Introduction

Historic approaches to management in the food, energy, and water (FEW) sectors are not sustainable. This is largely because resources traditionally have been managed from a sectoral perspective, often aimed at maximizing productivity. Focus on a single sector can make the other sectors more vulnerable, resulting in an overall decline in resilience1,2. Cascading impacts can also result in negative feedback loops, whereby efforts to maximize a sector’s productivity today threaten productivity tomorrow. In addition, population growth and climate change are increasing stress on, and exacerbating tensions among each of these sectors, contributing to disruptions and making resilience more difficult to achieve.

Since 2011, significant progress has been made in understanding FEW nexus dynamics3 including research on concepts, methods, and models and tools relevant for characterizing FEW nexus dependencies4,5,6. While understanding at the global scale is robust7 studies at the regional scale are limited. Recent examples applying this research at the regional level include work by Liang, et al. (2020), who quantified provincial-level interconnections of FEW system economic supply chains in China8 as well as Feng, et al. (2019), who quantified FEW flows in the Detroit, Michigan area9. These and other studies have focused on characterizing interactions of physical resources (e.g., through input-output models) associated with FEW sectors5. However, for the purposes of coordinating efforts to address FEW tensions, such characterizations are a valuable but often insufficient step.

Studies need to broaden understanding beyond physical resources to consider the breadth of dimensions shaping nexus interactions and dependencies. Context relating to the social, political, environmental, and economic dimensions surrounding FEW tensions is necessary10 including an understanding of institutional roles, community-level impacts, and social dynamics11,12. Moreover, studies can fall short of tracing and articulating the specific negative cascading effects and feedback loops central to regional FEW nexus tensions. Additionally, previous research has noted the failure to implement nexus-based approaches in practice13. Identification of specific actors—even if incomplete—and contextual information about FEW tensions can assist entities seeking to coordinate and affect positive change with knowing who the right individuals to involve are and tailoring discussions in more productive directions.

Transdisciplinary approaches and mixed methods are necessary to fill this gap and to move beyond FEW nexus characterizations toward implementation of nexus-based approaches5. Incorporating qualitative data—in conjunction with quantitative modeling and network characterization—may enhance the ability to explain the internal mechanisms of FEW nexus dynamics and provide needed detail and context5,14. In particular, news articles and grey literature may serve as valuable sources, particularly for regional analysis. For example, news articles can capture high-profile and emerging issues with local context, often mentioning specific stakeholders and stakeholder groups involved in tensions, including decision-makers and those affected by and driving tensions. Analysis of such articles can be used to identify relationships among topics, as well as their evolution15.

This paper fills these gaps by introducing a transdisciplinary and mixed-method approach applied at the regional scale to (1) identify and extract relevant contextual information on FEW nexus tensions from news articles and grey literature; (2) translate text-based cues into geospatial coordinates to create a geographically anchored network diagram of FEW dependencies; and (3) integrate both quantitative and qualitative information into a presentation format—ArcGIS StoryMaps. This format allows stakeholders to interactively engage with information about FEW nexus tensions, including the ability to view feedback loops existing at varying geographic scales, which allows for a more holistic picture of the FEW tensions confronting a region. Applying this relatively straightforward approach allows for early-stage identification of key players involved in FEW nexus tensions important to the region, assisting in efforts to develop shared contextual understanding and collaborative decision-making toward disrupting negative feedback loops. We leverage this approach to conduct a FEW analysis of the Gulf Coast region, consisting of the states of Florida, Alabama, Mississippi, Louisiana, and Texas (referred hereafter as the Region) and the Gulf waters. This Region has well-known hubs for food and agriculture and energy production, includes nationally critical natural bodies of water and waterways, and is a key region in current U.S. energy transition efforts16. To our knowledge, this study represents the first time such an analysis has been conducted on the Region.

Methods

Our approach follows guidelines for conducting an integrative literature review17 and consists of the following five steps:

  • Step 1: Conduct broad scanning.

  • Step 2: Perform thematic analysis.

  • Step 3: Supplement select themes with additional scanning.

  • Step 4: Extract network data and align data to geospatial coordinates.

  • Step 5: Construct the StoryMap.

Further details on each of these steps is provided in the following paragraphs.

