The review was guided by the PRISMA methodology (Page et al., 2021), capturing relevant studies from the Web of Science open repository and the Google Scholar database through a systematic procedure (Haddaway et al., 2015).

The search protocol ensures the presence of three elements in the search results: 1.

Subject – (“flood” OR “drought” OR “hydrology” OR “hydrological”) AND

The search results of the academic and grey literature scan found no documented ex-ante (pre-implementation) user-design procedures for complete research infrastructures or hydrological observatories (Adams et al., 2017), highlighting the novelty of this study. However, there were accessible examples of ex-ante user elicitations of more limited scope, such as the design of digital platforms integrating federated hydrological datasets (5 studies). Ex-post (post-implementation) evaluations of specific hydrological research infrastructures and monitoring observatories were more common (15 studies), from which we reviewed any references to user needs and priorities for enabling research and innovation. We also included literature that is non-project specific but identifies user information needs and infrastructure priorities for supporting research and innovation in hydrology (24 studies). In this article, we integrate evidence from the review with the interview analysis.

In 2021, we implemented an initial set of stakeholder consultation activities, including a scoping survey (127 completed), two workshops (81 attendees), and 20 further stakeholder group discussions (FDRI, 2022). These activities yielded evidence used to inform the design of the overall architecture of FDRI (FDRI, 2022), whilst identifying FDRI's main stakeholders and the key issues to be informed through a more detailed ex-ante (pre-implementation) elicitation of their perspectives. A snowball sampling approach was used to contact potential respondents, which benefitted from FDRI's network of key informants covering the science, industry, and civil society sectors (Gumucio et al., 2021; Saldana, 2021). The sampling was focussed as interviews progressed to represent the key expected organisational sectors of end-user, as identified during the prior consultation activities, which notably informed the need to sample a range of academics to cover different research areas (Fig. 2; Saldana, 2021). Within these groups, we specifically targeted individuals recognised by peers as knowledgeable about hydrological data systems and research infrastructures, with all respondents holding at least 5 years of relevant experience in their sector. Sampling experienced participants allowed for more substantive reflections on design elements and priorities. Interviews of 20 participants took place between November 2023 and March 2024, with sampling continuing until major organisational sectors were represented and where the amount of new information arising in the interviews was low (Saldana, 2021). The FDRI project intends to continue the interviews at more local scales and with more targeted questioning as the infrastructure design becomes more detailed. As such, these interview perspectives represent a first pass of user-priorities, upon which future elicitations and FDRI's corresponding local infrastructure design can be adapted. The participants have been pseudo-anonymised with labels representing their organisational sector and no further identifying information (Fig. 2).

A semi-structured interview approach ensured a consistent structure that addressed key questions, whilst leaving space for emergent information unfamiliar to the interviewer to be pursued through follow-up questioning (Galletta, 2013; Mojtahed et al., 2014). The interview template was informed by prior stakeholder consultations and iterative design within the FDRI team to ensure that questions reflected priority areas for user-input. The questions covered a range of topics, many of which targeted more detailed components of FDRI's operational design, such as training activities, the identification of existing partnerships and the scoping of long-term funding opportunities (FDRI, 2022, 2024). We present results from analysis of a sub-set of those questions, listed below, which more fundamentally interrogated the potential value of the infrastructure for research and innovation and how to optimise that value through a user-responsive design.

The first part of our analysis considers the value proposition of FDRI from the perspective of prospective users. Interviewees articulated four recurring themes of expected value: user community networks, data quantity & quality, testing spaces, and access to innovations (Table 1). In the analysis, we also draw on the literature review and indicate where comparable themes have been discussed in other international studies.

