Edited by: Christos Dimitrios Arvanitidis, Hellenic Centre for Marine Research, Greece
Reviewed by: Nils Teichert, National Research Institute of Science and Technology for Environment and Agriculture, France; Laura Uusitalo, Finnish Environment Institute (SYKE), Finland
*Correspondence: Jamie C. Tam
This article was submitted to Marine Ecosystem Ecology, a section of the journal Frontiers in Marine Science
†Present Address: Jamie C. Tam, Fisheries and Oceans Canada, Bedford Institute of Oceanography, Halifax, NS, Canada
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Ecosystem-based management (EBM) in marine ecosystems considers impacts caused by complex interactions between environmental and anthropogenic pressures (i.e., oceanographic, climatic, socio-economic) and marine communities. EBM depends, in part, on ecological indicators that facilitate understanding of inherent properties and the dynamics of pressures within marine communities. Thresholds of ecological indicators delineate ecosystem status because they represent points at which a small increase in one or many pressure variables results in an abrupt change of ecosystem responses. The difficulty in developing appropriate thresholds and reference points for EBM lies in the multidimensionality of both the ecosystem responses and the pressures impacting the ecosystem. Here, we develop thresholds using gradient forest for a suite of ecological indicators in response to multiple pressures that convey ecosystem status for large marine ecosystems from the US Pacific, Atlantic, sub-Arctic, and Gulf of Mexico. We detected these thresholds of ecological indicators based on multiple pressures. Commercial fisheries landings above approximately 2–4.5 t km−2 and fisheries exploitation above 20–40% of the total estimated biomass (of invertebrates and fish) of the ecosystem resulted in a change in the direction of ecosystem structure and functioning in the ecosystems examined. Our comparative findings reveal common trends in ecosystem thresholds along pressure gradients and also indicate that thresholds of ecological indicators are useful tools for comparing the impacts of environmental and anthropogenic pressures across multiple ecosystems. These critical points can be used to inform the development of EBM decision criteria.
Ecosystem-based management (EBM) of the ocean, which considers the management of the broad range of ecosystem services across ocean-use sectors (Slocombe,
Effective EBM requires the quantification of reference points to locate a balance between a healthy ecosystem and multiple human uses (Dearing et al.,
Ecological indicators are useful tools to interpret the complexity of ecosystems (Coll and Lotze,
Understanding the impacts of pressures on ecosystems is another key element of EBM (Jennings,
In many ecosystem studies, baseline reference points are often typically determined from comparisons of a measured value relative to the long-term average (or maximum/minimum) of a time series, from an expert-opinion derived value, or from estimates from presumed unexploited populations (Shears and Babcock,
Here, we aim to develop operational reference points by quantifying thresholds for a suite of ecological indicators along multivariate pressure gradients (both anthropogenic and environmental). We further compare these operational reference points among multiple marine ecosystems, recognizing the value in comparative ecosystem studies (Murawski et al.,
This study examined four Large Marine Ecosystems (LMEs: Alaska-Eastern Bering Sea, California Current, Northeast US and northern Gulf of Mexico) that are part of NOAA's IEA program (Levin et al.,
Study large marine ecosystem (LME; gray). Solid lines represent the US exclusive economic zone (EEZ) and dotted lines represent the integrated ecosystem assessment large marine ecosystem (IEA LME).
The indicator data used in this study were compiled from NOAA's fishery-independent surveys from Alaska (1982–2013), California Current (1981–2012), Northeast US (1964–2013), and Gulf of Mexico (1992–2010) which provide information regarding the ecology and oceanography of each respective LME (Table
Ecological indicators.
Mean length | Length | Mean length (cm) of individual fish for all species | Structural: size distribution Methratta and Link, |
Pelagic to demersal ratio | PDR | Ratio of biomass of pelagic to the biomass of demersal fishes | Structural: community structure, and energy flow Methratta and Link, |
Planktivore and benthivore to shrimp and fish eater ratio | LHTR | Ratio of the biomass of low trophic level to high trophic level fishes | Functional: trophic dynamics, energy flow Link et al., |
Mean trophic level | MTL | Mean trophic level of surveyed species weighted by abundance (biomass) | Functional: how energy flow within an ecosystem is processed Methratta and Link, |
Species richness | Rich | Number of species surveyed | Resilience: community status, biodiversity Downing and Leibold, |
Diversity (Effective number) | EN | Exponent of Shannon diversity index. Measure of species diversity. | Resilience: biodiversity accounting for sensitive species Jost, |
A variety of both anthropogenic and environmental variables were selected to reflect pressures on ecosystems (Table
Pressure variables from the Alaska (EBS), California Current (CC), Northeast US (NEUS), and Gulf of Mexico (GOMEX) ecosystems.
