betaFeedback

WorksComputer Science

Identifying companies and financial actors exposed to marine tipping points

Abstract

Climate change and other anthropogenic pressures are likely to induce tipping points in marine ecosystems, potentially leading to declines in primary productivity and fisheries. Despite increasing attention to nature-related financial risks and opportunities within the ocean economy, the extent to which these tipping points could affect investors has remained largely unexplored. Here we used satellite data to track fishing vessels operating in areas prone to marine regime shifts, as identified by their loss of resilience and vulnerability to marine heatwaves, and uncovered their corporate beneficial owners and shareholders. Despite some data gaps, we identified key countries, companies, and shareholders exposed to tipping risk. We also outline the potential challenges and opportunities that these actors may face if marine ecosystems shift to less productive states.

Introduction

Marine ecosystems are expected to undergo large biological reorganizations that can affect primary productivity and other ecosystem services[1]. These abrupt and nonlinear changes, known as regime shifts, pose major challenges for managers as they are difficult to predict and reverse[2]. For example, the collapse of Newfoundland cod in the 1990s is estimated to have caused large economic losses, over 40,000 job losses, and the crash of an iconic industry and cultural practice in the region[3]. Likewise, kelp forests have been observed to transition to urchin barrens, resulting in the loss of nursery habitats for commercially important fish species[4, 5]. Similarly, over 500 cases of oxygen-depleted zones have been reported around the globe, leading to harmful algae blooms and mass mortality events in fish which impact human health, local livelihoods and food provision[6, 7].

Exposure to regime shifts is mediated by pressures on marine ecosystems, such as climate change, overfishing, or pollution[1]. Marine heatwaves are expected to increase in frequency and intensity[8, 9], with stark economic consequences for companies, regions and nations that depend on ocean productivity[10]. For example, the Gulf of Alaska experienced a heat anomaly known as the Blob from 2014 to 2019, causing an estimated US$24M loss in a fishery valued at US$103M annually. Similarly, harmful algae blooms led to US$40M loss in tourist spending along the coast of Washington in 2015[10]. Other areas highly vulnerable to heatwaves include the tropical Pacific Ocean, the Caribbean Sea, northern Australia, the Eastern China Sea and the Western Pacific Ocean[10].

Forecasting the ecological and economic consequences of marine regime shifts is a difficult endeavor due to uncertainty. Increasing sea surface temperature can destabilize upwellings – currents that bring nutrient-rich water to the ocean’s surface – and affect some of the world’s most productive fisheries[11]. While insufficient nutrients can reduce primary productivity and cascade up through marine food webs, an excess can cause coastal eutrophication and hypoxia[6], with detrimental consequences to fish stocks. Climate change is expected to reduce fish biomass by 3 to 22% in tropical and temperate areas by the end of the century[12]. Overfishing and nutrient pollution from agriculture and urban waste can exacerbate this impact. A combination of oceanographic and climatic factors heightens the risk of regime shifts, putting both ecosystems and human livelihoods at risk[2, 13]. Yet, observing and measuring tipping points is a challenging task typically masked by low signal to noise ratios[14]. However, proactive management strategies that account for these risks could help mitigate exposure to tipping points[2].

Companies and financial institutions are key actors that contribute to the extraction of marine resources, while also being affected economically by changes in marine ecosystems. About 35.4% of fish stocks globally are currently depleted or overexploited[15], with only 13 companies controlling around 16% of wild seafood capture[16, 17]. High market concentration, a form of inequality, means that a few actors can have disproportional effects on the system for good or bad[18]. Fishing companies contribute to ecological degradation through harmful practices like overfishing[19], destructive fishing methods[20], and non-compliance with regulations[21, 22]. By removing biomass from the ocean, fishing has historically prevented the sequestration of 21.8M tons of carbon[15, 23]. In addition, the industry’s return on investment is becoming increasingly risky. Modern fishing fleets travel twice as far but catch only one-third of what they used to per kilometre travelled[24], it is estimated that 54% of high seas fishing would be unprofitable without public subsidies[15, 25]. Companies doing extractive activities other than fisheries are also projected to have impacts from climate change from extreme weather events and sea level rise[26, 27].

It should be in the best interest of companies and their shareholders to act as stewards of the seas and maintain a productive and healthy ocean[28, 29]. Financial actors – such as banks, asset managers, and pension funds – could play an important role in conservation by deciding not to finance economic activities that alter the ecological integrity of the biosphere[30, 31, 32, 33]. For instance, a small set of financial actors have an increasing influence over climate stability through their equity holdings in companies operating in the Amazon and boreal forests[30]. Likewise, large asset managers and other financial institutions invest in companies operating in areas vulnerable to the emergence of infectious diseases[31]. This leaves them exposed to the impacts of their investments[34].

Climate-change-driven extreme weather events are expected to cost up to USD 14 trillion annually by 2100[35], directly threatening the solvency of insurers and reinsurers[36]. Investors and asset managers face climate exposure being at risk of stranded assets, changes in commodity prices, scarcity of resources, disruption of supply chains, and infrastructure damages[37]. In response, financial actors are increasingly expected to disclose their risks through initiatives such as the Task Forces on Climate- and Nature-Related Financial Disclosures (TCFD and TNFD, respectively) or the EU non-financial reporting directive.

Analyses of financial exposure to biodiversity loss and ecosystem degradation are mostly focused on terrestrial ecosystems, with little research addressing ocean-related financial risks[38]. Yet, fishing companies, their shareholders, and the countries they operate in are exposed to potentially irreversible changes and consequent declines in ecosystem services, with consequences on the economy, investments and livelihoods. However, being exposed to regime shifts also means being in a position to take actions that reduce anthropogenic pressures that could drive the system across a potential tipping point. In this paper, we asked who are these corporate and financial actors exposed to marine tipping points?