The modally identified value theme of user community networks contrasts with traditional perceptions of monitoring infrastructures as largely generating their value to research and innovation through the datasets they provide. Instead, our respondents emphasise the value generated by creating and engaging in a community of monitoring infrastructure users and contributors. Respondents highlighted that, by creating a focal point to draw together stakeholders from different industries and research backgrounds, monitoring infrastructures can foster innovation when collaborations form among users with unique combinations of expertise [A1, A4, A6, A8, A9, S1, S2]. This emphasis on cross-sector collaboration is echoed in international experiences, where data infrastructures have been shown to support innovation by convening diverse communities of practice (Baron et al., 2017; Peek et al., 2020; Fleming et al., 2024; Holzer et al., 2019; Roy et al., 2020; Sartorius et al., 2024; Harrison et al., 2022; Averyt et al., 2018; Widdicks et al., 2024). Two start-up representatives highlight that these combinations can generate novel approaches that capitalise upon respective partner strengths to identify and address inter-disciplinary knowledge gaps [S1, S2] (Peek et al., 2020). Four interviewees suggested that partners in such collaborative projects address weaknesses by filling expertise gaps and cross-validating methods and results [A6, A2, S1, A4] (Averyt et al., 2018).

bringing in different opinions and ideas from different places is how to truly innovate  [S1]

The value of community collaboration is increasingly recognised by data infrastructure providers internationally (Baron et al., 2017; Peek et al., 2020; Fleming et al., 2024; Holzer et al., 2019; Roy et al., 2020; Sartorius et al., 2024; Harrison et al., 2022; Averyt et al., 2018; Widdicks et al., 2024), as reflected by trends towards investments aiming to facilitate “convergence” and “synthesis” research, supporting collaborations among stakeholders and researchers from different backgrounds (Fleming et al., 2024; Peek et al., 2020; Baron et al., 2017). Eight respondents recommended that FDRI set aside resources to sustain a community integrating data users, providers, and major stakeholders in research, innovation, and water resources management [T1, A2, A4, R1, A6, A10, C2, I1]. The importance of investing in such community-building has also been demonstrated in other infrastructure contexts, where sustained user engagement is critical to long-term scientific and operational impact (Holzer et al., 2019; Prokopy et al., 2017; Sartorius et al., 2024; Gaillardet et al., 2018; Cantor et al., 2021; Henriksen et al., 2018; Peek et al., 2020; Harrison et al., 2022; Tate et al., 2021; Kiese et al., 2018; Widdicks et al., 2024). A stakeholder elicitation for an integrated hydrological data system in California concludes that this community creation is critical even to the sustainability and long-term operation of the monitoring system beyond its initial capital investment (Cantor et al., 2021; Harrison et al., 2022):

Ensuring that an environmental data system is sufficient, accessible, useful and used hinges on meaningful, ongoing relationships with data users  – (from Cantor et al., 2021)

In the second theme, respondents identify the evident value of data quantity and quality for state-of-the-art research and innovation. Six interview respondents particularly highlight the value of open access to high-quality, long-term baseline monitoring [A1, A2, A3, T1, A5, N1] (Cantor et al., 2021; Widdicks et al., 2024). Co-locating a large range of hydrological parameters at high resolution enables interrogation of novel research questions enabled by unprecedented levels of data access and complementarity [A1, T1, N1] (FDRI, 2022). The presence of long-term benchmark datasets also creates ideal testing spaces for the deployment and validation of innovative methods, models, and technologies, which respondents believe can catalyse their development [A2, A3, A4, S1, S2, A7]. Three respondents suggested that, if access to innovations of hardware, software or methods can then be shared within enabled user communities and innovation spaces, synergistic value should be generated for researchers, innovators and other monitoring infrastructure users [A4, A7, S2]. Connected communities can share innovations [S2], jointly address mutual challenges such as land access or telemetry [A4, A7], and their collective research and innovation outputs can generate publicity, new partnerships and opportunities for funding [A2, A9, S1] (Widdicks et al., 2024).

Next, we identified the specific digital and monitoring products and services that prospective users identify as priorities to deliver on expected themes of value (Table 2). Because user elicitations are typically iterative across space and time, we treat these findings as a first cross-sectional input to national infrastructure design to be refined in subsequent rounds (Braud et al., 2020; Cantor et al., 2021; Widdicks et al., 2024). Interview respondents notably discussed whether monitoring should be provided by the infrastructure or collected by FDRI's user community with enabling support. We analyse this discussion point further as a key structural design principle in Sect. 3.3.