Population increase | Population.inc | Anthropogenic | EBS, CC, NEUS, GOMEX | Change in population from year to year derived from yearly census estimates | Halpern et al., |
Commercial fishing | Landings | Anthropogenic | EBS, CC, NEUS, GOMEX | Total biomass of commercial landings weighted by area of the LME (t/km2) | Link et al., |
Fisheries exploitation | Exploitation | Anthropogenic | EBS, CC, NEUS, GOMEX | Total landings by the total biomass estimated from the fishery-independent survey | Large et al., |
Commercial fishing 1 year lag | Landings_1 | Anthropogenic | EBS, CC, NEUS, GOMEX | Total biomass of commercial landings weighted by area of the LME (t km−2) with a 1 year lag | Large et al., |
Fisheries exploitation 1 year lag | Exploitation_1 | Anthropogenic | EBS, CC, NEUS, GOMEX | Total landings by the total biomass estimated from the fishery-independent survey with a 1 year lag | Large et al., |
Annual gross domestic product increase from fisheries and agriculture | GDP.inc | Anthropogenic | EBS, CC, NEUS, GOMEX | Changes in gross domestic product for coastal States from year to year for fisheries and agriculture measured by the Bureau of Economics | Mora et al., |
Pacific decadal oscillation | PDO | Environmental | EBS, CC | Multidecadal index of Pacific climate variability. | Mantua et al., |
Atlantic multidecadal oscillation | AMO | Environmental | NEUS, GOMEX | Multidecadal index of Atlantic climate variability. | Harris et al., |
Multivariate El niño index | MEI | Environmental | EBS, CC, NEUS, GOMEX | Index that characterizes El Niño Southern Oscillation events. | Litzow et al., |
Sea surface temperature | SST | Environmental | EBS, CC, NEUS, GOMEX | Mean surface temperature of the LME waters (degrees C) | Devred et al., |
Primary productivity/Chlorophyll |
Chlorophyll | Environmental | EBS, CC, NEUS, GOMEX | Annual mean chlorophyll from remote sensing data in mg m−3. 14 C primary productivity experiments in CC collected by CalCOFI. | Behrenfeld et al., |
Freshwater anomalies | Freshwater | Environmental | EBS, CC, NEUS, GOMEX | Annual discharge anomalies from major coastal catchments areas associated with the LME (cumecs km−2) | Carmack et al., |
North wall of the Gulf Stream | GS | Environmental | NEUS | Index of the position of the north wall of the Gulf Stream. | Taylor, |
Winter North Atlantic oscillation | NAO_w | Environmental | NEUS | Winter (Dec-Mar) mean of relative strength between subpolar low and subtropical high atmospheric pressure cells (index) | Link et al., |
Wind stress | Wind | Environmental | NEUS | Force of the wind on the surface of the ocean (N m−2) | Ecosystem Assessment Program, |
North Pacific Index | NPI | Environmental | EBS | The area-weighted sea level pressure of the region and measures interannual to decadal variations in the atmospheric circulation | Litzow et al., |
Ice retreat | Ice.Retreat | Environmental | EBS | Rate of ice retreat in the Eastern Bering Sea | Mueter and Litzow, |
Cold Pool | Cold pool | Environmental | EBS | Relative size of the area of cold, dense, salty water in the region. | Mueter and Litzow, |
Atlantic warm pool | AWP | Environmental | GOMEX | Size of the pool of warm water (>28.5 degrees C) that comprises the Gulf of Mexico and Caribbean. | Karnauskas et al., |
Currents | Currents | Environmental | GOMEX | Annual mean transport of the Loop Current, Florida Current and Yucatan Current | Leipper, |
Hypoxic area | Hypoxic Area | Environmental | GOMEX | Mean annual area of the hypoxic zone of the Gulf of Mexico | Rabalais et al., |
Total upwelling magnitude index | TUMI | Environmental | CC | Annual mean of upwelling magnitude (m3/s/100 m) | Levin et al., |
Northern oscillation index | NOI | Environmental | CC | Index of climate variability based on the difference in sea level pressure anomalies at the North Pacific High and a climatologically low sea level pressure region (Darwin, Australia). | Schwing et al., |
North Pacific Gyre Oscillation | NPGO_w | Environmental | CC | Index of climate variability in the northeast Pacific measuring change in the North Pacific gyres circulation | Di Lorenzo et al., |
Environmental variables that influence ecosystem circulation patterns, primary production, availability of nutrients, and vertical mixing were chosen for all LMEs, namely Sea Surface Temperature (SST), and broad scale climatological indicators such as, Pacific Decadal Oscillation (PDO; for the Pacific coast regions), Atlantic Multidecadal Oscillation (AMO; for the Atlantic coast regions), or Multivariate El Niño Index (MEI for all regions). A measure of system production was included (Chlorophyll