Methods

Areas at risk: We identified areas of the world vulnerable to marine regime shifts. We approximated these areas as places currently under the strong influence of marine heatwaves or already showing symptoms of resilience loss. Frölicher et al.[8] defined heatwaves as events when sea surface temperature exceeds the 99th percentile with respect to daily satellite observations for a 30-year period. Hobday et al.[39] further classified heatwave categories based on the 90th percentile, where moderate heatwaves are between 1x and 2x the 90th percentile, strong heatwaves between 2x and 3x, severe heatwaves between 3x and 4x, and extreme heatwaves above 4 times the 90th percentile. We computed the probability of severe and extreme heat waves based on daily sea surface temperature observations from the US National Oceanic and Atmospheric Administration (NOAA) using the optimum interpolation data from 1981 to 2022 at 0.25° resolution (Figure 1A-B). While several impacts of heatwaves have been documented in ecosystems and the economy[10, 39], a recent global assessment showed no systematic effect on fish biomass[40]. However, heatwaves by themselves are seldom the sole cause of fish populations decline, it is often the compound effect of temperature with other pressures such as low primary productivity or low oxygen zones[13].

Panel A showing mean number of days in extreme heatwaves

(a)

Panel B showing mean number of days in severe heatwaves

(b)

Panel C showing number of early warning signals

(c)

Figure 1. Areas at risk of tipping points. We computed extreme and severe heatwave events based on historical records of sea surface temperature (A, B). Areas in colour are places where the probability is non-zero, expressed as the mean number of days spent on a heatwave event. Ocean colour data was used to compute early warning signals of critical transitions (C). Note that there are no warnings computed for higher latitudes due to missing data. For the analysis of areas prone to marine regime shifts, we kept areas where the probability of extreme and severe heatwaves is non-zero, or where at least three early warning signals were found (see Methods). Supplementary figure Supplementary figure S1 provides additional empirical evidence of the occurrence of regime shifts by mapping 1448 case studies coded through literature review.

Thus we complemented heat waves detection with proxies of stability loss in primary productivity, which are in turn driven by oxygen depletion, salinity, and the mixing of the upper layer of the ocean[6, 7, 41]. We used a geographically explicit mapping of marine ecosystems showing symptoms of resilience loss[42]. Resilience is the ability of a system to withstand disturbances without losing its functions and identity[43]. When systems lose resilience and are close to tipping points, they leave statistical signatures in time series known as early warning signals[44]. The map was produced by computing temporal autocorrelation, standard deviation, skewness, kurtosis and fractal dimension on Chlorophyll-A concentration, a proxy of marine primary productivity, using data from the European Space Agency (weekly observations at 0.25∘0.25^\circ resolution from 1998-2018)[45]. The data are limited by the coverage of the ocean colour data (the 50∘50^\circ N-50∘50^\circ S region), excluding the Arctic and Antarctic oceans[42]. For this paper we used areas with at least 3 early warnings, since two of them are suitable for bifurcation tipping (autocorrelation and standard deviation), two of them are suitable for flickering and stochastic transitions (skewness and kurtosis), and the fractal dimension is robust against heteroskedasticity. At least three early warnings give us the chance to pick up more robust signals of the risk of tipping taking into account different ways of reaching the transition.

Evidence of regime shifts: We documented past regime shifts by reviewing scientific literature using the framework of the regime shifts database[46]. In addition, we performed change point analysis, a classical method to detect regime shifts in time series[47, 48]. We looked for break points on the Chlorophyll-A concentration data from the biological pump and carbon exchange process project[49] which offers monthly data at 9km pixel resolution, and is derived from the European Space Agency’s Ocean Colour Climate Change initiative. We also looked at break points on historical fish stock assessment data from the RAM legacy stock assessment database version 4.4[50, 51], similar to previous efforts identifying regime shifts on fish stocks[48]. Out of the 479 stocks for which there are time series reported, we assessed only stocks with more than 25 years with data (N=373N = 373). If there is an abrupt decline in primary productivity, one can expect that the foodweb depending on it can be prone to similar changes. However, linear changes in primary productivity can also generate abrupt decline on fish stocks. For that reason we also calculated the linear trend on Chlorophyll-A time series to show areas of the world with significant changes by using the Kendall tau statistic. For break point detection we used the Pettitt test of change in means[52], and the change point F test statistic for change on residuals[53].

Fishing pressure: We identified fishing vessels operating in pixels showing a high probability of heatwaves or symptoms of resilience loss. We used the dataset published by Kroodsma et al.[54] who identified >70,000 industrial fishing vessels in the size range from 6 to 146m in length, thanks to an automatic identification system (AIS). Vessels need to broadcast their position on the AIS to avoid potential collisions, so the data offers individual vessel trajectories from 2012-2020 (fishing effort dataset from Global Fish Watch version 2.0). We extracted vessels whose fishing activities were recorded in areas with high probabilities of heatwaves (severe or extreme events) or high probability of regime shifts. We only include pixels with at least three different early warning signals of resilience loss, that way we include multiple routes of tipping (e.g. noise-induced, flickering, not only the traditional slowing down), and keep pixels for which multiple lines of evidence suggest loss of resilience.

