Sustaining flood modeling systems as knowledge infrastructures: multi-level governance and inclusive stewardship—findings from Louisiana

Abstract

Hydrologic and hydraulic models are widely used to support flood risk assessment, watershed management, and infrastructure planning. While model accuracy and performance depend on technical advances and computing resources, the long-term usability and credibility of these models largely depend on how knowledge, authority, and operational responsibility are distributed within model governance systems. This study distinguishes governance arrangements, which shape institutional authority and coordination, from model stewardship, which refers to the operational lifecycle-management functions required to sustain long-term model use, including updating, storage, accessibility, maintenance, and version continuity. To examine the socio-technical and institutional dimensions of long-term flood model governance and stewardship, we conducted an in-depth case study of the Louisiana Watershed Initiative (LWI), a statewide effort that developed 48 watershed-scale models to support coordinated flood mitigation across Louisiana. The analysis draws on stakeholder focus groups, surveys, and interviews with 91 participants, including local and regional practitioners, state agencies, and system managers, alongside a comparative review of seven operational model management systems. Participants expressed strong support for regionally grounded model management arrangements, noting that they incorporate local and experiential knowledge, strengthen trust in model outputs, and support capacity building. At the same time, they emphasized the importance of shared infrastructure and broader-scale coordination to ensure consistency, data integrity, and resource efficiency. Comparative review of existing systems further revealed that many flood-model management platforms were originally designed primarily for flood-risk communication or short-term operational objectives rather than long-term stewardship continuity, resulting in fragmented lifecycle-management functions, evolving governance responsibilities, and challenges in sustaining model accessibility, updating, and institutional coordination over time. By conceptualizing flood models as knowledge infrastructures rather than purely technical tools, this study demonstrates how governance arrangements and stewardship structures shape the production, coordination, maintenance, and long-term usability of diverse forms of hydrologic knowledge. These findings highlight that sustaining flood modeling systems is not only a technical challenge but also an institutional one, with implications for designing inclusive and durable stewardship systems capable of supporting flood risk management and climate adaptation across diverse regional contexts.

1 Introduction

Flood hazards are among the most damaging natural risks in the United States and globally (Rashid et al., 2023; Rogers et al., 2025). Recent global risk assessments further identify climate-related hazards, including flooding, among the most significant long-term risks facing societies worldwide [World Economic Forum (WEF), 2026]. Projected increases in precipitation extremes, sea-level rise, and continued development in flood-prone areas are expected to intensify exposure and vulnerability to flood risk (e.g., Wing et al., 2018). In response, hydrologic and hydrodynamic (H&H) models have become central decision-support tools for infrastructure design and long-term flood risk management and adaptation. These models support a wide range of applications, including real-time flood forecasting and warning, flood hazard mapping [e.g., the U.S. Flood Insurance Rate Map program under the Federal Emergency Management Agency (FEMA) (2026a)], and the estimation of flood consequences and damages.

Louisiana represents a particularly important case for examining flood model stewardship and governance due to its long history of severe and recurrent flood hazards, extensive investments in flood mitigation infrastructure, and statewide institutional efforts to support watershed-scale flood modeling and planning. The state experiences multiple interacting flood processes, including riverine flooding, coastal flooding, storm surge, and intense pluvial flooding, often within interconnected low-gradient and coastal transition environments. These diverse flood hazards, together with the growing reliance on flood models for planning, permitting, infrastructure design, mitigation, and emergency management decisions, require coordination across governance institutions operating at local, regional, state, and federal scales. They also create increasing demand for sustainable stewardship and governance arrangements capable of maintaining model usability, credibility, accessibility, and operational continuity over time, particularly as flood risks, development pressures, and policy needs evolve. Although Louisiana represents a particularly distinctive and institutionally mature case due to its extensive history of severe flooding, large-scale investments in flood mitigation, and statewide watershed modeling initiatives, many of the governance, stewardship, and long-term sustainability challenges examined in this study are also increasingly relevant across flood-prone regions in the United States and globally.

Despite substantial advances in modeling techniques and flood risk mapping methodologies (e.g., Agonafir et al., 2023), comparatively less attention has been given to the institutional arrangements that determine whether models are sustained, maintained, updated, and made accessible to users (Alexander et al., 2016; Minano et al., 2021). In this paper, these institutional processes are collectively referred to as model stewardship, defined as the arrangements through which responsibility, authority, and resources for model use, updating, storage, and long-term maintenance are assigned and coordinated across institutions and actors. In practice, unclear, fragmented, or inconsistently implemented institutional processes for model use, storage, and maintenance can limit model accessibility and usability, reduce transparency regarding model assumptions and updates, and ultimately affect stakeholder confidence in how models are maintained and applied, particularly as watershed conditions, land use, and policy needs evolve (Evers et al., 2016). Over time, these challenges may erode stakeholder trust in the reliability and continued relevance of the models as decision-support tools. This issue is especially consequential in flood risk management contexts where decisions depend not only on technical model performance, but also on stakeholder confidence in how models are governed and updated (Srinivasan et al., 2018; Afzal et al., 2025; Alba et al., 2025). Prior research on flood risk governance highlights the importance of cross-sectoral coordination, data stewardship, and the integration of technical modeling infrastructures into broader governance frameworks as prerequisites for credible, usable decision-support systems (Pahl-Wostl et al., 2012; Challies et al., 2016). Yet the long-term institutional management of modeling systems remains underexplored (Morrison et al., 2018; McGlynn et al., 2023), despite their centrality to the sustained utility and legitimacy of flood modeling.

Recent developments in socio-hydrology emphasize that water systems are shaped by dynamic feedbacks between hydrologic processes and human institutions, behaviors, and decision-making structures (Sivapalan et al., 2012). Within this perspective, flood models are not merely computational tools but components of coupled human–water systems whose credibility and relevance depend on governance arrangements. Understanding model stewardship, therefore, requires attention to how modeling systems are embedded within multi-level governance regimes and adaptive learning processes (Pahl-Wostl, 2009).

Another related critical aspect is the governance scale—specifically, how to balance central coordination with local (or regional) autonomy and responsiveness (Arik et al., 2023; Dieperink et al., 2016; Fleischhauer et al., 2012). The tradeoffs are still not well understood in the context of sustaining large modeling investments. The literature on multi-level and adaptive water governance suggests that distributed yet coordinated arrangements can enhance adaptability, learning, and responsiveness in complex environmental systems (Ostrom, 2010; Pahl-Wostl, 2009). Governance scale influences not only administrative efficiency and data consistency, but also the extent to which locally grounded knowledge and practitioner experience are integrated into technical modeling systems. In this sense, questions of governance scale are also questions of knowledge recognition: whose expertise shapes model updates and informs revisions, and how different forms of hydrologic knowledge are coordinated within institutional systems.