Cross-cutting themes emerged from the interviews that extend beyond question-specific findings, supported by evidence from the literature review. These emphasised the need for hydrological monitoring infrastructures to move beyond traditional models where providers act mainly as data collectors, proprietors and distributors, towards designs that actively engage and support their user communities. Respondents and literature alike highlighted that such engagement can expand data availability, strengthen research and innovation outcomes, and improve long-term sustainability (Widdicks et al., 2024; Cantor et al., 2021). From this analysis, we identify three structural design priorities for hydrological data infrastructures, which are examined in the subsections that follow.

Firstly, our respondents emphasise that monitoring infrastructure requirements are local-context specific, influenced by, for example, pertinent issues in the local catchment, local climates, pre-existing stakeholder activities and local capacities [C1, C2, A7, A3, A5, A6, A7, A8, A9, N1]. As such, they recommend iterative, finer-scale user elicitations during their rollout to adapt the infrastructure design to local requirements. The recommendations from local user elicitations should be reviewed alongside the preferences of non-local researchers, who may prefer alternative monitoring or support arrangements towards more generalisable research themes. In such cases, having infrastructure-facilitated spaces for discussion (such as workshops and online forums) can discover and prioritise areas of mutual interests, as well as areas where suitable compromise is required in infrastructure design [S1, C2, A3, A7, A8]. Periodic evaluations should be continued indefinitely to respond to dynamic user needs and set up long-term “adaptive management cycles” (Braud et al., 2020; Cantor et al., 2021; Widdicks et al., 2024).

Second, respondents widely made recommendations to complement the provision of core datasets with additional services that are enabling of data collection where possible, through a suite of data collection support for its community of users and contributors [T1, A9, I2, A1, A2, A3, A4, R1, A7, S2]. An ex-ante elicitation of Nordic stakeholders for a web-based flood management tool reached a similar finding that, by supporting monitoring among an infrastructure's entire user community, data collection capacity can be expanded far beyond that of the central institution with its internal funding capacities alone (Henriksen et al., 2018; Kruczkiewicz et al., 2021). Respondents believe that community-led monitoring is also more likely than centrally-led monitoring to address relevant data gaps according to the dynamic data needs of local infrastructure user communities [C1, C2, A3, A5, A7], which is a view shared by recent studies (Kiese et al., 2018; Harrison et al., 2022; Widdicks et al., 2024). However, three respondents and the authors of this study emphasise that investments supporting data collection are contingent on having sufficient monitoring capacities, motivation and incentives to participate among stakeholders in each hydrological catchment [C1, R1, A6]. We also suggest that infrastructure providers consider whether expenditures on these enabling services will have opportunity costs, such as reducing the coverage of their provided datasets, when deciding how to allocate resources. Therefore, the extent to which monitoring responsibilities can be decentralised is context-dependent and in many cases the transition may be a gradual process, where infrastructure providers are expected to “take the lead” through demonstrative priority monitoring installations [R1, A8, S1, A5, A6, A7] that deliver local value and deepen user community engagement, while they work to gradually develop data collection capacities and incentives among local infrastructure users [A5, A6, A9, C1, C2, N1, I1]. Recommendations of how infrastructures can provide data collection enabling support, principally by addressing the barriers to field data collection, are outlined in Sect. 3.3.1.

Third, in line with the expected value generated by the creation of an active infrastructure user community (Table 1), there is a clear recommendation for active support that enables networking, sharing and collaborations to catalyse research and innovation among users. Recommendations for specific support enabling collaboration and innovation are analysed in Sect. 3.3.2.

Our results indicate that adding services enabling data collection and community innovation can substantially increase engagement, contributions, and the longer-term impact of hydrological monitoring and research infrastructures such as FDRI. In Fig. 4, we conceptualise this effect through a model visualising a user-enabling hydrological monitoring and research infrastructure.

