Original ArticlesGlobal thresholds in properties emerging from cumulative curves of marine ecosystems
Introduction
Several works on indicators in Marine Ecosystem (ME) ecology are moving into the exploration of indicator thresholds (Daan, 2005, Libralato et al., 2008, Fay et al., 2015, Large et al., 2015, Samhouri et al., 2017, Tam et al., 2017a). National and international regulations are increasingly calling for such indicator thresholds: examples include the Good Environmental Status in the Marine Strategy Framework Directive (European Parliament and Council of the European Union, 2008) context in Europe, Ecosystem-based Fisheries Management in the US (NMFS, 2016), and the Sustainable Development Goals in the UN context (United Nations, 2015). These regulations call for the quantification of ecosystem state, i.e., the capability of an ecosystem to carry out its main processes (ecosystem functioning) and maintaining its structure. Collectively, defining meaningful indicators and their thresholds represents an instance of making EBM more operational (Groffman et al., 2006, Shin et al., 2010, Tam et al., 2017b) by translating and condensing copious ecosystem information into decision criteria from which management measures can be triggered or evaluated (Coll et al., 2013).
There is also a growing body of work on the use of emergent properties in MEs that can represent useful indicators to synthesize marine ecosystem functionality. Among existing and well explored indicators (Christensen, 1995, Costanza and Mageau, 1999, Fath et al., 2001, Caddy, 2002, Marques and Jørgensen, 2002, Ulanowicz, 2004, Jørgensen and Müller, 2013, Longo et al., 2015) the cumulative biomass-trophic level and cumulative biomass-production curves (Pranovi and Link, 2009, Pranovi et al., 2014, Link et al., 2015) represent novel approaches.
In short, examining the cumulative curves of biomass and productivity across trophic levels (TLs) allow detecting new emergent properties that can provide insights on overall ecosystem functioning. As an advancement of classical ecology that proposes the analysis of trophic pyramids by looking at biomasses and production at integer trophic levels, the cumulative curves benefit from the identification of fractional trophic levels for species: biomasses and production are summed continuously across trophic levels to facilitate identification of emergent properties on resulting cumulative curves. Notably such emergent properties are based on a clear theoretical background of biomass accumulation that is (log-) normally distributed and transfer efficiency-limited up a food chain. Thus, if production at different trophic levels are always pyramids because the transfer efficiency is always lower than 1, cumulative curves of productions are monotonically asymptotic tending to plateau (near the sum of all system productivity). Fundamental trophodynamic features represented by primary production, turn-over of populations, average growth efficiency and growth in size are the overall system limits that influence the production curve (see for instance Fig. 1a in the work Link et al., 2015). Additionally, classical biomasses across trophic levels are not necessarily pyramidal but more often rhomboid in marine systems: the cumulative biomass curve across trophic levels (cumB-TL) is thus a sigmoidal curve, i.e. a curve with an inflection point. The cumB-TL curves exhibit a typical “S” pattern that seems to hold regardless of type of ecosystem or type of data used to construct them (Pranovi et al., 2014). Previous work has confirmed the existence and commonality of these curves from over 120 different MEs (Link et al., 2015) and demonstrated repeatable, consistent and predictable changes in curve shapes due to perturbations that can modify fundamental trophodynamic features of MEs. In a stylized example of the cumB-TL curve reported in Fig. 1 (adapted from Link et al., 2015), perturbations result in changes in the “S” curve over time that become less steep and move toward low TLs (Fig. 1 – Ib or IVb). Conversely, ecosystem recovery results in increased steepness and movement toward upper TLs of these curves (Fig. 1 – IIa or IIIa). These scenarios imply measurable changes on the major curve parameters, primarily determinants of “S” curve such as Biomass inflection point (B_Infl in the following), TL inflection point (TL_Infl) and steepness (Steep), which can be tracked over time to determine major shifts in condition in an ecosystem. These three simple curve parameters represent emergent properties of MEs with a surprising degree of insight into ecosystem structure and functioning. Thus the cumulative curves hold some promise in delineating regions of ecosystem state that require management action.
