Optimising NFM in headwater catchments to protect downstream communities

Lead Research Organisation: Durham University
Department Name: Geography

Abstract

Natural flood risk management (NFM) describes methods of modifying hillslope and catchment runoff through the modification of landscapes to restore natural hydrological behaviour which limits downstream flood risk. This proposal is for a focussed study of NFM benefits associated with wider ecosystem restoration work which is under way across the uplands of the UK. Headwaters comprise 60-80% of the length of most river systems and high slopes and high rainfall mean that they are important areas of hillslope runoff production. Across the UK there are communities which are prone to flash flooding from steep upland catchments. These headwater catchments are relatively small catchments and are areas where extensive upland restoration is occurring and so they are locations where positive impacts of NFM measures are likely to be observed. Vulnerable communities in headwaters are often small and dispersed with land values that rarely justify hard engineering flood defences through standard cost-benefit approaches. If relatively low cost upland restoration approaches can mitigate risk to communities such as this then it will be possible to provide some protection to communities where funding precludes hard engineering approaches.
This project will work with project partners Moors for the Future and Greater Manchester, Merseyside and Cheshire Environment Agency who have existing funded NFM work in the southern Pennines to undertake a series of field experiments. These will assess the potential impact of various forms of gully blocking, restoration of Sphagnum cover on moorlands, and establishment of upland woodlands on hillslope runoff production and channel flow. It will also assess the longer term evolution of woodland and gully blocking approaches through the study of mature woodland and well established gully blocked systems. This is an important consideration since investment in NFM works requires confidence in the long term impact of the restoration on runoff and knowledge of any ongoing maintenance costs for the interventions.

Installation of NFM schemes to mitigate flood risk requires careful planning and prediction of potential impacts. This project will develop conceptually sophisticated but computationally simple models which can run multiple scenarios in order to assess the catchment wide impacts on runoff of NFM measures implemented to a variety of designs and in a variety of spatial configurations.
The model will be developed (with input from project partner CH2M and input from potential users such as EA) and validated using data from the Glossop Brook catchment in Derbyshire which has a history of major flash flooding impacting households in the town of Glossop. The modelling approach will then be used to assess possible NFM interventions in the upland catchments draining to 21 communities at risk on the eastern edge of Greater Manchester. In each of these catchments we will model the optimum configuration of upland restoration measures for NFM benefit.

The project will also work with partners (Environment Agency, Natural Resources Wales, Scottish Environmental Protection Agency, International Union the Conservation for Nature) to identify existing headwater flow records across the UK which relate to areas of significant upland restoration. At these sites we will model expected impacts and interrogate the available flow data for evidence of these effects on runoff.

The project will work with its range of project partners which span England, Wales and Scotland and which comprise regulators, land managers and industry to develop guidelines to optimise future implementation of NFM measures in headwater catchments across upland Britain.

Planned Impact

Increasing frequency of floods with devastating impact on communities has meant that there is widespread interest in approaches to mitigating the impacts of heavy rainfall on downstream settlements. The efficacy of natural flood risk management (NFM) is therefore a pressing area of concern for a wide range of stakeholders including those charged with flood defence and policy (e.g. Environment Agency, Scottish Environmental Protection Agency, Natural Resources Wales and local councils), land managers, environmental consultancies, and communities at risk.
Successful completion of this project will provide a major advance in the understanding of the impact of upland landscape restoration (such as woodland planting and peatland restoration) on downstream flood risk. The project will also develop models to allow optimisation of NFM benefits at the planning stage of restoration works. The project will develop open source modelling approaches which will allow rapid assessment of multiple scenarios of NFM implementation for headwater catchments. These models, developed with input from project partners CH2M and the EA, will be freely available. They will be designed to support and supplement existing industry standard approaches, and could become a part of the standard toolkit for rapid assessment of projects by regulators and consultants. These advances will be of direct use to policy makers in terms of planning new flood defence works, and assessing the catchment scale impact of NFM measures in headwater catchments.
The project will develop guidelines for the optimum implementation of NFM works. We will work with project partners to co-produce these guidelines, and disseminate these to the widest possible range of stakeholders on completion of the project. The guidelines will cover both the design of NFM features and the spatial planning of these approaches. These planning principles will support land managers in the development of upland restoration schemes which will maximise NFM benefits alongside other ecosystem service benefits.
Quantification and prediction of NFM benefits will allow implementation of flood protection schemes based on NFM approaches in areas where hard engineering approaches are too costly, and will also provide the basis for assessing mixed schemes where NFM benefits are able to minimise the degree of hard engineering required for a given protection level. Therefore, there is potential for the implementation of low cost headwater NFM measures that will allow a greater degree of flood protection within limited budgets.
Ultimately, the aim of this work is to positively impact communities at risk of headwater flooding. Demonstration and quantification of NFM benefits associated with upland land restoration may allow the implementation of meaningful flood defence in communities where largescale hard engineering solutions will not be supported. This would empower communities to take local action to protect their homes. The project will produce a travelling exhibition of project findings which will be displayed in local settings to communicate the project's work to communities at risk.

