Fig 3 - uploaded by Jeroen Gerhard Steenbeek
Content may be subject to copyright.
Flow chart of Ecosampler batch run mode. Samples are loaded, one at the time, into Ecopath. Ecosampler first alters the base output path, after which it executes the Ecopath, Ecosim, and Ecospace models. Connected plug-ins automatically execute as well. Any component configured to save its output to drive will do so. Ecosampler will then clean up after itself by restoring the output path and restoring the Ecopath parameter set to its initial state.
Source publication
The widely used Ecopath with Ecosim (EwE) food web modelling approach has been extended with a new module to measure the impact of input parameter sensitivity on its results. Ecosampler records samples-alternate mass-balanced parameter sets for a food web model-from the built-in Monte Carlo routine, and replays these samples through all of EwE modu...
Context in source publication
Context 1
... run mode, Ecosampler loads a number of samples, one at the time. For each sample, the default EwE output folder is rerouted to a unique folder, and its Ecopath parameter values are written to that folder. Then, Ecosampler runs the mass-balanced model Ecopath, the time-dynamic model Ecosim (if loaded), and the temporal-spatial model Ecospace (if loaded), restores the initial Ecopath parameter sets, and restores the default EwE output lo- cation. Any EwE core module and any plug-in that is connected to these core modules will run, and, if configured to automatically write outputs, will write their predictions to the rerouted output folder values to drive (Fig. 3). At this point, the uncertainty analysis can then be performed using statistical software of ...
Similar publications
To understand the research status of ecological networks, a high-level analysis of ecological networks research domain is performed using scientometrics methods and visualization tools. We compiled the data of 2 782 articles during the period of 2007–2022 from Web of Science Core Collections (WOSCC) database, and conducted a scientometric analysis...
Citations
... EwE also has a large and collaborative user community with hundreds of models constructed to address an increasing array of issues (Colléter et al., 2013(Colléter et al., , 2015, multiple symposia and syntheses to document and evaluate technical advances and model uses (Christensen & Pauly, 2004;Coll et al., 2015;Villasante et al., 2016), free and easily accessible software with tested applications, modular subroutines, technical support (www.ecopa th.org), and diagnostic and best practices protocols (Ainsworth & Walters, 2015;Heymans et al., 2016;Lassalle et al., 2014;Link, 2010b;Steenbeek et al., 2018). Hence, EwE is well-positioned for operational use in fisheries and in natural resource management in general. ...
... to an array of model diagnostics Lassalle et al., 2014;Link, 2010aLink, , 2010bOlsen et al., 2016;Scott et al., 2016;Steenbeek et al., 2018Steenbeek et al., , 2021. Methods to assess uncertainty are increasingly applied to EwE (Essington, 2007;Gaichas et al., 2012;Guesnet et al., 2015;Whitehouse & Aydin, 2020) and to other ecosystem models (Bauer et al., 2019;Gårdmark et al., 2013;Spence et al., 2018), and approaches for effective decision-making in the face of uncertainty exist (Garrand et al., 2017). ...
The implementation of ecosystem management requires ecosystem modelling within the context of a natural resource management process. Ecopath with Ecosim (EwE) is the most widely used modelling platform for investigating the dynamics of marine ecosystems, but has played a limited role in fisheries management and in multi‐sector resource decision‐making. We review 10 case studies that demonstrate the use of EwE to support operational resource management. EwE models are being used to inform tactical decision‐making in fisheries and other ocean use sectors, as well as to identify key trade‐offs, develop appropriate policy objectives, and reconcile conflicting legislative mandates in a variety of ecosystems. We suggest the following criteria to enhance the use of EwE and other ecosystem models in operational resource management: (1) a clear management objective that can be addressed through modelling; (2) an important trade‐off and a receptive policy context amenable to trade‐off evaluation; (3) an accessible and well‐documented model that follows best practices; (4) early and iterative engagement among scientists, stakeholders, and managers; (5) integration within a collaborative management process; (6) a multi‐model approach; and (7) a rigorous review process. Our review suggests that existing management frameworks are as much or more of a limitation to the operational use of EwE than technical issues related to data availability and model uncertainty. Ecosystem models are increasingly needed to facilitate more effective and transparent decision‐making. We assert that the requisite conditions currently exist for enhanced strategic and tactical use of EwE to support fisheries and natural resource management.
... Thus, it is recommended to confirm the results through further analysis, before the model results are directly applied to policymaking. The EwE plug-in Ecosampler could be used to measure the impact of the input parameter sensitivity on the modelling results e.g. for the use of time-series simulation using Ecosim (Steenbeek et al., 2018). Other options to validate the results, could be stock assessment or bioeconomic and socio-economic analysis. ...