Step 1: Conduct broad scanning

The initial step involves a broad scan of news articles and grey literature, relying on reputable data sources from established media outlets and government agencies to maximize data quality. We emphasized identification of local news sources in the Region to maximize the inclusion of local context, as supplemented with state and national sources. To ensure more recent dynamics were captured, we set an initial bound of publication dates within 10 years from the time of analysis (2014–2024). The conceptual structure underpinning source collection was the FEW nexus—all articles addressed dynamics related to the nexus that affected resilience in the Region, regardless of topic. Efforts to understand the history and origins of tensions within the region (see “Discussion” section) helped inform the identification of keywords used in this initial stage of review (Table 1). Using the Google search engine, we searched various iterations of keywords independently and in combination, including in combination with Region states and major metropolitan cities.

Table 1 Key words searched individually and in combinations in initial broad scanning.
Full size table

Step 2: Perform thematic analysis

Thematic analysis was performed on the 84 sources identified in Step 1. Each researcher reviewed all articles, performing an initial round of inductive coding to extract themes related to the FEW nexus. Researchers conferred on the identification of a subset of 12 recurring themes that reflected key FEW tensions. Of these, we selected four to perform an additional round of review based on: (1) amount of coverage; (2) length of time an issue received; (3) breadth of stakeholders involved; and (4) impact on regional resilience. All sources not addressing the four selected themes were archived.

Step 3: Supplement select themes with additional scanning

As identified in Step 2, the four selected themes were as follows: (1) the dead zone, (2) the Permian Basin, (3) water markets and (4) biosolids. We compared regional context from the sources to the current understanding of the history and origins of tensions in the region (see “Discussion” section) to identify gaps in regional understanding. To capture regional FEW dynamics more comprehensively, we engaged in a second round of data collection by performing targeted searches of key words specific to each gap, including in combination with individual cities mentioned in sources, using the Google search engine. In total, 42 articles, including peer-reviewed sources, were identified for analysis after Step 3, with some articles providing context on more than one theme (Supplementary Table S1). For additional details on the final set of articles, see the “Results” section.

Step 4: Extract network data and align data to geospatial coordinates

To structure the data for geospatial analysis, we extracted references to sectors, key players, and locations from each source and identified them as driving or being impacted by tensions. For sectors, we identified the primary sector driving tensions and the primary and secondary sectors impacted by tensions. The secondary sectors impacted represent cascading impacts and were included to support identification of negative feedback loops.

We standardized locations to the county level. When an article referenced a general location (e.g., the Permian Basin), we assigned a centrally located county as a proxy. When an article referred to a water body, we used a single location within the water body for consistency. To facilitate identification of the roles of groups of key players, we assigned specific key players (e.g., an individual farmer) to general categories (as inductively identified during our review)—namely, academia, environmentalist, farmers and ranchers, federal government, fertilizer industry, fishing industry, foreign government, Gulf of Mexico stakeholders (i.e., stakeholders dependent on the Gulf of Mexico), (local) community, local government, nonprofit, oil and gas industry, state government, and water services. We joined latitude and longitude coordinates to each county entry to complete the extracted dataset.

Step 5: Construct the StoryMap

Data were mapped and analyzed in ArcGIS Pro (3.1.2). After validating the geolocated nodes visually, we used the “XY to Line” tool to construct directed interactions between pairs of nodes. Arrows originated at nodes driving tensions and terminated at nodes affected by tensions. Sizing of arrow thicknesses was accomplished using the “Count Overlapping Features” tool, which was based on the number of interactions occurring between each node pair.

For each of the four themes, we developed summaries based on the source content reviewed to provide context to the various network interactions displayed. Where possible, we validated statistics cited in each summary with other data sources. Information corresponding to interactions analyzed and tension contexts summarized were consolidated into a publicly-available digital report using ArcGIS StoryMaps on ArcGIS Enterprise (11.3) titled “Food-Energy-Water (FEW) Nexus in the Gulf of Mexico”.

Results

The final news and grey literature dataset consisted of 42 articles spanning the years 2012 to 2024 (Supplementary Table S1). All references to sectors, key players, and locations in these documents were tallied. Some articles addressed tensions in more than one theme, with 5, 11, 18, and 12 articles informing biosolids, dead zone, Permian Basin, and water markets themes, respectively.

Network summary and nexus tension points

Across all themes, a total of 200 links represented primary impacts and 102 links represented secondary impacts. The median number of primary impact links per source and per theme was 2 and 50, respectively; the median number of secondary impact links per source and per theme was 2 and 25.5, respectively (Fig. 1).

Fig. 1
figure 1

Number of links by impact and individual source or theme.