Owners and shareholders: We identified the individual and company beneficial owners of the vessels using the maritime mobile service identity to match the Global Fishing Watch data with vessel owners in the Lloyds database (www.seasearcher.com). We matched 13,537 records, but not all of them had available ownership information. We also used the dataset curated by Carmine et al.[55] where vessels fishing on the high seas were manually annotated. We added companies whose vessels matched our selection of vessels fishing in areas with marine tipping point risk. To identify shareholders and global ultimate owners of the companies responsible for fishing activities, we used Orbis, a database of companies around the world with coverage of >489>489 million companies. Where possible, we extracted data about their shareholder structure (e.g. direct and total ownership, shareholder type), annual revenues, number of employees, and shareholder’s countries of operation. Unfortunately, Orbis and Lloyds are not open access data sources and their terms of service forbids sharing the raw accessed data with third parties. Both of them were accessed through research subscriptions by the home institution.

Network models: We constructed an international network of foreign investments using the countries responsible for the fishing activities (vessel flags) and the shareholder’s country of origin. This network representation allowed us to explore the likelihood of investments (probability of a link) since investing in a particular country is not independent of network structure (e.g. investments already present in the network, or individual features of the countries involved). We used governance data from the World Bank (control of corruption, government effectiveness, political stability, regulatory quality, rule of law, and voice accountability) to investigate if foreign investments in fishing activities exposed to marine tipping points tend to occur in countries with lower effectiveness and weaker rule of law, as previously reported in the context of emerging zoonotic diseases[31]. To test this hypothesis, we used an exponential random graph model[56] where nodes are countries and links are investment flows between shareholders and companies running fishing operations.

Results

We found that 30% of global oceans have presented extreme heat waves, and 66% severe events (Figure 1). Most of the extreme events are located in high latitudes in the Arctic and Antarctic oceans. Severe events are also common in the tropical Pacific ocean (related to ENSO), coastal California, and the Indian ocean. Upwelling systems in the coast of California, Benguela or Canary islands show at least three early warning signals of the proximity of instabilities. 24% of the area analysed with early warning signals show symptoms of resilience loss[42]. The tropical north Atlantic is also a hotspot of resilience loss, a place where previous work has documented increase in summer time stratification of the water column driven by sea surface temperature and salinity[41]. With stronger stratification there is less mixing of water column and nutrient cycling, leading to a decline in primary productivity. We detect such signal in the linear decline of primary productivity (Figure S2).

Linear changes in primary productivity

Figure S2. Linear changes in primary productivity. We estimated Kendall tau on the full time series of Chlorophyll-A at 9Km resolution. 77.3% of the ocean shows a slope significantly different from zero, 41.9% with a declining trend (A). In (B) the same time series have been detrended using seasonal-trend decomposition using LOESS, and a linear model fit on the residuals showing that there are still negative and significant trends in the central Pacific and Indian oceans that are not related to seasonality.

Our change point analysis shows that there has been abrupt changes in primary productivity over time (Figure 2). The extend of the area varies depending on the statistical test used. The Pettitt test looks for changes in means, and detects significant changes in 50% of the oceans with break points in early 2000s in the south Pacific, and more recent changes in the rest of the world. The structural change test is more restrictive, seeking changes on the residuals of a lagged autoregressive model. We find again abrupt changes in the central Pacific, the tropical Atlantic, the coasts of California, and India; adding up to 16% of the oceans. For the test that looks if there is a decline on Chlorophyll-A regardless if it is abrupt or smooth (the slope of a linear regression), we found that 77.3% of the ocean area (50N-50S region) has a significant kendall tau, and for 41.9% of the area primary productivity is declining (Figure S2).

Abrupt changes in primary productivity

Figure 2. Abrupt changes in primary productivity. We used the structural change and Pettitt test to investigate abrupt shifts in Chlorophyll A data and computed the proportion of change before and after the shift (the mean after over the mean before). If the proportion is one there was no change in means, if the proportion is <1< 1 it means an abrupt decline (e.g. 0.5 is a 50% decline), or an increase if >1>1 (A). 16% of the oceans show significant abrupt changes with the structural change test (B) while 50% of the oceans show abrupt changes in primary productivity with the Pettitt test (C). Supplementary Figure S2 shows a similar test for linear trends in primary productivity.

Similarly, for the stock assessment data we found that 353 fish stocks show significant abrupt changes of which 271 are declining biomass while 82 are increasing (Figure 3). We find agreement between the two methods used for change point detection (256 stocks), but some differences in the time predicted. The mean decline between the mean before and the mean after the break point is of 50% of the biomass. Qualitatively, our results align with previous studies using different methods, smaller sample sizes but stricter

Abrupt changes in fish biomass

Figure 3. Abrupt changes in fish biomass. We tested abrupt changes in fish biomass data from the RAM legacy database using the Pettitt and structural change tests. We found 252 stocks having abrupt changes detected by both tests, 14 only by the structural change F statistic, 83 only by the Pettitt test, and 20 with no detection (A). The mean decline for all significant changes was 50% of the biomass with respect to the mean before the abrupt change (B). The bottom panel lists the top territories most affected (C), the geographic areas most impacted (D), and the species with most abrupt changes. Supplementary Figure S3 complements the analysis by comparing the timing of the shifts between tests and a linear test.

thresholds for collapse[48, 57, 58, 59], where the proportion of stocks collapsed are between 25% and 50%. Here we find that 119 stocks collapsed below 50% their historical average (31%), while 29 stocks (7%) collapsed below 15% their historical record (Figure 3B). Regardless if the changes are abrupt, the Sen slope test supports the direction of change identified through the change point analysis (Figure S2). The countries with more fish stocks showing abrupt changes are USA, Canada and Australia, although many occur in international waters; the fishing grounds where most abrupt changes reported are the Pacific coast, the Gulf of Alaska, central and southern Chile and the Gulf of Mexico. The species most impacted are herring, Atlantic cod, walleye pollock, red rock lobster and pacific herring (Figure 3).