The Gulf of Nicoya is a highly productive estuary located at the Pacific coast of Costa Rica. Previous studies have used trophic models to examine changes in the biomass of key species and in the food web in the last 25 years, revealing an overfished, degraded system with decreasing biomass of valuable target species. The ecosystem degradation was mainly driven by intensive fishing, while climate variations affected resource productivity additionally. This study tested the effectiveness of alternative scenarios including combinations of the current top-down fishing policy and the ban on shrimp trawling, together with a participatory management scenario developed in a previously performed stakeholder workshop. In parallel, the automated fishing policy search of the Ecopath with Ecosim (EwE) software was used to explore an optimized alternative management scenario. The analysis indicates that the ban on trawling is an important measure to allow for the recovery of certain target species, such as shrimps, demersal fish and their predators. However, this ban would not suffice to substantially rebuild the biomass of all key species in the system. Thus, two possible alternative management scenarios are proposed: in the first one, the economic losses are minimized, ecosystem health increases by 10% (by rebuilding target species biomass) and employment provided by fishing decreases (−15%). In the second scenario, higher economic losses are accepted (mainly for the semi-industrial fisheries sector) which allows for a higher increase in ecosystem health and biodiversity. Both scenarios call for additional reductions in fishing efforts, mainly by the semi-industrial purse-seine fleet and the artisanal longline fleet. This study exemplifies how holistic ecosystem models can be used for management advice, future policymaking and how stakeholders can be engaged in this process.
... The Ecosampler module in EwE model is used to analyze the impact of uncertainty of input parameters on model output parameters (Steenbeek et al., 2018). We used the Monte Carlo sampling method built in Ecosampler to set a 20% uncertainty for the B, P/B, Q/B and EE parameters of the functional group, and randomly generate 500 rebalanced XICR Ecopath models. ...
Global change has generated challenges for oceans, from individuals to the entire ecosystem, and has raised contemporary issues related to ocean conservation and management. Specifically, coral reef ecosystems have been exposed to various environmental and human disturbances. In this study, the Ecopath with Ecosim model was used to explore the impacts of ocean warming and fishing on Xisha Islands coral reef ecosystem in the South China Sea. The variables in this model included two ocean warming scenarios and three fishing scenarios. The model consisted of 23 functional groups including algae, coral, sea birds, and sharks. Our results showed that by the middle of the century, ocean warming and fishing led to a 3.79% and 4.74% decrease in total catch compared with 2009, respectively. In addition, the combined effects of ocean warming and fishing caused a 4.79% decrease in total catch, and the mean trophic level of catch was predicted to decrease by 6.01% under the SSP585-High fishing scenario. Reducing the fishing effort mitigates the effects of ocean warming on some species, such as large carnivorous fish and medium carnivorous fish; however, under low fishing effort, some functional groups, such as small carnivorous and omnivorous fish, have low biomass because of higher predation mortality.
... The CV allowed to vary the Ecopath input parameters for B, P/B and Q/B was obtained from the Ecopath Pedigree (Supplementary Material Table S6). For Diets, the Dirichlet distribution method was used, setting a multiplier equal to 30, which was selected after plotting the Dirichlet distributions for different values of the multiplier (between 1 and 100) (for details see Steenbeek et al., 2018). ...
Ecosystem models are important tools for the implementation of Ecosystem Based Fisheries Management (EBFM), especially in highly exploited ecosystems affected by climate change, such as the Mediterranean Sea. Herein, we present the development of an Ecopath ecosystem model for the North Aegean Sea (Eastern Mediterranean) in the early 1990s, as well as the parameterization of the temporal dynamic module (Ecosim) after fitting to catch and relative biomass time series for the period 1993-2020. The Ecosim model included as drivers (i) fishing, assuming a technology creep factor of 2% annual increase (0.79% for bottom trawls), (ii) Sea Surface Temperature, (iii) trophic interactions and (iv) a Primary Production (PP) Anomaly which was positively correlated with the North Atlantic Oscillation of the previous year, highlighting the synergistic effect of environmental and anthropogenic processes during the three-decades hindcast period. Trends in biomasses, catches and ecosystem indicators were characterized by a decline from 1993 to 2010 and a strong recovery thereafter. Sea warming scenarios for the period 2021-2050 indicated contrasting responses to increased temperature among the main commercial groups, while simulations of changes in productivity had relatively straightforward effects. Two scenarios of 10% and 25% reduction in fishing effort revealed quick increases in the biomass of most commercial species, though coupled with lower catches due to reduced fishing, except for few groups that their population increase was remarkably high. Although the 25% effort reduction resulted in high recoveries in the short term, it didn’t necessarily lead to constantly high biomasses resulting in reduced catches towards the end of the forecast period for some groups, in contrast to the 10% reduction. When impacts of reduced productivity were added to temperature effects, the model forecasted lower biomass increases for the winners of sea warming and even higher decreases for the losers. Biomass losses were compensated by a 10% reduction in fishing effort, but this was not overall enough to counterbalance losses in catches. The model developed here contributes to better elucidate observed changes in the past and to hind directions of change in future simulations, as well as to advance EBFM in the area.
