Supplying a nutrient-rich diet to a growing population without damaging the environment is a significant global challenge, and seafood plays an important role (Gordon et al., 2017; Troell et al., 2019; Fanzo et al., 2021; Naylor et al., 2023). According to the Food and Agricultural Organization of the United Nations (FAO), at least 767 million people lacked access to nutritious food in 2021, the equivalent of one out of every nine persons. Seafood, also called aquatic food (i.e., edible fish, shellfish, and crustaceans from fresh, brackish, or seawater), is an excellent source of protein, omega-3 fatty acids, and vitamins and minerals essential for good health (FAO, 2022). For 3.3 billion people, seafood provides at least a fifth of their animal protein intake (Golden et al., 2021; FAO, 2022). Seafood can reduce malnutrition and food insecurity, generally with an equivalent or lower environmental footprint than most terrestrial animal-based food sources (Gephart et al., 2021; Golden et al., 2021).

The food, energy, and water (FEW) nexus considers the interconnectedness of these three critical necessities and their dependence on one another (Albrecht et al., 2018; Abdi et al., 2020; Proctor et al., 2021). The nexus approach seeks to understand the tradeoffs within each resource while optimizing their synergies (Scanlon et al., 2017; Abdi et al., 2020). The interdependencies in the FEW nexus at the global level are striking. For example, 69% of freshwater withdrawals are used in the food production stage of the food system, including for irrigation, livestock, and aquaculture (UNESCO World Water Assessment Programme, 2021), and 10% of global withdrawals are used for energy (International Energy Agency, 2023). More than half (57%) of the water used in the global energy system is used to produce, process, and transport fossil fuels or during fossil fuel-powered electricity generation (International Energy Agency, 2023). The food system, including all stages from food production to retail and households, is estimated to be responsible for 13% of global net energy use (Usubiaga-Liaño et al., 2020). In North America, the following stages/uses comprise 75% of net energy use in the food system: agriculture (17%), food processing (19%), electricity/heat (19%), transport (4%), and households (16%; Usubiaga-Liaño et al., 2020). These are just a few high-level examples of the interdependencies in the FEW nexus.

Facilitating improvements in the FEW nexus requires the engagement of stakeholders whose actions define their status (Ghodsvali et al., 2019), but the integration of stakeholders in FEW nexus work is limited, except as end-users of the results (Hoolohan et al., 2018). The multifaceted nature of the FEW nexus requires multidisciplinary collaboration to effectively address system complexities (Bergendahl et al., 2018; Hoolohan et al., 2018; Ghodsvali et al., 2019), and qualitative research with stakeholders can provide critical insights on drivers and barriers that complement quantitative approaches (Bergendahl et al., 2018; Yung et al., 2019; Kropf et al., 2021). For example, comprehensive stakeholder engagement is an important element of marine spatial planning, as it is recognized that the long-term success of fisheries management demands consideration of human activities and ecological resources in decision-making (Pomeroy and Douvere, 2008; Nutters and Pinto da Silva, 2012). Most research to date on FEW nexus systems has focused on approaches that consider the nexus from a technological lens (Daher et al., 2017; Bergendahl et al., 2018; van Gevelt, 2020). Therefore, multidisciplinary knowledge informed by stakeholders who are connected to or impacted by FEW nexus realities is needed (Hoolohan et al., 2018; Yung et al., 2019; Kropf et al., 2021).

Multiple factors affect resource use in seafood production, with varying potential for adaptation. In capture fisheries, direct use of fuel and indirect inputs such as water are required to produce the fossil fuel energy to operate fishing vessels (Troell et al., 2019; Liu et al., 2020; Viglia et al., 2022a). In addition to being an indicator of environmentally unsustainable practices, the high reliance on fossil fuels has exposed fishery sectors to rising fuel costs and fuel price fluctuations that have challenged the viability of some fisheries (Parker et al., 2018). Aquaculture also relies heavily on fossil fuels, but aquaculture has a greater opportunity to shift to clean, renewable sources because electricity from major utilities comprises a significant share of energy used (Scroggins et al., 2022). At the same time, some forms of aquaculture are dependent on capture fisheries as a source of marine ingredients for fish feeds.

