Assessment tools are needed to support marine ecosystem-based management, but how to get them used practically?

Ecosystem-based management (EBM) is essential to maintain healthy, productive and resilient marine ecosystems while sustainably providing ecosystem services leading to the goods and benefits humans want and need. Ecosystem status assessment is essential to the EBM process and there are many and varied methods (or tools) to undertake that assessment in support of EBM. This paper analyses these tools against the characteristics that make them most suited for practical implementation. A total of 34 tools were identified, including 18 generic and 16 specific tools. Information on the characteristics of the available tools was obtained via a structured online survey that was completed by 45 experts. The survey focused on: (i) the purpose and context of the use of a tool (e.g., the EBM elements it addresses, who uses it or in which context it is applied, and its relevance for marine governance); (ii) the type of assessment that the tool provides (e.g., which components of the accepted cause-consequence-response sequence are involved, what spatial and temporal scales are relevant to the assessment); (iii) the requirements of the tool in terms of data (type and variables), expertise/skills and other resources, and (iv) any strengths and weaknesses, including barriers for practical implementation. Similarities and differences in the expert responses were explored between the tools. Each tool was shown to have a specific combination of characteristics, which may make it more or less suitable for practical use depending on the EBM context and elements to which it is applied (i.e., one-size-fits-all does not apply). The tool suitability is also determined by the user-specific requirements for the assessment and this study provides a valuable means to inform the user and guide their decision on which tool(s) to use in the case-specific implementation of the EBM.

1 Introduction

The sustainable management of marine ecosystems is the goal of most national and international governance instruments (Elliott et al., 2025a; Papadopoulou et al., 2025) as it allows for the supply of goods and benefits which satisfy basic human needs and wants, while ensuring the ecosystem delivers ecosystem services and is maintained in a healthy, clean, productive and resilient condition (Elliott, 2023; Olano-Arbulu et al., 2025). This requires adopting an ecosystem-based approach (EBA) to undertake ecosystem-based management (EBM; Kirkfeldt, 2019), where the ecosystem is considered in all its aspects (rather than focusing on individual components), including all natural assets and processes (including ecosystem services) and their interactions with humans as an integral part of the system (CSWD, 2020; Smith et al., 2022; Elliott, 2023).

The adoption of EBM is gaining momentum in many global commitments, such as the Agenda 2030 for Sustainable Development Goals, or regional ones, such as Regional Seas Conventions (such as the OSPAR and HELCOM). Indeed, EBM can be regarded as operationalising EBA which is the underpinning approach in the European Marine Strategy Framework Directive (MSFD; European Commission, 2008) or the Maritime Spatial Planning Directive (MSPD; European Union, 2014). These policies have different objectives, but usually have some common requirements, including governance instruments, monitoring networks, assessment status methods, targets to be achieved or management measures to be taken when the objectives are not reached (Elliott et al., 2015).

Addressing anthropogenic problems in the environment in essence requires and follows a cause-consequence-response framework (Patrício et al., 2016). To satisfy basic human needs and wants (Drivers), Activities are undertaken which generate Pressures on the environment; these are the mechanisms of change to the natural State of the ecosystem components (structure and functions) and of the ecosystem services they provide, resulting in Impacts on human Welfare (creating effects on the supply of societal goods and benefits), which require appropriate Responses (as management Measures). This framework, known as DAPSI(W)R(M) (pronounced dap-see-worm), following its development in the past two decades, is used here as the unifying framework for integrated marine management (Elliott et al., 2017; Elliott and O’Higgins, 2020). EBM uses the DAPSI(W)R(M) framework to recognise the full array of cause-effect interactions between the components of the marine ecosystem, integrating the natural (environment) and human (socio-economic) domains (Elliott, 2023; Papadopoulou et al., 2025).

The implementation of EBM requires the assessment and evaluation of the ecosystem, covering all of its components and interactions (including human activities and the success of management measures). There are many and varied tools to undertake such assessment in support of EBM; here a tool is defined as the instrument (e.g., a method, model) needed to carry out the assessment, to satisfy the concepts and principles and the outputs to achieve the desired outcome, i.e., EBA (Elliott et al., 2025a, b; Papadopoulou et al., 2025). Although there are still many challenges to effective EBM implementation, there are effective solutions (Haugen et al., 2024), and Papadopoulou et al. (2025) identified 18 specific aspects or issues inherent in an EBM process (EBM elements) and 19 types of tools that may be used to deliver these EBM elements. These encompass a range of methods including, for example, conceptual models, probabilistic models such as Bayesian Belief Networks (BBN), risk-based methods, cumulative impact spatial mapping, biogeochemical models and bioeconomic models. Papadopoulou et al. (2025) concluded that no single tool is likely to fully satisfy all aspects of EBM and hence a complement of tools (i.e., a toolbox) is needed. Based on this, Barnard et al. (in press) developed a decision support system (SEAS4GES: Selection of Ecosystem-based Approaches for Good Environmental Status; Barnard et al., 2025a) to guide the user in identifying the most suitable tool(s) to support EBM under a specific implementation scenario. For this, information on the characteristics of the different tools (in terms of the scope of their products or outputs, together with their operational requirements) was collected from experts. This information provides the evidence base that SEAS4GES uses to rank the tools suitability in relation to the specific requirements, resources and capabilities of the user (Barnard et al., in press).

Considering this background, this paper aims to describe and interrogate the variability and similarities in the characteristics of the different assessment tools in order to ascertain their multiple and variable capabilities in supporting EBM. In particular, the analysis presented here focuses on answering the following questions:

● What are the elements and principles of EBM most frequently delivered by the tools considered, and how frequently does the assessment require tools to be used in combination?

● What are the evidence types required, in general and specifically in relation to the different DAPSI(W)R(M) cause-consequence-response components included in the assessments?

● What are the spatial and temporal scales most commonly addressed?

● What are the skills or other resources most frequently required?

● How are the tools commonly used and by whom?

These questions were answered by collecting data on the relevant characteristics of the assessment tools through an online questionnaire and by analysing them through a frequency analysis, as detailed in the following sections.

2 Materials and methods

2.1 Data collection

Information about the tools was collected in June-July 2023 through an online questionnaire (see Supplementary Table S1 in Supplementary Material for details). This included 22 sets of questions (including main and follow-up questions for a total of 46 individual questions) regarding the capabilities and reasons for using a particular assessment method (also generically referred to as a tool), its requirements in terms of data and other resources (e.g., skills and expertise needed, software), and its assumptions and limitations, strengths and weaknesses. Most of the questions were closed, with pre-defined answer options provided for selection (see Supplementary Table S1). Open text answers were also requested for follow-up questions to encourage the respondent to add specific details, context or clarification. Here below, those parts of the questionnaire that were used to provide the information analysed in the current paper are described in more detail.

The reason for using a tool was ascertained by asking which element(s) of EBM it addresses, either in its current state and/or with further development or adaptation (question Q5 in Supplementary Table S1). A pre-defined list of 20 EBM elements was given, as modified from Papadopoulou et al. (2025) (Table 1; note that EBM18 ‘Other [non-MSFD] policy requirements’ from Papadopoulou et al. (2025) was differentiated into 3 elements according to the relevant policy). A subsequent question identified whether the assessment required by the selected EBM element(s) is fully delivered by the tool alone or whether it requires an in-combination approach together with other tools.

Table 1

Table 1. Ecosystem-based management (EBM) elements (from Papadopoulou et al., 2025) and their identification number (ID#) as used in this study.

To provide context about the tool use, a question (Q8) was asked about the relevance of the approach to marine legislation and administration and, specifically, whether the method is currently being used for (linked to the requirements of some policies): (i) Target setting/development; (ii) Delivery of monitoring programmes; (iii) Tracking progress against action plans; (iv) Remediation/management measures design; (v) Informing decision-making; (vi) MSFD thresholds development; (vii) Indirect use (supporting evidence for policy development), and/or (viii) Other. Information about who is using the tool was also gathered (Q7), with answers categorized as: National government or other sub-national competent authorities; Regional Sea Convention (RSC) assessment groups or for RSC reporting; European Environment Agency (EEA) assessments; Official MSFD reporting; EU Technical Expert Groups (TEG); ICES Working Groups; General Fisheries Commission for the Mediterranean (GFCM) groups; EU projects, and/or Other.

The scope and capabilities of a tool were ascertained by asking what type of information it requires (as input) and produces (as output) (Q9). The information types were defined according to the DAPSI(W)R(M) framework elements (Elliott et al., 2017) as: Activities, Pressures, Species (State change), Habitats (State change), Ecosystem Services (ES; State change), Societal Goods & Benefits (SG&B; Impact on welfare) and Response (management) measures. The question also required the respondent to indicate whether single or multiple variables could be handled by the tool in question for each information type.

Further information on tool capabilities was obtained by asking about the ability of the tool to produce spatial and/or temporal outputs (Q11), and the relevant scale(s) covered (Q12 and Q13). Spatial scales were categorised as: (i) Site, <10 km²; (ii) Local, 10–100 km²; (iii) National, 100-10,000 km²; (iv) Regional, 10,000-100,000 km², and (v) European, >100,000 km². It is emphasized here that any tools suitable at the European level would also be suitable for other large multinational regions. Temporal scales were categorised as: (i) Short term, <1 year (e.g., a season); (ii) Short-medium term, 1 year; (iii) Medium-long-term, 2–10 years, and (iv) Long term, >10 years.

Information on the resources required by the tool was obtained by asking which type of data is required to inform the input variables, with type categorised as: (i) quantitative, (ii) semi-quantitative and (iii) qualitative data, and also whether expert judgment can be used in place of data (Q14a-d). The expertise required to use the tool was also ascertained (Q15), with this being specified as: (i) None; (ii) Statistical; (ii) Analytical; (iii) Modelling; (iv) Programming (e.g., R); (v) GIS; (vi) Ecological knowledge, or (vii) Other. Multiple selections were allowed.