... Parameter uncertainty assessments were performed using the Monte Carlo and EcoSampler routines, which allow to assess the temporal uncertainty of model outputs (Steenbeek et al. 2018). For the sensitivity analysis in EwE performed using the Monte Carlo simulation, Monte Carlo alters in steps the input parameters (B, P/B, Q/B, EE) of EwE with a coefficient derived from the data pedigree (i.e., confidence intervals based on data origin (Pauly et al. 2000)) to obtain a best fit temporal model (Steenbeek et al. 2018). ...
... Parameter uncertainty assessments were performed using the Monte Carlo and EcoSampler routines, which allow to assess the temporal uncertainty of model outputs (Steenbeek et al. 2018). For the sensitivity analysis in EwE performed using the Monte Carlo simulation, Monte Carlo alters in steps the input parameters (B, P/B, Q/B, EE) of EwE with a coefficient derived from the data pedigree (i.e., confidence intervals based on data origin (Pauly et al. 2000)) to obtain a best fit temporal model (Steenbeek et al. 2018). The input data across all models was graded, which provided the error index (pedigree) for the parameters from low (1) to high uncertainty (8) ( Table S3, SI). ...
... Monte Carlo is limited by the number of parameters it can alter, and higher model complexity extends the time for the routine to run. The Monte Carlo routine was applied together with Ecosampler on the BSS model, following the steps to record, validate, and analyze samples based on the methods of Steenbeek et al. (2018). The other high latitude ecosystems could not be run independently through the Monte Carlo and EcoSampler routines, and the uncertainty analysis was limited to comparisons of the error index within the data pedigree. ...
Arctic ecosystems are at risk to climate impacts, challenging existing conservation measures such as protected areas. This study aims to describe the ecological dynamics of the Canadian Beaufort Sea Shelf (BSS) ecosystem and the Tarium Niryutait Marine Protected Area (TNMPA) under historical changes in sea surface temperature and sea ice extent. Using Ecopath with Ecosim, we compared the status of the BSS between two time periods, 1970-1974 and 2008-2012, and against four ecosystem models (Eastern Chukchi Sea, Barents Sea, Eastern Bering Sea, Gulf of Alaska) to inform the relative long-term health and status of Arctic marine ecosystems. We find that relative to the comparable ecosystems, the BSS had a greater proportion of biomass from pelagic primary and secondary producers, and limited production from higher trophic levels. Estimates of trophic structure indices for the BSS indicate temporal ecosystem stability, and no loss in diversity. While beluga whales are a focus of the TNMPA management plan, they are not considered a key component of the modeled food web. Rather, Arctic and polar cods (main beluga prey group), arthropods, large copepods, micro-zooplankton, and herring and smelt, were identified as keystone species and warrant attention as proxies for both beluga whales and ecosystem health.
... These models are now considered as promising tools to inform restoration actions, not solely for estuaries (Shephard et al. 2015, Vasslides et al. 2016) but for many ecosystems (Fraser et al. 2015, Kollmann et al. 2016). The models are moreover subject to many statistical improvements, with a wide community of users (Guesnet et al. 2015, Tecchio et al. 2016, Steenbeek et al. 2018. Access to processes related to trophic cascades, food web stability and diversity ecosystem function relationships offers a richer way to consider the functioning of each system in terms of top-down (consumer-control) and bottom-up (resource-control) forces. ...
This chapter presents the causes of physical and ecological degradation of estuaries in relation to human activities and climate change. The direct and indirect effects of degradation on ecosystem services and fish are listed, as well as the key questions that need to be answered in order to undertake rehabilitation and restoration actions. Ecohydrology and ecoengineering are indispensable tools to be mobilised alongside a variety of models to guide actions in favour of habitat and whole estuary functioning. Finally, several examples of restoration are presented to move from theory to practice. The restoration of estuaries has often become essential to ensure sustainable fish communities and requires a holistic view of the problems and a coordination of efforts to ensure the success of the actions undertaken.
