003e\u003cspan address=\"https://www.theguardian.com/environment/2024/nov/23/catastrophic-marine-heatwaves-are-killing-sealife-and-causing-mass-disruption-to-uk-fisheries\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOsgood GJ, White ER, Baum JK (2021) Effects of climate-change-driven gradual and acute temperature changes on shark and ray species. J Anim Ecol 90:2547\u0026ndash;2559\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCline TJ, Schindler DE, Hilborn R (2017) Fisheries portfolio diversification and turnover buffer Alaskan fishing communities from abrupt resource and market changes. Nat Commun 8:14042\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmaya DJ et al (2023) Marine heatwaves need clear definitions so coastal communities can adapt. Nature 616:29\u0026ndash;32\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHertz T et al (2024) Eliciting the plurality of causal reasoning in social-ecological systems research. Ecol Soc 29\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eReusch TBH (2014) Climate change in the oceans: Evolutionary versus phenotypically plastic responses of marine animals and plants. Evol Appl 7:104\u0026ndash;122\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoore JW, Schindler DE (2022) Getting ahead of climate change for ecological adaptation and resilience. Science 376:1421\u0026ndash;1426\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZeller D et al (2016) Still catching attention: Sea Around Us reconstructed global catch data, their spatial expression and public accessibility. Mar Policy 70:145\u0026ndash;152\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePauly D, Zeller D (2016) Catch reconstructions reveal that global marine fisheries catches are higher than reported and declining. Nat Commun 7:10244\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMercator Ocean International (2023) Global Ocean Ensemble Physics Reanalysis (GLOBAL_MULTIYEAR_PHY_ENS_001_031). EU Copernicus Marine Service \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48670/moi-00024\u003c/span\u003e\u003cspan address=\"10.48670/moi-00024\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMercator Oc\u0026eacute;an International (2024) Global Ocean Biogeochemistry Hindcast (GLOBAL_MULTIYEAR_BGC_001_029). Copernicus Marine Service \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48670/moi-00019\u003c/span\u003e\u003cspan address=\"10.48670/moi-00019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerruche C, Szczypta C, Paul J, Dr\u0026eacute;villon M (2024) \u003cem\u003eQuality Information Document for GLOBAL_MULTIYEAR_BGC_001_029\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48670/moi-00019\u003c/span\u003e\u003cspan address=\"10.48670/moi-00019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e doi:10.48670/moi-00019\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8109072/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8109072/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eClimate extremes are increasingly disrupting marine ecosystems and fisheries. However, evidence on extreme ocean temperature effects on fish stocks is mixed, and the combined impacts of heat and productivity extremes remain unclear. Using three decades of global data encompassing 6,659 time-series for 1,246 species across 254 regions, we conducted a risk-based analysis to quantify how local extreme high temperatures and low ocean productivity that these species were exposed to, alone and together, affect local fisheries catches. These events, particularly when compounded, sharply increase the likelihood of local extreme low catch events, especially in tropical and subtropical regions. Species critical to food security and conservation are disproportionately affected, with widespread risks in socio-economically vulnerable countries. Without adaptation, extreme events\u0026rsquo; risks are projected to strongly intensify by the mid-21st century. Strengthening monitoring and climate-responsive fisheries management is urgently needed to build resilience.\u003c/p\u003e\u003cp\u003eMain Text\u003c/p\u003e","manuscriptTitle":"Large Fisheries Declines Linked to Compound and Extreme Climate Events","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-19 06:14:00","doi":"10.21203/rs.3.rs-8109072/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"2da55692-1ca2-4c17-910e-8eb5d3d26166","owner":[],"postedDate":"November 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":58141462,"name":"Earth and environmental sciences/Ecology/Climate-change ecology"},{"id":58141463,"name":"Earth and environmental sciences/Climate sciences/Ocean sciences/Marine biology"}],"tags":[],"updatedAt":"2026-03-15T18:25:07+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-19 06:14:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8109072","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8109072","identity":"rs-8109072","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}