We used random forest and gradient forest methods on time series of a suite of ecological indicators (Table
While random forests are useful for quantifying the ability of pressure variables to predict response variables, gradient forests integrate individual random forest analyses over many response variables and are also used to identify thresholds in those indicator responses along anthropogenic and environmental pressure gradients (Ellis et al.,
Because gradient forest analysis can detect thresholds in a multivariate context, this method is particularly useful for examining thresholds at the ecosystem level (see Pitcher et al.,
We used a set of complementary analyses to further examine the multivariate ecosystem trends across pressure variables and to confirm that detected thresholds are robust. We first distilled all of the ecological indicators used in the gradient forest analysis in each LMEinto ecosystem trends using Dynamic Factor Analysis (DFA; R package MARSS, R Core Team,
Using the best model for the ecosystem, we then used Generalized Additive Models (GAM; R package mgcv, R Core Team,
where
A potential strength and weakness of this approach is that it does not rely on
The total model prediction performance from the gradient forest analysis (the proportion of variance explained in a random forest) averaged across the suite of ecological indicators from each LME ranged from 0.01 to 0.07 (
Mean Model Performance (
Alaska | Length, Rich | 0.01 | 0–0.06 |
California current | Length, Rich, EN | 0.04 | 0–0.25 |
Northeast US | Length, PDR, LHTR, EN | 0.07 | 0–0.29 |
Gulf of Mexico | MTL, Length, PDR, Rich, EN | 0.05 | 0–0.30 |
Both the gradient forest analyses and GAMs did not identify a single driver that was consistently dominant across the four ecosystems, though fisheries landings was an important predictor in models for all systems (Figure
Importance of human and environmental pressure variables across ecological indicator outputs (
Deviance explained for the generalized additive model results for each ecosystem trend (DFA Trend) and pressure variable (using variables common in all ecosystems).
Trend 1 | Landings | ||||
Trend 2 | 0.01 | 0.00 | 0.25 | 0.00 | |
Trend 3 | 0.11 | 0.00 | 0.38 | NA | |
Trend 4 | 0.13 | NA | NA | NA | |
Trend 1 | Landings_1 | 0.07 | |||
Trend 2 | 0.05 | 0.00 | 0.31 | 0.00 | |
Trend 3 | 0.00 | 0.42 | NA | ||
Trend 4 | NA | NA | NA | ||
Trend 1 | Exploitation | 0.00 | 0.13 | 0.14 | |
Trend 2 | 0.00 | 0.14 | 0.61 | ||
Trend 3 | 0.00 | NA | |||
Trend 4 | 0.08 | NA | NA | NA | |
Trend 1 | Exploitation_1 | 0.00 | 0.18 | 0.17 | |
Trend 2 | 0.00 | 0.18 | 0.11 | ||
Trend 3 | 0.00 | NA | |||
Trend 4 | 0.18 | NA | NA | NA | |
Trend 1 | Population.inc | 0.08 | 0.03 | ||
Trend 2 | 0.00 | ||||
Trend 3 | 0.60 | NA | |||
Trend 4 | NA | NA | NA | ||
Trend 1 | SST | 0.04 | 0.34 | 0.10 | |
Trend 2 | 0.29 | 0.00 | 0.00 | ||
Trend 3 | 0.00 | 0.00 | NA | ||
Trend 4 | 0.42 | NA | NA | NA | |
Trend 1 | AMO/PDO | 0.00 | 0.56 | 0.09 | |
Trend 2 | 0.01 | 0.12 | 0.39 | ||
Trend 3 | 0.08 | 0.32 | NA | ||
Trend 4 | 0.03 | NA | NA | NA | |
Trend 1 | MEI | 0.14 | 0.17 | 0.00 | 0.10 |
Trend 2 | 0.09 | 0.00 | 0.00 | ||
Trend 3 | 0.01 | 0.14 | 0.02 | NA | |
Trend 4 | 0.18 | NA | NA | NA | |
Trend 1 | GDP.inc | 0.04 | 0.10 | 0.02 | |
Trend 2 | 0.15 | 0.17 | 0.00 | ||
Trend 3 | 0.22 | 0.21 | NA | ||
Trend 4 | 0.00 | NA | NA | NA | |
Trend 1 | Freshwater.anom | 0.16 | 0.00 | 0.01 | |
Trend 2 | 0.14 | 0.00 | 0.12 | ||
Trend 3 | 0.00 | 0.26 | 0.00 | NA | |
Trend 4 | 0.00 | NA | NA | NA | |
Trend 1 | Chlorophyll | 0.09 | 0.06 | 0.27 | 0.09 |
Trend 2 | 0.06 | 0.11 | |||
Trend 3 | 0.05 | 0.14 | 0.37 | NA | |
Trend 4 | 0.08 | NA | NA | NA |
Common landings and exploitation thresholds of ~2–4.5 t km−2 landings and ~20–40% exploitation of the total estimated biomass were detected with the gradient forest analyses (Figures
Mean thresholds and 95% CI ranges for
Cumulative shifts (in
Ecosystem trend (Trend 1 for all LMEs) responses to landings (t km−2) from Alaska (Bering Sea), California Current, Northeast US, and Gulf of Mexico ecosystems. Dotted lines are the smoothed GAM line, gray polygons surrounding the trend line are 95% CI, solid black lines indicate a significant threshold region.
