Natural River Recovery- thoughts from the virtual beaver and Stage Zero workshops, RRC Conference 2020

Guest blog - Fiona Bowles, RRC Vice-Chair

As usual, the parallel workshop sessions at the ‘all new online’ 2020 RRC conference were frustrating - there were at least three that I really wanted to attend. Luckily, delegates can watch the others now and I just sat down to compare the two newest approaches to restoring natural river processes, spurred on by a visit to Otterhead Lakes. We have been talking about floodplain reconnection for years but the new Stage Zero approach (from NW USA) really delivers that - as do the Eurasian beavers that have been gradually expanding their range throughout the UK since 2008 (under licence and on their own paws) when the Scottish trial was licensed. Both workshops were well attended and attendees shared their experience of planning and delivering projects with those interested in trying them out. Both approaches effectively hand over river processes to nature - allowing the joint processes of flow, geomorphology (e.g. gradient, geology) and biology to determine the river corridor, rather than an engineer. Both effectively reset the clock on river valley landscapes that have altered since mankind started reclaiming the floodplains for agriculture.

What will they look like?

Otterhead Lakes, was historically developed as a series of ornamental lakes in the estate, on the source of the River Otter in the Blackdown Hills, Somerset. It was then developed as a water resource for Wessex Water, with just two lakes remaining. A ‘Stage Zero’ restoration of this small and well wooded valley might involve removing the redundant lake walls and blocking the river channel and some tree felling so as to encourage restoration of wet woodland and anabranching channels in the flatter reaches. The beavers, who moved upstream from the River Otter Beaver Trial are busy restoring some of the old lakes by damming the broken weirs and hatches and rapidly re-flooding the valley to its multi-lake historic form. Being a none arable headwater, there is perhaps less silt available here than on other trial sites but the Alder and Willow woodland which developed as the lakes dried out will provide organic matter. Wessex Water and University of Exeter are monitoring the changes and benefits and the rate of change by one family of beavers is astounding. The site already highlights some of the common features of these approaches; the uncertainty of the outcome and the challenges of monitoring a rapidly evolving mosaic of habitats. Standard channel metrics won't wash.

Ellen Wohl,  Professor of geoscience in Colorado University hosted a webinar last week on ‘Messy Rivers - the role of spatial heterogeneity in sustainable river ecosystems’. She is researching the impact of Stage Zero schemes in NW USA and you may have read her book on beavers too. She mentioned her surprise at the lack of trees in the UK. Here it is largely through 2000 years plus of farming, whereas in Colorado it followed Elk population explosions after predators died out and is more recent. The resultant drying out of Rocky Mountain river valleys is exacerbating their terrible forest fires at present. In the UK, our tree free cultural norm may be a serious handicap to our accepting ’natural’ river valley landscapes. Unlike the USA, we now have few natural reference sites remaining and water is seen only in the context of the river channel, rather than in it's broader continuum within the floodplain and hyporheic zones of valleys. Hence both workshops identified a need for demonstration sites. These, like the Devon enclosed Beaver trial and the Porlock Stage Zero scheme will help to develop evidence of the environmental benefits, inform our management of dis-benefits and engage people with the very different landscapes that result. 

Regulating these natural processes

Leaving design to nature raises issues for ‘permitting' these restoration approaches and for assessing their impact on important habitats and species. Luckily LiDAR-based modelling approaches are being developed which predict the likely extent of habitat change. For beavers there is Habitat Suitability plus Impact modelling and for Stage Zero, the EA Stage Zero feasibility project is developing techniques for catchment wide use. Key to both will be using an adaptive management approach, rather than a classical project delivery plan. 

Pros and cons

Since both approaches recreate a mosaic of wet habitats within a river valley, the workshops identified that they offer the same benefits - increases in carbon sequestration, biodiversity, nutrient capture, reducing flood risk by slowing the flow and raising groundwater levels. Both are primarily effective in the low gradient valleys, and although beaver or logjams can develop in steeper narrow valleys, they tend to be transient there, whereas in U-shaped valleys, they become self sustaining, pushing water out of channel and thus reducing the eroding stream power. The key potential conflicts however are also the same, since very little of our natural floodplains are still available to river systems, especially in lowlands. In addition to the general fear of rivers breaching their banks, there is the potential conflict if these natural processes spread water onto our built environment. The key barrier to both approaches was identified as ‘Agriculture’ (78% of beaver workshop voters), being the primary use of river valley floors today in lowland Britain. The economics of land use change will be key, as will funding long term management in the case of beavers. Fisheries concerns were identified too, exacerbated by the state of salmonid populations and the general concern that logjams or dams act as barriers to migration.

Whats in a name?

One of the final questions in the Stage Zero workshop was whether there is a more suitable name that describes this restoration approach? I liked ‘Messy’ as it describes exactly the heterogeneity that provides a resilient and sustainable river system. However, like ‘Re-wildling’, it may alienate those who like their countryside neat and safe. The top vote went for 'full floodplain reconnection’ although 'Valley Floor reset or restoration' was also suggested. The question will be how much of the valley floor remains today as ‘floodplain’ and whether we can open up more space for water in our river valleys. 'Catchment Rehydration' seems like a good name for those communicating with Drought or Water Resource management planners battling with climate change and growing populations. There seemed to be no need for discussion on what to call beaver projects and the engagement in the NT Purbeck Beaver project and the River Otter Trial has shown that the public are generally very supportive.


River restoration over the 21 years of RRC started small and is still usually reach based so a great plus for both of these approaches is their scale of delivery. That, with the inherent sustainability gained by restoring natural processes, means that they offer techniques to meet the new wider ambitions for Nature Recovery and the new Environmental Land Management Schemes. They both provide flood storage in the uplands and mitigate climate change. Both also need a rich valley floor flora to be effective - for beavers to eat and a rough floodplain to slow flows initially for Stage Zero schemes. Somewhere for the UK’s target 11 million trees perhaps?

The workshops showed great enthusiasm for widening our scales of restoration and a need for more pilots to test the effects across a range of river types. Innovative monitoring is essential to understand the outcomes. Whilst I am keeping my fingers crossed for meeting for real in 2021, in the meantime I will be enjoying the rest of the conference talks and waving the flag for 'making space for water’ during the lockdown webinars on natural capital, ELMS and Nature Recovery .

With thanks to Ben Eardly, National Trust, for Porlock photos



Add new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.