Out of the 114,191 vessels reported in the Global Fish Watch database, we found that 15 % of all vessels tracked (N=16,878N = 16{,}878) fish in areas prone to marine tipping points (Figure 1). The bulk of these vessels do not report a flag or country of registry; however, for those who do, the largest fleets are associated to China, the United States, Japan and Norway (Figure 4A, see Figure S4 for vessels types ). China’s fleet is at least three times larger than the next larger fleet, making it a dominant actor. Out of the 16,878 vessels, we identified 2,981 companies, for 855 of them we identified 1,825 unique shareholders and 2,783 ownership relationships between shareholders and companies. Most of the shareholders are based in Norway, China, Russia, Spain, the United Kingdom, and Denmark (Figure 4B); the most common shareholder types are corporate entities and individual families.

Descriptive statistics of vessels, shareholders, ownership shares, and investments

Figure 4. Descriptive statistics Number of vessels per flag reported (A), shareholders’ country of origin (B), shareholder type (C), and the probability density of their ownership shares (D). Most fishing companies analyzed have only a few shareholders (E). Countries ranked by their number of foreign investments (out-degree) and the number of investments received (in-degree), the size of the dot is proportional to the number of domestic investments (F). Countries in the yellow line have the same number of incoming and outgoing links in the network, above the line are countries with disproportionally more investments, and below countries which receive the investments.

Vessel types

Figure S4. Vessel types

The network of ownership and shareholder investments is relatively sparse, lacking the presence of major financial giants common to multiple companies. Unlike previous studies that identified “keystone actors” – central corporations and financial entities with disproportionate influence in the market[16] and potential exposure to tipping elements of the Earth system[30]– our findings reveal a more diffuse network with no central actors. This is similar to the pattern found by Jouffray et al. when looking at shareholders of publicly listed seafood companies[33]. One possible explanation is the ownership structure where most fishing companies in our dataset are owned by only a few shareholders who in turn own on average half, if not all, of the company’s shares. (Figure 4D, E). This is reflected by a large proportion of our shareholders being individuals, families, or small-scale businesses (Figure 4C). Another possibility is that keystone actors would emerge in the network only when considering not just vessel owners, but also retailers and traders involved in the supply chain.

Despite the sparsity of the ownership network, there were actors with disproportional weights on the network (Figure 5). The top company in our data set controlled over 60 vessels, and most top companies operated in China, South Korea or Russia. The top shareholders were also predominantly based in China with several of them being corporate companies (e.g. all companies that are not banks, financial or insurance companies). However, in terms of the number of companies invested in, public authorities and government offices emerged as the most important actors. This emphasizes the key role that public authorities play in not only regulating fishing operations, but also in investing and subsidizing activities that can be detrimental to the long-term interest of the citizens they represent. Notably, our findings show that this influence is not restricted to national governments, and that regional and city-level authorities could also be key actors in fisheries management.

Actors exposed to marine tipping

Figure 5. Actors exposed to marine tipping Top companies (A) and shareholders (B) exposed to areas with high risk of regime shifts. Colours correspond to country and contours to shareholder types (B). The rank for global ultimate owners is reported in Figure S5

Global Ultimate Owners

Figure S5. Global Ultimate Owners

The network of international investments revealed a different set of relevant actors (Figure 6). Most shareholders were based in Norway but their investments were largely domestic and focused on Norwegian fishing companies. International investments were dominated by Spain, the UK and France as countries where investments originated, while Russia seemed to disproportionately receive investments (Figure 6, Figure 4F). Using this network of international investment we asked what features increased the likelihood of an investment.

Network of shareholder investments

Figure 6. Network of shareholder investments While most investments are domestic, a network of foreign investments emerge featuring European countries, China and Russia as central actors (A). Results from exponential random graph models on the World Bank governance indicators (B) reveals that investments tend to occur when differences in governance efficiency and political stability exist. Two letter codes were used to identify countries (ISO 3166-1 alpha 2 standard), for territories without a code we used numbers as follows: Curaçao: 1, Cook Islands: 2, Falkland Islands: 3, French Southern Territories: 4, Mayotte: 5, Reunion: 6, St. Pierre and Miquelon: 7, St. Helena: 8, and Azores: 9.

To answer this question we fit three exponential random graph models of the network as response variable and governance indicators as explanatory variables (Figure 6B). The null model simply assumes that investment relationships form randomly. The difference model only takes into account the difference in governance statistics between the country where the investor is based minus the statistic where the fishing company is located. The full model uses both the differences and the raw statistics per country to predict the likelihood of investments. The full model outperformed the other two by maximizing the log likelihood and minimizing the Akaike Information Criterion (see SM table S1).

We found that government effectiveness was the best predictor of foreign investments (Figure 6). Countries with effective governments tended to invest in governments that have less effectiveness (hence the positive difference in the full model). We also found a smaller, yet significant and negative effect on political stability. Countries with low political stability in our data set invested in companies located in countries with slightly higher political stability (hence the negative difference on the full model). The implications of this result is that the relevant decisions to manage tipping point risk are not only in the hands of fishing vessels, the companies that own them or their shareholders. The countries where these activities take place (fishing and investing) can design regulations that restrain or disincentivise investments in other regions of the world where fishing activities are happening in high risk areas.