The original characterization and presentation of the cumB-TL curves (Link et al., 2015) hypothesized that there may be thresholds associated with these curve parameters that could signify important shifts in ecosystem state. In particular, (Link et al., 2015) proposed the inflection point of the cumB curve at 40% of the total biomass (B_Infl = 40%) as a threshold that would indicate a critical condition, and conversely if crossed a recovering ecosystem state. Defined on the basis of results from a meta-analysis of several aquatic ecosystems, such a threshold is connected with the maximum of accumulation of biomasses across TLs. The cumulative biomasses below such an accumulation point is in normal condition when larger than 40% of total biomass while lower values are indicative of an unbalanced distribution of biomasses (Link et al., 2015). A more thorough examination of the value of that threshold, along with others, is warranted. Data from over 60 Large Marine Ecosystems (LMEs; see Sherman et al., 1993) can help emphasize not only the importance of cumB-TL curve parameters, but trajectories thereof. Changes across time of curve parameters, in fact, are expressions of pressures (natural or anthropogenic) that an ecosystem has experienced which have resulted in modified ecosystem functioning and structure. Tracking the trajectories of curve parameters can thus provide an indication of changes that have occurred. The dynamics of these parameters, particularly regarding recent history and direction of change, are as important as the specific value of the parameters and can help defining “areas of transition” between LMEs or within the same LME through time (see Fig. 1).
Here we wanted to delineate thresholds in these “S” curve parameters from 62 LMEs to determine if there are common curve parameter thresholds and to ascertain how those have changed over time. The objective of the work is to answer the following questions:
- -
are there consistent thresholds for parameters B_Infl, TL_Infl and Steep of the CumB vs TL curves as estimated for all LME data?
- -
Can these thresholds be linked to cumB curve theory?
- -
Can we track changes in curve shape over time to understand ecosystem state with respect to thresholds?
Here we identify the thresholds using an objective methodology, use them to distinguish different conditions of the ecosystems, and ultimately use them to contribute towards the delineation of ecosystem state for each of these LMEs.
Section snippets
Data used
We used landings data for 62 Large Marine Ecosystems (LMEs) obtained from the Sea Around Us Project (SAUP) database (http://www.seaaroundus.org/). These data are the widest sampling of the species and, although not covering the whole spectrum of species in the ecosystem, are used here as a proxy for ecosystem status definition. Data consists in landings weight (ton) by taxa caught each year for the period (1950–2010 included, i.e., 61 years). Taxa represent species or genera for the most
Results
The plot of all curve parameter pairings (Fig. 2) resulting from the analysis of the CumB-TL curves for all years in all 62 LMEs highlights a few interesting, and even peculiar, patterns. In particular, Steep vs TL_Infl showed two peaks, with neither clear nor obvious patterns, and with most LME-years with very low Steep (Fig. 2a). TL_Infl vs B_Infl showed a wide spread of data but primarily a positive linear relation between the two parameters (Fig. 2b). Finally, Steep vs B_Infl showed a clear
Discussion
What does it mean to shift the “S” curve shape as evinced by the dynamic curve parameters? In essence an ecosystem is degrading, recovering, or (relatively) stably functioning. This confirms the theoretical and empirical descriptions of curve parameters as linked to major ecosystem drivers—and responses—as seen in prior works (Pranovi et al., 2014, Link et al., 2015). The individual ecosystems shown here exhibit clear patterns of decline to a new ecosystem state (e.g., Baltic Sea and Humbolt
Acknowledgements
We acknowledge funding sources of Flagship RITMARE—The Italian Research for the Sea—coordinated by the Italian National Research Council and funded by the Italian Ministry of Education, University and Research within the National Research Program 2011–2013 through grants provided to S.L. to support this work. We thank the Sea Around Us Project (http://www.seaaroundus.org/) for their compilation and availability of these catch data. We want to particularly note that this work is the result of
References (44)
Limit reference points, traffic lights, and holistic approaches to fisheries management with minimal stock assessment input
Fish. Res.
(2002)Ecosystem maturity - towards quantification
Ecol. Modell.
(1995)- et al.