In the model, the infrastructure provider's inputs of funding, coordination and operational resources sets up a range of services to catalyse data collection (as in Sect. 3.3.1) and research and innovation (Sect. 3.3.2) among the infrastructure's user community. We show these integrated digital, monitoring and support services within the Venn diagram (summarised from Sect. 3.2 and 3.3), which deliver value towards the community members' objectives (as defined in Sect. 3.1). Benefits from these services incentivise a range of return inputs considered by other studies to be critical to the infrastructure's long-term sustainability (Cantor et al., 2021; Peek et al., 2020; Gaillardet et al., 2018; Holzer et al., 2019; Harrison et al., 2022; Widdicks et al., 2024). These include contributions of data and equipment by the user community to expand the monitoring network, as well as new results, methods and technologies from associated research and development activities. Over time, some users will have a willingness to pay for appropriate services such as data storage, telemetry or data analytics to support the infrastructure's cost recovery. Evidence of value will also attract additional finance options, such as research grants, public funding, private industry contributions, private equity for innovations, and options for debt finance if revenues approach or exceed operational expenditures. Collectively, this is expected to support a sustainable financial model for continuing long-term operation, which may be a combination of public and private funding, supported by revenues from paid services.

These priority areas reflect a growing demand for monitoring infrastructures that better enable two-way engagement with their user communities. This demand for “enabling” support and two-way exchange reflects the improving capacities of decentralised hydrological stakeholders, who want to take more active roles in monitoring and associated research and innovation. Our findings reflect UK-based key informant recommendations from a range of professional and locational contexts, as well as references from international case studies in high-income countries. They are based on a relatively small number of national stakeholders (n=20), purposively sampled for breadth and expertise during this formative design phase, and should therefore be interpreted as a first iteration of user priorities to be complemented by future rounds of engagement at more local scales. As such, we caution that specific infrastructure design priorities may differ significantly in other contexts, especially in low- and middle-income countries or elsewhere where there is less external capacity available for user community-led monitoring, research and innovation activities. This underlines the need to conduct unique user-centred design activities prior to the design and implementation of any new hydrological monitoring infrastructure to tailor services to contextual requirements.

Once operational, a mutual realisation of value for infrastructure users and providers improves the infrastructure's sustainability through continued respective contributions. These inputs can generate multiplier effects, whereby contributions towards the infrastructure's growth and improvement increase its value offer, engagement and subsequent contributions over time (Cantor et al., 2021). However, this is contingent on a continuous incorporation of user feedback to keep the value offer relevant and adapted to temporally and spatially evolving user requirements. Channels of feedback should be built into operational services for their periodic evaluation and adaptation (Braud et al., 2020; Cantor et al., 2021).

Given the potential for enabling monitoring infrastructures to grow, and the capacities of their user communities to increase over time, infrastructure providers should consider options for eventual decentralisation of services operation to user community members (Cantor et al., 2021; Widdicks et al., 2024). For the infrastructure provider, this will alleviate the staffing and cost burdens of service provision, whilst for decentralised stakeholders, adopting new responsibilities can improve the quality of local infrastructure services, improve organisational reputations, increase local user engagement and generate similar multiplier effects (Fogarty et al., 2025). The extent to which different infrastructure services can be decentralised, the benefits, and the associated risks of doing so require further research. Subsequently, we now plan to complete more localised and longitudinal user elicitations for FDRI, as well as catchment-scale pilot projects, to generate evidence and recommendations for the longer-term evolution of its operational structure and governance.

These future developments will also consider how FDRI positions itself within the wider ecosystem of research infrastructures, now that its objectives and major design principles are becoming better defined. Broader initiatives such as eLTER, Horizon and the European Strategy Forum on Research Infrastructures (ESFRI) are examples that offer important opportunities for alignment, particularly through shared protocols, data sharing and interoperability (Ohnemus et al., 2024). While FDRI's initial remit diverges by being more targeted towards hydrological extremes, which addresses a specific monitoring, research and practical challenge in the UK, its design principles resonate with broader international debates on monitoring and research infrastructure design (Nasta et al., 2025; Widdicks et al., 2024). By engaging with similar projects and aligning with broader frameworks where appropriate, FDRI can deliver on its immediate national priorities, while retaining the flexibility to evolve its role and integrate more closely with international research agendas over time towards addressing shared research agendas in the future (Brantley et al., 2017).