Despite data incompleteness, our results show that if marine tipping points were to synchronize globally, they could impact an industry generating over 6.6 billion dollars in revenue and affect more than 100,000 jobs (Figure 7). The vessel flags producing the highest revenues in our data were Ecuador, Panama, Russia, Spain and the UK, while those employing the largest workforces were led by Russia, Japan, Ecuador, Panama and Argentina (Figure 7). Vessel flags are not necessarily where profits go, but data gaps prevented a more accurate approximation. We were able to retrieve revenue and employment data for only 2,216 companies, with roughly 70% of the data missing values. As a result, our figures likely underestimate the true potential economic and social impacts of possible declines in primary productivity from marine heatwaves and marine regime shifts. Much higher levels of transparency in company reporting are needed to improve these initial estimates.

Potential impacts of marine tipping points

Figure 7. Potential impacts of marine tipping points. Countries ranked by the revenues reported in our dataset (A). Only countries with revenues higher than 1M US$ shown. Countries ranked by workforce employeed (B), only countries with more than 500 employees shown. The data however has high percentage of missing values (70%), so it should be interpreted as an under estimation.

Discussion

Recent studies showed that heatwaves alone are unlikely to decrease fish biomass[40] despite some events have been already observed[8, 10, 39]. Yet, it is often the effect of compound events or the interaction of multiple drivers that can drive abrupt fish declines or other marine regime shifts[2, 13]. Their frequency and intensity are predicted to increase in the near future with climate change scenarios[1, 8, 9, 13]. In this paper, we examined which corporate and financial actors are exposed to marine tipping points by identifying vessels and companies operating in areas that are simultaneously exposed to marine heat waves and instabilities in primary productivity. We focused on heat waves and changes in primary productivity because of data availability, acknowledging that marine ecosystems can tip due to other causes as well[2].

Here we reported lists of top countries, companies, and their shareholders associated with vessels fishing in areas at high risk of marine regime shifts. In summary, we found that fishing fleets associated with China, the US, Japan and Norway are particularly exposed. The shareholders of companies exposed to tipping point risks are based in Norway, China, Spain, Russia, the UK and Denmark. Spain, UK and France are home to particularly vulnerable shareholders investing in companies who fish in high-risk areas, while Russia is headquarters for fishing companies getting such investments. The implications are that the management is not only in the hands and decisions of companies. Countries could also design laws and regulations that restrain or disincentivise investments in other regions of the world where fishing activities are occurring in risk areas. We developed a method to track vessels and companies at risk which depends on a mix of publicly available[55] and non-public datasets.

Our method and findings are limited by data completeness and transparency. For example, corporate ownership often involves shell companies, offshore entities and indirect holdings that may not be fully documented in Orbis, or risk misidentifying or omitting ultimate owners. Another example is vessel flags, many vessels operate with flags of convenience obscuring the true ownership structure in the data. The absence of financial data at vessel level further weakens the link between company ownership and fishing activities.

While transparency and accountability are values sought after by the financial sector, our analysis highlights the complexity and opacity of global fisheries, even with access to state-of-the-art data[54, 55] and databases (e.g. Lloyds, Orbis). We first identified close to 17,000 vessels fishing in high-risk areas, but as we traced ownership up the chain –from vessel owner to company beneficiaries and shareholder– data availability dropped by an order of magnitude at each step. A high proportion of vessels have unidentified flags (20%) and lack registered owners (81.41%). Thus our results are biased towards the vessels and countries that report ownership data, that is primarily publicly listed companies operating in legal fisheries. Further research is needed to correct this bias[60]. Nonetheless, our study serves as a benchmark for what is achievable with current databases and is the first study to connect ocean tipping points to financial risks.

Another limitation stems from known problems with the AIS data on which our vessel identification is based. The data is known to be less reliable before 2017 and not all countries follow the same standards in reporting. For example, while AIS is primarily used by vessels over 300 gross tons operating beyond national jurisdiction, a country like China mandates AIS for a broad range of vessels also within the Exclusive Economic Zone (EEZ). This has led previous studies to restrict their analyses to the high seas[55].

Here, we nevertheless included EEZs in our analysis because they are often prone to marine heatwaves and regime shifts (Figure 1). In fact, over 500 observed hypoxic and anoxic zones –a regime shift that impacts primary productivity and hence fishing– have been observed in national jurisdictions[6, 7].

Despite these limitations, our study offers key insights for managing the financial risks associated with marine tipping points. In contrast to previous work that identified keystone actors dominating the exploitation of natural resources[17, 30, 31], we found that the fisheries sector does not seem to be highly concentrated, at least not at the level of decision-making that is exposed to risk of regime shifts – where to go fishing and when. Market concentration probably occurs further down the supply chain, at the level of intermediaries[33], but this level of aggregation makes it harder for financial institutions and regulators to discriminate whether the fish traded comes from areas prone to tipping points.

Financial actors and corporations are increasingly required to report their climate and nature-related risks, as seen in initiatives like the Task Force on Climate-Related Disclosures (TCFD), or the European Corporate Reporting Directive (CSRD). Central to these reporting efforts is the concept of double materiality: companies need to assess what is their vulnerability to ecological or climate change; but also how their activities impact climate change or ecological degradation[61]. This feedback loop heightens the risks faced by both the company and its investors. Our maps of tipping point risk combined with readily available georeferenced information on fishing fleet location enable companies and financial investors to assess their exposure, adjust their fishing areas (companies), and manage their investment portfolio (investors) to reduce exposure. Heatwaves and cold spells forecasting tools[9], as well as near-real-time monitoring of ecosystem resilience[42] can support adaptive strategies that minimize risk exposure.