The free-text provided by the respondents to supplement the responses above and about the tool strengths and weaknesses (Q22) were also used to interpret the results.

The respondents were asked to complete separate questionnaires for the individual tools with which they were familiar, with a pre-defined list including the 19 generic assessment methods (or tool groups/types) identified in Papadopoulou et al. (2025) (Q2). Respondents were also required to indicate whether their assessment was based on the characteristics of the generic tool or those of a specific tool implementation (Q3). Their level of familiarity with the tool assessed was always recorded (Q4) as: (i) Expert, i.e., the respondent has used the tool and is proficient with its operation, and/or they have been directly involved in the tool design and/or development; (ii) Very familiar, i.e., the respondent has used the tool, has experience of using it and understands its operation; and (iii) Familiar, i.e., the respondent has used the tool but only ‘at arms’ length’ and they have not been too engaged in the detail.

The online survey was developed using the Jisc Online Survey system (https://beta.jisc.ac.uk/online-surveys) and it was widely circulated via email to over 300 recipients including participants (researchers and practitioners) in multiple ongoing Horizon Europe research projects and also including 70 targeted experts which were known to develop or use the tools (as identified from the preliminary review in Papadopoulou et al., 2025).

The completed questionnaires were initially screened for inconsistencies in the responses of an individual respondent across questions (e.g., highlighting possible misinterpretation). Where these were identified, the relevant respondent was contacted via follow-up email with an ad-hoc request for clarification so that such inconsistencies could be corrected. It is of note that the contact details of the respondents were only collected for this purpose, and that their responses were anonymised for the subsequent analysis.

2.2 Data analysis

Multiple responses received for an individual tool were merged before the analysis. Where there was disagreement between such multiple responses to a certain question, the frequency of the different responses obtained for the tool and the level of expertise of the respondents were considered. The product of these two variables (with level of familiarity scored as 0.75 for ‘expert’, 0.31 for ‘very familiar’, 0.06 for ‘familiar’, according to a rank normalisation procedure; Sureeyatanapas, 2016), summed across multiple respondents, was used as an indicator of the confidence associated with the resulting assessment of each tool. The range of confidence values was divided into three equal quantiles (corresponding to natural breaks in the data distribution) and the values categorised accordingly as Higher, Medium or Lower confidence.

The processed responses for the different tools were combined into a single dataset, which was analysed to identify the distribution of the characteristics across the tools supporting EBM. A frequency analysis was used on the assessment questions of the study.

3 Results

A total of 60 questionnaire returns were obtained from 45 respondents familiar with the tools. The respondents were mostly experts from universities or research institutes (41%) and from governmental agencies and organizations (including their technical advisors) (43%), with also environmental consultancies (SMEs) represented (16%). They were distributed across 17 countries, including 40 respondents from 14 European countries and the remaining from Australia (1 respondent), Canada (1) and United Arab Emirates (2). Most (56%) of the respondents from Europe were from countries in the North Eastern Atlantic Ocean region, followed by the Mediterranean Sea (31%), Baltic and Black Seas (6% each), with a minimum of 3 respondents from each of these regions. One respondent contributed anonymously so their location and affiliation could not be ascertained.

The questionnaire returns provided full coverage of 34 assessment methods, identified as either generic assessment methods (or tool groups) and/or specific tools (implementation examples within a tool group) (Table 2, also reporting the tool reference number (#) as cited in the text). These included the original 19 tools identified in Papadopoulou et al. (2025) and also including an additional method (Size spectrum models) and specific tools, as identified by the respondents. The detailed characteristics of the tools were summarised as factsheets in Barnard et al. (2025b), where the relevant illustrative references are also given for each tool.

Table 2

Table 2. Tools assessed (including generic tool groups, identified by integer numbers, and specific tools within) and number of questionnaire returns received by familiarity level.

Almost all tools were assessed as a generic method (or tool group), except for Semi-quantitative mental models and Food web models, for which the responses only regarded specific tools (namely Fuzzy Cognitive Modelling developed in Mental Modeler, and Ecopath with Ecosim (EwE), i.e., tools #2.1 and #10.1 in Table 2, respectively).

The maximum number of responses for a single tool (7) were received for Cumulative impact spatial mapping (#6), all with high level of familiarity, thus identifying this method as being in the highest confidence class as regards its initial assessment (Table 2). Most of the other tools (27 out of 34, 79%) received either 1 or 2 responses, but most frequently by respondents with high level of knowledge of the tools (experts or very familiar), thus increasing the confidence on their assessment despite the low number of responses. Only Specific biogeochemical models (#9.1) and Generic single species models (#8) were assessed by respondents which were, at best, only familiar with the tool, therefore being in the lowest confidence class (Table 2).

3.1 Tool ability to address EBM elements

The EBM elements most frequently addressed by the tools (28 out of 34 tools, 82%) included the assessment of activities-pressures footprints (EBM7), of single MSFD Descriptors or single issues (EBM10; e.g., eutrophication, Non-Indigenous Species, Harmful Algal Blooms), of the state change in single species or ecosystem components (EBM11), and of threatened habitats and species (EBM12), also in the context of complying with requirements of the EU Birds and Habitats Directives (EBM18.2) (Figure 1). In turn, whole ecosystem assessments (EBM3), special biotic effects/impacts (EBM5), climate change (EBM13) and uncertainty (EBM16) were addressed by fewer tools (21–22 out of 34).

Figure 1

Figure 1. Frequency with which the Ecosystem-Based Management (EBM) elements are addressed by the tools assessed, currently (block colour) and potentially (i.e., after further development and adaptation; patterned colours). See Tables 1 and 2 for the reference numbers for tools and EBM elements, respectively.

Bow-tie analysis (a specific tool for the Risk-based methods assessing exposure-effect-hazard-vulnerability, #5.1), tools of the MARXAN family (as part of Systematic conservation planning, #16.1) and Ecosystem models (#11) were also identified as potentially being able to address the full set EBM elements, from the 19 out of 20 EBM elements they currently address. Their improvement would allow uncertainty (EBM16) to be also addressed by tool #5.1, risks (EBM17) by tool #16.1 and requirements of the MSPD (EBM18.1) by tool #11 (Figure 1). Further improvement was also required for the PlanWise4Blue tool for Cumulative impact spatial mapping (#6.1) to be able to address all EBM elements (from the 16 currently addressed) and in particular to include the assessment of whole ecosystem structure and functioning aspects (EBM3), climate change (EBM13), and to address policy requirements of the EU Birds and Habitat Directives (EBM18.2) and Biodiversity Strategy (EBM18.3).

Of the other tools, only the MAMBO conceptual model (Environmental Matrix for the Management of Blooms, #1.1), BBN probabilistic models (#4) and Size spectrum models (#20) were not considered to be able to deliver any of the EBM elements in their current status. However, this indicates potential if there was further tool development, e.g., for all three tools to address GES MSFD assessments (EBM2) and Links activities pressures impacts (EBM9). This possibly reflects the fact that these models provide insights into the theoretical background of deviations or likely causes but do not provide assessment answers (e.g., MAMBO) or they need to be tailored to specific questions and analysis (e.g., BBN) and this would affect which EBM elements the tool may potentially address.

3.2 DAPSI(W)R(M) components and data requirements

Overall, all the DAPSI(W)R(M) cause-consequence-response components were considered in the assessments undertaken by the tools, either as inputs and/or outputs and most often represented by multiple variables in a tool (Figure 2). Activities, pressures and their effect on the state of ecosystem components (species and habitats) are most often used as input by the tools, while the consequent effects on ecosystem services, societal goods and benefits and responses as management measures are most frequently provided as outputs (Figure 2). In addition, some tools (e.g., specific applications of Knowledge Graphs, #3.1; Natural capital accounting tools #13 and #13.1; Bioeconomic/socioeconomic models, #14) may also require evidence on the Drivers (human needs and wants) that dictate what activities are required.

Figure 2

Figure 2. DAPSI(W)R(M) components used by the assessed tools as input (A) or output (B), as single or multiple variables (ES, ecosystem services; SG&B, societal goods & benefits). The % represents the overall relative frequency of use of a DAPSI(W)R(M) component across all 34 tools.

Only 38% of the assessed tools (13 out of the 34) were indicated to individually use input data for all of the DAPSI(W)R(M) components, but this percentage increases to 76% (26 tools) when input data for only activities, pressures and their effect on the state of ecosystem components are considered (see Supplementary Table S2 in Supplementary Material for details). In general, Conceptual models (tool #1) can use input on any DAPSI(W)R(M) component, although this may vary with the specific application (e.g., MAMBO (#1.1) only uses input data on activities and species state). Other tools that can use all DAPSI(W)R(M) components as input include Fuzzy cognitive models (#2.1), Knowledge graphs (#3 and #3.1), Risk-based methods assessing exposure-effect-hazard-vulnerability (generic #5 and Bow-tie analysis #5.1), Food web models (EwE #10.1), and Natural capital accounting and ecosystem services valuation (generic #13). Cumulative impact spatial mapping, Impact risk ranking through linkage-chain-frameworks and Overarching assessment tools also include all DAPSI(W)R(M) components as input, in general (tools #6, #7 and #18, respectively) and for some but not all of their specific applications (PlanWise4Blue #6.2 and Aquacross #7.3). For example, CIMPAL (cumulative impact of invasive alien species, #6.1) does not use input data on ecosystem services, societal goods & benefits and response measures but it does provide outputs on response measures; the ICES/Mission Atlantic variation tool (#7.4) does not use input data on ecosystem services and societal goods & benefits, with also SCAIRM (#7.1.) and NEAT (Nested Environmental status Assessment Tool, a specific application of Overarching assessment tools, #19.1) not using data on societal goods & benefits as input. In turn, tools such as generic Bio-/socio-economic and societal goods and benefits valuation (#14) only require input data for societal goods and benefits, while Biogeochemical models (#9 and #9.1) only require this for pressures and the effects on the state of ecosystem components (as species or species groups).