Stakeholder engagement and participatory modeling are widely recognized as ways to improve model credibility and adoption, and to facilitate mutual learning and improve the usability of model outputs within institutional routines (Voinov and Bousquet, 2010; Lane et al., 2011; Lema, 2025; Thaler and Levin-Keitel, 2016; Gupta et al., 2023; Coletta et al., 2024). Empirical research further demonstrates that flood information tools are more likely to be trusted and used when they are anchored in community experience, local knowledge systems, and participatory practices such as citizen science and community-based data collection, rather than introduced solely as technical products. These approaches highlight the importance of recognizing diverse forms of hydrologic knowledge within flood governance and modeling processes (Bonney et al., 2016; Vohland et al., 2021; Skilton et al., 2022; Habib et al., 2023).

Research on co-production and participatory modeling emphasizes the importance of engaging stakeholders across the full modeling lifecycle, including model development, application, and revision (Lemos et al., 2018; Lane et al., 2011). However, stakeholder involvement often remains concentrated in later interpretive stages focused on communicating or interpreting model outputs, limiting opportunities for shared model ownership and long-term governance (Landström et al., 2023). This narrow engagement constrains model adoption and prevents modeling systems from becoming embedded within organizational routines and practices, regardless of their technical quality.

In this study, governance refers broadly to the institutional structures, authority systems, and coordination mechanisms that shape decision-making regarding flood modeling systems (Ostrom, 2010; Pahl-Wostl, 2009). Governance configurations refer more specifically to the organizational arrangements through which stewardship responsibilities are distributed across centralized, regional, or hybrid institutional scales (Alexander et al., 2016; Dieperink et al., 2016). Model stewardship, by contrast, refers to the operational lifecycle functions required to sustain long-term model usability, including storage, updating, maintenance, version control, accessibility, and institutional continuity (Evers et al., 2016; Minano et al., 2021). Stewardship is therefore conceptualized here as the operational implementation dimension of governance, embedded within broader socio-technical governance systems. This study also conceptualizes flood modeling systems as knowledge infrastructures—durable socio-technical systems through which technical, institutional, and experiential forms of hydrologic knowledge are generated, maintained, coordinated, and operationalized over time. Drawing from knowledge infrastructure scholarship (Edwards et al., 2013; Deslatte et al., 2024; Nahiduzzaman et al., 2026), such systems extend beyond computational tools by embedding data standards, governance processes, institutional roles, and stakeholder practices into long-term infrastructures that shape how knowledge is sustained and mobilized for decision-making.

These governance and stewardship challenges become particularly visible in large-scale flood modeling initiatives that require coordination across multiple institutions, governance levels, and watershed jurisdictions over extended operational timeframes. This paper examines these governance and stewardship challenges through a case study of the Louisiana Watershed Initiative (LWI) (2025), a statewide program that has developed a suite of 48 watershed-scale H&H models covering the entire state. Like many large-scale flood modeling efforts, the LWI faces challenges associated with establishing sustainable systems for model management (e.g., use, storage, and maintenance), a core aspect of model stewardship. The present study builds upon earlier technical and planning activities conducted under the LWI Model Use, Storage, and Maintenance (MUSM) Plan (Habib et al., 2021), which primarily focused on evaluating approaches to the operational implementation of model use, storage, and maintenance within the LWI program. However, rather than focusing on the development of a specific implementation strategy, the present study examines flood modeling systems through a broader analytical lens centered on governance configurations, stewardship sustainability, stakeholder coordination, and knowledge infrastructure theory. Rather than proposing a specific implementation strategy, this study investigates how institutional arrangements, stakeholder roles, and long-term stewardship responsibilities shape the sustainability, usability, and legitimacy of watershed-scale flood modeling systems. In doing so, the study aims to contribute broader conceptual and transferable insights relevant to flood-prone regions beyond Louisiana.

2 Methodology

This study employs a mixed-methods research design that integrates qualitative stakeholder engagement activities with a comparative review of existing flood model management systems. The stakeholder engagement component included focus groups, surveys, and semi-structured interviews designed to capture perspectives on governance preferences, stewardship responsibilities, institutional constraints, and model use practices across governance scales. The comparative review component combined desk-based examination of publicly available system documentation, governance materials, and technical descriptions with follow-up interviews with system managers to evaluate lifecycle management functions, stewardship approaches, and governance structures across existing operational systems. By integrating these complementary data sources, the methodology enables examination of how governance arrangements, institutional roles, and stewardship responsibilities shape the long-term sustainability of watershed-scale hydrologic and hydrodynamic models, while linking stakeholder perspectives to practical governance and technical design choices in operational model management systems.

The stakeholder outreach and comparative system review activities were originally conducted as part of the broader Louisiana Watershed Initiative Model Use, Storage, and Maintenance (MUSM) assessment effort (Habib et al., 2021), but are re-examined here through a broader governance and stewardship analytical framework. The Louisiana Watershed Initiative (LWI) was organized to coordinate flood risk mitigation across 48 Hydrologic Unit Code (HUC-8) watersheds and 64 parishes statewide. LWI delineated eight provisional watershed regions that cross over political jurisdictions (Figure 1; a ninth region was established after the conclusion of this study), each supported by a regional steering committee composed of representatives from local governments and planning entities. These committees served as key governance bodies through which regional perspectives on model stewardship, use, and maintenance were collected. The regional structure provided a natural setting for examining governance-scale tradeoffs between centralized and distributed regional responsibility.

Figure 1

To ensure representation of decision-relevant actors across regions, the regional stakeholder engagement was conducted primarily with steering committee members and their regional coordinators (75 participants in total) using structured focus groups, surveys, and follow-up interviews. Additional focus groups were conducted with modeling consultants and practitioners affiliated with regional or local entities (7 participants), providing perspectives on model implementation and maintenance. To complement stakeholder-based evidence, the study also conducted a comparative review of existing flood model management systems in other U.S. states, based on desk-based document analysis and semi-structured interviews with system managers (5 participants). Finally, a focus group with state agency representatives (4 participants) was used to capture perspectives related to statewide coordination, oversight, and long-term institutional responsibility.

Across all stages of data collection, 91 participants representing multiple levels of model use and governance contributed data on model management practices, challenges, and needs. This multi-scalar engagement supported analysis of how stewardship responsibilities and knowledge integration differ across governance levels. Data from focus groups and interviews were synthesized to identify recurring governance patterns, areas of convergence and divergence across scales, and perceived tradeoffs between centralized and regionally distributed model management and governance arrangements. Table 1 summarizes participation across data collection activities.

2.1 Regional stakeholder outreach

Regional stakeholder engagement was conducted between October 2020 and January 2021, with participants identified through the LWI. Data collection employed a combination of online focus groups, web-based surveys, and semi-structured follow-up interviews to capture perspectives on model use, storage, and maintenance across governance scales. Focus groups were approximately 90 min in length and were integrated into scheduled regional steering committee meetings to ensure participation by decision-relevant actors.

Participants in the regional focus groups included representatives from city and parish governments, police juries, planning commissions, watershed organizations, and community-based entities. This diversity of participants enabled the inclusion of regulatory, operational, and locally grounded knowledge relevant to model stewardship decisions. Pre-focus group surveys were administered to collect individual-level input before and after group discussions; approximately 67% of focus group participants completed surveys. Pre-survey results were reviewed during each focus group to inform discussion and to surface shared priorities and points of divergence across regions. Focus groups were used to identify common themes between regions, explore differences in regional capacity and preferences, and elicit perspectives on governance options for model stewardship.