When fisheries collapse (e.g. cod in the 1990s) or when stocks decline abruptly (e.g. the Blob in 2014-19), governments often need to step up and pay the economic and social costs. At the same time, governments are subsidizing high seas fisheries, without such subsidies 54% of their catch would have been unprofitable otherwise[15]. Our study shows that governments also have a role to play as shareholders in some of the biggest fishing fleets. This is particularly evident in China where national, regional and municipal government organizations sit on the boards of fishing companies exposed to tipping point risks. These governmental and public organizations have decision power to mitigate risk by directing fisheries to less risky grounds or by holding the fleet accountable for avoiding high-risk areas.

As heatwaves and marine regime shifts unfold, there is an additional risk that governance frameworks may become outdated. While benthic species tend to have low mobility and be impacted by these events, pelagic species can more readily relocate if the area affected is within their mobility range. The problem arises when a stock quota that was assigned in an economic zone to a particular set of actors moves to another economic jurisdiction, generating conflicts between fisheries and their managing institutions[62, 63]. Many fish stocks are already expected to move polewards due to climate change[64], and marine regime shifts may exacerbate these tensions. Governments and fishing authorities should put mechanisms in place to prevent, mediate or resolve these emerging conflicts, as non-linear environmental changes can not only alter the “rules of the game” but also shift the playing field.

Conclusion

In this paper, we asked who are these corporate and financial actors exposed to marine tipping points? We first identified areas of the ocean where tipping points can manifest because they are already showing a high intensity of extreme and severe heat wave events, or they are already showing statistical signatures of resilience loss in primary productivity. We found that about 30% of the oceans are at risk of regime shifts (extreme marine heat waves and signals of instability in primary productivity), and provided evidence of past regime shifts from literature reviews, and abrupt change observed in primary productivity and stock assessments data. 16% of the oceans have shown abrupt declines in primary productivity, 42% if accounting only the linear negative trend; while 252 fish stocks resulted in abrupt declines on historical data. The literature already showcase past events with impacts on employment and economic losses on the millions of dollars. So who is at risk today?

We found that 15% (16,878) of the vessels tracked in Global Fishing Watch do actually fish in areas susceptible to marine regime shifts. We did not find the levels of market concentration found in other economic sectors such as aquaculture, in fact the ownership structure is quite sparse with most companies owned by one or two shareholders. Nonetheless, we find that Norway and China are important actors in hosting many of the vessels and shareholders in our data, but they are not strongly involved on foreign investment. Spain, the UK and France are central countries of origin for foreign investment, while Russia is a hotspot attracting these investments. The likelihood of these financial flows seems to be related to differences in government effectiveness and political stability.

We provided lists of top shareholders and top companies with vessel ownership fishing in areas prone to marine regime shifts. We discussed the limitations of our study mainly related to data transparency. Nonetheless, we hope these lists of companies and shareholders help to identify actors with agency to manage the risk of marine regime shifts. We discussed potential interventions that can reduce regime shifts risks. Redesigning subsidy programs, lobbying company boards to fish in low-risk areas, or supporting the implementation of international agreements that take into account potential range shifts and conflicts are just a few examples of how these actors can incentivise practices that are aligned with long term sustainability values.

Acknowledgements

This work has been supported by the project Networks of Financial Rupture funded by FORMAS grant 2020-00198; and the Finance to Revive Biodiversity (FinBio) program, financed by Mistra – the Swedish Foundation for Strategic Environmental Research (DIA 2020/10), and the Swedish Research Council (VR grant 2022-04122). Our work benefited from discussions with Robert Blasiak.

References

Supplementary Material

World map of documented marine regime shifts and bar chart of regime shift types

Figure S1. Empirical evidence of regime shifts. Over 1440 marine regime shifts are documented on the regime shifts database, 788 cases are of hypoxia, followed by marine eutrophication (245) and fisheries collapse (181). For each dot in the map there is a paper in the scientific literature reporting regime shifts in marine ecosystems. Source: www.regimeshifts.org (beta version at www.regimeshifts.netlify.app).

modellogLikAIC
1 null-537.221076.43
2 difference-527.781073.56
3 full-516.751065.5

Table S1. Model statistics

Linear changes in fish stocks

Figure S3. Linear changes in fish stocks. We first compared the timing of the break points detected with the Pettitt test and the structural change test (A). Despite both test agree on the existence of the shift, there are some differences in timing. The median decline is 50% (red dotted line in B-C), and most shifts occurred between the 1970s and early 2000s. A comparison with the Sen slope test show that linear trends agree with the direction of change reported on the abrupt changes, the units of the slope are in biomass.