When considering the specific data requirements for the different DAPSI(W)R(M) components (see Supplementary Table S3 in Supplementary Material for details), only Biogeochemical models (#9 and #9.1), specific applications of Natural capital accounting & ecosystem services valuation (Ocean Accounts #13.1) and Size spectrum models (#20) were reported to have strict requirements for quantitative data. For most of the other tools, quantitative data may still be preferred to ensure a higher detail/quality of the resulting outputs, but they are not essential, and, where not available, semi-quantitative or qualitative data may be used instead to characterise the different DAPSI(W)R(M) components; for example, all quantitative inputs in Cumulative impact spatial mapping (#6) may be replaced by the use of indices or scores on a Likert scaling. Many tools, with few exceptions, are also able to use expert judgment to replace (or partially replace) data including for example Single species models (#8) and Biogeochemical models (#9 and #9.1), Species distribution models (#12), Simple assessment index (#17), MSFD Descriptor- or theme-specific combination of indices and models (#18), Overarching assessment tools (#19 and 19.1) and Size spectrum models (#20).

3.3 Spatial and temporal scales

In total, 24 out of 34 tools (71%) have the potential to perform spatial assessments, providing spatially resolved outputs but consequently also requiring spatially-explicit data as an input (e.g., the spatial distribution of the ecological features of interest, environment data, and pressures). These include, for example, Species distribution models (#12), Spatial planning models (#15), Systematic conservation planning tools (#16, including MARXAN), Cumulative impact spatial mapping tools (#6, #6.1 and #6.2), Food web models (#10.1, when EwE is implemented with its spatial-temporal module Ecospace), and Marine ecosystem models in general (#11). Overarching assessment tools (#19) may also perform spatial assessments in general terms, but their data requirements depend on the specific tool application (e.g., Ocean Health Index (OHI) does require spatial input, while NEAT needs data that are referred to a spatial assessment unit for which the tool is applied but it does not need these data to be spatially resolved within the area).

Approximately 65% of the tools (22 out of 34) also provide temporal assessments hence also requiring the input of time-series data. For example, the output of Ecosystem (end-to-end) models (#11) is a 3D spatially-resolved time-series for each model variable. Food web models (#10.1, where Ecopath is used with Ecosim) explicitly use time-series as inputs (e.g., catch rates, biomass of components, etc.) and can then output modelled trends. NEAT (#19.1) requires temporal data to produce indicators with their associated standard errors, and, although they do not produce time-series as outputs, the combination of multiple outputs (referred to different specific time periods, e.g., years) may be used for temporal assessment. The same is valid for Species distribution models (#12). Other tools which may provide temporal assessments include Size spectrum models (#20) and Single species models (#8; the tool may use spatially explicit information, but the assessment is not spatial and only temporal outputs are provided).

Other tools do not require or produce spatially or temporally resolved inputs/outputs; these include, for example, Conceptual models (#1), Fuzzy cognitive models (#2.1), Knowledge graphs (#3, #3.1), BBN probabilistic models (#4), Risk-based methods such as Bow-tie analysis (#5.1). However, although not a requirement, some tools may accommodate spatial or temporal input data (e.g., #4, #5.1, Impact risk ranking through linkage-chain-frameworks #7). For example, Impact risk ranking through linkage-chain-frameworks such as ODEMM (#7.2) may consider some spatial or temporal distributions of data to inform the reasoning behind the assessment. However, the assessment is for the whole area and period as a unit, although the expert indicated that the tool is being developed to address quantitative spatial and temporal aspects.

Where spatial assessments are undertaken, most of the tools are able to assess at multiple scales, with those most frequently represented being between site and regional scale (up to 100,000 km²), whereas assessments at the larger European/continental scales are less frequent (Figure 3). Generic risk-based methods (#5), specific Biogeochemical models (#9.1) and Ecosystem models (#11) are the only tools for which assessments were reported to focus only on a few of the smaller spatial scales (#5: site, local and national; #9.1: local and regional; #11: local and national).

Figure 3

Figure 3. Frequency of spatial scales of the assessments undertaken by the tools. Whether the individual scale is assessed on its own by a tool, or in combination with 1, 2 or more other scales is also shown. Spatial scales: Site level (e.g., < 10 km²), Local (e.g., 10–100 km²), National (e.g., 100-10,000 km²), Regional (e.g., 10,000-100,000 km²), European (e.g., > 100,000 km²).

With regard to the temporal dimension, most of the tools are able to assess at multiple scales; the ones most frequently represented being short-medium and medium-long term (i.e., between one individual year and multiple years within a decade), whereas assessment on short and long terms is less frequent (Figure 4). The only tools for which assessments were reported to focus only on a few of the temporal scales were the generic Simple assessment indices (#7: short to short-medium term), specific Biogeochemical models (#9.1) and MSFD Descriptor- or theme-specific combination of indices and models (#18) (both: short-medium to medium-long term) and Single species models (#8) and Size spectrum models (#20) (both: medium-long to long term).

Figure 4

Figure 4. Frequency of temporal scales of the assessments undertaken by the tools. Whether the individual scale is assessed on its own by a tool, or in combination with 1, 2 or more other scales is also shown. Temporal scales: Short-term (e.g., seasonal), Short/medium-term (e.g., 1 year), Medium/long-term (e.g., 2–10 years), Long-term (e.g., > 10 years).

3.4 Resources required

Most of the tools (> 90%) require some ecological knowledge to be implemented (Figure 5). Only PlanWise4Blue (a specific tool for Cumulative impact spatial mapping; #6.2), GIS spatial planning models (#15.1) and Single species models (#8) do not have this specific requirement according to the respondents (although the latter tool assessment has lower confidence; Table 2). Analytical, statistical and modelling expertise, as well as programming and GIS skills, are also a common requirement (in 40-62% of the tools). The need for additional knowledge of the social, economic and governance system, often associated with expertise in stakeholder engagement (e.g., through running workshops) was also indicated for a few tools (12-18%; Figure 5), such as, for example, Fuzzy cognitive modelling (#2.1) or Impact risk ranking through linkage-chain-frameworks (e.g., the ICES/Mission Atlantic variation, #7.4). PlanWise4Blue (#6.2) and SCAIRM (a specific Impact risk ranking tool through linkage-chain-frameworks, #7.1) were the only tools for which no specific expertise requirement was indicated. However, it was highlighted that this result for the latter tool is only valid for its most basic (qualitative/semi-quantitative) application, as some expertise with GIS and/or R programming and some ecological knowledge to quantify effects is needed where quantitative data are used in SCAIRM.

Figure 5

Figure 5. Frequency of types of expertise required in using the assessed tools.

Most (70%) of the tools use publicly available software, although purchase and/or licence costs may be required (e.g., MS Office, some GIS software, Bow-tie XP proprietary software). Some of this software is available through tool-specific applications (Mental Modeller, EwE, MARXAN, ZONATION, M-AMBI), but several tools (such as Bow-tie) may also be implemented using common (non tool-specific) software such as MS Office (Excel, Access, PowerPoint, etc.), R or Python-based codes, or GIS-based software (e.g., ArcGIS, QGIS). This is the case of Conceptual models (#1 and MAMBO #1.1), Fuzzy cognitive mapping (#2.1), Risk-based methods (#5 and Bow-tie #5.1), Cumulative impact spatial mapping (#6 and PlanWise4Blue #6.2), Impact risk ranking through linkage-chain-frameworks Aquacross (#7.3) and ICES/Mission Atlantic variation (#7.4) and some Simple assessment indices (#17). However, in some cases, the use of specialist and proprietary software may make the tool implementation easier (e.g., using Mental Modeller for Conceptual and fuzzy cognitive mapping (#1, #2.1)) or be required for specific implementations of the tool (e.g., M-AMBI Simple assessment index in #17). Only 3 tools require software that was reported as not yet publicly available because of its specificity (the proprietary Knowledge graph system custom-built for the DAPSI(W)R(M) for the Abu Dhabi Environment Agency, #3.1), or because of the tool being still under development (SCAIRM (#7.1), running on R code and associated databases; Ocean accounts (#13.1) for which the ARtificial Intelligence for Environment & Sustainability (ARIES) platform is still being developed for application to the marine environment). Some tools may also require specialised equipment, e.g., Biogeochemical modelling (#9), which may require high-performance computers where large-scale 3D models are applied (but for 1D modelling, a PC-based approach is possible).

In general, all tools are expected to succeed when in data- and skills-rich situations. Some tools are particularly data-/skills-hungry and are considered unsuitable for implementation in data- and/or skills-poor situations. For example, the use of Food web models (#10.1), Species distribution models (#12) and Natural capital accounting & ecosystem services valuation (#13) seem to be limited, especially by data availability (i.e., they are poorly suited to data-poor situations even when skills are available). In turn, skills availability appears to be a limiting factor, e.g., for Systematic conservation planning tools (#16 and #16.1), Ecosystem models (#11) and BBN probabilistic models (#4), which all require, as a minimum, ecological knowledge and programming skills, with the addition of statistical, analytical or modelling skills (and GIS expertise for #4, #16 and #16.1). Some tools are well suited for implementation even in data- and skills-poor situations, even though this may lead to higher uncertainty and lower confidence associated with the outputs. These include Conceptual models (#1), Fuzzy cognitive mapping (#2.1), Risk-based methods assessing exposure-effect-hazard-vulnerability (#5), and Impact risk ranking through linkage-chain-frameworks (#7, #7.1, #7.3 and #7.4).