... Tools like the Monte Carlo routine, MultiSim, and Ecosampler in the EwE platform could be used to provide a more robust assessment. Ecosampler could capture parameter sensitivity of our Ecosim calibration and Ecotracer parameters in future versions of this MCBs model for management applications (Steenbeek et al., 2018). ...
With increased global production of plastics since the 1950s, marine environments have experienced an increase in plastic pollution. This pollution has the potential to contaminate marine organisms with microplastics, which, in turn, may have deleterious effects on humans that consume seafood. Plastic pollution is often presented as a global issue; however, its sources are often based on local actions and potential health effects occur at an individual level. Environmental management to control this problem also can occur on a local scale. To draw attention to the issue and demonstrate the need to take management actions to reduce plastic inflow, we have developed a proof-of-concept model that connects inflow of plastic in a small-scale marine environment to a contaminants-based food web model. We use Ecotracer in the Ecopath with Ecosim modeling suite to estimate current organism concentrations of microplastics and then use model outputs to calculate human health effects. The model is used to project future microplastic concentrations in marine organisms and human health effects under different environmental plastic inflow rate scenarios. The model is parameterized to simulate the Maryland Coastal Bays ecosystem, which is adjacent to Ocean City, Maryland (USA) a region dependent on the tourism and seafood industries. We consider this a proof-of-concept model, because data for the system are limited. This approach helps to illustrate local consequences of a global problem. In addition, it provides a summary of pertinent regional data on the issues and helps identify gaps for future monitoring and research.
... In the last decade, various methods have been implemented to account for this uncertainty (as well as for other trophic parameters; e.g. Guesnet et al., 2015or Steenbeek et al., 2018 such as Monte-Carlo exploration of a range of possible values for each predator-prey interactions. However, the range of possible values still remains mainly compiled from literature concerning other systems or timescales. ...
... This article is protected by copyright. All rights reserved in of Ecopath (Steenbeek et al., 2018), ESCROpath is clearly one of the first models to be able to combine both food-web flow representation (i.e., biomass, productivity, and consumption rates are included) and diet estimation; and that it is a great opportunity for studying the uncertainty in food-web model outputs coming from diet information. In particular, compared to Ecosampler, ESCROpath is a way of propagating realistic uncertainties, not guesstimates of uncertainty around parameter guesstimates. ...
Food‐web modelling is a key tool to provide a global and comprehensive knowledge on community structure, biodiversity and ecosystem processes and functioning. In particular, it allows computing integrative and holistic indices describing food web characteristics, topology and functioning. However, one of the main sources of uncertainty in most food‐web models is the estimation of diet matrices. In the present work, we propose an innovative approach that combines (a) a Bayesian mixing model using both isotopes and contaminants as chemical tracers with (b) classical mass‐balance equations. This dual approach allows the simultaneous estimation of diet composition, isotopic enrichment, contaminant biomagnification, and contaminants and biomass flows in the whole food web. This original model named ESCROpath also provides food‐web indices derived from ecological network analysis (ENA). As a case study, the approach was implemented in the Gironde estuarine food web (SW France) for which isotopes, contaminants and trophic data exist. Two sets of priors were constructed accounting for more or less uncertainty in trophic parameter estimates. Outputs were compared with previous published Ecopath results. A constrained calibration led to very similar outputs as Ecopath (which shows that the method is able to find the initial set of parameters if it is forced to do so), whereas a free calibration led to slight differences in trophic parameters and ENA indice estimations (which shows that the Ecopath solution was not fully optimal). Quite different diet matrices and estimations of flows distribution within the food web can thus be obtained. ESCROpath is an original flexible food‐web modelling tool that, for the first time, makes it possible to go from a model mainly built on an ‘estimate’ of the parameters based on expert knowledge (which constitutes the main criticism formulated against Ecopath) to a statistical Bayesian framework for the estimation of the trophic parameters. It thus provides a very integrated framework for food‐web modelling by estimating simultaneously trophic parameters, diet compositions and trophic enrichment/magnification factors. By doing this, it notably provides reliable and robust uncertainty estimations for output parameters. La modélisation des réseaux trophiques est un outil essentiel permettant lier la structure des communautés et la biodiversité aux processus au fonctionnement des écosystèmes. En particulier, elle permet de calculer des indices intégratifs et holistiques décrivant les caractéristiques, la topologie et le fonctionnement des réseaux trophiques. Cependant, l'une des principales sources d'incertitude dans la plupart des modèles trophiques réside dans