References

  1. [1]Beaugrand, G. et al. Prediction of unprecedented biological shifts in the global ocean. Nature Climate Change 9, 237–243 (2019).DOI
  2. [2]Rocha, J., Yletyinen, J., Biggs, R., Blenckner, T. & Peterson, G. Marine regime shifts: Drivers and impacts on ecosystems services. Philosophical Transactions of the Royal Society B 20130273 (2015). doi:10.1098/rstb.2013.0273DOI
  3. [3]Richard L. Haedrich, Lawrence C. Ha. The Fall and Future of Newfoundland’s Cod Fishery. Society & Natural Resources 13, 359–372 (2000).DOI
  4. [4]Ling, S. D. et al. Global regime shift dynamics of catastrophic sea urchin overgrazing. Philosophical Transactions of the Royal Society B: Biological Sciences 370, 20130269 (2015).DOI
  5. [5]Ling, S. D., Johnson, C. R., Frusher, S. D. & Ridgway, K. R. Overfishing reduces resilience of kelp beds to climate-driven catastrophic phase shift. Proceedings of the National Academy of Sciences 106, 22341–22345 (2009).DOI
  6. [6]Breitburg, D. et al. Declining oxygen in the global ocean and coastal waters. Science 359, eaam7240 (2018).DOI
  7. [7]Diaz, R. J. & Rosenberg, R. Spreading dead zones and consequences for marine ecosystems. Science 321, 926–929 (2008).DOI
  8. [8]Frölicher, T. L. & Laufkötter, C. Emerging risks from marine heat waves. Nature Communications 9, 650 (2018).DOI
  9. [9]Frölicher, T. L., Fischer, E. M. & Gruber, N. Marine heatwaves under global warming. Nature 560, 360–364 (2018).
  10. [10]Smith, K. E. et al. Biological impacts of marine heatwaves. Annual Review of Marine Science 15, 119–145 (2023).DOI
  11. [11]Bakun, A., FIELD, D. B., Redondo-Rodriguez, A. & WEEKS, S. J. Greenhouse gas, upwelling-favorable winds, and the future of coastal ocean upwelling ecosystems. Global Change Biology 16, 1213–1228 (2010).DOI
  12. [12]Ariza, A. et al. Global decline of pelagic fauna in a warmer ocean. Nature Climate Change 12, 928–934 (2022).DOI
  13. [13]Le Grix, N., Cheung, W. L., Reygondeau, G., Zscheischler, J. & Frölicher, T. L. Extreme and compound ocean events are key drivers of projected low pelagic fish biomass. Global Change Biology 29, 6478–6492 (2023).
  14. [14]Hillebrand, H. et al. Thresholds for ecological responses to global change do not emerge from empirical data. Nature Ecology & Evolution 4, 1502–1509 (2020).DOI
  15. [15]Andersen, N. F. et al. Good fisheries management is good carbon management. npj Ocean Sustainability 3, 17 (2024).DOI
  16. [16]Osterblom, H. et al. Transnational corporations as ‘keystone actors’ in marine ecosystems. PLoS ONE 10, e0127533 (2015).DOI
  17. [17]Folke, C. et al. Transnational corporations and the challenge of biosphere stewardship. Nature Ecology & Evolution 3, 1396–1403 (2019).DOI
  18. [18]Hamann, M. et al. Inequality and the Biosphere. Annual Review of Environment and Resources 43, 61–83 (2018).DOI
  19. [19]Daskalov, G. M., Grishin, A. N., Rodionov, S. & Mihneva, V. Trophic cascades triggered by overfishing reveal possible mechanisms of ecosystem regime shifts. 104, 10518–10523 (2007).
  20. [20]Amoroso, R. O. et al. Bottom trawl fishing footprints on the world’s continental shelves. 17, 201802379–E10282 (2018).DOI
  21. [21]Park, J. et al. Illuminating dark fishing fleets in north korea. Science Advances 6, eabbl197 (2020).DOI
  22. [22]Park, J. et al. Tracking elusive and shifting identities of the global fishing fleet. Science Advances 9, (2023).DOI
  23. [23]Mariani, G. et al. Let more big fish sink: Fisheries prevent blue carbon sequestration—half in unprofitable areas. Science Advances 6, (2020).DOI
  24. [24]Tickler, D., Meeuwig, J. J., Palomares, M.-L., Pauly, D. & Zeller, D. Far from home: Distance patterns of global fishing fleets. Science Advances 4, eaar3279 (2018).DOI
  25. [25]Sumaila, U. R. et al. WTO must ban harmful fisheries subsidies. Science 374, 544–544 (2021).
  26. [26]Cruz, A. M. & Krausmann, E. Vulnerability of the oil and gas sector to climate change and extreme weather events. Climatic Change 121, 41–53 (2013).DOI
  27. [27]Brown, S., Hanson, S. & Nicholls, R. J. Implications of sea-level rise and extreme events around Europe: a review of coastal energy infrastructure. Climatic Change 122, 81–95 (2013).DOI
  28. [28]Osterblom, H., Jouffray, J.-B., Folke, C. & Rockström, J. Emergence of a global science-business initiative for ocean stewardship. Proceedings of the National Academy of Sciences 114, 9038–9043 (2017).DOI
  29. [29]Österblom, H. et al. Scientific mobilization of keystone actors for biosphere stewardship. Scientific Reports 12, 3802 (2022).DOI
  30. [30]Galaz, V., Crona, B., Dauriach, A., Scholtens, B. & Steffen, W. Finance and the earth system – exploring the links between financial actors and non-linear changes in the climate system. Global Environmental Change 53, 296–302 (2018).DOI
  31. [31]Galaz, V. et al. Financial influence on global risks of zoonotic emerging and re-emerging diseases: An integrative analysis. The Lancet Planetary Health 7, e951–e962 (2023).DOI
  32. [32]Crona, B., Folke, C. & Galaz, V. The Anthropocene reality of financial risk. One Earth 4, 618–628 (2021).DOI
  33. [33]Jouffray, J.-B., Crona, B., Wåssénius, E., Bebbington, J. & Scholtens, B. Leverage points in the financial sector for seafood sustainability. Science Advances 5, (2019).DOI