3.5 Use of the tools

The assessed tools appear to be most frequently used within national governments (for national reporting, e.g., of Fisheries Management Plans, and relevant competent authorities) and for EU project work, followed by the use within Regional Sea Conventions, ICES Working Groups and EU Technical Expert Groups (Figure 6). An individual tool is most often employed in a combination of these different uses. The use for official MSFD reporting, often associated with EEA assessments, is limited to fewer tools, namely Cumulative impact spatial mapping (generic (#6) and as the specific tool PlanWise4Blue (#6.2)), Single species models (#8), Species distribution models (#12), Spatial planning models (#15 and #15.1), Simple assessment indices (#17), MSFD Descriptor- or theme-specific combination of indices and models (e.g., HEAT, BEAT, CHASE; #18) and Overarching assessment tools (e.g., NEAT, OHI; #19).

Figure 6

Figure 6. Frequency of uses for the assessed tools. Blue lines at the bottom indicate the most frequent (from top to bottom) combinations of uses for an individual tool.

Almost all tools have been reported as relevant to marine governance and management, being most frequently used to inform decision-making and to track progress (e.g., against action plans), followed by the design of remediation and management measures (in some cases being specifically used to analyse, evaluate and implement the most appropriate measures, as for risk-based methods such as Bow-tie (#5.1)) (Figure 7). Less frequent roles are in the design of monitoring programmes, the development or setting of targets or MSFD thresholds, and policy development (including providing supporting evidence). Additional roles were identified in assessing the effectiveness of Programmes of Measures (PoM), as required by the EU MSFD implementation, the development of pressure indicators for Cumulative impact spatial mapping (#6), assessing the achievement of overarching policy goals (in particular in the Common Fisheries Policy and/or as elements of the MSFD Descriptor 3 (commercial species) assessment for criteria D3C1 (fishing mortality rates) and D3C2 (spawning stock biomass)) for Single species models (#8). Only BBN probabilistic models (#4) were indicated as not relevant for implementing marine governance instruments. However, it might be argued that, where developed to address environmental management problems, the probabilistic representation of correlative and causal relationships that BBNs provide is likely to constitute valuable supporting evidence for policy development and decision-making (even if this was not the direct experience of the expert who provided the response for this tool).

Figure 7

Figure 7. Frequency of roles of the assessed tools in marine governance. Blue lines at the bottom indicate the most frequent (from top to bottom) combinations of roles for an individual tool.

4 Discussion

Operationalising the Ecosystem-based Approach (EBA), the overarching policy, by implementing Ecosystem-based Management (EBM) requires a set of tools and approaches, recently termed Ecosystem-Based Technical Measures (EBMT) (Elliott et al., 2025b). Hence it is important to know not only what are those tools but what are the pros and cons of their use.

As shown here, a variety of assessment methods are available to support the multiple assessment requirements throughout the EBM process (O’Higgins et al., 2020; Chust et al., 2022; Papadopoulou et al., 2025). In this study, the tools outlined in Papadopoulou et al. (2025) were considered and their overall characteristics explored from expert input.

4.1 Expert input

It is acknowledged that the results of this study reflect the views of the experts consulted, with only a few experts (up to 7, but 1 or 2 in most cases) contributing to the assessment of each individual tool. However, their knowledge of, and familiarity with, the tools was high in most cases (often also including researchers involved in the design or development of the tool or who use it routinely), thus increasing the confidence in their assessment and the results obtained. Notably, only two tools, namely Single species models (#8) and specific Biogeochemical models (#9.1), received input from researchers with a moderate level of familiarity and thus lower confidence, and therefore the results for these tools must be treated with caution. It is also highlighted that the assessment in this study is based on the characteristics of the tools as known at the time of data collection (summarised as factsheets in Barnard et al., 2025b), but that some tools are in continuing development (e.g., Piet et al. (2023) and Tamis et al. (2024) for SCAIRM (#7.1); Sagarminaga et al. (2023) for MAMBO (#1.1); Chiappi et al. (2025) for CIMPAL (#6.1)).

4.2 Who uses the tools

A primary concern in interrogating the available tools was to discern which are not merely academic exercises but that can be used by policy implementers or at least their specialist advisors. Borja et al. (2025) concludes that while policy implementers and policy developers will be charged with employing systems for achieving the Ecosystem-based Approach and enacting Ecosystem-based Management, they would not personally use the tools described here. More likely they would employ advisors who may even be an intermediary between the tool developers and the practitioners thereby linking EBA to EBM to Ecosystem-based Technical Measures (Elliott et al., 2025b). Hence, as starting point, in interrogating the tools here, it is necessary to be clear who would be their user. Of course, many are developed by scientists and researchers for scientists and researchers, but to be valuable in marine management and governance, they need to be suitable for use by policy advisors, who may also have a science background (Borja et al., 2025). It is therefore important to acknowledge the degree of skills and expertise required in implementing the tool. Despite this, policy implementers and developers - as well as their political masters - acknowledge that EBM underpins all existing policies (EBA) as enacted by statutory instruments as well as the functioning of environmental administrative bodies (e.g., Elliott et al., 2025a, b).

4.3 Tool capabilities

4.3.1 EBM elements addressed

The first step in undertaking the assessment needed to support EBM in a specific implementation case is to identify which specific tasks or objectives are to be delivered by the assessment, namely which EBM elements are to be addressed (as defined in Papadopoulou et al., 2025). This includes, for example, whether: cumulative effects are to be assessed; MSFD GES assessment is to be undertaken; the footprint of pressure, activities or impacts is to be determined; the cause-effect links between these components are to be analysed; the measures to reduce pressures or mitigate impacts are to be identified, or the effects of climate change are considered. For increased efficiency (and economy of effort), tools that are able to address multiple EBM elements are likely to be most valuable and to give the most comprehensive assessment. For example, Generic conceptual models, Risk-based methods assessing exposure-effect-hazard-vulnerability, Cumulative impact spatial mapping, Impact risk ranking through linkage-chain-frameworks, Food web models, and Overarching assessment tools (e.g., NEAT) appear to be able to address all the elements of EBM considered. These tools exemplify the central essence of EBM – to consider all aspects, holistic solutions, links between the natural and societal aspects, and indicating the science-policy links. Indeed, some of these tools were designed from specific wide-ranging policies such as the NEAT tool and EU MSFD implementation. It is emphasized that underpinning policy implementation with good science should start with defining the testable question to be addressed and summarizing the situation with a conceptual model. The latter not only will indicate the important elements to assess but also indicate which aspects are unnecessary, thereby creating a cost-effective assessment.

As also seen in Papadopoulou et al. (2025), the fact that a tool may address an EBM element does not necessarily mean that it is able to fully deliver on that EBM element. In fact, a tool might provide outputs that are only relevant to specific aspects or applications of the EBM element, or that need to be used in combination with other tools (or their outputs) to fully deliver the EBM element and satisfy the EBM principle(s). It is also emphasized that while an output from a tool is important, it is the outcome of the use of that tool that is paramount; in this way the overall vision (i.e., to achieve EBA) is fulfilled (Elliott et al., 2025a). Several examples are present in the literature where multiple tools are combined in support of EBM, including for example integrating food web models with biogeochemical models (Lynam et al., 2016) or with Bayesian network models (Uusitalo et al., 2022) to support GES assessment and the implementation of the MSFD (Borja et al., 2024). In particular, nested models, and especially end-to-end models, may cover several or many elements thereby giving a greater holistic approach (Peck et al., 2018).

4.3.2 DAPSI(W)R(M) components covered and associated data requirements for social-ecological systems

In considering the desires of society and the way they are fulfilled, the resulting pressures and effects on the natural and human systems, and the management measures addressing those problems, the DAPSI(W)R(M) framework summarises the essence of the EBA to EBM to EBTM continuum; as such the findings here show the importance of tools necessary for the cause-consequence-response approach (Elliott et al., 2025b). Hence, the tools considered here incorporate multiple components of the framework (and their inter-relationships) into the assessment, thus determining the data requirements for the implementation of such tools. Different tools may incorporate different combinations of DAPSI(W)R(M) components, and therefore, the specific data requirements will depend on the tool being applied. The ability of an assessment method to incorporate multiple DAPSI(W)R(M) components is essential considering the importance of using an integrated, multi-sectoral approach to EBM as highlighted by Marshak et al. (2016). Including all such elements then reflects the marine ecosystem as a combined natural and human system and thus requires a social-ecological systems analysis, again reinforcing the central underlying principles of EBM (Smith et al., 2025). As such, whichever tool is used, it is likely that its practical implementation will require the collection of data on multiple activities and/or pressures in the case study area and on the state of the ecosystem components.

As a minimum data requirement for the tools assessed, and as the primary cause of any non-achievement of EBM, activities and pressures need to be categorized and interrogated based on their occurrence or, where a spatial assessment is undertaken, their distribution in the study area. However, such features may need to be quantified in terms of their frequency or intensity (e.g., the concentration of chemicals for contamination pressure, the concentration of nutrients leading to eutrophication, catch or by-catch mortality pressure for certain biota) (Elliott et al., 2020). The latter may also be expressed semi-quantitatively (e.g., as score on a Likert scale), for example when using risk-based methods on exposure-effect-hazard-vulnerability, impact risk ranking through linkage-chain-frameworks or cumulative impact spatial mapping (such as the Bow-tie method or SCAIRM). Input data on activity or pressure intensity and distribution may also be used by some tools (e.g., EwE food web models; Corrales et al., 2017) to generate and assess the implications of management responses to future or plausible scenarios, such as exploring spatio-temporal changes in fishing pressure (as mortality to key stocks) in order to predict changes in the food web. Alternatively, they can explore the sensitivity of changes in pressures over space and time with cumulative impact spatial mapping (Clarke Murray et al., 2015).

As for the characterization of state (and changes to it) of ecosystem components, data on individual species or species groups (e.g., identified by taxonomy, function, size or a combination of these) and on habitats are commonly required as input to the assessment tools. Some tools may require input data for one or few specific species or species groups (e.g., MAMBO, single species models, simple assessment indices focusing on a particular eco-component as M-AMBI (macrobenthos); Borja et al., 2019; Sagarminaga et al., 2023). Others may be able to integrate data from multiple ecological components for a better representation of the ecosystem (e.g., food web models, ecosystem models, cumulative impact spatial mapping) (Borja et al., 2016).