Technological innovations have enabled intensification of aquaculture (Asche et al., 2022) which results in greater yields per unit cost and provides economies of scale (Kumar et al., 2020). However, it has also resulted in more water and energy use to provide feed, exchange water, and maintain water quality during production (Wilfart et al., 2013; Viglia et al., 2022b). In addition, processing of seafood has become more automated in some sectors (Asche et al., 2018), and thus energy and water demands have changed (Brown et al., 2022). Seafood production and processing methods are highly variable within both aquaculture and wild-capture fisheries (Naylor et al., 2021) and also vary significantly by species (Love et al., 2022), resulting in heterogeneous resource use profiles and environmental impacts, and bidirectional interactions between fisheries and aquaculture (Natale et al., 2013; Troell et al., 2019; Bohnes and Laurent, 2021).

The United States (US) sources seafood from all over the world and is the world’s leading importer of seafood by value and second in terms of quantity (Shamshak et al., 2019). Over the last thirty years, the US seafood supply has consistently delivered ~5.8 oz. per person weekly (National Marine Fisheries Service (NMFS), 2022). This is largely supplied from imports, and increasingly from aquaculture, as US and global capture fisheries are fully exploited while aquaculture production is growing (Garlock et al., 2020). However, although there is significant variation between demographic groups, almost 90 percent of Americans do not meet US Department of Agriculture (USDA) dietary intake recommendations (U.S. Department of Agriculture and U.S. Department of Health and Human Services, 2020; Love et al., 2022, 34). Recommendations from US Department of Health and Human Services (HHS) and USDA to double seafood consumption among Americans raise sustainability concerns related to resource use (U.S. Department of Agriculture and U.S. Department of Health and Human Services, 2020). Resource inefficiencies, including waste of seafood, and increasing stress on resources due to changing global diets and climate change highlight the need for practical, cost-effective interventions to improve existing supply chain practices and resource use efficiency (Halpern et al., 2019; Love et al., 2020).

This paper contributes to an improved understanding of the FEW nexus and sustainable food systems, with a focus on seafood. In-depth exploration of seafood is essential due to its distinctive and variable production practices, long supply chains, perishability, and nutritional value. Studies on the FEW nexus and/or seafood sustainability tend to have a quantitative approach and many focus on one stage of a supply chain (e.g., fishing or farming) and/or one type of seafood. In addition, studies that cover multiple stages and/or supply chains often use secondary data from multiple sources (for example, see Hallström et al., 2019 and Koehn et al., 2022). This paper adds to, and complements, the existing literature by providing highly detailed descriptions of (i) how water and energy are used in production and processing stages across six seafood supply chains, (ii) how use of these resources has changed over time, and (iii) perspectives of supply chain actors on potential strategies to improve efficiencies. The results are based on primary data collected via qualitative interviews. The insights cover wild-caught, farmed, domestic (US), and international seafood supply chains, thus enabling analysis of synergies and tradeoffs regarding use of energy and water within and across seafood supply chains.

The FEW nexus was the overarching framework used for this study, and we applied it to the six seafood supply chains using a multiple-case study design. A multiple-case study design allows researchers to study cases that exist in different contexts and compare and contrast them (Yin, 2009). The FEW nexus informed the research questions and methods for data collection and analysis. At the same time, we used an inductive, descriptive approach to data analysis instead of developing and testing hypotheses within the FEW nexus (additional details on data analysis are below).

This work was part of a larger study that involved primary and secondary quantitative data collection, consumer surveys, and separate case studies. The larger effort spanned additional supply chain stages and included waste of seafood. This study provides qualitative results from stakeholders that compliment quantitative aspects of the overall study. The quantitative data we collected on the six seafood supply chains are not reported in this paper, but just as the qualitative data contextualized the quantitative data, the quantitative data was used to check and improve our understanding of the qualitative data. Mixed-method and quantitative results from the larger study have been published (Brown et al., 2022; Scroggins et al., 2022; Viglia et al., 2022a,b; Love et al., 2023a) and other results are forthcoming.

In this paper, water use includes all freshwater that serves a purpose at an operation. We include some information about sea water to fully describe certain supply chain stages, but we did not explore strategies to reduce use of sea water. Other parts of our overall research effort used a lifecycle inventory approach to quantify water use and defined water use consistent with the concept of “blue water” (Chapagain and Hoekstra, 2008), as water that has been sourced from surface or groundwater resources and has either evaporated, been incorporated into a product, or taken from one body of water and returned to another (e.g., surface water pumped onto a farm and discharged into the same body of water would not be considered “water use”; Viglia et al., 2022a). Using a broader definition in this study was consistent with how interviewees interacted with water and enabled exploration of important issues identified by interviewees that did not fit the narrower definition.