Full size image

Tables 2, 3 and 4 summarize sectors, key players, and locations, respectively, that drove and were impacted by tensions. While results vary by theme, totals serve to elucidate dominant trends in the Region. Several key players dominated the role of driving tensions, including farmers and ranchers (n = 31); fertilizer industry (n = 26); oil and gas industry (n = 36); and state (n = 44) and federal (n = 22) government. Key players impacted by tensions were dominated by broad stakeholder groups such as community stakeholders (n = 71) and Gulf of Mexico stakeholders (n = 30). Farmers and ranchers (n = 41) and oil and gas industry (n = 23) were also largely impacted.

Table 2 Count of media Article references to sectors driving and impacted by Food-Energy-Water nexus tensions.
Full size table
Table 3 Count of media Article references to the top three key player categories driving and impacted by Food-Energy-Water nexus tensions.
Full size table
Table 4 Count of media Article references to the top three locations driving and impacted by Food-Energy-Water nexus tensions.
Full size table

Florida (n = 48) and Texas (n = 68) were locations most frequently referenced as driving tensions. Florida and Texas also ranked high for locations impacted by tensions along with the Gulf of Mexico (n = 31, 79, 33, respectively). Florida’s influence largely stemmed from the fertilizer industry while influence from Texas stemmed from state government and the oil and gas industry.

The water sector was central to tensions. First, it was the dominant driver of tensions (n = 102). The water sector also received the most primary impacts from tensions (n = 125) that largely cascaded to the food sector, which experienced the most secondary impacts (n = 82). Finally, water plays a role in each of the following negative feedback loops identified in the analysis:

Food-Water-Food

  • Overapplying fertilizers in agriculture degrades water quality locally, limiting quality water available for irrigation, reducing agriculture productivity.

  • Overapplying fertilizers in agriculture degrades water quality in distant regions, making water uninhabitable for marine life, reducing fishing productivity.

  • Fertilizer plants supporting agriculture pollute local water resources, reducing fishing productivity.

  • Over-pumping water for irrigation limits water availability and degrades the quality of remaining water, limiting quality water available for irrigation, reducing agriculture productivity.

  • Farmers selling water rights to the energy sector and not irrigating potential crops reduces agriculture productivity.

  • Application of contaminated soil amendments (i.e., biosolids) degrades local water quality, impacting the health of livestock dependent on local water sources and limiting quality water available for irrigation, reducing agriculture productivity.

Energy-Water-Energy

  • Contaminating water via fracking degrades water quality, limiting quality water available for fracking, reducing energy productivity.

  • Overusing water for fracking limits water availability and degrades the quality of remaining water, limiting quality water available for fracking, reducing energy productivity.

Water-Water

  • Over-allocating water limits water availability, desalination plants that mitigate water scarcity degrade water quality.

  • Water exports from regions facing water scarcity limit water availability in the exporting region and to stakeholders dependent on the water source (i.e., aquifer).

  • Water trade negotiations among water-scarce regions result in water provision commitments not met and less water available in the receiving region than planned for in the water budget.

  • Water exports upstream limit water availability downstream and degrade quality of remaining water.

Discussion

Examining the history and origins of FEW nexus tensions in the Region provided valuable context for the initial stage data collection and review.

Sectoral management has been employed dating back to the 1950s to meet increasing food and energy demands of the growing global population, which catalyzed the Green Revolution and Great Acceleration18,19. In the Green Revolution, agricultural inputs (i.e., irrigation water, fertilizers, and chemicals) were increased to maximize outputs (i.e., yields)20. Increased dependence on irrigation water resulted in depletion of water resources, inducing and worsening water scarcity21. As water levels decline, contaminants concentrate, degrading the quality of remaining water. Further, nutrients in agriculture contribute to water quality issues. For example, nutrients are applied at high rates (e.g., fertilizers, manure)22,23 with more than half of those applied regularly unused by crops24,25 and susceptible to entering water resources26,27,28 causing algal blooms, eutrophication, and loss of habitat29,30,31 with legacy effects enduring up to 35 years32. Important for the Region is that these impacts can be delocalized – choices made outside of the Region can affect resilience of FEW sectors in the Region.

In the Great Acceleration, increasing energy demands were largely met by coal, oil, and gas33. As the US depleted the most accessible, conventional fossil fuels, hydraulic fracturing (fracking) was implemented to extract unconventional fossil fuels (e.g., oil sands, shale oil, shale gas). Fracking requires significantly more water than extraction of conventional sources34,35 and water demand is climbing year over year with water scarcity impacts in the largest fossil fuel source in the US, the Permian Basin oil field35,36. Additionally, fracking generates toxic wastewater (produced water), which contributes to groundwater and land contamination. Finally, while interest in alternative energy resources has grown in response to the association of fossil fuels and greenhouse gases37some new technologies, such as biofuels, have higher water footprints than fossil fuels7.