The gradient forest analysis identified the relative size of the cold pool in Alaska as the only region specific pressure variable that ranked within the top five important pressures in explaining ecosystem shifts (Figure
Our results demonstrate that there are consistent patterns in ecosystem response from common pressures across four large marine ecosystems, and despite multiple potential mechanisms, the detected trends and thresholds to such pressures in these ecosystems were remarkably repeatable. Although each ecosystem examined has different socio-economic histories (Hollowed et al.,
One key result is that, at an ecosystem-level, removals of biomass (via landings-based exploitation) do have repeatable and consistent thresholds. There are different ecological mechanisms in which such ecosystem-level responses can be observed, but consistently there is an impact to overall size, congruent with overfishing theory, as well as tendencies toward smaller organisms with hyper-exploitation (Pauly et al.,
Another commonality was that when examining the impacts of the pressure variables to cumulative ecosystem responses, anthropogenic pressures rank high. Due to the pressure-response relationship of these analyses, this does not necessarily reflect the current status of a given ecosystem, but rather implies that certain pressures have heavily impacted these ecosystems within the history of the time series analyzed. This is not to say that environmental pressures are not important; rather that the anthropogenic pressures tended to more consistently emerge as clearer features that impact observed ecosystem dynamics.
As a particular example of such anthropenic pressures, in all the ecosystems, landings have decreased since the 1970s and 1980s, and some fish stocks have experienced a phase of rebuilding (Rosenberg et al.,
Lower population increases and higher GDP in coastal communities were related to ecosystem shifts where population density is highest (population density of Northeast US: ~300 indv km−2). These threshold values increased with decreasing population densities (Gulf of Mexico: ~56 indv km−2, California Current: ~50 indv km−2. Alaska: ~0.5 indv km−2; U.S.Census Bureau,
In terms of environmental pressures, another commonality was evidence that all ecosystems appear to influenced by multi-annually varying climate drivers, albeit seen via different indices in each region. The discovery of large-scale climate patterns have been an important step in connecting climate to ecological, biological, and oceanographic patterns in marine ecosystems (Mantua and Hare,
The analyses in this study were not used to examine specific mechanistic links between specific pressures and responses, but rather to identify and compare significant threshold ranges of ecosystems (represented by indicators) along pressure gradients. Individual ecological indicator thresholds can also be examined across individual pressure gradients to determine specific reference points at the single indicator level (e.g., Samhouri et al.,
Mechanistic links have been made between fisheries production and multiple drivers including fishing, trophodynamics, and the environment (Gaichas et al.,
Each ecosystem in this study has, at some stage in the last half century, experienced overfishing (Bakkala et al.,
The thresholds developed here can also be used to build proactive strategies to avoid regime shifts due to overfishing, population increase and climate change, particularly when explored through simulation modeling (Samhouri et al.,
While the location of ranges of ecosystem-level thresholds along both human and environmental pressure gradients are easily interpreted, these insights are best made against a backdrop of dynamic biological and environmental conditions. While it is easy for many to agree that benefits to human well-being correlate positively with ecosystem services, it is often difficult to incorporate these ideas into management (Arkema et al.,
There is a sense of urgency to develop management and policy that supports ecosystem-level sustainability and conservation given the current global demand for living marine resources and marine ecosystem services (Pauly and Palomares,
JT authored/drafted, analyzed, provided final approval, and is accountable for this manuscript. JL and SL provided intellectual content, revised, provided final approval, and is accountable for this manuscript. RS provided intellectual content, revisions, final approval and is accountable for this manuscript. KA, KF, JG, EH, KH, MK, NT, SZ and JS provided data, revisions, final approval, and are accountable for this manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We thank the research vessel crews and scientific staff at NOAA-Fisheries, whose hard work make such studies possible. We thank S. Benjamin at the NEFSC (Social Sciences Branch) for map creation. We also thank S. Lucey (NEFSC) and I. Kaplan (NWFSC) for assisting in data procurement and helpful comments. We also thank internal reviewers S. Gaichas, K. Craig, and K. Osgood for their helpful comments and suggestions. This work was supported by a NOAA Postdoctoral Fellowship to JT and funding from the IEA Program. The findings and conclusions in the paper are those of the authors and do not necessarily represent the views of the National Marine Fisheries Service, NOAA. Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA.
The Supplementary Material for this article can be found online at: