Flooding and cherry picking


I don’t get flooded, living on the side of a hill. However, the River Aire some 145m below me flooded its valley at the end of October 2000 (see photos (1)) when the flow rose dramatically to 250 cubic metres per second, over 100 cubic metres per second more than its average flow (2). A day after the flooding, and the impact reached even us hill dwellers when two of the nearest bridge crossings in the valley were closed. I had to get over into the next river valley down to teach an evening class and so playing safe, I went upriver towards Keighley before reaching one of the bigger crossings, and then headed south. The people of Stockbridge next to the river where I turned were less lucky. At 3.00am the day before, the river had broken its banks, eventually flooding 292 households (3). A few had warning of the flooding, most did not. People were shocked at the power, speed and depth of the floodwaters. By 10am, the local authority had set up an emergency response hub at a nearby Leisure Centre, and the Social Services Manager for older people in the area took the lead role in responding to the crisis.

Civil society worked very effectively over the first days and weeks: finding accommodation for any who needed it; meals provided for the first three weeks; meetings held every day for the first four weeks with attendance as high as 400 people; eight editions of a information bulletin produced and distributed over the first month; and many staff and volunteers brought in, principally from the local authority, but also from a range of service providers and public services. The aftermath, and how the neighbourhood coped with it has become an exemplar, as has the approach of the local authority in recognising the need to review the flooding, its causes and its potential remediation (3).

Independent inquiry on flooding

The rural affairs policy team in the Council made much of the early running in that review process. Chris Baines was a keynote speaker at the Rural Renaissance conference held in January 2002 (4). He talked about the rural landscape being redesigned to bring about better water management, so that the flooding of the previous years could have been avoided by paying for the land to function as a part of flood protection. Thus he advocated that broadleaved woodlands could be planted in uplands, and farmland in lowlands could be allowed to flood as seasonal and permanent wetland. At the time, I had left landscape design and was a casual researcher with rural affairs, carrying out a study of rural perceptions, listening to the District’s rural land users as they described their business, aired their pressing issues and looked to their future (5). They were not entirely used to being asked what help they needed from the Council, but nevertheless welcomed the attention, especially since farming was still depressed from the outbreak of foot and mouth disease in 2001. I gave a presentation on the local rural economy at that Rural Renaissance conference, followed by a workshop.

The following year, the Council commissioned an inquiry by an independent expert to undertake an analysis of information from agencies involved in water management, and the public affected by flooding, in order to support future policy development and implementation (6). The inquiry was led by Professor Richard Ashley, professor in urban drainage at Bradford University. It ran from January to May 2004, taking oral and written evidence, as well as holding four stakeholder meetings. After a period as a rural business adviser with Business Link in the Chamber of Commerce, I was back working in rural affairs on rural proofing at the time of the inquiry, and found myself volunteered to be on the contact list of staff to be called in, in event of there being another flooding emergency. I went along to the Land Managers meeting of the inquiry, and I asked Prof. Ashley about the effectiveness of tree planting as a means to reduce flooding. He said that there were increased infiltration rates from the presence of trees, and that there were known calibrations available to assess the impact on flooding of new areas of tree planting.

The work on rural proofing coming to an end, I left rural affairs to return to landscape design, but the lessons of land use and flooding stayed with me. A second rural conference came up in October 2004, and so I put together a view of the District’s rural landscape and its public goods and services (7). The subtitle of the presentation was LAND WATER TREES. I used spatial data (drawn by hand!) on access land; agricultural land classification, less favoured areas, and predominant farming land use (dairy holdings on the better valley bottom land, sheep having precarious viability on hillsides and moors); average annual rainfall (highest into the west of the District – the Pennine upland); the reservoirs and watercourses, and showing flood risk areas; and woodland coverage in relation to those flood risk areas, noting a low (<5%) woodland cover in the District and flooding risk in all riparian habitats. I also made a detailed proposal for controlling bracken on my local moor through rewooding of those areas with native species, an approach I had seen used up in the Bolton Abbey Estate. The feedback from the workshop was positive, exploring new woodland as a way in to developing a view and strategy for the public goods and services of the rural landscape of the District, by considering the examples of tree planting in riparian habitats for flood mitigation and for the control of bracken on moorland. We also talked about exploring the potential of planning obligations as a means of raising funds in the District for organisations to carry out tree planting identified in action plans arising from the strategy.