More recent management approaches also contribute to tensions, such as the use of biosolids – organic materials resulting from treatment of domestic sewage in water treatment plants. Biosolids are rich in nutrients, making them an attractive option for use as fertilizers. Yet, biosolids can pose significant health risks. They are a recognized source of antibiotic-resistant pathogens38 and contain more than 700 pollutants39. Of particular concern are per- and polyfluoroalkyl substances (PFAS) or “forever chemicals”. These manufactured chemicals break down slowly because they are resistant to water, oil, grease, and heat. They are known to increase risks of certain cancers, modify responses of the immune system, alter liver enzymes, and increase health risks to pregnant women and their babies, among others40. PFAS contaminate farms when they are present in applied soil amendments, harming soil and farm animals. Further, they can leach into water resources, contaminating irrigation and drinking water. Finally, they can be present in food commodities because PFAS can be taken up by crops through the water, soil, or air41.

Table 5 provides a high-level, qualitative summary of dynamics across each theme derived from the analyzed sources (Supplementary Table S1). As an example of detailed context generated by this approach, Fig. 2 illustrates the link analysis results for the dead zone accompanied by the following description, which was generated by paraphrasing the content in the original sources. Similarly detailed results and narratives for the remaining three themes are accessible on the publicly-available StoryMap (Food-Energy-Water (FEW) Nexus in the Gulf of Mexico).

Table 5 Qualitative summaries of analyzed themes.
Full size table
Fig. 2
figure 2

Food-Energy-Water nexus tensions related to the dead zone in the Gulf of Mexico. Geospatial layers pertinent to dynamics driving tensions are represented in panel A). Remaining panels reflect network analysis of tensions largely related to B) agricultural practices; C) the fertilizer industry; and D) the energy sector. Circular nodes represent drivers of tensions. Square nodes represent direct and cascading impacts of tensions. Blue, green, and orange arrows and icons represent the water, food, and energy sectors driving tensions, respectively. Blue, green, and orange outlines of square nodes represent the water, food, and energy sectors impacted by tensions, respectively. Numerals sequence dynamics and cascading impacts across stakeholders and regions according to the narrative.

Full size image

The Gulf of Mexico dead zone ranked among the top three largest in the world in 202442 (Fig. 2A). It results from excess nutrients, which drive harmful algal blooms and lead to hypoxia (low dissolved oxygen) making water uninhabitable. The Mississippi River serves as the dominant corridor transporting nutrients, discharging them into Gulf waters. Excess nutrients from agricultural activity serve as the primary source. While local agriculture plays a role, decisions made in distant agricultural regions of aquifers connected to the Mississippi River and its tributaries, such as the Great Plains and Midwest regions, contribute significantly to the dead zone.

Network analysis in Fig. 2B depicts interactions among government, the agriculture industry, and the fishing and restaurant industries. The government (1), acting through the Mississippi River / Gulf of Mexico Hypoxia Task Force established in 1997, was charged with minimizing the dead zone. However, it is not on track to meet its goals43. A major critique of the government’s approach is reliance on voluntary conservation practices in agriculture (2) rather than pursuing enforceable limits on nutrients to minimize those reaching the Gulf to swell the dead zone (3). Importantly, the dead zone leads to contaminated seafood and reductions in quantity and size of catches, which come at great financial cost to local fishing and restaurant industries (4).

Figure 2C illustrates dynamics related to the fertilizer industry (1), which plays a central role in these tensions as important fertilizer plants are located in the Region. In addition to selling fertilizer products to local and distant farmers (2), of which a portion of applied nutrients leach and return to swell the dead zone (3), fertilizer plants have contaminated local water resources, killing fish, damaging marine life, exacerbating the dead zone, and resulting in legal issues with state and federal governments (4). Florida plays a key role in the fertilizer industry as one of the country’s top producers of soil amendments sourced from biosolids that are applied on land throughout the state and leach to impact water quality. While the government encourages land application, concerns of impacts on the community (5) are increasingly common. Further, Florida is home to one of the richest phosphate deposits in the world and the largest ammonia plant in the world is located in Louisiana. Production of these nutrients is water intense and contributes to water scarcity issues.