  34. [34]Boissinot, J., Goulard, S., Salin, M., Svartzman, R. & Weber, P.-F. Aligning financial and monetary policies with the concept of double materiality: Rationales, proposals and challenges. (2022).
  35. [35]Caldecott, B., Clark, A., Koskelo, K., Mulholland, E. & Hickey, C. Stranded Assets: Environmental Drivers, Societal Challenges, and Supervisory Responses. Annual Review of Environment and Resources 46, 417–447 (2021).DOI
  36. [36]Bolton, P., Després, M., Pereira da Silva, L., Samama, F. & Svartzman, R. Green swans’: Central banks in the age of climate-related risks. Banque de France Bulletin 229, 1–15 (2020).DOI
  37. [37]Abramskiehn, D., Wang, D. & Buchner, B. The landscape of climate exposure for investors. San Francisco: Climate Policy Initiative (2015).
  38. [38]Sanchez-Garcia, P., Galaz, V. & Rocha, J. Finance, climate and ecosystems: A literature review of domino-effects between the financial system, climate change and the biosphere. (2022).
  39. [39]Hobday, A. et al. Categorizing and naming marine heatwaves. Oceanography 31, (2018).DOI
  40. [40]Fredston, A. L. et al. Marine heatwaves are not a dominant driver of change in demersal fishes. Nature 621, 324–329 (2023).DOI
  41. [41]Sallée, J.-B. et al. Summertime increases in upper-ocean stratification and mixed-layer depth. Nature 591, 592–598 (2021).DOI
  42. [42]Rocha, J. C. Ecosystems are showing symptoms of resilience loss. Environmental Research Letters 17, 065013 (2022).DOI
  43. [43]Folke, C. Resilience (republished). Ecology and Society 21, art44 (2016).DOI
  44. [44]Scheffer, M. et al. Early-warning signals for critical transitions. Nature 461, 53–59 (2009).DOI
  45. [45]Sathyendranath, S. et al. An ocean-colour time series for use in climate studies: The experience of the ocean-colour climate change initiative (OC-CCI). Sensors (Basel, Switzerland) 19, (2019).DOI
  46. [46]Biggs, R., Peterson, G. D. & Rocha, J. C. The Regime Shifts Database: a framework for analyzing regime shifts in social-ecological systems. Ecology and Society 23, (2018).DOI
  47. [47]Andersen, T., Carstensen, J., Hernández-García, E. & Duarte, C. M. Ecological thresholds and regime shifts: approaches to identification. Trends in Ecology & Evolution 24, 49–57 (2009).DOI
  48. [48]Vert-pre, K. A., Amoroso, R. O., Jensen, O. P. & Hilborn, R. Frequency and intensity of productivity regime shifts in marine fish stocks. Proceedings of the National Academy of Sciences 110, 1779–1784 (2013).
  49. [49]Kulk, G. et al. BICEP / NCEO: Monthly global Marine Phytoplankton Primary Production, between 1998-2020 at 9 km resolution (derived from the Ocean Colour Climate Change Initiative v4.2 dataset). (2021). doi:10.5285/69B2C9C6C4714517BA10DAB3515E4EE6DOI
  50. [50]RAM Legacy Stock Assessment Database. RAM Legacy stock assessment database v4.66. (2024). doi:10.5281/ZENODO.1404031
  51. [51]Ricard, D., Minto, C., Jensen, O. P. & Baum, J. K. Examining the knowledge base and status of commercially exploited marine species with the RAM Legacy Stock Assessment Database. Fish and Fisheries 13, 380–398 (2011).
  52. [52]Pettit, A. N. A non-parametric approach to the change-point problem. Applied Statistics 28, 126 (1979).
  53. [53]Zeileis, A., Leisch, F., Hornik, K. & Kleiber, C. strucchange: An RPackage for Testing for Structural Change in Linear Regression Models. Journal of Statistical Software 7, (2002).
  54. [54]Kroodsma, D. A. et al. Tracking the global footprint of fisheries. Science 359, 904–908 (2018).DOI
  55. [55]Carmine, G. et al. Who is the high seas fishing industry? Science of the Total Environment 3, 730–738 (2020).
  56. [56]Robins, G., Pattison, P., Kalish, Y. & Lusher, D. An introduction to exponential random graph (p*) models for social networks. Social networks 29, 173–191 (2007).DOI
  57. [57]Essington, T. E. et al. Fishing amplifies forage fish population collapses. Proceedings of the National Academy of Sciences 112, 6648–6652 (2015).DOI
  58. [58]Pinsky, M. L. & Byler, D. Fishing, fast growth and climate variability increase the risk of collapse. Proceedings of the Royal Society B: Biological Sciences 282, 20151053 (2015).DOI
  59. [59]Durant, J. M., Holt, R. E. & Langangen, Ø. Large biomass reduction effect on the relative role of climate, fishing, and recruitment on fish population dynamics. Scientific Reports 14, (2024).DOI
  60. [60]Coalition, F. T. Fishy networks: Uncovering the companies and individuals behind illegal fishing globally. (2023).
  61. [61]Bebbington, J. et al. Shaping nature outcomes in corporate settings. Philosophical Transactions of the Royal Society B: Biological Sciences 379, (2024).DOI
  62. [62]Pinsky, M. L. et al. Preparing ocean governance for species on the move. Science 360, 1189–1191 (2018).
  63. [63]Spijkers, J. & Boonstra, W. J. Environmental change and social conflict: the northeast Atlantic mackerel dispute. Regional Environmental Change 17, 1835–1851 (2017).DOI
  64. [64]Pinsky, M. L., Worm, B., Fogarty, M. J., Sarmiento, J. L. & Levin, S. A. Marine Taxa Track Local Climate Velocities. Science 341, 1239–1242 (2013).DOI

Paper details

vlnc:W_xtz2uthw
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 6

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Computer Science

Not recorded for this paper

View statement
Looking for related papers…
Contents