The inquiry report was published in April 2005. Disconcertingly, it concluded that there were, at that time, no statutory duties for the provision of flood defence development and management (6). Worse still, the key agencies of the Council, the Environment Agency, and the Internal Drainage Board, all operated their functions under permissive powers that required them to raise funds first if they sought to use these powers for new developments, a difficult path under budget constraints. There was also a reluctance to use these permissive powers to take necessary actions when private landowners for example, were placing others at flood risk. The report observed that the then practice of moving water through the landscape quickly in order to prevent flooding would need to be reconsidered, given the recent experiences in the District.

More rewarding was that the discussions that there had been over land use had made it into the report. It noted lessons from the Upper Wharfedale Best Practice Project about high concentrations of sheep and cattle grazing reducing vegetation height and root mass, accompanied by surface compaction and puddling of soils, leading to increased water runoff, soil erosion, river bank erosion and an increased risk of flooding. On the other hand, tall vegetation reduced water run off in a number of ways: it stops soil capping, protects the soil from freezing or drying out, increases porosity and encourages earthworm populations, and impedes water flow therefore slowing down runoff and increasing infiltration.

In its recommendations for land use, the report noted that changes in land management on the hills and valleys within the District could help reduce intensity of rainfall runoff during periods of intense or prolonged rainfall. Examples would be reducing animal stock numbers; the planting of woodlands in the upper catchments to create habitats that will hold more water; planting of wet woodland in the floodplain, including narrow valley bottoms in the uplands and the wider main valleys. Such woodland could help retain the floodwaters as the tree trunks, buttress roots, and deadwood on the valley floor slow down the movement of water; and the development of wetland habitats including meadows, lakes and ponds to store water during times of flood.

On river management, the report recommended managing the accumulation of gravel and debris at key points on the river or watercourse, as well as reducing erosion of river and gill banks. Erosion is a natural process that is accelerated by a reduction in bankside vegetation caused by livestock grazing, removal of trees or heavy trampling of footpaths and rights of way. The remedies given were the provision of specific stock watering access points on riverbanks; the use of fencing and hedges to separate off buffer zones; and footpath repair work using natural techniques such as woven willow. In time the willow will grow and the roots will help provide long-term bank stability.

Sad to say, I’m not aware of any land use changes that have occurred in the District in response to the inquiry report, although it is still an aim in the recent Aire Catchment Flood Management Plan that work still needed to be done with landowners and others to change the way land is managed to reduce runoff (8). Unbeknownst to me though at the time, wild nature, through the aegis of birds, had already been planting trees into the bracken on my local moor, the annual seeding-in of rowan becoming more apparent as the saplings grew larger each year. It continues now, a stark testament to the cessation of sheep grazing that happened after a thousand sheep were slaughtered in July 2001 from being identified as potentially having had "dangerous contact" with a foot and mouth infected area (9). The dreadful Friends group now see this natural tree regeneration as a threat, and are already manufacturing an excuse for its destruction (10).

Yorkshire in transition

I kept an interest in flooding and river management after the inquiry, following schemes in Yorkshire that sought land use changes in mitigation. A useful morning was spent in Buckden with Martin Davies, property manager for the Upper Wharfedale and Malham Tarn area of the National Trust's Yorkshire Dales Estate. Martin was keen to reduce the build up of gravel shoals in the Wharfe that affected bank erosion, as well as replacing the hard engineering on banksides, allowing greater dynamism of the river, but reducing erosion in the longer term. One approach to the gravel was to tackle it at source, reducing its passage down into the river by planting up gill woodlands in an attempt to trap it. Thus 1.5ha of Buckden Beck Gill were fenced off and planted with native broadleaves (and see (11)). The effectiveness of this would need to be monitored over the timescale of woodland establishment (5-25 years). A more immediate exigency in allowing greater dynamism was the spatial unpredictability of erosion, anathema to farmers, and which would require some bankside fencing to keep livestock back. Martin’s solution was to offer tenants additional land elsewhere in compensation if they signed up, but at the time he had only succeeded in getting one to agree.

