Community-based ('citizen science') monitoring for catchment characterisation, modelling and management

Abstract

Despite there being well-established meteorological and hydrometric monitoring methods, many smaller UK catchments remain ungauged. This leaves characterisation, modelling, forecasting and management activities a challenge when working on a local level. Many ‘citizen science’ projects are encouraging the public to participate in data collection activities and generate new knowledge across a range of environmental disciplines, but they have not been fully investigated within catchment science. This project has designed and implemented an innovative community-based monitoring scheme within the 42km2 Haltwhistle Burn catchment (Northumberland, UK) to explore the feasibility, reliability, value and sustainability of citizen science within the catchment management process. Like many rural UK catchments, the Haltwhistle Burn responds rapidly, experiences flash flooding, and does not benefit from any traditional monitoring networks. Various simple, lowcost and internet-based methods have enabled the public to collect and share rainfall, river level, water quality and flood-related observations successfully over a 29-month period. This generated a patchwork of heterogeneous catchment information. Although a wide range of people actively participated, 73% of the total number of observations were generated by just four dedicated individuals or households. Despite monitoring efforts being sporadic and unpredictable, rainfall and river level observations were favoured. Participation levels also intensified during high flows or flood events; web-based tools, particularly Twitter, then played an important role in sharing these real-time observations. However, spatial and temporal monitoring efforts are biased towards individual capabilities and interests, and should therefore fill data gaps rather than replace traditional monitoring schemes. Training, ongoing facilitation and feedback help to generate meaningful and good quality data. A traditional hydrometric monitoring network was installed to aid in assessing the quality and value of community-based observations. Examples presented here verify that citizen science can generate high quality data, provided that robust validation and verification measures are in place. Evidence suggests that participants were conscious of collecting consistent datasets, but this does not guarantee reliable data from every citizen scientist. The value of community-based observations have been demonstrated by using them to build and run a physically-based, spatially-distributed hydrological model. Results reveal how the local network of community-based observations, when used alongside traditional sources of hydro- Abstract iv information, supports the characterisation of catchment response more accurately than when using traditional observations alone. Community-derived datasets appeared to be most valuable during local flash flood events, particularly towards peak discharge. Such information is often missed or poorly represented by ground-based gauges, or significantly underestimated by rainfall radar, as this study clearly demonstrates. Community-based observations were also used to tailor the design of a natural flood management (NFM) scheme above the town of Haltwhistle. Post-installation monitoring has revealed that image-based observations collected using simple monitoring methods can provide concerned locals with meaningful and relatable (therefore valuable) information. Such outcomes are important when relieving common barriers affecting the widespread uptake of NFM. It is acknowledged that the long-term retention of volunteers and the sustainability of citizen science is a challenge. The full monitoring period exposed that participation levels escalated, peaked and then tailed off within Haltwhistle. However, the winter 2015/16 widespread floods reactivated mass data collection. Driven by an existing community-led group, an additional case study site in Northumberland (Acomb) also demonstrates how the public want to monitor, acknowledge the benefits of local datasets, and are capable of initiating and funding their own monitoring scheme. Sustainable (long-term) citizen science therefore requires strong leadership and pertinent (flood risk) motivations. Raising volunteers’ awareness on how to maximise the value of their own monitoring efforts will also reduce monitoring fatigue. Furthermore, options have been explored to demonstrate how citizen science can be scaled up to a regional and national level, and be integrated into the existing flood risk and catchment management process. Although the co-production of environmental knowledge is not a new phenomenon, evolving technology and communications provides a timely and cost-effective solution to mass data collection. Without this data, very little information would be available to characterise catchments and implement localised management measures with confidence. This participatory approach also offers the public an exciting opportunity to share valuable local knowledge, gain ownership, and be actively part of the catchment management process. Overall, it is concluded that citizen science and the wider community-based monitoring toolkit should now be seen as a fundamental component of any catchment study. The findings and impact generated as a result of this Ph.D. have therefore made a significant contribution to research in this area, and lay the foundations for future community-based projects.