During each focus group, participants responded to questions organized around three analytical themes: (1) identification of regional modeling stakeholders; (2) model update needs and preferred update frequency; and (3) interest in and capacity for regional model storage and maintenance (e.g., multi-watershed scales), as opposed to a state-centralized approach. These themes were derived from the conceptual framing of model management systems and were designed to analyze stewardship across institutional, technical, and capacity dimensions. Online polling and chat functions were used during the focus groups to collect structured input and to provide additional opportunities for attendee engagement throughout the discussions. As an engagement strategy, the focus groups did not include direct interaction with modeling or visualization platforms. Instead, participants were briefly introduced to illustrative examples of regional and local flood modeling systems, including DOTD and USACE HEC-RAS applications, as well as the Harris County Flood Control District tools, to provide context regarding the types of information and stewardship functions such platforms can support and to facilitate discussion regarding long-term needs for model use, maintenance, and governance. A summary of the research questions guiding focus groups and surveys is provided in Table 2, with the complete set of questions documented in the LWI technical report (Habib et al., 2021). After each focus group, participants were asked to complete a post-survey with additional questions about model management processes. The post-surveys enabled the research team to cover more topics than the limited focus group time allowed and gave participants an opportunity to share more information after reflecting on what they had learned during the focus group discussion.

To further explore design considerations for model management systems, semi-structured follow-up interviews were conducted with seven focus group participants representing five LWI watershed regions (Regions 1, 2, 4, 5, and 7). These interviews clarified governance preferences and enabled a deeper examination of proposed institutional arrangements. Interview participants were selected to reflect variation in regional capacity and governance preferences identified during focus groups.

2.2 Analysis of existing model management systems

Seven existing flood model management systems were analyzed to provide a comparative institutional context (Table 3). The systems reviewed included Harris County’s Model Management and Mapping (M3) platform operated by the Harris County Flood Control District (HCFCD) (n.d.), FEMA’s Integrated Flood Risk Management (InFRM) platform [Federal Emergency Management Agency (FEMA), 2026b], the North Carolina Flood Risk Information System (FRIS) (North Carolina Floodplain Mapping Program, n.d.), Delft-FEWS forecasting systems (Werner et al., 2013), the Iowa Flood Information System (IFIS) (Krajewski et al., 2017), Charlotte-Mecklenburg flood information systems (Charlotte-Mecklenburg Storm Water Services, n.d.), San Antonio River Authority systems San Antonio River Authority (SARA) (n.d.), and USACE HEC modeling platforms [U.S. Army Corps of Engineers Hydrologic Engineering Center (USACE HEC), n.d.]. Systems were identified through professional experience, targeted web searches, consultation with peers and LWI partners, and recommendations from system managers. Selection criteria prioritized systems with varying lifecycle management functions, governance documentation, or established institutional roles related to model maintenance. The selected systems also represented variation in institutional scale, governance structure, operational objectives, and system maturity, enabling comparison of how stewardship priorities and governance arrangements shape lifecycle-management capabilities.

Analysis began with a desk-based evaluation in which each system was assessed against stages of a typical hydrologic and hydrodynamic model lifecycle, including development, use, sharing, updating, and maintenance (Figure 2). Assessing systems across lifecycle stages enabled evaluation of how system design supports or constrains the continuity of knowledge, version control, and long-term stewardship.

Figure 2

Each system was qualitatively evaluated according to the extent to which it supported key stages of the hydrologic and hydraulic model lifecycle, including model development, execution, sharing, updating, approval, storage, version control, and long-term maintenance. Evaluations were based on desk-based review of publicly available documentation, technical materials, system descriptions, and follow-up interviews with system managers when available. Lifecycle support scores were assigned using a three-level qualitative classification framework consisting of: no capability (NC), partial or limited capability (C), and substantial or well-developed capability (C+). Scores reflected the degree to which systems operationalized lifecycle management functions within institutional workflows rather than software sophistication alone. Comparative interpretations were developed through iterative cross-review among members of the interdisciplinary research team to improve consistency across system evaluations.

To complement the desk-based analysis, semi-structured interviews were conducted with managers of five of seven systems. Interviewees were selected based on the relevance of their systems to model maintenance functions or governance design (e.g., model catalog, download/check-out, modification, upload/check-in, and notification of proposed/approved model updates). The interviews included the Harris County HCFCD system, the North Carolina FRIS system, the Charlotte-Mecklenburg Storm Water Services, the Iowa IFIS system, and the Delft-FEWS system. Interviews were conducted via video conference, lasted between 45 and 70 min, and were guided by a common set of questions (Table 4), with additional probing as appropriate. Comparative insights from these interviews were used to contextualize findings from the specific system reviews and identify transferable governance design principles.

2.3 State agency outreach

State-level perspectives were collected through a semi-structured focus group with representatives from four Louisiana state agencies: the Coastal Protection and Restoration Authority (CPRA), the Department of Transportation and Development (DOTD), the Department of Environmental Quality (LDEQ), and the Department of Wildlife and Fisheries (LDWF). The focus group was conducted via Zoom, lasted approximately 75 min, and was guided by a standardized set of questions (Table 5). This engagement provided insights into statewide coordination needs, institutional oversight responsibilities, and long-term stewardship considerations. State-level input was essential for understanding centralized governance capacities and institutional constraints.

2.4 Analytical techniques and qualitative coding methods

All focus groups and interviews were recorded and transcribed, and contemporaneous notes were taken to support analysis. Qualitative data from transcripts, notes, and survey responses were analyzed using a grounded theory approach (Strauss and Corbin, 1998) to identify recurring concepts and themes across questions and regions. The analysis emphasized pattern identification, including areas of convergence and divergence across governance scales (Miles and Huberman, 1984).

Data were coded manually through an iterative process that incorporated deviant case analysis (Kitzinger, 1995) to ensure that minority perspectives were represented. Coding and interpretation were conducted collaboratively within an interdisciplinary research team to enhance analytical robustness and reduce individual bias. Triangulation across focus groups, interviews, surveys, and comparative system analysis strengthened the validity of findings and supported cross-scale interpretation of governance tradeoffs.

3 Results

The results are organized in three parts. First, we synthesize regional stakeholder perspectives on model stewardship, governance responsibilities, update practices, and institutional capacity constraints identified through focus groups, surveys, and interviews. Second, we compare design features across existing model management systems to assess how technical structures reflect underlying governance assumptions and stewardship priorities. Third, we examine state agency perspectives on hosting, oversight, mission-specific needs, institutional constraints, and centralized versus regional governance arrangements. Across data sources, the analysis illustrates how model stewardship operates as an institutional system rather than a purely technical function. Taken together, the findings highlight how governance scale shapes the integration of local hydrologic knowledge and the long-term sustainability of stewardship arrangements.