The inevitability of the intransigence of farmers to consider what are small losses in productive area is a constant theme. This was the case for a study designed to evaluate and demonstrate the contribution of floodplain woodland to flood alleviation on the River Laver that drains into the River Skell near Ripon in North Yorkshire (12). Modelling results demonstrated that planting woodland at four sites in the River Laver catchment, totalling an area of 40 ha, delayed the progression of a 1-in-100 year flood by almost one hour. The aim was thus to facilitate the establishment of a sizeable area of floodplain woodland (15ha) to help reduce flood risk, but the outcome was frustrated:
“Despite the positive findings from the modelling work, the landowners proved unwilling to submit an application for planting floodplain woodland at any of the identified sites and a decision was taken to close the project (after 15 months). The main reasons given by the landowners are described, the most important of which was the lack of sufficient payments/incentive to compensate for the perceived reduction in capital value of the land and loss of agricultural income, as well as for the increased risks associated with land use change”

Another project has been Slow the Flow, finding ways of alleviating the serious flooding that took place in Pickering in 2007 (13). A key factor in the selection of the Pickering Beck catchment for the project was the nature of land ownership, as the experience gained from the Ripon work had shown that land owners were generally resistant to implementing the identified measures at specified sites due to the reasons given above. It was essential to achieve changes on the ground at locations where they could be expected to deliver the greatest benefits if the project was to demonstrate how land management interventions could work to reduce flood risk. The advantage of the Pickering Beck catchment was that around half of the land was either in public ownership (by the Forestry Commission and the North York Moors National Park Authority) or owned by the Duchy of Lancaster Estates, providing greater flexibility and likelihood of success. The project has been able to go ahead recently, with the start of works on the construction of a water storage reservoir on Duchy land that will hold back over one hundred thousand cubic metres of water in Newtondale at times of peak flow, the Forestry Commission planting trees on its own land, and the use of woody debris dams, timber bunds and other means to block the water running off the moors (14). It was less than hoped for, as the final report of the project explains (13):
“The greatest challenge has been creating new riparian and floodplain woodland with only 4.1 ha (riparian woodland) being delivered in the Pickering Beck Catchment. Planting of riparian woodland was significantly constrained due to landscape and biodiversity factors, while financial considerations were key in the lack of landowner interest in floodplain woodland”

Recent opportunity mapping in Yorkshire identified 35,328ha of the fluvial floodplain that is potentially available for woodland planting (15). A further 18,730ha were identified where riparian woodland would help reduce flood flows through planting in the upper reaches of most rivers, including those catchments in the Southern Pennine Fringe at high flood risk but with limited potential for planting floodplain woodland.

Lack of riverine woodland

All this really begs the question of the lack of floodplain woodland in Britain. Historical clearance of riverine woodland has eliminated most true alluvial forests. Forests dominated by ash, elm and oak can be found on the alluvial floodplains along the courses of the larger rivers in many countries in Europe, including Belgium, France (Allier, Garonne, Loire, Rhône) Germany (Elbe, Danube, Rhine) Italy (Tagliamento) Poland and the Netherlands (16). Both France and Germany have about 14,000ha of this woodland type, whereas Italy has 11,600ha. I asked Keith Kirby, lately of Natural England about this, and he said it is a type of riparian woodland that may once have occurred quite widely in the lowlands of England, probably along the Trent, Ouse and Thames Rivers, but was cleared away for agriculture even more efficiently than woodland at the tree-line level in the uplands!

Keith suspected that if there may have been some remnant of it in places, such as the New Forest, then it would likely have been counted in to a different, more variable riparian woodland that compromises ash and alder with willow, and which is found along lowland and hill water courses where they are periodically inundated (17). Many surviving patches of this type are fragmentary and often of recent origin. The New Forest has about 260ha of this ash-alder riverine woodland, less than 1% of its area (18,19). The total area of this woodland type for England is around 2,500ha, but there is little confidence in the accuracy of that figure (20) which is rather depressing when a small country like Belgium has 20,800ha, France in its lowland areas has 87,000ha, and Germany has 67,800ha (17). The scarcity, small stand size, fragmentation and consequent isolation of this alluvial woodland type in England reduces the opportunities for interaction between sites, creating doubt that all of the component species can perpetuate themselves (20):
“Concern about fragmentation is heightened by the scale of loss of this habitat, compared to its natural status and given that its area is at an historical low. Alluvial forests undoubtedly did once cover a much greater area than at present. Map 3.3.1 shows the extent of the riverine drainage system in part of southern Britain. At least small stands of the type would have occurred along most of the courses of rivers and major streams. Larger expanses would have occurred on major floodplains. The resource would have been largely connected, albeit that it often formed narrow, linear stands”