From 2019 to 2021, we conducted 47 semi-structured interviews. Interviewees were recruited through industry and academic contacts, and using chain sampling (i.e., snowball sampling). Chain sampling involves one interviewee referring the research team to one or more contacts who may be willing to be interviewed. Recruitment methods for the overall study, including the qualitative interviews, are described in more detail elsewhere (Brown et al., 2022; Scroggins et al., 2022; Viglia et al., 2022a,b). The majority of the interviewees (44) were employees or owners of feedmills, hatcheries, commercial fishing boats, fish farms, and processing plants (Table 1). These interviewees were mostly business owners or middle/upper management, so they generally provided business-level perspectives. In addition, we were referred to and interviewed three experts based on their role in a relevant government agency or as a representative of a trade group supporting one of the supply chains in the study. These interviewees provided complementary sector-level perspectives for wild-caught sockeye salmon and farmed pangasius (Table 1). Several interviewees represented more than one stage of the supply chain; for example, interviewees whose business included multiple stages provided information about hatchery and grow-out operations, or grow-out and processing. The study was approved by the Johns Hopkins Bloomberg School of Public Health Institutional Review Board (IRB no. 8345). Key aspects of data collection, data analysis, and the structure of the results section are summarized in Figure 2.

The interviews were semi-structured; interviewers used a list of questions and potential probes, and also asked follow-up questions that were not pre-written. The first set of questions focused on water use (Appendix), including how water is used by the business, how water use has changed and drivers of any changes (if applicable), current and potential strategies to reduce water use, and motivations and barriers related to adopting water conservation practices. The second set of questions focused on energy use and covered the same topics. Interviewers also asked questions about the amount of seafood that is wasted at each supply chain stage, and those results are reported elsewhere (Love et al., 2023a). The final questions were about general challenges facing the business or sector and how the business prepares for future challenges. Questions about the impacts of the COVID-19 pandemic were added to the questionnaire in 2020.

Interviews were conducted in person, over the phone, or via Zoom (Zoom Video Communication, Inc., San Jose, CA, United States). Notetakers participated in all interviews to accurately capture responses. The audio of phone and Zoom interviews was recorded to allow the research team to check their notes, except for two interviewees who were not comfortable being recorded.

We analyzed the data using a combination of deductive and inductive approaches. A deductive approach to qualitative data analysis involves applying pre-determined codes to the data, and this approach is often used to test whether data is consistent with an existing theory (Creswell, 2007). An inductive approach does not have pre-determined codes, and instead involves developing codes based on the collected data and then applying those codes across the data. Combining deductive and inductive data analysis techniques is beneficial because it allows researchers to analyze the data using codes created based on an existing theory (i.e., “top down”) and codes that emerge from the collected data (i.e., “bottom up”), resulting in a robust understanding of the cases that is not limited to one approach. We developed codes prior to data analysis that aligned with the subtopics and concepts in the research questions (section 1.2). This part of the analysis process applied the FEW nexus framework to the data and was deductive (Creswell, 2007). Codes included: current energy use, changes in energy use, current energy conservation strategies, potential energy conservation strategies, motivating factors, etc. Similar codes were developed for water, as well as codes for interactions between water and energy that involve synergies and/or tradeoffs. We also used inductive analysis techniques (Creswell, 2007); we developed additional codes based on the collected data and focused on creating rich descriptions for each seafood supply chain, or “case.” Codes were applied to the detailed notes from each interview to organize and analyze the text. Data analysis involved examining text relevant to synergies and tradeoffs, and identifying similarities and differences across stages and supply chains. Coding and data analysis were completed using MAXQDA (VERBI Software, Berlin, Germany) and Excel (Microsoft Corp., Redmond, WA, United States).

Results from qualitative interviews are used to better understand how issues impact people and groups, including businesses, and how related issues interact. Qualitative research methods are not designed to generate results that are quantifiable, and due to our sampling methods, our results cannot be interpreted as representative of a larger population (i.e., supply chain actors who were not interviewed). Therefore, the results are reported using descriptive language and do not include the number of interviewees who gave a certain response.

As summarized in Figure 2, study results are presented in tables and a narrative. Descriptive summaries for each supply chain, by stage, are presented in Tables 2, 3. Information in the tables include the source of water or type of energy, the main uses of each resource, stakeholders’ perceptions, trends, challenges, and current and potential conservation strategies. The text below focuses on findings across the stages and supply chains, including synergies, tradeoffs, trends, and factors driving trends.

Using the FEW nexus as a framework, we collected qualitative data from a variety of seafood supply chain actors across six supply chains that are important for the US seafood supply. Understanding how water and energy are used in these settings, and supply chain actor motivations, barriers, and perspectives regarding efficiency strategies, is critical for improving resource use.