3.1 Regional stakeholder outreach

3.1.1 Model use

Participants across nearly all regional focus groups, supported by survey responses and follow-up interviews, identified flood models as critical decision-support tools used by both institutional and technical actors. Institutional actors included parish and municipal governments, planning commissions, drainage and levee districts, and watershed organizations involved in regulatory review, planning, and watershed management processes. Technical actors included engineers, consultants, model developers, and floodplain specialists responsible for model implementation, technical analysis, and regulatory evaluations. Across stakeholder groups, models were consistently characterized as operational governance tools supporting infrastructure planning, permitting, flood mitigation, and long-term watershed management rather than as purely analytical or research-oriented products.

Across regions, models were identified as critical inputs for reviewing development permit applications to ensure compliance with local drainage ordinances and FEMA requirements. They were also used to support floodplain analyses at both regional and local scales, including development review, drainage studies, climate adaptation planning, and project prioritization for grant applications. Beyond regulatory applications, models were viewed as institutional tools for public education and outreach, enabling communities to understand how development patterns and watershed changes influence flood risk. Participants further emphasized the role of models in informing long-term development strategies, evaluating policy changes, implementing master drainage plans, and supporting emergency response and mitigation efforts. In these discussions, models were framed not simply as computational tools but as shared resources that support negotiation, coordination, and decision-making across regions, watersheds, and institutions.

Recurring barriers to model use and stewardship were identified across nearly all regional discussions. The most frequently cited challenges included limited computational resources, insufficient staffing to support routine model operation and updates, software expertise gaps, and long-term funding constraints. Participants from several regions also emphasized broadband limitations and data-transfer difficulties, particularly in areas with weaker digital infrastructure. These constraints were discussed not only as technical limitations, but also as institutional barriers influencing which stakeholders could effectively access, interpret, and operationalize model information within governance and planning processes.

3.1.2 Model update needs and frequency

Across regions, the analysis highlights a shared recognition that flood models have a finite operational lifespan and require structured update cycles to remain credible decision-support tools. Stakeholders consistently identified continuous updates to structural flood mitigation measures—such as regional detention facilities, pumping systems, and control structures—as the highest priority, given their substantial influence on flood behavior. Non-structural and adaptive measures, including changes to stormwater regulations, conservation easements, buyouts, and other planning interventions, were also identified as important but secondary priorities for updates.

Across stakeholder groups, participants generally distinguished between major periodic model revisions and more frequent incremental updates associated with ongoing development activity. Most respondents indicated that major updates—such as incorporation of new mitigation infrastructure, revised topographic datasets, updated land-use conditions, or transition to newer model versions—would likely occur at multi-year intervals, depending on funding availability and staffing capacity. Less frequent updates were also expected for planning-oriented applications intended to support long-term flood mitigation planning, infrastructure prioritization, scenario evaluation, and policy development. In contrast, more frequent updates were associated with permitting-related activities and rapidly changing land development conditions, including site-specific development review, drainage compliance assessments, and infrastructure modifications. These differences suggest that governance context influences not only update expectations, but also model precision and operational stewardship requirements.

Importantly, stakeholders framed model updates as opportunities to incorporate locally observed changes in watershed conditions, development patterns, drainage behavior, and mitigation practices. Participants described how observations from local engineers, planners, watershed managers, and community stakeholders—such as undocumented drainage modifications, recurring localized flooding areas, newly constructed detention systems, changing land-use conditions, and observed discrepancies between modeled and experienced flood behavior—can trigger review of model assumptions, hydraulic structure representation, boundary conditions, calibration priorities, and validation datasets. These observations are evaluated through coordination among regional entities, technical consultants, and model managers during scheduled model review and update processes, allowing locally grounded hydrologic knowledge to inform subsequent model revisions. In this sense, stewardship was described as an ongoing institutional learning process through which experiential and institutional knowledge is translated into evolving formal modeling systems.

3.1.3 Interest and capacity for regional model storage and maintenance

Findings from regional stakeholder engagement indicate strong support for regionally grounded stewardship arrangements over fully centralized state-hosted approaches. Across focus group polling responses and follow-up discussions, decentralized regional housing models consistently received the strongest support, while no focus group responses favored a fully centralized governance structure. Participants emphasized that regional entities are better positioned to manage day-to-day model access, incorporate locally grounded hydrologic knowledge, and respond to evolving watershed and development conditions.

Follow-up interviews reinforced this perspective, particularly among participants representing watershed-scale planning and management organizations. Stakeholders frequently associated regional stewardship with stronger institutional ownership, greater responsiveness to local priorities, and improved integration of experiential knowledge into model updates and validation processes. Regional autonomy was therefore viewed not only as an administrative preference but also as a mechanism for aligning stewardship investments with locally perceived flood risks, political feasibility, and ongoing watershed management activities.

Participants identified several organizational structures that can support regional stewardship functions, including planning commissions, metropolitan planning organizations, regional universities, drainage districts, and levee authorities. Regional universities received particularly strong support due to their technical expertise, educational missions, and access to computational resources, followed by regional planning commissions and watershed organizations.

At the same time, concerns regarding long-term funding and institutional sustainability emerged across nearly all regional discussions. Participants consistently emphasized that stable governance arrangements and dedicated funding mechanisms are prerequisites for sustaining stewardship of the regional model over time. Several stakeholders also noted that governance structures spanning multiple watersheds within a single region may be more difficult to sustain institutionally than stewardship organized at the individual watershed scale, particularly in regions with uneven technical or financial capacity.

Most stakeholder groups expressed preference for regional coalitions or planning committees that could leverage existing technical expertise, institutional partnerships, and funding resources. Frequently discussed funding mechanisms included sales taxes, property tax millage, permitting fees, and stormwater utility fees, with permitting fees viewed as particularly viable in rapidly developing regions. However, participants also emphasized that variability in political support and voter willingness across regions reinforces the importance of maintaining flexibility and regional autonomy within broader governance arrangements.

Table 6 synthesizes the major governance, stewardship, and institutional themes identified across the regional stakeholder engagement activities, including recurring barriers, preferred governance arrangements, and long-term sustainability considerations.

3.2 Review of existing model management systems

3.2.1 Analysis of existing systems

The comparative analysis results (Table 7) demonstrate that none of the reviewed model management systems provided comprehensive support across the full lifecycle of hydrologic and hydraulic models. Instead, systems emphasize different lifecycle stages, reflecting the institutional objectives and operational priorities present at the time of their development. Delft-FEWS implementations prioritize early-stage functions such as data preparation and model execution, while systems operated by Harris County and the San Antonio River Authority emphasize later-stage functions related to model discovery, approval, and version control. Other systems supported only limited portions of the model lifecycle.

Across systems, visualization and dissemination of model results were widely supported, whereas comprehensive model setup and development functions were generally limited. Capabilities for calibration, validation, and model modification were largely absent, except for select Delft-FEWS implementations and USACE HEC tools that support more advanced forecasting and model development workflows. FEMA InFRM, FRIS North Carolina, Charlotte-Mecklenburg, and the Iowa Flood Information System primarily emphasized public-facing functions for flood-risk communication, mapping, visualization, and information dissemination. Both FEMA InFRM and San Antonio supported model discovery and download capabilities. Among the systems reviewed, Harris County’s M3 platform provided the most extensive support for lifecycle stages related to model access, check-out, modification, approval, and update notification.

