Appendix 4: Reducing risk of harm from environmental hazards

Reducing flood risk


i) Research suggests that surface roughness which is enhanced by maintaining or restoring the ‘hummock and pool’ micro-topography and re-vegetating bare peat surfaces is key to ‘slowing the flow’ and improving ‘within-event’ water storage21,71. Drain/grip blocking may also reduce overland flows but outcomes are highly site-specific and related to the orientation of the drain/grip21. Avoiding bare peat and blocking drains are now common management actions on grouse moors which extend to c423,000 hectares across England’s uplands.


ii) Modelled data suggests that over short time scales and small areas controlled burning may allow easier surface flow72. However, there is no consistent empirical evidence for the impacts of controlled burning on surface micro-topography and the two studies that suggested controlled burning led to an increase in overland flow have been criticised for methodological flaws73.

iii) Overgrazing has been shown to reduce flood management capability72 so sheep stocking levels for GMM must be appropriate for burning intensity, rate of habitat regeneration and local likelihood of severe weather events.


iv) To date it has not been possible to supply a total ‘value’ for the protection and mitigation of flooding from upland peatlands because of the uncertainty of the contribution that peatland, wetted or unwetted, makes to natural flood management (NFM)20.

v) Short term or limited site data shows peat can store large quantities of water (c85-95% of its volume) leading to the misconception that peatlands definitely a beneficial role in NFM when intact peatlands actually have minimal additional capacity to absorb water as water-tables are already close to the surface21,74. No study has fully established a causal link between peatland vegetation management and flooding downstream because defining how much controlled burning, cutting, or grazing is too much in a particular area is difficult given the complexity of factors at the catchment scale73.

vi) Once peatland soils and surface vegetation are saturated there is a risk of over-land flow causing downstream flooding24 suggesting that full rewetting may not always be appropriate in flood-prone catchments given projected increases in rainfall intensity with climate change.

vii) Run-off (and therefore downstream flood risk) may be affected by managing the vegetation by mowing/cutting as it affects the plot micro-topography by removing the tops of hummocks and leveling the peat surface and therefore reducing water retention at least for periods immediately after cutting30.


ix) Water table levels are affected by evapo-transpiration and changes in surface cover and vegetation alter soil water storage and the prevalence of overland flow generation within storm events75. Research in lowland situations identifies water table as the over-riding control on GHG emissions76. Research is needed to determine the height of water table in upland peatland and associated management approaches that best balance a) climate mitigation objectives with b) flooding and c) biodiversity ambitions.

x) Some studies suggest that NFM via peatland restoration is applicable to small (<20 km2) catchments but the evidence at larger catchments is from modeling only with inconsistent evidence across restoration approaches21.

xi) Research21 suggests that peatlands subject to intensive and extensive wildfires have ‘flashier’ responses to rainfall events due to wildfires removing surface vegetation, therefore reducing surface roughness, and creating smooth, hard peat ‘crust’ surfaces.

Reducing wildfire


i) Wildfire threat can be mitigated by reducing fuel load (mainly determined by the volume of vegetation) through controlled burning, particularly on peat dominated upland heath, the creation of fire breaks (by burning or cutting with brash removal77), and perhaps by rewetting (where hydrology and geology permit)8.

ii) Gamekeepers are involved in many local wildfire groups which bring together a variety of stakeholders to address wildfire risk and issues at local level and support the Fire & Rescue Service (FRS) at wildfires. A proportion of wildfires, including out-of-control managed burns, are brought under control by gamekeepers without the need to involve the FRS.


iii) A Natural England Evidence Review78 on wildfires stated that there is “Strong evidence …. that, in the minority of cases when a more specific cause was assigned [in the uplands]... (62, only 10% of all upland fires), …. the majority were assigned to land manager burns (68%), followed by campfires (9%) and barbeques (8%)”. Thus, very few wildfires were certainly caused by farming, conservation or GMM. The review went on to state that interpretation of this data requires care given the small number in the sample and the bias and subjectivity in the assignment as to cause as a result of land managers rarely causing wildfires but typically remaining to fight and control them, while the gross majority of wildfires are caused wilfully or unknowingly by persons unknown.


iv) Alongside increased land-use and recreational pressures, hotter and drier summers will increase the risk of conditions favourable for the ignition and spread of wildfire79. Whilst the chance of a summer as hot as 2018 (when the Saddleworth and Winter Hill wildfires occurred) was low (<10%), by 2050 hot summers could become about 50% more common and by 2080 the combination of higher temperatures, decreased summer rainfall and drier soils could led to a 30-50% increased risk of wildfire in the UK80. Wildfires are episodic, generally occurring in late spring, when the potential fuel consists of dry fine fuels from dead vegetation after winter, and during hot summers81. Regularly removing combustible plant material may be particularly important for drier, shallower peats (less than 1m deep), which can be more vulnerable to wildfire damage than deeper peats82.

v) Wildfires also occur in upland grasslands and forests (Table 2).