The results indicate that the indirect costs of water use is a priority area for developing efficiency strategies, including instances where water and energy use are coupled. Availability and direct costs of water were not significant concerns across the respondents, but interviewees were aware of indirect costs, including cost of energy, that could be reduced if they were able to reduce the volume of water used. For example, reducing the volume of water used on a pond-based farm could reduce energy used to run pumps, and using less ice in product shipments (e.g., due to improved packing materials that keep products cold) would decrease water and energy used to make ice at processing plants. These strategies, that target coupled use of energy and water instead of areas that involve tradeoffs, could result in less use of water and energy per unit of production and thus lower costs for businesses.

Development and/or adoption of efficiency strategies that address commonalities across multiple supply chains, such as wastewater treatment and preservation technologies, should be prioritized. For example, more efficient and/or cheaper ways to aerate, treat, and/or recirculate water in aquaculture systems are cross-cutting and can impact multiple supply chains. Similarly, innovations in cooling and preservation technologies that reduce the volume of ice used during shipping is relevant across supply chains.

We found numerous differences across production methods and settings. The industry is fractured by geographic and sociopolitical settings, varying priority areas needing innovation due to different stressors, and competition within the industry. The heterogeneity likely results in less investment in and support for identifying, developing, and scaling-up strategies and innovations that would improve efficiencies and lower costs than might be expected considering the overall size of the seafood production industry.

Fisheries research has shown that supply chain actors are an important source of ideas for innovative strategies to solve problems (Jenkins, 2010), and also that substantial government support is needed for many innovations to become commercially viable (Dreyer et al., 2019). These fisheries studies, and the results of the current study, support development and/or strengthening of partnerships between industry, government agencies, and academic researchers focused on identifying or developing innovative solutions and working together to scale-up adoption. The benefit of partnerships, and varying levels of current effectiveness, were brought up by interviewees across supply chains. With the goal of improving resource use across the seafood industry, concerted efforts should extend beyond country borders.

There are limitations of this study. We focused on collecting information that was most salient to the FEW nexus, and many aspects of sustainability of food supply chains and social responsibility are not included in the study, such as holistic ecosystem effects and worker health/wellbeing. In addition, prioritizing strategies based on the greatest potential to reduce resource use or improve cost-effectiveness was beyond the scope of the study. Similarly, the study did not attempt to evaluate the prevalence of various practices. Future research should examine these factors. Also, the research relies on perceptions and recall of interviewees, and, given the number of supply chains and stages studied, the research often included a relatively small sample in specific stages of the different supply chains. Lastly, most interviewees were in management and/or ownership roles, and their perceptions may differ from those of frontline workers.

This study contributes to an in-depth understanding of six seafood supply chains, including wild and farmed, and international and domestic supply chains. We characterized the operations of multiple stages in each supply chain using the FEW nexus framework, and the descriptions of trends, resource use, and synergies and tradeoffs provide extensive context that should inform future research, interventions, and policies. The study illustrates the importance of engaging with supply chain actors in order to learn from their expertise and perspectives and to develop practical solutions. The supply chain actors shared firsthand knowledge of their operations and explained which aspects of resource use at the operations they have been working to improve and/or are most interested in addressing. These perspectives are critically important for stakeholders to understand because current perceptions will impact rates of adoption of an intervention. In particular, we found interest among many supply chain actors in reducing water use to lower associated energy and/or water treatment costs, and common resource use issues in multiple seafood supply chains that would benefit from innovations that are scaled-up in multiple settings. Future studies can build on these results by developing inquiries informed by the details we provide on how the FEW nexus describes the supply chain stages, including quantifying resource use and modeling costs and impacts of potential interventions.

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

The studies involving humans were approved by Johns Hopkins Bloomberg School of Public Health IRB. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin because of a low risk of harm to participants.

JF: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. RS: Investigation, Writing – review & editing. TG: Writing – original draft, Writing – review & editing. DL: Conceptualization, Funding acquisition, Investigation, Writing – review & editing. FA: Conceptualization, Funding acquisition, Investigation, Writing – review & editing. MB: Funding acquisition, Investigation, Writing – review & editing. EN: Investigation, Writing – review & editing. LN: Investigation, Writing – review & editing. LJ: Funding acquisition, Writing – review & editing. JA: Conceptualization, Funding acquisition, Writing – review & editing. RN: Conceptualization, Funding acquisition, Investigation, Project administration, Writing – review & editing.