It is axiomatic that while monitoring is not a management tool, it is needed to indicate whether management is required or indeed has achieved its aims. Hence the number of tools that rely on monitoring data and the number of questions asked here regarding those data. The analysis here has shown that several tools accommodate data on ecosystem components in a qualitative (species or habitat occurrence in an area, spatial distribution) or semi-quantitative way (e.g., abundance score). However, quantitative data (e.g., counts/abundance, density, sightings, biomass) are required to provide more robust assessments with increased confidence. Several tools indicate that many of the basic data required for ecosystem components concern the distribution and/or quantity (abundance or extent) of these natural assets. These data are commonly obtained from standard monitoring programmes, which are usually required in most countries and locations, albeit with variable coverage in space/time or ecosystem components (e.g., Katsanevakis et al. (2023) highlighted the lack of specific/standardised monitoring programmes for marine invasive alien species in most countries).

In addition, those tools involving models are required where users have been charged with giving either mathematical descriptions and/or predictions of the EBM processes and adequate and appropriate data are required to populate those models, as indicated by the questionnaire here. Those models often use the above data but some models may require further characterisation of the ecosystem components, e.g., in terms of size, age-distribution, growth, and maturity (e.g., food web models, single species models, some simple assessment indices). In this way, the tools are not only considering the structural ecosystem aspects but also the functional aspects, again relating to the questions being asked of the policy advisors. Hence, the models may require the parametrisation of the relationships between ecosystem components (e.g., species diet matrices in food web models) or with other DAPSI(W)R(M) components. For example, this includes the links from pressures to state changes and impacts on human welfare such as dose-effect relations, specific recovery durations, and the sensitivity to pressures. It also emphasizes the major questions being asked of policy makers such as whether structural biodiversity is as it should be (e.g., Descriptor 1 for biodiversity in the MSFD) or the system should have an appropriate trophic functioning (e.g., Descriptor 4 on food webs for the MSFD). In most cases, these aspects may be characterised in a semi-quantitative way (e.g., using species-habitat specific impact weights), as in cumulative impact spatial mapping (e.g., CIMPAL, Katsanevakis et al., 2016) or in impact risk ranking methods (e.g., ODEMM, SCAIRM; Piet et al., 2023). Such evidence may be derived from literature review or expert judgement, although assumptions may be made that may not accurately represent reality (e.g., additive impacts and linear response model in cumulative impact assessments). Through the use of the SCAIRM tool (for impact risk ranking), linkages with the ecosystem services delivery are investigated, i.e., how multiple pressures and impact risk affect the capacity of ecosystem services delivery (Piet et al., 2024).

Policy implementers and their advisors are increasingly required to apply EBM holistically, i.e., by including socio-economic aspects as well as ecological ones and, as shown here, require tools to complete this. As such, the assessment tools in EBM are increasingly required to provide end-to-end assessments, i.e., from the natural ecosystem structure and functioning to the production of societal goods and benefits (SG&B) supplied by the ecosystem (Peck et al., 2018; Elliott, 2023). Data for the latter are less frequently required as inputs by many of the assessment tools considered here which may have an over-emphasis on the natural system. The societal information may be included as qualitative information on the occurrence (delivery) of the SG&B in the studied system, such as where tools are used to map a problem (conceptual models, knowledge graphs, fuzzy cognitive modelling). Their quantification in monetary terms (e.g., market prices, exchange values) may also be required, especially for socio-economic valuation in natural capital accounting, goods and benefits valuation, bio- and socio-economic models (Vallecillo et al., 2022). However, these tools need further development as some societal benefits are not directly amenable to monetary valuation (e.g., because they are not directly associated with an economic activity, for example cultural benefits), and semi-quantitative or qualitative values may be used to characterise them (e.g., well-being indices, welfare values assigned to SG&B based on stated preference methods; Potschin et al., 2015).

As shown here, especially for mathematical descriptions and predictive modelling, some tools require time-series for the data mentioned above to allow the hindcasting and forecasting of the effects of activities/pressures on the ecological components (e.g., food web models, BBN models). These will allow integrated assessments over a period of time while accounting for uncertainty (e.g., as standard deviation around mean values over the study period, as in overarching assessment tools as NEAT), or to allow the full assessment of natural capital accounts (to be opened and closed over a defined time period). This shows that the directions of the development of natural and societal tools are coming together as required by implementing EBM to achieve EBA. Despite this, the flexibility in data requirements mentioned above (i.e., the ability to accommodate qualitative or semi-quantitative data where quantitative data are not available) is an advantage as it widens the applicability of the method to cases where data availability may be poor. However, the user should be aware that using lower quality data may lead to less informative outputs (i.e., less detailed in the type of information they provide) which may be less trustworthy (i.e., with higher uncertainty or lower confidence).

4.3.3 Tool capabilities for the assessment of cumulative effects

While it is acknowledged that describing and predicting the effects of single activities and their single pressures is now straightforward, perhaps a major challenge in marine management is to tackle cumulative impacts, accommodating and quantifying the pressures created by several activities occurring at one place and time and the pressures across sea basins (Elliott et al., 2020; Piet et al., 2024). As managing cumulative effects is a central principle of EBM, as shown here and in Long et al. (2015) and Papadopoulou et al. (2025), then tools addressing this aspect are becoming increasingly important. As shown here, while there are some tools (e.g., SCAIRM) which are designed for cumulative effects, users may require a combination of tools. As such, the significance of the suitability of the tools in supporting EBM (through addressing one or more of its elements) relies on the understanding of the combination of the characteristics of the tools, rather than considering them separately (e.g., the EBM elements addressed, the data requirements, the spatial-temporal scales of the assessment). Given this, it is necessary to consider the combined characteristics of the tools in assessing cumulative effects.

Given its name, as expected, Cumulative impact spatial mapping, a tool designed specifically to assess cumulative effects/impacts, is the best fit to deliver this element of EBM, as also assessed by Papadopoulou et al. (2025). This tool spatially assesses the potential effects of combined activities and pressures on ecological components (receptors). The method developed by Halpern et al. (2008), to consider co-occurring activities, has been widely applied in its original form or with further developments and expanded into the Ocean Health Index although this may reflect the presence of activities rather than adverse effects per se (Borja et al., 2024). In general, Cumulative impact spatial mapping is sufficiently flexible to accommodate evidence on any and multiple activities, pressures and receptors. Although the tool may use evidence of variable quality (e.g., all quantitative inputs may be replaced by the use of indices or scores on a Likert scale, or by presence/absence records), the most robust outputs are obtained when supported by quantitative data (e.g., from monitoring the distribution and intensity of pressures/activities or presence/abundance of receptors), thus increasing its ability to confidently assess cumulative effects/impacts. It is a spatial tool which can operate at a range of spatial scales, from site level (< 10 km²) up to broader European scale (> 100,000 km²) (e.g., Fernandes et al., 2017; Menegon et al., 2018; Andersen et al., 2020; Hammar et al., 2020; Vaher et al., 2022). It does not require temporal data, although the experts who assessed this tool indicated that it is being extended into temporal outputs, from a short-term (seasonal variability within a year) to a long-term scale (> 10 years). In addition, specific applications of this type of tool are sufficiently flexible to use expert judgment where data are unavailable. However, they deliver the EBM element only partially by focusing on a specific issue, and do not account for ecosystem services (e.g., CIMPAL, Katsanevakis et al., 2016; although CIMPAL-ES is currently being developed to include ecosystem services; Dr S. Katsanevakis, University of the Aegean, pers. comm.).

Despite its highest-ranked suitability to assess cumulative impacts/effects (Papadopoulou et al., 2025), Cumulative impact spatial mapping has limitations in its capabilities. For example, the tool narrative here suggests that it does not allow to determine the status and distance-to-target or confidence, nor to identify tipping points or thresholds of pressures resulting from multiple activities (although it can map and display these perhaps inferring state or distance-to-target). It still needs to be used in combination with other tools to provide spatial information and incorporate a risk-based method (e.g., GIS, Species Distribution Modelling, Statistical Modelling; see Stelzenmüller et al. (2018) for a review of used tools), and on monitored state/impact data to assess the accuracy of the predicted impact maps.

There are several other promising tools for cumulative effects/impacts assessments. Papadopoulou et al. (2025) identified impact risk-linkage-chain-frameworks (e.g., ODEMM; Borgwardt et al., 2019) and risk-based methods (e.g., Bow-tie; Cormier et al., 2018, 2019) as the next best options (after Cumulative impact spatial mapping) to deliver this element of EBM. Furthermore, as an ISO-accredited technique, and therefore used by industry, the Bow-tie analysis is intimately linked to risk management as well as risk assessment (Cormier et al., 2019). However, these are tools for overall risk identification, and they operate in a semi-quantitative way at best (even when quantitative data are available, they are used to derive qualitative or semi-quantitative variables as needed by the model). This increases their suitability for implementation even in data- and skills-poor situations, although this limits the use of their outputs compared to where quantitative outputs may be required. In addition, these tools do not readily produce spatially or temporally resolved outputs, only providing an assessment for the whole study area and period as a unit (although one expert reported that ODEMM is currently being developed to address quantitative spatial and temporal aspects for marine planning in England, as part of the Marine Spatial Prioritisation (MSPri) programme; Defra, 2023). Hence, they have a reduced ability to fully deliver the cumulative effects element of EBM compared to cumulative impact spatial mapping. As a promising development, the semi-quantitative impact risk-linkage-chain-framework method is being taken forward with the quantitative method by Piet et al. (2021); this results in spatial and temporal mapping of impact risk and more recently incorporating ecosystem services and EwE modelling (SCAIRM; Coll et al., 2024; Piet et al., 2023).

