Marine regime shifts around the globe: theory, drivers and impacts

1. Background

Sudden, dramatic, long-lasting shifts in ecosystem structure and function have been increasingly documented in multiple, diverse marine ecosystems around the globe [1–21]. Such regime shifts can be explained with theory on nonlinear systems crossing critical thresholds [2,22]. Both theory and observations highlight that a particular regime may not represent a stable, unchanging condition. Rather it may be characterized by fluctuations of ecosystem state around an attractor. A regime shift thus does not simply involve a large change in an ecosystem, but rather a persistent shift in the structure and dynamics of the system, often involving a change in the internal feedbacks of the system [1]. However, linking observed abrupt changes in the field to the mathematical theory of complex systems is still a challenging endeavour [22–27].

Abrupt transitions between differing states in complex systems are also widely discussed in social sciences, because they often have large impacts on human economies and societies [1]. Regime shifts usually lead to drastic changes in the provision of ecosystem services, as typically exemplified by worldwide fisheries collapses [28,29]. The interest in the theory of critical transitions in ecology and especially for marine ecosystems is large, not only because of interest in unravelling the functioning of natural communities and ecosystems, but also because of its potential management implications [30,31]. Under increasing anthropogenic pressure [32], either globally through climate change or more locally through overexploitation, biodiversity loss and change, eutrophication and contamination, abrupt and catastrophic changes are anticipated to increase in strength and frequency [33,34]. Theory on abrupt transitions brings up questions on the reversibility of change, and costly and still potentially failing management measures and conservation efforts. Furthermore, theory may be instrumental in developing approaches to leading indicators that may be able to anticipate abrupt changes while predictive models to do so are still lacking [35].

Despite prominent publications, special issues and reviews that have tried to clarify the theory on abrupt transitions [2,3,22,26,36–38], there is still considerable scientific debate and in particular the marine community is largely divided into ‘believers’ and ‘sceptics’ [22] of the regime shift context. Scientists are trapped in their specific mental models and discussions are mainly focused around the inter-related questions on the predominance of regime shift drivers and the potential existence of alternative, and hence difficult to reverse, stable states.

The terminology used to describe the theory often varies between disciplines and research communities, as can be seen in the papers of this theme issue. The terms ‘regimes’ or ‘states’ are used to describe potentially differing configurations of populations, communities or whole ecosystems. Switches between regimes/states are then called ‘regime shifts’ or ‘phase transitions’, or a combination of both. While the history of the terminology in different scientific fields is reviewed elsewhere [2,26], we for the present theme issue adopted the term ‘regime shift’. A major reason for our choice was first the higher popularity of ‘regime shifts’ in contrast to the alternatives, but especially because this term initiated and still initiates a number of scientific debates (but see below). More scientifically, we based our choice on the observation that ecosystem (or other) states are always dynamic, which is better reflected by using ‘regimes’ instead of ‘states’, which implies a more static perspective [22].

This theme issue ‘Marine regime shifts around the globe: theory, drivers and impacts’ has the goal to make a step change towards a more unified understanding of regime shifts in marine ecosystems. Towards this purpose we define ecological regime shifts as ‘dramatic, abrupt changes in the community structure that are persistent in time, encompassing multiple variables, and including key structural species—independently from the mechanisms causing them’.

Our definition deliberately includes regime-like changes without evidence of multiple alternative stable states (or multiple basins of attraction), as we think emphasis on this theoretical aspect often hinders progress in considering abrupt changes in marine ecosystem-based management. Our definition is hence more practical for marine management purposes and can be used for both benthic and pelagic regime shifts, even where the link with the mathematical theory is not yet fully established (see [39]).

This special issue brings together experts from different marine science disciplines and trophic level expertise (i.e. benthic ecology, pelagic ecology, fisheries, marine conservation and management), from diverse marine ecosystems, and from a mixture of geographical areas around the globe. More than 80 authors from six continents have contributed to the 16 papers in this issue, around the following themes: (i) advances in marine regime shift theory, (ii) drivers of marine regime shifts and (iii) management of marine regime shifts. Below, the papers are introduced, and the major findings of the special issue are synthesized in Conversi et al. [39].