As network analysis illustrates in Fig. 2D, the energy sector (1) and government regulations (2) impact Gulf water quality (3) and communities (4) throughout the region. Notably, pollution from the Deepwater Horizon oil spill directly impacted water quality and decimated the oyster population. Not only did this directly impact the food sector (5), but it also indirectly impacted the water sector through loss of a key ecosystem service – oysters filter water, removing algae and excess nutrients. The energy sector also has permission to expel wastewater in the Gulf. In Corpus Christi, a single oil and gas company has permission to expel large quantities of polluted wastewater daily.

Results demonstrate that consolidating qualitative data and translating into quantitative data geographically yielded a more holistic understanding of nexus tensions in the Region than could be obtained in a single source or by qualitative or quantitative analysis alone. Analyzing numerous sources comprehensively that address a specific theme reveals abundant (Fig. 1), widespread links at various spatial scales (i.e. within and outside the region; Fig. 2). This contrasts with analysis of single sources, which tend to include only a small number of links (Fig. 1) and, therefore, cannot provide comprehensive insight. The geospatial analysis and visualization facilitate interpretation of the roles of sectors – when they drive and are impacted by tensions – and identification of regional hotspots and key players. Making interpretation more comprehensive, the qualitative data elucidates the broader social, political, environmental, and economic context.

Regional hotspots and key players were identified (Tables 3 and 4) and were evident in mapping products (Fig. 2). For example, in the dead zone, the fertilizer industry in Florida played a significant role driving tensions with primary impacts to farmers and ranchers locally and outside the region as well as cascading impacts to the waters of the gulf, which include environmental – water quality issues – that cascade to economic – the fishing industry. The influence of the Florida fertilizer industry extended to biosolid tensions, with qualitative data in sources identifying Florida as one of the primary producers and distributors of biosolids in the US. The oil and gas industry was frequently identified as a driver of tensions across the qualitative data, often related to impacts on water scarcity and water quality, particularly in the Permian Basin. As investments continue amidst depleting fossil fuel resources, sources highlighted concern over tensions being exacerbated. These dynamics also drove tensions in the water market, particularly between Texas and New Mexico.

Farmers and ranchers were central to tensions with impacts from within and outside of the Region, highlighting the importance of flexibility in the approach related to spatial scale. Management choices of farmers and ranchers within the region impact all four themes. Fertilizer runoff contributes to the dead zone and the application of soil amendment products that contain biosolids (including when the presence of biosolids is unknown to the landowner) exacerbates tensions stemming from contamination. Further, selling water rights in lieu of producing agricultural commodities impacts food and energy sectors in the Permian Basin and contributes to tensions in the water market. Importantly, farmers and ranchers represent one of the most important players driving delocalized impacts. Because of the expansive footprints of the Ogallala Aquifer and Mississippi River basin, management choices throughout the Great Plains and Midwest regions have important implications in the water sector directly and the food and energy sectors indirectly in the Region. Relatedly, adjusting spatial scales facilitates identification of policy implications across different levels of the government, which played an important role across all themes. State- and federal-level guidance played critical roles in the production and distribution of biosolids, energy production, land management, and water use. Importantly, federal-level guidance was critical for the dead zone and Permian Basin, permitting overuse and pollution of water resources via the delocalized influence of management choices by farmers and ranchers mentioned above. Local governments drove impacts related to water scarcity by over allocating water resources, especially to the oil and gas industry in coastal cities where new infrastructure development is concentrated.

Regional hotspots and key players impacted by tensions were also identified. Not surprisingly, broad stakeholder groups were most frequently impacted. Interestingly, farmers and ranchers and the oil and gas industry were also frequently impacted. This points to the important dynamic of negative feedback loops. For example, decades of overallocation within the Region and throughout the Great Plains have led to water being consumed from the Ogallala Aquifer 10 times faster than can be naturally recharged44. Water scarcity is further exacerbated by rapidly increasing water needs for fracking45. Consequently, the food and energy sectors increasingly face limited access to quality water to continue productivity, in addition to impacts to communities in general.

Analysis and visualization of the roles of the FEW sectors (Table 2; Fig. 2) also reveal negative feedback loops, in which the water sector played a central role. Feedback loops in the food and energy sectors resulted from cascading impacts, by which management in the food or energy sectors negatively impacted the water sector and, in turn, negatively impacted the food or energy sector. In contrast, feedback loops resulting from management choices in the water sector had direct impacts on the water sector. Importantly, the water sector was most frequently identified as the primary driver. This was largely influenced by tensions related to biosolids, which are sourced from waste at water treatment plants. Additionally, many articles cited the current state of the water sector – widespread water scarcity and degraded quality – as the primary driver of tensions. It is important to note the historic contributions of the food and energy sectors to vulnerabilities in the water sector. Had more articles included these original drivers, an increase in the roles of food and energy sectors as drivers and in the water sector as experiencing primary impacts may have been observed. This observation highlights the fact that analytic results are limited by the topics subjectively covered in the qualitative data.