Another example of tools that may address cumulative impacts/effects is food web models (specifically those undertaken using EwE). However, it is not as highly ranked as the ones mentioned above regarding its ability to deliver this element of EBM (Papadopoulou et al., 2025). This is a quantitative tool that potentially gives a complete ecosystem perspective when included in end-to-end modelling (Peck et al., 2018). The extent of the impact assessment, in terms of ecological components covered, depends on the formulation of the individual model being applied (i.e., which groups and functions are included/highlighted). When used to its full extent (as Ecopath with Ecosim and Ecospace), the tool provides spatially and temporally explicit outputs, thus allowing the forecast of cumulative impacts of key components over time, from short-term to long-term predictions, and over space, from site level to broader European scale (Serpetti et al., 2025). However, it is limited in the types of pressures that can be currently included in the model to generate the combined impact. EwE started as a fisheries assessment tool, but has subsequently been applied to contaminants and HABs, and it may also account for climate change (see for example Stock et al., 2023).

Food web models using EwE as the main application was the only tool consistently reported to be able to deliver all the EBM elements in full in this study, although it is noted that the 3 respondents who provided information about this tool were very familiar with it at best. Furthermore, these are models that may need substantial amounts of data if applied in their full spatial-temporal capacity (Stock et al., 2023). Food web models, and other marine ecosystem models, have been historically mostly used to address fishery-related issues in EBM (e.g., fishing activity - pressure - state links, individual fish stock assessment). Indeed, the first applications of the Ecosystem–based Approach were sectoral and related to fisheries (FAO, 2003) although it is emphasised here that EBM should not be sectoral but by definition should be holistic. The application of food web models has been extended to incorporate a spatial perspective, e.g., their use in evaluating Marine Protected Areas and spatial management measures in general (de Mutsert et al., 2023); they have also been used to determine the impacts of cumulative impacts including human induced climate change (Stock et al., 2023). Depending on the link between the pressure(s) and the ecosystem element, functional relationships are inputs to the EwE models (e.g., climate change effect on recruitment). An example in such a direction is provided by Scotti et al. (2022) who consider the impact of warming on the recruitment of cod, Spring-spawning herring and sprat in the western Baltic Sea.

4.4 Limitations and further work

While the study recruited as many respondents as possible and available, it is acknowledged that sample size was a limitation, especially when the number of respondents is compared with the amount and variety of tools being assessed. As indicated above, this has led to an imbalanced data coverage which has likely affected confidence variability across tools. Although this was partly compensated by the high degree of expertise of the respondents for most tools, there are acknowledged gaps for some tools (see tools with Low confidence marked in Table 2) and further data collection to fill these gaps would certainly increase confidence in these assessments.

Furthermore, the possible geographic distribution of the experts may also have influenced some of the results. Although distributed over a wide geographic range, the expertise location was skewed towards Europe and especially the North-East Atlantic region. It is emphasised that geographic location was not a criterion for the data collation but rather it reflected the distribution of the expertise used to inform this study and perhaps in Europe as a whole. It is acknowledged that this may have affected the geographical representativity of the specific tools assessed by the experts, especially where they have been used in regional or local applications (e.g., tool #9.1 covering biochemical models used in the North East Atlantic by OSPAR). In such cases there is a clear advantage of having included tool groups in the assessment, in addition to specific tool examples. In fact, the assessment of tool groups provides information on the methodological approach as a whole rather than a specific implementation, thus ensuring the wider capabilities of the method are represented. There is likely to be a trade-off between the detail and accuracy of the tool characterisation in the data used and the specificity of the tool, whereby the specific tools are likely to be characterised in more detail than the general tool group, thus providing a more accurate but possibly less representative overview of its capabilities.

Future developments of the assessment undertaken here could expand on the respondent sample size while also including more examples of specific tools within each tool group, and with as wider geographic representation as possible of their applications. The characteristics of these specific tools could be used cumulatively to represent the range of capabilities of the related tool group. In addition, as indicated above, considering the continued development of some the tools assessed here to expand their application and capabilities, it is envisaged that regular revisions of this assessment will be needed to ensure it provides an up-to-date representation of the tools and their characteristics available at the time.

5 Conclusions

It is emphasised here that the essence of EBM as a way of operationalizing EBA requires a holistic approach, covering both the natural and social sciences, and its ability to consider all human activities and their pressures and effects (on the natural and human systems) in an area. Accordingly, it requires assessment tools which can satisfy as many of the EBM principles/elements as possible, whether using single tools or tools in combination, and that any tool which fulfils the most elements will be more suitable for addressing the continuum from an approach (EBA) to management (EBM) to the methods used (EBTM) (Elliott et al., 2025b). The tools should then indicate what EBTM need to be used, for example a tool indicating a risk to an ecological component then requires a management measure to prevent or mitigate that risk. This study has illustrated a large array of available tools addressing the EBA-to-EBM-to-EBTM continuum.

The analysis indicates that tool use requires the tool developers to be aware of the science-policy interface and the constraints of undertaking EBM and achieving EBA as the outcome (in essence, why a tool is needed, by whom and what it will achieve). It has been suggested that while several of the tools were developed as academic outputs, i.e., without a precise EBM principle/element in mind, successive use has shown their value and, in some cases, their limitations in use to undertake EBM. Generic, often qualitative tools (such as conceptual models and mind-mapping) should and do provide the starting point for defining the challenges faced by policy-makers and implementers who can then, with the help of specialist advisors, use more precise, often quantitative, tools.

It is clear that each tool has advantages and limitations for its practical use and that these are to be known and understood by the user in light of the answers they require to support EBM. Because of the complexity of some tools, often (usually) the user is a specialist advisor to policy implementers rather than the policy implementer themselves. As shown in this study, the advantages and limitations of a tool are reflected in the multiple aspects of its capabilities (e.g., the scales at which it operates, the components of the social-ecological system it can assess, whether the assessment is qualitative or quantitative). It is up to the user to make a decision on which tool provides the answers needed and to the required specifications, also considering the data and skill available to them. This study provides a valuable means to inform that decision, which may be further aided by a decision support system such as the one (SEAS4GES) developed by Barnard et al. (2025a, 2025b, in press). As long as the user is clear about what outputs and outcomes are required then finding the most suitable tool to satisfy those is straightforward. Conversely, if the user is not clear about why they require a tool, then they require additional guidance to choose the appropriate tool.

It is acknowledged that the data and skills available to the user may be limiting in their ability to implement a particular assessment tool. Capacity building initiatives such as data sharing (e.g., through platforms such as EMODnet) and training (e.g., tutorials on tools, as for example the one available online on CIMPAL, Teixeira and Radoux, 2025) may mitigate this limitation and thus empower the user in implementing the tool(s).

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors upon request, without undue reservation.

Author contributions

AF: Writing – review & editing, Methodology, Investigation, Conceptualization, Writing – original draft, Visualization, Formal analysis, Data curation. ME: Funding acquisition, Writing – review & editing, Writing – original draft, Validation, Methodology, Conceptualization, Supervision. EA: Methodology, Writing – review & editing, Writing – original draft, Investigation, Data curation. SB: Writing – review & editing, Formal analysis, Methodology, Writing – original draft, Investigation, Data curation, Conceptualization. CS: Conceptualization, Validation, Writing – review & editing, Methodology, Writing – original draft, Investigation, Funding acquisition. ÁB: Conceptualization, Funding acquisition, Validation, Writing – original draft, Writing – review & editing. RC: Validation, Writing – review & editing, Writing – original draft, Conceptualization. NP: Methodology, Validation, Conceptualization, Supervision, Writing – original draft, Funding acquisition, Writing – review & editing, Investigation.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This article is based on work from the project GES4SEAS (Achieving Good Environmental Status for maintaining ecosystem services, by assessing integrated impacts of cumulative pressures). The project is funded by the European Union under the Horizon Europe program (grant agreement no. 101059877, for EU partners) and UK Research and Innovation (grant agreement no. 10050522, for IECS Ltd). The views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or UK Research and Innovation.

Acknowledgments

We wish to thank the many experts for promoting and evaluating their own and other tools and those from the GES4SEAS and other projects who answered the questionnaire.

Conflict of interest

Authors AF, ME, EA, SB, were employed by the company International Estuarine & Coastal Specialists IECS Ltd.

The remaining 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.

The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmars.2025.1643943/full#supplementary-material

Supplementary Table 1 | Outline of the topics addressed by the questions (Q) in the online questionnaire used to gather information about the assessment tools supporting EBM. The type of response required is also shown (open or close, with indication of the response options available in the latter case). Section A (question Q1), not shown here, collected information about the respondent.

Supplementary Table 2 | DAPSI(W)R(M) components used as input by the individual tools. Column headings: A= activities; P = Pressures, S = effects on the state of species (spp), habitats (hab) or ecosystem services (ES); I = Impact on welfare (societal goods & benefits, SG&B); R = response and management measures. Values in the table indicate use as single (1) or multiple variables (2). No use is marked by blank cells.

Supplementary Table 3 | Data requirements for DAPSI(W)R(M) components to be used as input by the assessed tools. Column headings: DAPSI(W)R(M): A= activities; P = Pressures, S =effects on the state of eco-components (eco) or ecosystem services (ES); I = Impact on welfare (societal goods & benefits, SG&B); R = response and management measures. Data type: Quant = quantitative; semi-quant = semi-quantitative; qual = qualitative. The tools for which each variable was reported by respondents are indicated by their ID#.

References

Andersen J. H., Al-Hamdani Z., Harvey E. T., Kallenbach E., Murray C., and Stock A. (2020). Relative impacts of multiple human stressors in estuaries and coastal waters in the North Sea–Baltic Sea transition zone. Sci. Total Environ. 704, 2020. doi: 10.1016/j.scitotenv.2019.135316

PubMed Abstract | Crossref Full Text | Google Scholar

Barnard S., Franco A., Amorim E., Papadopoulou N., Smith C. J., Elliott M., et al.“The practical application of Ecosystem-Based Management tools to marine ecological assessment: optimising tool selection for specific applications – the SEAS4GES Decision Support Tool,” in Encyclopedia of Ecology, 3rd Edition. Ed. Fath B. D. (Elsevier, Amsterdam, NL). In press.