Given the events of the last few weeks (21) there is a certain painful irony that the map referred to shows the streams and rivers, major floodplains and fens of the Somerset Levels. You may not be surprised to learn that the River Parret catchment that drains the Somerset Levels, was one of the earlier areas in which opportunity mapping was carried out for floodplain woodland (22).The greatest scope for woodland was found in the middle and upper reaches, with a total of 27 large (>50 ha) 67 medium (10-50 ha) and many small sites (<10 ha) being identified, equating to 4,880ha or 21% of the entire fluvial floodplain.

More recently, I came across the doctoral studies of Simon Dixon at Southampton University, funded by the Environment Agency (23). Simon has used the New Forest for his modelling studies, as there is access to long-term hydrometric data, as well as a fine detail digital terrain data. His River Management blog has a very useful article about how wood in rivers affects flood risk, a method that is being used by Slow the Flow project at Pickering (see above). He has tracked individual pieces of wood over 3 years, finding that small pieces of wood can move a very long way and are highly mobile. However very large pieces of wood are very stable, with wood 2-2.5 times the channel width in length functionally immobile, and thus acting as an effective site for trapping smaller pieces of large wood moving in the flow. His modelling suggests that just inserting wood and/or logjams into rivers within a catchment, as part of flood control for a downstream location, is highly unpredictable. However, he says there are real benefits for flood control coming from restoring floodplain forests. Complex forested floodplains dramatically slow water moving over them as they have an irregular surface covered by tree roots, upright tree trunks and dead wood (23):
“The most promising scenarios, and the real take home message is the restoration of floodplain forests to entire “sub-catchments” of the main catchment (a tributary of the main river and all streams draining to it) always decreases flood peak height after 25 years growth, and can have dramatic effects. If this is done for areas of 20-35% of the catchment reduction in flood peak height of 10-15% are modelled after 25 years of forest growth”

Claims and counterclaims

It’s been an unedifying spectacle watching the claims and counterclaims about flooding, especially about the efficacy of dredging rivers (24). Few have been immune from the temptation of cashing in – see the Wildlife Trusts (25) but it has not helped when academic colleagues take it upon themselves to be the all-knowing arbiters, a presumption of wisdom when their knowledge of ecological restoration is non-existent. Their accusations of cherry picking and conflation fall hard when they arise from prejudice in support of their naked attempt to justify funding on the back of an issue that only became alive because of the efforts of others at least engaged in the issues, rather than in the silo mentality of academic arrogance.

On a wider point, I fear the conservation industry has made a cross for itself (and us) to bear in its fetishising of species and habitats in degraded landscapes, such that large areas of the uplands that are Special Protection Areas for wading birds will preclude any loss of open-ness through woodland planting. More proof, if needed, of how the conservation industry has blighted any real debate about our uplands. But it is not just the uplands – the opportunity mapping for woodland in the fluvial floodplain of the Parret River catchment was severely constrained by the presence of a large chunk of the Somerset Levels and Moors SPA and the North Moor SSSI, open wetland sites designated for breeding waders and wintering Bewick swan (26,27). While you would think that the New Forest would be predominantly woodland, over half of it is an open landscape of predominantly heathland (34%) followed by grassland (10%) and wetland (10%) (18). The designation of the New Forest SPA for heathland birds, such as Nightjar, woodlark and Dartford warbler, as well as the presence of waders in the wetland areas (28) will ensure a restriction on woodland expansion there too.

I saw Prof. Richard Ashley popping up on television recently, commenting on the flooding. He’s since moved to the Pennine Water Group at Sheffield University, but is still knee deep in floodwater, so much so that he is fed up with Governments not listening to the science, and then cutting budgets at great risk to people in communities at risk of flooding: “For them to rubbish the Environment Agency is frankly disingenuous and stupid" (29). Its not just the ignoring of science that is giving us hard lessons, it is the lack of a policy future that is also making us stupid if we don’t, like Martin Varley, writing about the flooding in the Lake District in 2009, consider relinquishing our total hold on landscapes (30):
“Currently, upland management is contributing to the problem of lowland flooding, but the remedies may be neither palatable or practical. Perhaps it’s time for a radical rethink of what our uplands are more valuable for – food production or flood protection – so that we can prioritise our policies to help them give us what we want”