Because content covered in qualitative data is subjective and biases may exist, this work cannot be considered an exhaustive representation of regional nexus dynamics. Rather, this approach serves as an initial step to understanding the most high-profile dynamics in a region and can provide direction for next steps when conducting additional research and analysis to fill identified gaps. For example, articles gave attention to agriculture impacting the water sector. However, references generally applied to row crops while information specific to animal agriculture was lacking. Similarly, virtual water transfers (tracking water footprints of commodities and regional transfers through the supply chain) were not addressed in the media, though they are recognized as critical in FEW nexus dynamics7. Additionally, care must be taken to ensure quality data that captures the most up-to-date state of the nexus. However, not all content can be validated with existing literature and rapidly evolving circumstances may not yet be captured in available sources. Finally, interpreting the quantitative results in the context of qualitative data shortcomings is key. For example, it is important to note that an expansive region of stakeholders was often denoted as a single reference in an article and, therefore, as a single node in analysis. This was observed frequently for farmer and rancher key players in the Great Plains and Midwest regions. The low number of references outside the Region may be an artifact of this accounting. Therefore, the perceived level of influence of key stakeholders and regions may be low relative to others addressed with more specificity in articles. As such, interpreting single nodes as indicators of wider regions is critical. With this context, it is clear that choices made outside the Region impacted resilience within the Region.

Conclusion

This work demonstrates an approach for holistic horizon scanning that integrates qualitative and quantitative methods to identify regional aspects of FEW nexus tensions including key players, regional hotspots, and roles of FEW sectors, which elucidate important dynamics and negative feedback loops across all three sectors. The approach is both flexible and reproducible across regions as well as able to move through spatial scales and integrate new insights and local context as the nexus evolves.

While this work aimed to improve understanding of tensions, the approach can similarly be applied to improve understanding of the state of solution development. Articles reviewed in the broad scanning phase highlighted several integrated approaches being implemented in the Region by an array of stakeholders. For example, communities are working to support recovery of oyster populations, thereby supporting the fishing industry and improving water quality46. Farmers are transitioning to regenerative agricultural practices to mitigate impacts on water quality and scarcity while improving soil health for agricultural production47. Ranchers are adopting agrivoltaics, which can have benefits across all three sectors and beyond48,49,50.

A critical next step will involve engaging regional stakeholders and subject matter experts in an equitable manner51 to validate findings and continue filling knowledge gaps around tensions as well as to ensure paths forward optimize tradeoffs, reflect local priorities, and maximize potential for successful implementation. Because stakeholder engagement begins with identifying, categorizing, and investigating relationships among key stakeholders52 this approach can be leveraged to convene stakeholders important to regional resilience. This is because sources frequently referred to individual stakeholders (e.g., individual people, specific companies, specific government offices). While an exception to this was often observed when referencing groups of stakeholders outside the region, as in the case of farmers and ranchers in the Great Plains and Midwest regions, regional stakeholders were regularly referenced with specificity and our results have identified, categorized, and analyzed relationships among them.

Affecting change will require a combination of top-down and bottom-up approaches. Because of delocalized impacts, top-down policies that improve land management and water use outside the Region would enhance resilience in the Region. Further, because the pursuit of integrated approaches in the Region has already begun, it may be an opportune time to unite local stakeholders in a bottom-up approach to build momentum. Supporting the organization of such local communities may benefit from collaborations with organizations such as the Gulf Research Program, which works to improve resilience through the application of science, and which excels at convening stakeholders.

Making efforts from both approaches complimentary may improve the success of implementation. For example, creating funding opportunities such as zero- or low-interest loans to install agrivoltaic systems may promote adoption of this nexus approach that benefits all three FEW sectors, while inclusion of a requirement to qualify for the loan program that agricultural management practices transition from conventional to regenerative would multiply positive impacts on the food and water sectors and requiring proper disposal or recycling of solar panels at end of life would mitigate potential future contamination. Because stakeholders local to the Region are currently pursuing both agrivoltaics and regenerative agriculture, this may suggest that these approaches reflect local priorities. Including such local context in the development of top-down approaches may, therefore, increase the likelihood of adoption, compliance, and improved resilience.