Sustainability implications of honouring the Code of Conduct for Responsible Fisheries
Global Environ. Change
(2013) - et al.
Complementarity of ecological goal functions
J. Theor. Biol.
(2001) - et al.
Emergent properties delineate marine ecosystem perturbation and recovery
Trends Ecol. Evol.
(2015) - et al.
Three selected ecological observations interpreted in terms of a thermodynamic hypothesis. Contribution to a general theoretical framework
Ecol. Modell.
(2002) - et al.
Ecosystem exploitation and trophodynamic indicators: a comparison between the Northern Adriatic Sea and Southern New England
Prog. Oceanogr.
(2009) - et al.
How well can FMSYand BMSYbe estimated using empirical measures of surplus production?
Fish. Res.
(2012) Quantitative methods for ecological network analysis
Comput. Biol. Chem.
(2004)- Blasco, J., Chapman, P.M., Campana, O., Hampel, M., 2016. Marine ecotoxicology: current knowledge and future issues,...
What is a healthy ecosystem?
Aquat. Ecol.
An afterthought: ecosystem metrics and pressure indicators
ICES J. Mar. Sci.
Why the complex nature of integrated ecosystem assessments requires a flexible and adaptive approach
ICES J. Mar. Sci.
Directive 2008/56/EC of the European Parliament and of the Council
Off. J. Eur. Union
Management performance of ecological indicators in the Georges Bank finfish fishery
Ices J. Mar. Sci.
Hypoxia and acidification in ocean ecosystems: coupled dynamics and effects on marine life
Biol. Lett.
Ecological thresholds: the key to successful environmental management or an important concept with no practical application?
Ecosystems
Chemical contaminant effects on marine ecosystem functioning
J. Appl. Ecol.
Ecosystems Ecology
Quantifying patterns of change in marine ecosystem response to multiple pressures
PLoS ONE
Complex ecological data sets
Numer. Ecol.
Cited by (14)
Emergent properties of free-living nematode assemblages exposed to multiple stresses
2024, Science of the Total EnvironmentA simulation-based approach to assess the stability of marine food-webs and inform Good Environmental Status
2024, Ecological IndicatorsSimulations and interpretations of cumulative trophic theory
2022, Ecological ModellingCitation Excerpt :The application of cumulative trophic theory for the wise management of marine ecosystems has higher utility knowing we can more repeatedly and comfortably predict their responses to a wide range of conditions. Thus the applications that may follow might prove useful for the even wiser management of marine ecosystems (Link et al., 2015, 2020; Libralato et al., 2019; Pranovi et al., 2020). More so, that a relatively simple equation can depict, capture and predict such a wide range of marine ecosystem dynamics across a broad array of situations is not trivial, and further suggests the robustness of the cumulative trophic theory.
Measuring left-tail risk of fish species
2021, Ocean and Coastal ManagementComparative production of fisheries yields and ecosystem overfishing in African Large Marine Ecosystems
2020, Environmental DevelopmentCitation Excerpt :Of the five out of eight African LMEs that have exhibited symptoms of EOF, the results are not surprising and generally consistent with what we know about the Guinea Current (e.g., McGlad et al., 2002; Mensah and Quaatey, 2002; Ukwe et al., 2006, FAO, 2018a), Benguela Current (e.g., Shannon et al., 2006; Cochrane et al., 2009, Shin et al., 2010b, Shannon et al. 2014; Blamey et al., 2015, Jarre, 2016; FAO, 2018a), Mediterranean Sea (e.g., Tudela, 2004; Pranovi et al., 2014; FAO, 2018a; Stergiou et al., 2016; FAO, 2018a; 2018b; Russo et al., 2019), and Arabian Sea (e.g., Siddeek, 1999; FAO, 2018a; Lam and Pauly, 2019) LMEs. These is also consistent with six of these 8 LMEs being classified as in a transition state with respect to the “S” curves (Pranovi et al., 2020; Libralato et al., 2019). All have exhibited histories of notable stock overfishing and changes to the ecosystem; hence that we see evidence for ecosystem overfishing is not surprising.
Probabilistic indicators for soil and groundwater contamination risk assessment
2020, Ecological Indicators