Collectively, these findings indicate that most systems were designed primarily to facilitate model access, visualization, or execution rather than long-term institutional stewardship, leaving important gaps in lifecycle continuity, version governance, operational coordination, and sustained maintenance responsibility.

Comparative differences in governance and stewardship characteristics across reviewed systems are summarized in Table 8. More broadly, the comparative analysis suggests that observed differences across systems reflect not only technical design decisions but also differing institutional priorities, governance structures, operational mandates, and funding models. Systems developed primarily for public communication and flood-risk visualization emphasized accessibility and the dissemination of model outputs, whereas systems associated with operational forecasting and watershed-management agencies generally placed greater emphasis on lifecycle governance and stewardship continuity. The findings, therefore, indicate that lifecycle-management capabilities are closely linked to broader institutional governance arrangements rather than determined solely by software architecture or computational functionality.

3.2.2 Results of interviews with managers of existing systems

Interviews with managers of existing systems provided additional insight into how institutional priorities, governance structures, and operational mandates shaped platform design and lifecycle-management capabilities. Following on from the earlier results summarized in Table 8, the interviews further showed that most systems were originally developed 10–15 years ago, primarily as public-facing flood-risk communication platforms that emphasize visualization and dissemination of model outputs rather than comprehensive stewardship functions such as structured version control, update governance, and lifecycle continuity. As a result, model management capabilities were often added incrementally, leading to complexity and maintenance challenges. Harris County’s M3 system was a notable exception, intentionally incorporating model management components to protect significant investments in hydraulic and hydrologic modeling and to support systematic model updating. Flood risk communication functions served a broad audience—including citizens, real estate agents, planners, and engineers—while model management capabilities were primarily used by technical professionals and state or federal agencies. This functional bifurcation illustrates how many systems prioritized public communication objectives over long-term stewardship continuity and governance coordination.

System managers highlighted several operational challenges when initial designs often attempted to combine flood risk communication with model management, resulting in overly complex systems that proved difficult to sustain. HCFCD’s case exemplifies this evolution. Initially, its M3 system attempted to support a wide array of features. However, after finding that additional features increased complexity without commensurate value, the system was redesigned to focus on three core functions: model discovery and download, model update and upload, and model notification mechanisms. This redesign illustrates a shift from feature accumulation toward modular stewardship architecture.

The interviews revealed that, while some systems can track users who access models for specific purposes (e.g., flood risk studies), none of the reviewed systems supported formal check-in/check-out processes for model revisions, despite respondents identifying these as desirable features to improve version control. Similarly, systems—except for FEWS—do not currently offer extensibility features, thus limiting interoperability and integration with other institutional systems.

Development costs ranged from approximately $580,000 to $1.4 million (2020 USD), and annual operation and maintenance typically required two to three full-time equivalent staff. Funding mechanisms varied across systems, including deed recording fees (FRIS), property tax millage (HCFCD), and stormwater utility fees (Charlotte-Mecklenburg). These findings underscore that stewardship sustainability depends on stable institutional revenue streams that anchor governance responsibilities over time, rather than one-time capital investments.

In terms of Information Technology (IT) architecture, managers indicated that earlier systems were often developed as monolithic applications but are increasingly being updated or replaced with more modular designs. This shift reflects growing recognition that technical systems must accommodate evolving governance roles, user groups, and update processes. Managers also emphasized that architectural flexibility has become increasingly important for supporting distributed stewardship responsibilities and long-term system adaptability.

Managers reported that decisions regarding system enhancements are generally informed by stakeholder input through formal or informal engagement. Several managers described a shift toward developing related but separate applications that share underlying data, rather than maintaining fully integrated platforms. These modular applications would serve distinct user groups (e.g., home buyers, engineers, planners), allowing simplified interfaces tailored to specific use cases. This shift suggests growing recognition that technical systems must remain adaptable to diverse users and evolving governance arrangements.

3.3 State agency outreach

State agency representatives emphasized mission-specific uses for watershed models, including coastal–inland transition zone analysis, permitting support for levee districts, dam-breach analysis, hydraulic-structure design, environmental-impact assessment of culverts and fish passage, and Total Maximum Daily Load (TMDL) analyses. Some agencies reported limited in-house modeling capacity to perform routine model modifications, underscoring reliance on external stewardship arrangements.

Agency representatives emphasized the need for periodic model reviews to prevent error accumulation, recommending comprehensive updates every 5 to 10 years, supported by dedicated funding and expert oversight. This reflects an institutional preference for structured, periodic validation cycles and formal oversight mechanisms to maintain credibility across agencies.

Discussions of storage and housing options revealed concerns about fully regional governance arrangements, particularly regarding model and data integrity and increased risk of human error due to multiple points of management. However, concerns were also expressed regarding centralized state-hosted systems, including high internal data storage costs (often exceeding commercial cloud rates) and restrictive IT control requirements that could limit flexibility.

Regarding system functionality, agency representatives favored open access for model downloads with restricted upload permissions and expert review processes for revisions prior to public release. Opinions were mixed on whether systems should execute model runs on behalf of users, given the increased cost of storing multiple model versions. Instead, several participants suggested web service interfaces to enable model editing and execution on host systems, thereby reducing data transfer burdens and improving version control.

Overall, state-level perspectives point to an ongoing tension between maintaining data integrity through centralized oversight and enabling broader access and regional participation in model use and revision.

4 Discussion

Building on the findings above, this section discusses how institutional arrangements influence the long-term sustainability and operational continuity of flood models. In particular, it considers how different governance arrangements affect institutional capacity and the incorporation of locally grounded knowledge.

4.1 Regional stewardship and local knowledge integration

Regional outreach revealed a clear preference for governance arrangements in which model use, storage, and maintenance are managed at the regional level (e.g., watershed scales that connect political jurisdictions), rather than a centralized state scale. Stakeholders consistently emphasized that regional entities possess detailed knowledge of watershed dynamics, development pressures, and drainage systems, and are therefore well positioned to interpret and update models in ways that reflect on-the-ground conditions. They argued that regions have the greatest stake in ensuring that models remain current and defensible, especially when they are directly responsible for watershed management and development review. This finding aligns with prior evidence showing that the perceived legitimacy and usability of flood information tools depend on how well they are embedded within community knowledge systems and institutional routines (Habib et al., 2023).

Participants noted that routine stewardship activities—such as check-in/check-out procedures, periodic updates, incorporation of mitigation projects, and review of revisions prepared by consultants and city engineers—are likely to be performed more effectively at the regional scale and within regional institutions (e.g., watershed districts). Familiarity with local hydrologic complexities, direct knowledge of flood-prone areas, and regulatory authority over development decisions position regional agencies to manage models more responsively and contextually than a centralized, state-led system would allow.

Beyond administrative efficiency, regional stewardship practices also influence how hydrologic knowledge is maintained, validated, and incorporated into broader institutional decision-making systems over time. By situating model updates within institutions that routinely engage with watershed dynamics and development pressures, locally grounded knowledge is more likely to inform model revisions. In this way, institutional arrangements influence not only operational continuity but also how local hydrologic knowledge is produced, validated, coordinated, and incorporated into evolving modeling systems. However, participants stressed that the regional governance scale alone does not guarantee effectiveness and that it must be matched with institutional resources and long-term staffing commitments; otherwise, it risks fragmentation, uneven implementation capacity, and inconsistent stewardship practices across watersheds.