Publications

10 25 50
 
Description The project has completed pre-intervention monitoring of runoff from erosion gullies on peat moorland on circa 20 sites in the peak District to the east of Manchester. The study of a large number of sites has demonstrated that despite considerable natural variability the degree of change in runoff behaviour in response to moorland restoration seen in a previous study by the team is beyond the range of variability seen at the sites in this study. This validates the conclusions of the previous work and allows us to hypothesise that the restoration measures will generate measurable change in runoff behaviour and so potentially contribute to downstream flood resilience. Installation of the restoration works was competed just ahead of lockdown 1 so that there is almost a year of post-intervention data currently being analysed.
Our empirical work has demonstrated that re-vegetation of bare peat sites leads to slower runoff and reduced peak discharge from degraded peatland sites. We hypothesised that this was because water ran off more slowly across rough re-vegetated surfaces.
Our recent modelling work has confirmed that this is the most probable mechanism underlying this observation which has clear implications for NFM for downstream communities.
Modelling and observations of individual cobble dams suggests that they have little hydraulic impact on peak discharge but that small modifications to their design could yield large increases in their ability to attenuate peak flows.
Work on restored sites on Kinder Scout has demonstrated that the spread of sphagnum on restored sites leads to significantly longer lag times.
Our recent paper on the pre intervention data from our Stalybridge site demonstrated an important control of rainfall intensity on runoff generation.
Exploitation Route Too early to say, final data collection has taken place for the intervention study and the results are currently being analysed
Sectors Environment

URL https://research.reading.ac.uk/nerc-nfm/
 
Title A generalised continuous time implementation of Dynamic Topmodel 
Description TOPMODEL is an efficient and widely used rainfall-runoff model that assumes, given a rainfall, some parts of the catchment will manifest similar runoff behaviour independent of their specific spatial location (and groups them together to form Hydrologically Similar Units allowing massive simulation speed-up. Dynamic TOPMODEL relaxed one of the strongest and most limiting assumptions within TOPMODEL (steady-state recharge) by incorporating a time-dependent kinematic formulation of the governing equations of the subsurface storage and fluxes. However, this improvement comes at a relatively high computational cost. We have made two contributions to TOPMODEL design. First, we have implemented Dynamic TOPMODEL in MATLAB, reformulated the governing equations and used off-the-shelf solvers to create a continuous-time implementation of Dynamic TOPMODEL that is both more accurate and faster than existing discrete-time versions. Second, we have generalised dynamic TOPMODEL by introducing a number of modifications: iso-basin spatial discretisation, diffusion wave routing, depth-dependent overland flow velocity, relaxing the assumption of water-table parallelism to the ground surface, a power-law hydraulic conductivity profile, new unsaturated zone flux, and a reference frame adjustment. The code is published on GitHub with an accompanying journal article (Goudarzi et al., 2023, Water Resources Research) that demonstrates the value of the approach. 
Type Of Material Improvements to research infrastructure 
Year Produced 2023 
Provided To Others? Yes  
Impact This tool is enabling us to perform faster more accurate model simulations to examine the influence of Natural Flood Management interventions on catchment discharge. We have performed these numerical experiments at the micro-catchment (i.e. sub hectare) scale to understand how interventions alter hydrological processes and at a flood relevant scale for the Glossop catchment (40 km2). We are now expanding the number and scale of catchments that we examine using the same approach. 
URL https://github.com/SalimGoudarzi/Generalised-Multistep-Dynamic-TOPMODEL
 