l'estimation des matrices de régimes alimentaires. Dans le présent travail, nous proposons une approche innovante qui combine (1) un modèle de mélange bayésien utilisant à la fois des isotopes et des contaminants comme traceurs chimiques avec (2) des équations classiques de conservation de la masse. Cette double approche permet l'estimation simultanée de la composition du régime alimentaire, de l'enrichissement isotopique, de la bioamplification des contaminants et des flux de contaminants et de biomasse dans l'ensemble du réseau trophique. Ce modèle original, baptisé ESCROpath, permet également de calculer des indices de réseau trophique dérivés de l'analyse des réseaux écologiques (ENA). A titre illustratif, l'approche a été mise en œuvre dans le réseau trophique de l’estuaire de la Gironde (sud‐ouest de la France) et comparée aux résultats d’un modèle Ecopath publié précédemment. Une calibration « contrainte » a conduit à des résultats très similaires à ceux d'Ecopath, tandis qu'une calibration « libre » a conduit à de légères différences dans les estimations des paramètres trophiques et des indices ENA (suggérant que la solution d'Ecopath n'était pas totalement optimale). Des matrices de régimes alimentaires et des estimations de la distribution des flux au sein du réseau trophique assez différentes peuvent ainsi être obtenues. ESCROpath est un outil original et flexible de modélisation du réseau trophique qui fournit un cadre intégré pour la modélisation des réseaux trophiques en estimant simultanément les paramètres trophiques, les compositions des régimes alimentaires et les facteurs d'enrichissement/amplification trophiques. Ce faisant, il fournit notamment des estimations fiables et robustes de l'incertitude pour les paramètres de sortie.
... Indeed, we acknowledge that the lack of a complete biomass time series for more functional groups, as well as the lack of multi-stanza consideration, add to the uncertainty of the model results, which was actually put forward through Monte Carlo simulations ( Supplementary Fig. A5). In fact, all models have an inherent level of uncertainty that stems from the quantity and quality of input data (Steenbeek et al., 2018). Nevertheless, despite these limitations, we maintain that the present ecosystem model of the Thermaikos Gulf provides a useful tool in the data-poor eastern Mediterranean (Dimarchopoulou et al., 2017) and adds to the EwE models of other exploited areas that have been developed in the Aegean Sea (Tsagarakis et al., 2010;Dimarchopoulou et al., 2019b). ...
... Ecosystem models like EwE can make use of time series data to perform and validate projections in the future (that are plausible given the initial model configuration), which can then be helpful in assessing ecosystem status and adjusting management practices to successfully meet future conservation and sustainability targets (Brasier et al., 2019). Obviously though, since models are mathematical abstractions of real complicated systems and therefore hold an inherent level of uncertainty that has to do with the quality and reliability of the input data, they should be treated and analyzed with attention (Steenbeek et al., 2018). ...
Ecosystem modelling constitutes a useful tool for the ecosystem approach to fisheries management, which demands a shift to more complex models that include multi-species trophic interactions, environmental and anthropogenic forcing. The Thermaikos Gulf is a shallow gulf in the northwestern Aegean Sea (Greece) and one of the major fishing grounds of the northeastern Mediterranean concentrating high fishing effort of trawlers and purse-seiners and producing more than 20% of the total Greek catches. In the present work, we developed an Ecopath base model and ran Ecosim simulations for 26 years (2000–2025), including the calibration period (2000–2016), aiming to describe the food web structure and function of the Thermaikos Gulf, identify main components and interactions among the 33 functional groups, assess the ecosystem impacts of fishing over time and compare ecosystem properties with other Mediterranean areas. Overall ecosystem degradation with biomass and catch decline was observed at the end of the calibration period due to the impact of environmental factors and fishing activities. The ecosystem seemed to stabilize in an intermediate state by the end of the projection years, but with an overall biomass and catch decline. Fishing effort reduction after 2016 resulted in higher biomass and catches compared to 2014–2015, that could not however reach the 2000 levels in most cases. The examined fishing effort reduction scenarios clearly showed that the more the fishing effort is reduced, the higher the biomass in the ecosystem and the lower the catches obtained compared to the baseline scenario. In a nutshell, since environmental drivers may be harder to predict or control, lowering the exploitation levels is an important step towards the rebuilding of overfished marine resources and more resilient ecosystems.
... These models are now considered as promising tools to inform restoration actions, not solely for estuaries (Shephard et al. 2015, Vasslides et al. 2016) but for many ecosystems (Fraser et al. 2015, Kollmann et al. 2016). The models are moreover subject to many statistical improvements, with a wide community of users (Guesnet et al. 2015, Tecchio et al. 2016, Steenbeek et al. 2018. Access to processes related to trophic cascades, food web stability and diversity ecosystem function relationships offers a richer way to consider the functioning of each system in terms of top-down (consumer-control) and bottom-up (resource-control) forces. ...
