Assessing Statistical models of Temporary River Intermittence for Decision makers (ASTRID)

Temporary rivers (TRs) are dynamic features of the landscape that transition between hydrological and terrestrial states, providing a range of habitats and ecosystem services. They are important for ecological diversity, nutrient processing and water resources management and adversely impacted by the pressures of climate change and local anthropogenic activities. Despite their importance, their likely prevalence in headwaters and groundwater-fed catchments, and the role of drying as the primary determinant in ecological diversity, they are underrepresented in monitoring networks and mapping. Furthermore, where traditional datasets of gauged flows or network contraction exist, they overlook the identification of ponded water as a distinct and ecologically important habitat. The lack of data means that there is an inadequate understanding of the number of TRs in the UK, their distribution and characteristics.

There is therefore a need to map the spatial extent of TRs, to quantify their behaviour in a way that is relevant for decision makers, and to promote monitoring with optimal use of resources. This project aims to address these needs with three objectives: 1) Engage stakeholders in co-designing metrics relevant for decision-makers; 2) Statistical modelling of intermittence in UK TRs through training and validating; 3) Mapping the characteristics of intermittence in UK TRs.

Each objective will be delivered by a work package. In Work Package 1 (WP1) the core activity is a workshop to allow key stakeholders, most notably Environment Agency water resource hydrologists responsible for decision making on the management of TRs, to engage in the direction of the project and the detailed design of useful deliverables. A literature review to establish the state-of-the-art of modelling intermittence on TRs will be delivered as a brief synopsis. In Work Package 2 (WP2), hydrological state data in England and France, and independent variables including rainfall, geology, topography, river flow will be collected and pre-processed. Statistical models, including parametric and non-parametric approaches, for estimating metrics of intermittence identified in discussion with the stakeholders will be assessed, and uncertainty analysis conducted to assess model performance. In addressing the third objective, WP3 will explore the transferability of selected models to unmonitored catchments using sensitivity analysis. Deliverables will comprise estimated intermittence metrics for TRs throughout the UK, hierarchy maps spanning a number of confidence scenarios and a summary report of the uncertainty analysis.

Knowledge exchange activities will target two distinct groups of stakeholders in regions identified by the modelling as priorities for data collection. Maps and datasets will be shared and discussed with decision makers to seek views on the next steps for TR-appropriate drought and water resource assessment. Secondly, local interest groups will be invited to engage in data collection activities using the existing University of Zurich citizen science app, CrowdWater. Two workshops, one in each priority area will take place, with morning and afternoon agendas tailored to the two groups of stakeholders. Research findings will be published in a high impact journal, and presented at two academic meetings. The first, with a national focus, will enable engagement with ecologists, hydroecologists and river managers with an interest in the study and management of TRs. The second, with an international focus, will be the General Assembly of the European Geoscience Union.

The ASTRID project includes two work packages with a focus on stakeholder engagement (WP1) and knowledge exchange (WP3). A workshop at the start of the project (WP1) will involve stakeholders, notably water resource hydrologists responsible for making decisions on the management of temporary rivers (TRs), in discussion surrounding the direction of the project and the detailed design of useful deliverables. In WP3, we will travel to priority regions, identified by the statistical modelling as priorities for data collection. At one day workshops, we will spend the morning meeting with stakeholders to discuss the mapping results and seek views on the next steps for drought and water resources assessment appropriate for TRs. In the afternoon, we will meet with local interest groups to engage communities in data collection through an existing citizen science app, addressing the lack of a hydrological observations on TRs that has been a limiting factor in their study.

The research will be presented to ecologists, hydroecologists and river managers at a national workshop on TRs and streams (WP3, TRS) and internationally, at the Annual Assembly of the European Geosciences Union. We will publish a paper in a high impact journal, making a significant contribution to the very limited literature on the modelling of hydrological intermittence on TRs.

The potential exists for future mapping that covers a greater diversity of temporary river types with socio-economic benefits in improved balancing of the needs of the environment and people. Models that separate natural and artificial causes of drying for water resource assessment and tailored environmental standards, enforced in legislation, will improve upon the protection implicitly offered by the Water Framework Directive. Following ASTRID, this will provide incentive for the future policy improvements required, tailored to deliver effective management of ecosystem services, drought, water resources and ecological status on temporary rivers.