Table 2: Size of wildfire incidents (area burnt) by land cover class in hectares, England, 2009 to 2017

  2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17
woodland (NFI)
34 123 62 18 66 55 79 23
Semi-natural grassland 336 884 872 298 658 54 212 656
Mountain, heath
and bog
202 2,824 11,481 318 654 769 823 538
Total semi-natural 572 3,831 12,415 634 1,378 878 1,114 1,216

Source: ONS/Forest Research

vi) Natural bog is not immune from wildfire during periods of drought due to the reduced moisture content in the surface vegetation and peat (physiological drought) and leaf damage83,84 and once ignited even damp peat can smoulder for a long period85.

vii) The negative ecological impact of preventing all fire has been noted in America8 but there has been no detailed research in the UK on the impacts of fire exclusion or the reduction/cessation of other land management60. The removal of or reduction in management under some rewilding approaches (such as the impacts of reducing deer86,87 and sheep numbers) is of concern as this will lead to an increasing vegetative fuel load88. It is speculated that reintroducing megafauna can moderate fire regimes by controlling the vegetation and engineering the soil and litter layer89.

viii) A common approach to restoring peatland is rewetting, usually accompanied by the cessation of vegetation management. Except under rare permanent bog and pool conditions, this will result in a significant increase in above-ground biomass90 and a higher fuel load over extensive areas with the potential to increase the likelihood of a severe wildfire given current predictions91. Rewetted peatland may not prevent the horizontal and downward spread of smouldering peatland wildfires, particularly as they usually occur during the summer when there is reduced moisture content24,32.


ix) Cutting can be used to create firebreaks but the brash should be removed otherwise it can enhance wildfire risk as well as smothering regrowth30. The results of the ongoing research by York University will provide data to guide future cutting approaches92.

x) The costs of wildfires are seldom quantified. As events occur more data is being collected but even these may not fully reflect the total cost per wildfire. Some individual estimated costings have been collated for the Saddleworth Moor wildfire in 2018:

    • £21.1 million cost to economy from health impacts (PM2.5)2.
    • £3.6 million lost through loss of 15,400 tonnes of carbon sequestration capacity93.
    • £1.1-1.6m of carbon emissions (based on estimates of soil carbon losses only (no calculation of carbon from surface biomass was included))94.
    • Up to £1m estimated cost to Fire & Rescue Service (FRS) from managing a typical single, large moorland fire95.
    • £205,000 impact on the local economy of fewer tourist visits93.

xi) The additional ‘costs’ of the Saddleworth fire (1100ha) in 2018 could be:

    • Surface combustion on near natural bogs could release 18t CO2/ha in a wildfire alone96. During the Saddleworth Moor wildfire this could have equated to another 19,800t CO2.
    • Legacy emissions from a degraded site. Assuming the Saddleworth Moor site is the equivalent of drained bare peat, which it is estimated emits 13.28tCO2e/ha/yr97, this suggests that the Saddleworth Moor wildfire event would lead to in excess of 14,500t CO2e/yr being released, at least initially.
    • Post-wildfire effects will include increased erosion. Exposed peat can be lost by wind-blow, surface water flow and frost heave, resulting in losses of 0.8-1.0cm per year98.

Controlling tick-borne disease


i) Graziers in many parts of upland UK need to treat sheep with chemical acaricides to reduce tick worry and tick borne diseases99,100. Tick biting and tick-borne diseases also have an impact on red grouse survival, breeding success and grouse densities thereby impacting on the potential shooting surplus101. Grouse shooting estates monitor tick burdens on shot grouse and test for louping-ill prevalence.  Where tick density is high, an acaricide treatment (and when available vaccination) programme on the sheep grazing the moor may be introduced.

ii) Using sheep as a means of controlling tick-borne diseases has been explored as a policy to support upland sustainability102,103.

iii) Tick populations are found in a wide range of habitats where a moist vegetation layer prevents desiccation and death and vegetation is tall enough to climb up and latch onto a host.

4.3 Figure 2Figure 2: Ticks distribution in England based on Public Health
England (PHE) and Biological Records Centre (BRC) data.


iv) Public Health England, the Health & Safety Executive and some national park authorities provide advice on the risk to human health of Lyme Disease (LD), a high-profile tick-borne disease. The number of LD cases has increased rapidly with an incidence rate of 12.1 per 100,000 persons per year and an estimated total for the UK of 7,738 in 2012; although numbers are probably under-reported104.

v) The number of tick-borne diseases is increasing dramatically, with seven diseases currently posing serious health risks to birds, mammals, and people in the UK. Tick-borne diseases include arborvirus (which includes tick-borne encephalitis and the Flavivirus group); protistans; bacteria (including Lyme disease); tick paralysis; and alpha gal syndrome105. In 2020 Public Health England confirmed the first case of babesiosis and a probable second case of tick-borne encephalitis in England.

vi) Societal costs for LD relate to healthcare costs, but these are not fully understood106. No data exists for England but a 2003 analysis estimated a cost of £331,000 per annum in Scotland107 with case numbers increasing since then108.

vii) GWCT research109 has shown that regular treatment of sheep with an acaricide can reduce the prevalence of LIV in red grouse. The annual number of acaricide treatments is higher if managing for red grouse and sheep rather than just for sheep. GWCT estimates £2/adult sheep/yr based on data from its demonstration farm in Aberdeenshire. It is expected that the number of treatments to protect recreational visitors from tick bites and LD would be similar to that for red grouse tick management.