Google Scholar

Barnard S., Franco A., Amorim E., Papadopoulou N., Smith C. J., Elliott M., et al. (2025a). SEAS4GES: Selection of Ecosystem-based ApproacheS for Good Environmental Status (GES) (Geneva, Switzerland: Zenodo). doi: 10.5281/zenodo.14765919

Crossref Full Text | Google Scholar

Barnard S., Franco A., Amorim E., Papadopoulou N., Smith C. J., Elliott M., et al. (2025b). SEAS4GES: Tool Factsheets (Geneva, Switzerland: Zenodo). doi: 10.5281/zenodo.14760469

Crossref Full Text | Google Scholar

Borgwardt F., Robinson L., Trauner D., Teixeira H., Nogueira A. J. A., Lillebø A. I., et al. (2019). Exploring variability in environmental impact risk from human activities across aquatic ecosystems. Sci. Total Environ. 652, 1396–1408. doi: 10.1016/j.scitotenv.2018.10.339

PubMed Abstract | Crossref Full Text | Google Scholar

Borja Á., Berg T., Gundersen H., Hagen A. G., Hancke K., Korpinen S., et al. (2024). Innovative and practical tools for monitoring and assessing biodiversity status and impacts of multiple human pressures in marine systems. Environ. Monit. Assess. 196, 694. doi: 10.1007/s10661-10024-12861-10662

PubMed Abstract | Crossref Full Text | Google Scholar

Borja Á., Bremner J., Kopke K., Gruber S., Alessandrini M., Estrela A., et al. (2025). Bridging the gap between marine science and policy: communicating for an informed society and decision-making. Front. Communication. 10. doi: 10.3389/fcomm.2025.1641970

Crossref Full Text | Google Scholar

Borja Á., Chust G., and Muxika I. (2019). Chapter Three - Forever young: The successful story of a marine biotic index. Adv. Mar. Biol. 82, 93–127. doi: 10.1016/bs.amb.2019.05.001

PubMed Abstract | Crossref Full Text | Google Scholar

Borja Á., Elliott M., Andersen J. H., Berg T., Carstensen J., Halpern B. S., et al. (2016). Overview of integrative assessment of marine systems: the Ecosystem Approach in practice. Front. Mar. Sci. 3. doi: 10.3389/fmars.2016.00020

Crossref Full Text | Google Scholar

Chiappi M., Stranga Y., Kalloniati C., Tsirintanis K., Tsirtsis G., Azzurro E., et al. (2025). CIMPAL expanded: unraveling the cumulative impacts of invasive alien species, jellyfish blooms, and harmful algal blooms. Front. Mar. Sci. 12. doi: 10.3389/fmars.2025.1631423

Crossref Full Text | Google Scholar

Chust G., Corrales J., Gonzalez F., Villarino E., Chifflet M., Fernandes J. A., et al. (2022). “Marine biodiversity modelling study,” in European Commission, Directorate-General for Research and Innovation, Food, Bioeconomy, Natural Resources, Agriculture and Environment (Publications Office of the European Union, Luxemburg). doi: 10.2777/213731

Crossref Full Text | Google Scholar

Clarke Murray C., Agbayani S., Alidina H. M., and Ban N. C. (2015). Advancing marine cumulative effects mapping: An update in Canada’s Pacific waters. Mar. Policy 58, 71–77. doi: 10.1016/j.marpol.2015.04.003

Crossref Full Text | Google Scholar

Coll M., Lynam C., Piet G., Carstensen J., Hemraj D., Borja Á., et al. (2024). Selected methods and potential applications in Learning Sites. (Geneva, Switzerland: Zenodo). doi: 10.5281/zenodo.13740831

Crossref Full Text | Google Scholar

Cormier R., Elliott M., and Kannen A. (2018). IEC/ISO Bow-tie analysis of marine legislation: A case study of the Marine Strategy Framework Directive Vol. 342 (Copenhagen, Denmark: ICES Cooperative Research Report), 56 pp. doi: 10.17895/ices.pub.4504

Crossref Full Text | Google Scholar

Cormier R., Elliott M., and Rice J. (2019). Putting on a Bow-tie to sort out who does what and why in the complex arena of marine policy and management. Sci. Total Environ. 648, 293–305. doi: 10.1016/j.scitotenv.2018.08.168

PubMed Abstract | Crossref Full Text | Google Scholar

Corrales X., Coll M., Ofir E., Piroddi C., Goren M., Edelist D., et al. (2017). Hindcasting the dynamics of an Eastern Mediterranean marine ecosystem under the impacts of multiple stressors. Mar. Ecol. Prog. Ser. 580, 17–36. doi: 10.3354/meps12271

Crossref Full Text | Google Scholar

CSWD (2020). “Commission Staff Working Document: Background document for the Marine Strategy Framework Directive on the determination of good environmental status and its links to assessments and the setting of environmental targets,” in Accompanying the Report from the Commission to the European Parliament and the Council on the implementation of the Marine Strategy Framework Directive (Directive 2008/56/EC) (European Commission, Brussels, Belgium). European Commission SWD(2020) 62 final.

Google Scholar

de Mutsert K., Coll M., Steenbeek J., Ainsworth C., Buszowski J., Chagaris D., et al. (2023). “Advances in spatial-temporal coastal and marine ecosystem modeling using Ecospace,” in Reference Module in Earth Systems and Environmental Sciences, vol. 2023. (Amsterdam: Elsevier). doi: 10.1016/B978-0-323-90798-9.00035-4

Crossref Full Text | Google Scholar

Department for Environment, Food & Rural Affairs (Defra) (2023). Three-year report on the east inshore and east offshore Marine plans: For the period 2 April 2020 to 1 April 2023 (London: Defra).

Google Scholar

Elliott M. (2023). Marine Ecosystem Services and Integrated Management: “There’s a crack, a crack in everything, that’s how the light gets in”! Mar. Pollut. Bull. 193, 115177. doi: 10.1016/j.marpolbul.2023.115177

PubMed Abstract | Crossref Full Text | Google Scholar

Elliott M., Borja Á., and Cormier R. (2020). Activity-footprints, pressures-footprints and effects-footprints – Walking the pathway to determining and managing human impacts in the sea. Mar. Pollut. Bull. 155, 111201. doi: 10.1016/j.marpolbul.2020.111201

PubMed Abstract | Crossref Full Text | Google Scholar

Elliott M., Borja Á., and Cormier R. (2025a). Managing marine resources sustainably – But how do we know when marine management has been successful? Ocean Coast. Manage. 265, 107623. doi: 10.1016/j.ocecoaman.2025.107623

Crossref Full Text | Google Scholar

Elliott M., Borja Á., McQuatters-Gollop A., Mazik K., Birchenough S., Andersen J. H., et al. (2015). Force majeure: will climate change affect our ability to attain Good Environmental Status for marine biodiversity? Mar. Pollut. Bull. 95, 7–27. doi: 10.1016/j.marpolbul.2015.03.015

PubMed Abstract | Crossref Full Text | Google Scholar

Elliott M., Burdon D., Atkins J. P., Borja Á., Cormier R., de Jonge V. N., et al. (2017). And DPSIR begat DAPSI(W)R(M)!” - A unifying framework for marine environmental management. Mar. Pollut. Bull. 118, 27–40. doi: 10.1016/j.marpolbul.2017.03.049

PubMed Abstract | Crossref Full Text | Google Scholar

Elliott M., Cormier R., and Borja Á. (2025b). Making sense of marine management – the ten-tenets revisited. Mar. Pollut. Bull. 221, 118580. doi: 10.1016/j.marpolbul.2025.118580

PubMed Abstract | Crossref Full Text | Google Scholar

Elliott M. and O’Higgins T. G. (2020). “From DPSIR to DAPSI(W)R(M) emerges … a butterfly – ‘protecting the natural stuff and delivering the huma stuff,” in Ecosystem -Based Management, Ecosystem Services and Aquatic Biodiversity. Theory, Tools and Applications. Eds. O’Higgins T. G., Lago M., and DeWitt T. H. (Springer, Cham, Switzerland), 61–86.

Google Scholar

European Commission (2008). Directive 2008/56/EC of the European Parliament and of the Council establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework Directive). Off. J. Eur. Union L 164, 19–40.

Google Scholar

European Union (2014). Directive 2014/89/EU of the European Parliament and of the Council of 23 July 2014 establishing a framework for maritime spatial planning. Off. J. Eur. Union L 257, 135–145.

Google Scholar

Fernandes M., Esteves T. C., Oliveira E. R., and Alves F. L. (2017). How does the cumulative impacts approach support Maritime Spatial Planning? Ecol. Indic. 73, 189–202. doi: 10.1016/j.ecolind.2016.09.014

Crossref Full Text | Google Scholar

Food and Agriculture Organization of the United Nations (FAO) (2003). “Fisheries management - 2,” in The Ecosystem Approach to Fisheries. FAO Technical Guidelines for Responsible Fisheries, vol. 4. (FAO, Rome), 2. Available online at: https://www.fao.org/4/Y4470E/y4470e00.htm (Accessed November 22, 2025).