Description Environment Agency 
Organisation Environment Agency
Country United Kingdom 
Sector Public 
PI Contribution Protect team are working with EA to monitor impact of EA funded NFM work
Collaborator Contribution EA are providing advice, access to data and field support for hydrometry, as well as attending regular project meetings
Impact None to date
Start Year 2018
 
Description Moors for the Future 
Organisation Moors for the Future Partnership (MFF)
Country United Kingdom 
Sector Public 
PI Contribution The team are working closely with Moors for the Future on selecting sites for field experiments which Moors fo the future will undertake restoration on. The team are providing expertise to Moors on experimental design and data analysis for this project and other work. The team are working with Moors to jointly publish data from this and previous projects
Collaborator Contribution Moors are playing a major role in planning field experiment locations and managing relations with landowners.
Impact Shuttleworth, E., Evans, M.G., Pilkington, M., Milledge, D. Walker, J. & Allott, T.E.H. (2018) Blanket peat restoration delays flows from hillslopes and reduces peak discharge. Journal of Hydrology X Goudarzi, S., Milledge, D.G., Holden, J., Evans, M.G., Allott, T.E., Shuttleworth, E.L., Pilkington, M. and Walker, J. (2021) Blanket peat restoration: Numerical study of the underlying processes delivering natural flood management benefits. Water Resources Research, 57(4), p.e2020WR029209.
Start Year 2012
 
Description Gully block location mapping 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Third sector organisations
Results and Impact Salim Goudarzi and David Milledge undertook a project to identify optimum locations for gully blocking in the South Pennine Moors SAP. This work was undertaken separately from the NERC PROTECT project and was funded by the Moors for the future partnership (MFFP) as a discrete piece of consultancy. However, our experience in the PROTECT project directly influenced both our decisions in developing tools to identify block locations and our description of how these block locations should be interpreted. In that respect PROTECT had impact through us as beneficiaries and knowledge intermediaries as we passed on the benefit/knowledge to practitioners through a series of outputs (optimum block location maps, a report on how these were generated and how they should be interpreted and a series of meetings to co-produce an approach and report its results). It is our understanding that these outputs have influenced operational decisions in locating gully blocks: 7800 blocks will be installed as part of this project and MFFP hopes to install a further 100,000 over the next 10-20 years. The MFFP team gave a webinar summarising what this research meant to them and how they plan to use it that can be accessed here: https://www.moorsforthefuture.org.uk/our-work/our-projects/weg-building-blocks-next-steps-in-gully-blocking

Testimony by email from Sam Dixon at MFFP on 12/4/2021: "We are absolutely using the outputs from the model we installed almost 7000 gully blocks under the project and plan to do about another 4000 this year! The model has been fundamental in guiding us to areas for gully blocking"
Year(s) Of Engagement Activity 2019
URL https://www.moorsforthefuture.org.uk/our-work/our-projects/weg-building-blocks-next-steps-in-gully-b...
 
Description NFM Webinar 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Salim Goudarzi presented our research findings to an audience of 40-80 people in a webinar that was part of the NERC NFM Webinar series. The presentation was focussed on our findings specifically and was advertised as such. The attendees were an even mix of academic researchers and professional practitioners, some from third sector organisations (based on affiliations or the nature of their comments) but may also have inclued interested members of the general public. The presentation sparked questions that were shared in the chat and addressed afterwards.
Year(s) Of Engagement Activity 2020