2. Advances in marine regime-shift theory

A major theoretical finding when it comes to regime shifts is that ecosystems recover slowly from small perturbations in the vicinity of tipping points. Hence, indicators of critical slowing down have been developed that can provide an early warning signal of a nearby tipping point, or may offer a possibility to rank reefs, lakes or other ecosystems according to their resilience. However, indicators of critical slowing down are not manifested in all cases where regime shifts occur, because not all regime shifts are associated with tipping points. Dakos et al. [40] provide a review of the related literature and develop guidance based on critical transition theory on what to expect and what not to expect when it comes to early warning signals.

Whether alternative stable states exist in marine ecosystems is of key importance for the prospects of ecosystem recovery and management. Answering this question requires experimental evidence and therefore is particularly difficult in the pelagic realm. Gårdmark et al. [41] show how mechanisms underlying alternative stable states caused by predator–prey interactions can be revealed in the pelagic ecosystems using time-series analyses guided by theory on size-structured community dynamics. Based on Baltic cod (Gadus morhua) field data, they discuss and distinguish two types of feedback mechanisms that can cause alternative stable states: abundant versus collapsed, unable to recover, piscivorous fish populations. Under ‘cultivation-depensation’, predatory fish cannot recover from low densities because their fish prey (then released from predation) outcompetes the juvenile predatory fish. Under ‘overcompensation’, when predatory fish abundance is low, the lack of top-down control of the prey population dynamics results in a different prey size structure, with less individuals vulnerable to predation, and the predatory fish cannot recover owing to food shortage. In contrast, Beaugrand [42] proposes a theory, based on planktonic systems, showing that regime shifts can originate from the interaction between climate-induced environmental changes and the species' ecological niche. This theory negates the necessity of alternative stable states, at least in some of the marine regime shifts, and offers a way to predict future climate-induced community shifts and their potential associated trophic cascades and amplifications.

3. Drivers of marine regime shifts

Three case studies from benthic systems show that multiple drivers usually interact in reducing ecosystem resilience to change, and hence in causing regime shifts. Bozec & Mumby [43] investigate on how global warming can affect coral reefs. They show that Caribbean coral reefs changed by epizootics are vulnerable to both acute and chronic effects of temperature. They find that stress types act additively on reef state, but chronic temperature stress decreases the size of the coral basin of attraction (i.e. its resilience) by reducing coral calcification and growth, whereas acute temperature stress induces coral bleaching and can push the system closer to an algal attractor. Jouffray et al. [44] demonstrate the existence of three distinct reef regimes in the Hawaiian Archipelago. They further investigate on the possible anthropogenic drivers of these regimes and show that herbivore biomass is the key regime-shift driver, acting together with the effects of land-based pollution and land-use change. Another dramatic regime shift in benthic ecosystems is known from temperate rocky reefs, where productive macroalgal beds transition to impoverished urchin barrens as a result of sea urchin overgrazing. Ling et al. [45] explore the generality of regime shift dynamics across reefs worldwide and find a globally coherent pattern in which the hysteresis is approximately one order of magnitude in urchin abundance between critical thresholds of overgrazing versus recovery. Anthropogenic stressors such as overexploitation add to positive feedback mechanisms, by eroding the resilience of desirable macroalgal beds while strengthening the resilience of the urchin barrens, thus making these unwanted regime shifts virtually irreversible.

While current benthic studies of regime shifts tend to focus on stressor effects on ecosystem resilience, the relative importance of top-down (biotic) or bottom-up (physical) processes is a central issue in the discussion of regime shifts in marine pelagic systems. Pershing et al. [46] review the literature and provide a general methodology for distinguishing top-down and bottom-up drivers and apply this methodology to time series from ecosystems in the Black and Baltic Sea as well as the Scotian Shelf off Canada. They show that the importance of top-down control decreases with the openness of the systems, and challenge the assumption that negative correlation between consecutive trophic levels implies top-down forcing. The importance of bottom-up (physical) forcing for pelagic ecosystem dynamics is investigated by Beaugrand et al. [47] through the comparative analysis of 11 (mainly zooplanktonic) ecosystems from two oceans and three regional seas in the Northern Hemisphere. They show a quasi-synchronicity in marine pelagic regime shifts across the Northern Hemisphere in the late 1980s and propose that temperature and arctic atmospheric circulation patterns are likely responsible for this synchronicity. Fisher et al. [48] review the presence/absence of regime shifts in multiple ecosystems in the Northern Hemisphere, considering spatial variability in key biotic and abiotic drivers. They highlight the importance of understanding the scale-dependent spatial dynamics of climate influences and key predator–prey interactions to unravel the dynamics of regime shifts, and the potential of spatial downscaling as a means of evaluating hypotheses that emerge from among-system comparisons.