4.2 Governance scale and institutional tradeoffs

The stakeholder preferences discussed above can be further understood by examining governance scale more systematically, particularly the institutional dimensions that structure model maintenance and IT infrastructure. The structure of a model management system is shaped by two key dimensions: (1) the scale of model maintenance—who performs updates, revisions, and reviews; and (2) the scale of IT infrastructure administration—where hosting, data storage, and technical support are managed. Each dimension can be organized at regional (e.g., watershed districts that connect multiple parishes and municipalities), central (e.g., state agencies), or blended levels, resulting in distinct governance configurations. These configurations differ in how they balance local knowledge, consistency, staffing requirements, funding stability, and stakeholder engagement. The choice of governance model has direct implications for the credibility, usability, and long-term sustainability of models as decision-support tools.

Centralized governance promotes standardization and efficiency by leveraging shared resources, but may weaken regional ownership and capacity building. In a fully centralized model, state-level staff oversee both model maintenance and IT infrastructure. The principal risks include reduced regional buy-in, weakened local capacity, and dependence on centralized funding and administrative priorities that may shift over time.

Regional governance emphasizes local expertise, responsiveness, and integration with existing operations, but risks fragmentation and uneven capacity. A fully regionalized approach maximizes local control over both modeling and IT systems. This can improve responsiveness and integration into local workflows but requires substantial regional staffing, coordinated communication across regions, and mechanisms to ensure compatibility and version control.

Blended approaches combine centralized support with regional leadership in key functions. For example, model maintenance responsibilities may be assigned to regional entities while IT infrastructure is centralized. This preserves the benefits of regional expertise while leveraging shared IT platforms for consistency, security, and reduced duplication. Blended approaches offer a pragmatic middle ground in which regions retain stewardship of model content while relying on centralized infrastructure for technical reliability. Flexible governance models that allow regions to gradually assume greater responsibility while benefiting from shared infrastructure are likely to produce more durable outcomes. Such incremental transitions may enable capacity building without sacrificing system integrity, thereby strengthening both institutional learning and long-term sustainability of stewardship.

The institutional configurations identified in this study represent a continuum of tradeoffs rather than discrete institutional choices. Decisions about institutional configurations extend beyond administrative design; they shape how authority and responsibility for model stewardship are distributed across institutions. In doing so, they influence whether modeling systems remain adaptive and locally responsive or become administratively efficient but institutionally distant from on-the-ground decision-making contexts.

4.3 Technical design implications of governance choices

The governance configurations outlined in Section 4.2 have direct implications for the design of model management systems and their IT infrastructure. Choices about who is responsible for updates, validation, and oversight must be reflected in the technical architecture that supports those functions. Systems designed without explicit consideration of stewardship responsibilities often struggle with version control, inconsistent documentation, and unclear update authority. By contrast, when governance roles are clearly defined, technical systems can be designed to reinforce those roles through formalized workflows, transparent version histories, and differentiated access permissions.

The comparative review indicates that stewardship systems are more durable when designed using modular and service-oriented architectures, particularly in distributed or blended governance contexts. Modular architectures allow responsibilities to be allocated across institutional levels while maintaining shared data standards and review practices. For example, regional entities may be responsible for initiating updates, while centralized infrastructure ensures standardized storage, review procedures, and archival continuity. In this way, technical design becomes an institutional instrument: it operationalizes governance decisions and shapes how knowledge is revised, validated, and circulated.

Importantly, system architecture does not determine governance outcomes on its own. However, poorly aligned technical systems can constrain institutional learning and create friction between agencies. Conversely, architectures that incorporate structured review processes, clear metadata practices, and transparent update histories can reduce ambiguity regarding model ownership and authority. Technical design, therefore, should be understood as part of stewardship strategy rather than as a separate engineering exercise.

Taken together, these findings suggest that flood models function not simply as analytical tools but as institutional infrastructures whose durability depends on the alignment of governance roles, funding commitments, technical design, and stakeholder engagement. When stewardship functions are treated as peripheral rather than central within governance systems, ambiguity over authority and responsibility can erode trust, marginalize local knowledge, and reduce long-term usability.

4.4 Limitations and future work

This study has two main limitations that could be addressed in future research. First, the analysis is primarily based on stakeholder feedback and focuses on governance scale and stakeholder roles throughout the flood-model lifecycle. While this approach provides important insight into institutional preferences, coordination challenges, and operational constraints, long-term model stewardship also involves a broader set of governance policies, technical procedures, and institutional mechanisms related to model storage, retrieval, updating, accessibility, and version continuity. Future research is needed to examine the development and implementation of formal stewardship policies and governance mechanisms for model management, including funding structures, ordinance integration, institutional accountability, and compliance procedures. In addition, longitudinal studies examining how stewardship arrangements evolve over time would help clarify the relationship between governance design, institutional learning, and sustained modeling capacity.

5 Conclusion

This study examined how governance arrangements and stewardship structures shape the long-term sustainability of flood modeling systems through a case study of the Louisiana Watershed Initiative. These models play a central role in watershed management, flood risk mitigation, drainage planning, infrastructure design, emergency response, and long-term adaptation decisions. By integrating results from focus groups with regional stakeholders, interviews with state agency representatives, and a comparative assessment of existing model management systems, the analysis demonstrates that sustaining model usability and credibility depends as much on governance design and institutional capacity as on technical performance.

The findings highlight that regional involvement in model stewardship enhances relevance, trust, and responsiveness by leveraging local expertise, institutional knowledge, and historical familiarity with watershed conditions. However, regional approaches also require stable funding mechanisms to avoid fragmentation or uneven capacity, underscoring the need for governance systems adaptable to diverse regional capacities and institutional contexts. Centralized governance can provide consistency and shared standards, but may risk weakening local ownership and adaptability. Blended governance arrangements appear particularly well suited to balancing these tradeoffs and competing demands, enabling both regional and local engagement with system integrity.

The comparative review of existing model management systems highlights recurring gaps in long-term stewardship. Many platforms were initially designed for communication rather than lifecycle management and therefore lack structured update workflows, clear version control practices, or defined review processes. Systems that align governance roles with technical design—through transparent documentation, structured revision procedures, and consistent hosting arrangements—are better positioned to support long-term continuity.

More broadly, this study conceptualizes flood models as socio-technical knowledge infrastructures embedded within multi-level governance systems. Their long-term durability depends not only on hydrologic accuracy and technical performance, but also on how institutional responsibilities for stewardship and lifecycle management are distributed and sustained over time. In this sense, sustaining flood models is not only a technical challenge, but also an institutional challenge involving how diverse forms of hydrologic knowledge—technical, institutional, and locally grounded—are recognized, coordinated, maintained, and operationalized within governance systems. By examining how governance scale, stakeholder engagement, incorporation of local knowledge, and system design interact, this study contributes to emerging scholarship on knowledge infrastructures while also providing practical insight for agencies seeking to develop flood modeling systems that remain credible, adaptable, and inclusive under evolving environmental, institutional, and policy conditions.