Google Scholar

Halpern B. S., Walbridge S., Selkoe K. A., Kappel C. V., Micheli F., D’Agrosa C., et al. (2008). A global map of human impact on marine ecosystems. Science 319, 948–952. doi: 10.1126/science.1149345

PubMed Abstract | Crossref Full Text | Google Scholar

Hammar L., Molander S., Pålsson J., Schmidtbauer Crona J., Carneiro G., Johansson T., et al. (2020). Cumulative impact assessment for ecosystem-based marine spatial planning. Sci. Total Environ. 734, 139024. doi: 10.1016/j.scitotenv.2020.139024

PubMed Abstract | Crossref Full Text | Google Scholar

Haugen J. B., Link J. S., Cribari K., Bundy A., Dickey-Collas M., Leslie H. M., et al. (2024). Marine ecosystem-based management: challenges remain, yet solutions exist, and progress is occurring. NPJ Ocean Sustainability 3, 7. doi: 10.1038/s44183-024-00041-1

Crossref Full Text | Google Scholar

Katsanevakis S., Olenin S., Puntila-Dodd R., Rilov G., Stæhr P. A. U., Teixeira H., et al. (2023). Marine invasive alien species in Europe: 9 years after the IAS Regulation. Front. Mar. Sci. 10. doi: 10.3389/fmars.2023.1271755

Crossref Full Text | Google Scholar

Katsanevakis S., Tempera F., and Teixeira H. (2016). Mapping the impact of alien species on marine ecosystems: the Mediterranean Sea case study. Diversity Distributions 22, 694–707. doi: 10.1111/ddi.12429

Crossref Full Text | Google Scholar

Kirkfeldt T. S. (2019). An ocean of concepts: Why choosing between ecosystem-based management, ecosystem-based approach and ecosystem approach makes a difference. Mar. Policy 106, 103541. doi: 10.1016/j.marpol.2019.103541

Crossref Full Text | Google Scholar

Long R. D., Charles A., and Stephenson R. L. (2015). Key principles of marine ecosystem-based management. Mar. Policy 57, 53–60. doi: 10.1016/j.marpol.2015.01.013

Crossref Full Text | Google Scholar

Lynam C. P., Uusitalo L., Patrício J., Piroddi C., Queirós A. M., Teixeira H., et al. (2016). Uses of innovative modeling tools within the implementation of the marine strategy framework directive. Front. Mar. Sci. 3. doi: 10.3389/fmars.2016.00182

Crossref Full Text | Google Scholar

Marshak A. R., Link J. S., Shuford R., Monaco M. E., Johannesen E., Bianchi G., et al. (2016). International perceptions of an integrated, multi-sectoral, ecosystem approach to management. ICES J. Mar. Sci. 74, 414–420. doi: 10.1093/icesjms/fsw214

Crossref Full Text | Google Scholar

Menegon S., Depellegrin D., Farella G., Gissi E., Ghezzo M., Sarretta A., et al. (2018). A modelling framework for MSP-oriented cumulative effects assessment. Ecol. Indic. 91, 171–181. doi: 10.1016/j.ecolind.2018.03.060

Crossref Full Text | Google Scholar

O’Higgins T. G., Lago M., and DeWitt T. H. (2020). “Ecosystem-based management, ecosystem services and aquatic biodiversity,” in Theory, Tools and Applications (Springer, Cham, Switzerland). doi: 10.1007/978-3-030-45843-0

Crossref Full Text | Google Scholar

Olano-Arbulu A., Uyarra M. C., Pouso S., and Borja Á. (2025). Does the marine functioning link with the supply of ecosystem services and human benefits? A practical application in the Bay of Biscay. Front. Mar. Sci. 12. doi: 10.3389/fmars.2025.1497521

Crossref Full Text | Google Scholar

Papadopoulou N., Smith C. J., Franco A., Elliott M., Borja Á., Andersen J. H., et al. (2025). ‘Horses for courses’ – an interrogation of tools for marine ecosystem-based management. Front. Mar. Sci. 12. doi: 10.3389/fmars.2025.1426971

Crossref Full Text | Google Scholar

Patrício J., Elliott M., Mazik K., Papadopoulou K.-N., and Smith C. J. (2016). DPSIR—Two decades of trying to develop a unifying framework for marine environmental management? Front. Mar. Sci. 3. doi: 10.3389/fmars.2016.00177

Crossref Full Text | Google Scholar

Peck M. A., Arvanitidis C., Butenschon M., Melaku Canu D., Chatzinikolaou E., Cucco A., et al. (2018). Projecting changes in the distribution and productivity of living marine resources: A critical review of the suite of modeling approaches used in the large European project VECTORS. Estuarine Coast. Shelf Sci. 201, 40–55. doi: 10.1016/j.ecss.2016.05.019

Crossref Full Text | Google Scholar

Piet G., Bentley J., Jongbloed R., Grundlehner A., Tamis J., and De Vries P. (2024). A cumulative impact assessment on the marine capacity to supply ecosystem services. Sci. Total Environ. 948, 174149. doi: 10.1016/j.scitotenv.2024.174149

PubMed Abstract | Crossref Full Text | Google Scholar

Piet G., Grundlehner A., Jongbloed R., Tamis J., and De Vries P. (2023). SCAIRM: A spatial cumulative assessment of impact risk for management. Ecol. Indic. 157, 111157. doi: 10.1016/j.ecolind.2023.111157

Crossref Full Text | Google Scholar

Piet G. J., Tamis J. E., Volwater J., De Vries P., Van Der Wal J. T., and Jongbloed R. H. (2021). A roadmap towards quantitative cumulative impact assessments: Every step of the way. Sci. Total. Environ. 784, 146847. doi: 10.1016/j.scitotenv.2021.146847

PubMed Abstract | Crossref Full Text | Google Scholar

Potschin M., Haines-Young R., Fish R., and Turner R. K. (2015). Routledge Handbook of Ecosystem Services (Abingdon, UK: Routledge), 629, ISBN: 978-1-138-02508-0.

Google Scholar

Sagarminaga Y., Garcés E., Francé J., Stern R., Revilla M., Magaletti E., et al. (2023). New tools and recommendations for a better management of harmful algal blooms under the European Marine Strategy Framework Directive. Front. Ocean Sustainability 1. doi: 10.3389/focsu.2023.1298800

Crossref Full Text | Google Scholar

Scotti M., Opitz S., MacNeil L., Kreutle A., Pusch C., and Froese R. (2022). Ecosystem-based fisheries management increases catch and carbon sequestration through recovery of exploited stocks: The western Baltic Sea case study. Front. Mar. Sci. 9. doi: 10.3389/fmars.2022.879998

Crossref Full Text | Google Scholar

Serpetti N., Piroddi C., Akoglu E., Garcia-Gorriz E., Miladinova S., and Macias D. (2025). State of the art modelling for the Black Sea ecosystem to support European policies. PloS One 20, e0312170. doi: 10.1371/journal.pone.0312170

PubMed Abstract | Crossref Full Text | Google Scholar

Smith C. J., Papadopoulou N., Elliott M., Franco A., Barnard S., Borja Á., and Uyarra M. C.(2022).Marine Strategy Framework Directive Terminology Definitions and Lists. GES4SEAS Project. Milestone 2.1 Report. 299. Available online at: https://www.ges4seas.eu/wp-content/uploads/2022/11/GES4SEAS__Report-Definitions-and-Lists-17112022final.pdf.

Google Scholar

Smith G., Atkins J. P., Gregory A., and Elliott M. (2025). The minimum complexity necessary: the value of a simple Social-Ecological systems analysis in holistic marine environmental management. Sustain. Futures 9, 100476. doi: 10.1016/j.sftr.2025.100476

Crossref Full Text | Google Scholar

Stelzenmüller V., Coll M., Mazaris A. D., Giakoumi S., Katsanevakis S., and Portman M. E.(2018). A risk-based approach to cumulative effect assessments for marine management. Science of The Total Environment 612, 1132–1140. doi: 10.1016/j.scitotenv.2017.08.289

PubMed Abstract | Crossref Full Text | Google Scholar

Stock A., Murray C. C., Gregr E. J., Steenbeek J., Woodburn E., Micheli F., et al. (2023). Exploring multiple stressor effects with Ecopath, Ecosim, and Ecospace: Research designs, modeling techniques, and future directions. Sci. Total Environ. 869, 161719. doi: 10.1016/j.scitotenv.2023.161719

PubMed Abstract | Crossref Full Text | Google Scholar

Sureeyatanapas P. (2016). Comparison of rank-based weighting methods for multi-criteria decision making. Eng. Appl. Sci. Res. 43, 376–379. Available online at: https://ph01.tci-thaijo.org/index.php/easr/article/view/70803 (Accessed November 22, 2025).

Google Scholar

Tamis J. E., Jongbloed R. H., Rozemeijer M. J. C., Grundlehner A., de Vries P., Van Gerven A., et al. (2024). Assessing the potential of multi-use to reduce cumulative impacts in the marine environment. Front. Mar. Sci. 11. doi: 10.3389/fmars.2024.1420095

Crossref Full Text | Google Scholar

Teixeira H. and Radoux J. (2025). Biotope - Cumulative IMPacts of invasive ALien species (CIMPAL) version (A tutorial available on the LifeWatch ERIC Virtual Research Environment). Available online at: https://training.lifewatch.eu/user-manuals-and-tutorials/resources/?category=17 (Accessed August 15, 2025).

Google Scholar

Uusitalo L., Blenckner T., Puntila-Dodd R., Skyttä A., Jernberg S., Voss R., et al.(2022). Integrating diverse model results into decision support for good environmental status and blue growth. Science of The Total Environment 806, 150450. doi: 10.1016/j.scitotenv.2021.150450

PubMed Abstract | Crossref Full Text | Google Scholar

Vaher A., Kotta J., Szava-Kovats R., Kaasik A., Fetissov M., Aps R., et al. (2022). Assessing cumulative impacts of human-induced pressures on reef and sandbank habitats and associated biotopes in the northeastern Baltic Sea. Mar. Pollut. Bull. 183, 114042. doi: 10.1016/j.marpolbul.2022.114042

PubMed Abstract | Crossref Full Text | Google Scholar

Vallecillo S., Maes J., Teller A., Babí Almenar J., Barredo J. I., Trombetti M., et al. (2022). EU-wide methodology to map and assess ecosystem condition: Towards a common approach consistent with a global statistical standard Vol. 2022 (Luxembourg: Publications Office of the European Union). doi: 10.2760/13048,JRC130782

Crossref Full Text | Google Scholar