Eventually, Rocha et al. [49] assess the vulnerability of marine ecosystems to regime shifts and their potential consequences by reviewing the scientific literature for 13 types of marine regime shifts. Using network analysis of the co-occurrence of drivers and ecosystem service impacts, they demonstrate that regime shifts are in general caused by multiple drivers and have multiple consequences that co-occur in a non-random manner. Drivers related to food production, climate change and coastal development were the most common co-occurring causes of regime shifts, whereas cultural services, biodiversity and primary production are the most common cluster of ecosystem services affected.

4. Management of marine regime shifts

Clearly, marine regime shifts are a challenge to society because they are largely unpredictable, often occurring unexpectedly and carrying large economic implications. Accounting for regime shifts in management hence clearly requires integrative, ecosystem-based management (EBM) approaches [50]. Yet, despite the rapid and intense development of EBM theory, its implementation has languished; in addition, most implemented or proposed EBM schemes largely ignore the special characteristics of regime shifts. Levin & Möllmann [50] explore key aspects of regime shifts that are of critical importance to EBM, and then suggest how regime shifts can be better incorporated into EBM using the concept of integrated ecosystem assessment. For the more sectorial approach of ecosystem-based fisheries management (EBFM), King et al. [51] review efforts to incorporate regime shifts and states into EBFM and outline a potential framework to include regime shifts and changes in ecosystem states into fisheries management.

In the past decade, as stressor–response relationships have become better understood, using thresholds has become increasingly relevant in the context of environmental management related to regime shifts. As a consequence, there is increased interest in identifying pre-emptively thresholds to inform decision-making. Kelly et al. [52] evaluate the success of management measures implemented in ecological systems with well-characterized biophysical thresholds. They found that management is most effective when it is explicitly using science to avoid thresholds or to reverse ecosystem change after a threshold has been crossed. Furthermore, they demonstrate that management success is associated with routine monitoring of the system on a temporal and spatial scale relevant to the ecological threshold, and with local- and regional-scale management rather than decision-making at larger spatial scales.

Eventually, ecosystem dynamics and regime shifts are not just biophysical phenomena, and humans are strongly involved over multiple spatial and temporal scales. O¨sterblom & Folke [53] investigate the timing of pelagic marine regime shifts in relation to the emergence of regional and global fishing activities of the Soviet Union. They show that the Soviet Union was a major fishing actor in all large marine ecosystems where regime shifts have been documented and propose that a deeper understanding of the role of global players (here the fishing sector) is central for improved management of marine ecosystems.

5. Synthesis

The collective work presented in this theme issue shows a progression in reconciling observed ecosystem reorganizations with the mathematical theory of complex systems. Conversi et al. [39] note that the process of linking regime shift observations to theory is moving at a faster rate in the benthic community than in the pelagic community, in part because alternative states are easier to observe and experiments easier to undertake on reef populations, and in part because of the difficulty in discerning temporal from spatial (i.e. biogeographic) shifts in the pelagic realm. Conversi et al. [39] hence propose a general framework for studying marine regime shifts, in which multiple, interacting exogenous stressors impact the endogenous trophic interactions of the food web and the overall state and resilience of the ecosystem. They propose leaving behind the false dichotomy between top-down (or biotic) versus bottom-up (or physical) drivers of marine regime shifts and rather than focusing on single drivers, move towards identifying mechanisms, and specifically the combination of processes that may cause regime shifts or affect the resilience of the ecosystem. Furthermore, future studies should use nested approaches and/or multi-ocean-basin comparisons, following this scientific framework. As early warnings of regime shifts are not yet possible, precautionary management should entail enhancing system resilience and EBM to withstand the multiple pressures marine ecosystems increasingly face in our globalized world.

Footnotes

†These authors contributed equally to this study.