References

• 1

  AfzalN.SwensonL. J.ZipperS.ZwickleA.WardropperC. B. (2025). Using values-informed mental models to understand farmer, water manager, and scientist use and perceptions of hydrologic models. J. Hydrol.658:133171. doi: 10.1016/j.jhydrol.2025.133171

  • CrossRef
  • Google Scholar

• 2

  AgonafirC.LakhankarT.KhanbilvardiR.KrakauerN.RadellD.DevineniN. (2023). A review of recent advances in urban flood research. Water Sec.19:100141. doi: 10.1016/j.wasec.2023.100141

  • CrossRef
  • Google Scholar

• 3

  AlbaR.ter HorstR.TranB. N.KleinA.UnverzagtK.Godínez-MadrigalJ.et al. (2025). Situating hydrological modeling: a proposal for engaging with the power of models. WIREs Water12:e70030. doi: 10.1002/wat2.70030

  • CrossRef
  • Google Scholar

• 4

  AlexanderM.PriestS.MeesH. (2016). A framework for evaluating flood risk governance. Environ. Sci. Pol.64, 38–47. doi: 10.1016/j.envsci.2016.06.004

  • CrossRef
  • Google Scholar

• 5

  ArikA. D.ChionneD.RenouY.BrochetA.BlanchetJ.KharboucheM.et al. (2023). The limits of scalability: uncovering friction between levels of flood risk governance in the French Alps. Int. J. Disaster Risk Reduct.97:104044. doi: 10.1016/j.ijdrr.2023.104044

  • CrossRef
  • Google Scholar

• 6

  BonneyR.PhillipsT. B.BallardH. L.EnckJ. W. (2016). Can citizen science enhance public understanding of science?Public Underst. Sci.25, 2–16. doi: 10.1177/0963662515607406,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  ChalliesE.NewigJ.ThalerT.KochskämperE.Levin-KeitelM. (2016). Participatory and collaborative governance for sustainable flood risk management: an emerging research agenda. Environ. Sci. Pol.55, 275–280. doi: 10.1016/j.envsci.2015.09.012

  • CrossRef
  • Google Scholar

• 8

  Charlotte-Mecklenburg Storm Water Services (n.d.). Flood information and mapping services. Available online at: https://charlottenc.gov/StormWater/Pages/DataDownloads.aspx (Accessed June 29, 2026).

  • Google Scholar

• 9

  ColettaV. R.PaganoA.ZimmermannN.DaviesM.ButlerA.FratinoU.et al. (2024). Socio-hydrological modelling using participatory system dynamics modelling for enhancing urban flood resilience through blue-green infrastructure. J. Hydrol.636:131248. doi: 10.1016/j.jhydrol.2024.131248

  • CrossRef
  • Google Scholar

• 10

  DeslatteA.AdamsJ. A.CheemaF. S.Alonso VicarioS.BarnesJ. L.KoebeleE. A. (2024). How state-reinforced knowledge infrastructure influences adaptive urban water governance. Ecol. Soc.29:28. doi: 10.5751/ES-15454-290428

  • CrossRef
  • Google Scholar

• 11

  DieperinkC.HeggerD. L. T.BakkerM. H. N.KundzewiczZ. W.GreenC.DriessenP. P. J. (2016). Recurrent governance challenges in the implementation and alignment of flood risk management strategies: a review. Water Resour. Manag.30, 4467–4481. doi: 10.1007/s11269-016-1491-7

  • CrossRef
  • Google Scholar

• 12

  EdwardsP. N.JacksonS. J.ChalmersM. K.BowkerG. C.BorgmanC. L.RibesD.et al. (2013). Knowledge Infrastructures: Intellectual Frameworks and Research Challenges. Ann Arbor. Deep Blue. Available online at: http://hdl.handle.net/2027.42/97552 (Accessed June 29, 2026).

  • Google Scholar

• 13

  EversM.JonoskiA.AlmoradieA.LangeL. (2016). Collaborative decision making in sustainable flood risk management: a socio-technical approach and tools for participatory governance. Environ. Sci. Pol.55, 335–344. doi: 10.1016/j.envsci.2015.09.009

  • CrossRef
  • Google Scholar

• 14

  Federal Emergency Management Agency (FEMA) (2026a). National Flood Insurance Program (NFIP). Available online at: https://www.fema.gov/flood-insurance (Accessed June 29, 2026).

  • Google Scholar

• 15

  Federal Emergency Management Agency (FEMA) (2026b). Integrated flood risk management (InFRM) https://webapps.usgs.gov/infrm/ (Accessed June 29, 2026).

  • Google Scholar

• 16

  FleischhauerM.GreivingS.FlexF.ScheibelM.SticklerT.SereinigN.et al. (2012). Improving the active involvement of stakeholders and the public in flood risk management. Nat. Hazards Earth Syst. Sci.12, 2785–2798. doi: 10.5194/nhess-12-2785-2012

  • CrossRef
  • Google Scholar

• 17

  GuptaN.BearupL.JacobsK.HalperE.CastroC.ChangH.-I.et al. (2023). Stakeholder-informed hydroclimate scenario modeling in the Lower Santa Cruz River Basin for water resource management. Water15:1884. doi: 10.3390/w15101884

  • CrossRef
  • Google Scholar

• 18

  HabibE.MilesB.SkiltonL.ElsaadaniM.MeselheE.HuK. (2021). Analysis of Approaches and Requirements for LWI model use, Storage, and Maintenance – Part I. Louisiana Watershed Initiative Technical Design and Quality Team Report. Available online at: https://d10zxfp0rexahe.cloudfront.net/docs/LWI_MUSMPlan-5-31-2022.pdf (Accessed June 29, 2026).

  • Google Scholar

• 19

  HabibE. H.MilesB.SkiltonL.ElSaadaniM.OslandA. C.WillisE.et al. (2023). Anchoring tools to communities: insights into perceptions of flood informational tools from a flood-prone community in Louisiana, USA. Front. Water5:1087076. doi: 10.3389/frwa.2023.1087076

  • CrossRef
  • Google Scholar

• 20

  Harris County Flood Control District (HCFCD) (n.d.) Model management and mapping (M3) platform. Available online at: https://www.hcfcd.org/Interactive-Mapping-Tools/Model-and-Map-Management-M3-System (Accessed June 29, 2026).