Mark Fisher 16 February 2014

(1) Flood! Multi-Story Water. Celebrating Shipley and its waterways. 23 June 2013


(2) A New Year’s Eve Flood History, Multi-Story Water. Celebrating Shipley and its waterways. 31 December 2012


(3) Wilkinson, D., Warburton, D., Porter, S. & Colvin, J. (2005) Joining Up: Stockbridge pathfinder, Science Report E2-057/ Environment Agency September 2005


(4) Baines, C. (2002) Future Connections. In Rural Renaissance Conference – Post Conference Pack, Sustainability Team, Policy Development Service, CBMDC, February 2002

(5) Rural Aspirations: some impressions and observations from Bradford District’s rural land use community A report by the Rural Affairs Development Team, Policy Development Service City of Bradford Metropolitan District Council March 2002


(6) Review to Consider the Future of Water Management and the Associated Problems of Flooding in the Bradford District. Bradford Council 2005


(7) Fisher, M. (2004) A view of Bradford District’s rural landscape and its public goods and services. Workshop at the Rural Conference, 21 October 2004


(8) Aire Catchment Flood Management Plan Summary Report. Environment Agency December 2010


(9) Baildon Moor sheep to be culled, telegraph & Argus 14 JULY 2001


(10) Friends of Baildon Moor, Facebook 2014


(11) River Management Techniques. 1.2. Gill habitat regeneration. Upper Wharfedale Best Practice Project (Information Series No 6)


(12) Nisbet, T.R. & Thomas, H. (2008) Project SLD2316: Restoring Floodplain Woodland for Flood Alleviation. Final Report Department for Environment, Food and Rural Affairs Flood Management Division, 10th June 2008


(13) Project RMP5455: Slowing the Flow at Pickering. Final Report April 2011 Defra FCERM Multi-objective Flood Management Demonstration project


(14) Work to start on long-awaited project to protect flood-hit town, The Northern Echo 31 December 2013


(15) Broadmeadow, S. and Nisbet, T.R. (2010) Opportunity mapping for woodland to reduce flooding in the Yorkshire and the Humber region, Forest Research Monograph, 1, Forest Research, http://www.forestry.gov.uk/pdf/York_and_Humber_flooding_final_report_2009.pdf/$FILE/York_and_Humber_flooding_final_report_2009.pdf

(16) Habitat code:91F0. Habitat name: Riparian mixed forest of Quercus robur, Ulmus laevis and Ulmus minor, Fraxinus excelsior or Fraxinus angustifolia along the great rivers. Habitats Directive Article 17 Reporting. EEA


(17) Habitat code:91E0. Habitat name: Alluvial forests with Alnus glutinosa and Fraxinus excelsior. Habitats Directive Article 17 Reporting. EEA


(18) The New Forest. NATURA 2000 Standard Data Form. JNCC http://jncc.defra.gov.uk/ProtectedSites/SACselection/n2kforms/UK0012557.pdf

(19) New Forest Wetland Management Plan 2006 – 2016, Forestry Commission LIFE02/NAT/UK8544 April 2006


(20) Conservation status assessment for : H91E0: Alluvial forests with Alnus glutinosa and Fraxinus excelsior (Alno-Padion, Alnion incanae, Salicion albae) Second Report by the United Kingdom under Article 17 on the implementation of the Directive from January 2001 to December 2006


(21) Winter Floods 2013/14. House of Commons Library Standard Note: SN/SC/06809


(22) Nisbet, T.R. & Broadmeadow, S. (2003) Opportunity mapping for trees and floods. Final Report to Parrett Catchment Project. Wet Woodland Group, December 2003. Forest Research


(23) Dixon, S. (2014) How wood in rivers affects flood risk. The River Management Blog. 17 January 2014


(24) Flooding and dredging - a reality check. Chartered Institution of Water and Environmental Management February 2014


(25) Time to restore nature's flood defences. The Wildlife Trusts


(26) Somerset Levels and Moors SPA


(27) North Moor SSSI


(28) New Forest SPA


(29) UK floods: urban water professor urges Government to listen to science. University of Sheffield News 12 February 2014


(30) Flooding, farming and the future. Martin Varley. Geographical 11 August 2011



www.self-willed-land.org.uk  mark.fisher@self-willed-land.org.uk