  • Google Scholar

• 21

  KitzingerJ. (1995). Introducing focus groups. BMJ311, 299–302. doi: 10.1136/bmj.311.7000.299,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22

  KrajewskiW. F.CeynarD.DemirI.GoskaR.KrugerA.LangelC.et al. (2017). Real-time flood forecasting and information system for the state of Iowa. Bull. Am. Meteorol. Soc.98, 539–554. doi: 10.1175/BAMS-D-15-00243.1

  • CrossRef
  • Google Scholar

• 23

  LandströmC.SarmientoE.WhatmoreS. J. (2023). Stakeholder engagement does not guarantee impact: A co-productionist perspective on model-based drought research. Soc. Stud. Sci.54, 210–230. doi: 10.1177/03063127231199220

  • CrossRef
  • Google Scholar

• 24

  LaneS. N.OdoniN.LandströmC.WhatmoreS. J.WardN.BradleyS. (2011). Doing flood risk science differently. Trans. Inst. Br. Geogr.36, 15–36. doi: 10.1111/j.1475-5661.2010.00410.x

  • CrossRef
  • Google Scholar

• 25

  LemaM. W. (2025). Sustaining rural livelihoods through participatory water governance: a review of community-driven water resource management models in east and Central Africa. Front. Water7:1704600. doi: 10.3389/frwa.2025.1704600

  • CrossRef
  • Google Scholar

• 26

  LemosM. C.ArnottJ. C.ArdoinN. M.BajaK.BednarekA. T.DewulfA.et al. (2018). To co-produce or not to co-produce. Nat. Sustain.1, 722–724. doi: 10.1038/s41893-018-0191-0

  • CrossRef
  • Google Scholar

• 27

  Louisiana Watershed Initiative. (2025). Creating a more resilient Louisiana. Available online at: https://watershed.la.gov/

  • Google Scholar

• 28

  McGlynnB.PlummerR.GuerreroA. M.BairdJ. (2023). Assessing social-ecological fit of flood planning governance. Eco. Soc.28:23. doi: 10.5751/ES-13863-280123

  • CrossRef
  • Google Scholar

• 29

  MilesM. B.HubermanA. M. (1984). Qualitative data analysis. SAGE.

  • Google Scholar

• 30

  MinanoA.van den BrinkM.MeesH. (2021). Governance of flood risk data. Environ. Sci. Pol.120, 42–52. doi: 10.1016/j.envsci.2021.02.004,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  MorrisonA.WestbrookC. J.NobleB. F. (2018). A review of the flood risk management governance and resilience literature. J. Flood Risk Manag.11, 291–304. doi: 10.1111/jfr3.12315

  • CrossRef
  • Google Scholar

• 32

  NahiduzzamanK. M.KayesA. N. M.HewageK.SadiqR. (2026). Beyond infrastructure: re-conceptualizing flood resilient governance through systems thinking, indigenous knowledge, and proactive planning. J. Urban Manag.15, 512–518. doi: 10.1016/j.jum.2026.04.004

  • CrossRef
  • Google Scholar

• 33

  North Carolina Floodplain Mapping Program (n.d.) Flood risk information system (FRIS). Available online at: https://fris.nc.gov/fris/ (Accessed June 29, 2026).

  • Google Scholar

• 34

  OstromE. (2010). Polycentric systems for coping with collective action and global environmental change. Glob. Environ. Change20, 550–557. doi: 10.1016/j.gloenvcha.2010.07.004

  • CrossRef
  • Google Scholar

• 35

  Pahl-WostlC. (2009). A conceptual framework for analysing adaptive capacity. Glob. Environ. Change19, 354–365. doi: 10.1016/j.gloenvcha.2009.06.001

  • CrossRef
  • Google Scholar

• 36

  Pahl-WostlC.LebelL.KnieperC.NikitinaE. (2012). From applying panaceas to mastering complexity. Environ. Sci. Pol.23, 24–34. doi: 10.1016/j.envsci.2012.07.014

  • CrossRef
  • Google Scholar

• 37

  RashidM. M.WahlT.VillariniG.SharmaA. (2023). Fluvial flood losses in the contiguous United States under climate change. Earth's Future11:e2022EF003328. doi: 10.1029/2022EF003328

  • CrossRef
  • Google Scholar

• 38

  RogersJ. S.ManetaM. P.SainS. R.MadausL. E.HackerJ. P. (2025). The role of climate and population change in global flood exposure and vulnerability. Nat. Commun.16:1287. doi: 10.1038/s41467-025-56654-8

  • CrossRef
  • Google Scholar

• 39

  San Antonio River Authority (SARA). (n.d.). Floodplain mapping and flood risk management tools. Available online at: https://gis.sara-tx.org/D2MR/ (Accessed June 29, 2026).

  • Google Scholar

• 40

  SivapalanM.SavenijeH. H. G.BlöschlG. (2012). Socio-hydrology: a new science of people and water. Hydrol. Process.26, 1270–1276. doi: 10.1002/hyp.8426

  • CrossRef
  • Google Scholar

• 41

  SkiltonL.OslandA. C.WillisE.HabibE. H.BarnesS. R.ElSaadaniM.et al. (2022). We don’t want your water: broadening community understandings of and engagement in flood risk and mitigation. Front. Water4:1016362. doi: 10.3389/frwa.2022.1016362

  • CrossRef
  • Google Scholar

• 42

  SrinivasanV.SandersonM.GarciaM.KonarM.BlöschlG.SivapalanM. (2018). Moving socio-hydrologic modelling forward: unpacking hidden assumptions, values and model structure by engaging with stakeholders: reply to “What is the role of the model in socio-hydrology?”Hydrol. Sci. J.63, 1444–1446. doi: 10.1080/02626667.2018.1499026

  • CrossRef
  • Google Scholar

• 43

  StraussA.CorbinJ. (1998). Basics of Qualitative Research. 2nd Edn. Thousand Oaks, CA: SAGE.

  • Google Scholar

• 44

  ThalerT.Levin-KeitelM. (2016). Multi-level stakeholder engagement in flood risk management. Environ. Sci. Pol.55, 292–301. doi: 10.1016/j.envsci.2015.04.007

  • CrossRef
  • Google Scholar

• 45

  U.S. Army Corps of Engineers Hydrologic Engineering Center (USACE HEC). (n.d.). HEC-RAS River Analysis System. Available online at: https://www.hec.usace.army.mil/software/hec-ras/ (Accessed June 29, 2026).

  • Google Scholar

• 46

  VohlandK.Land-ZandstraA.CeccaroniL.LemmensR.PerellóJ.PontiM.et al. (eds.) (2021). The Science of Citizen Science. Cham, Switzerland: Springer International Publishing. doi: 10.1007/978-3-030-58278-4

  • CrossRef
  • Google Scholar

• 47

  VoinovA.BousquetF. (2010). Modelling with stakeholders. Environ. Model. Softw.25, 1268–1281. doi: 10.1016/j.envsoft.2010.03.007

  • CrossRef
  • Google Scholar

• 48

  WernerM.SchellekensJ.GijsbersP.van den DijkM.AkkerO.HeynertK. (2013). The Delft-FEWS flow forecasting system. Environ. Model. Softw.40, 65–77. doi: 10.1016/j.envsoft.2012.07.010

  • CrossRef
  • Google Scholar

• 49

  WingO. E. J.BatesP. D.SmithA. M.SampsonC. C.JohnsonK. A.FargioneJ.et al. (2018). Estimates of present and future flood risk in the conterminous United States. Environ. Res. Lett.13:034023. doi: 10.1088/1748-9326/aaac65

  • CrossRef
  • Google Scholar

• 50

  World Economic Forum (WEF). (2026). The Global Risks Report 2026 21st ed. Geneva, Switzerland. World Economic Forum. Available online at: https://www.weforum.org/publications/global-risks-report-2026/ (Accessed June 29, 2026).

  • Google Scholar