Review of muirburn impacts on soil carbon and biodiversity
Authors: Willie Towers, Alison Hester, Steve Chapman, Robin Pakeman, Nick Littlewood and RebekkaArtz (MLURI); Colin Legg (University of Edinburgh)
- On balance and based on scant evidence, well managed muirburn can be responsible for limited losses of soil carbon in the long term.
- Some studies suggest that the physical changes to the peat after severe wildfire (not managed burning) can lead to higher rates of erosion
- Modelling suggests that burning in September may carry an increased risk of severe fire and greater carbon loss in some years and in some regions, though carefully managed burning would still be possible in most years. The limited available data did not suggest that autumn fires caused higher carbon consumption than the overall average.
- Little is known about geographic variation in the impact of burning but from first principles and modelling, differences could be expected based on climatic and site characteristics.
- There are no suitable data to assess the impact of extending the burning season on either plant diversity or heather performance and regeneration, but we consider it unlikely to make a big difference to plant regrowth/diversity if muirburn guidelines are followed.
- Burning earlier in the autumn would involve burning heather with a higher moisture content, which can slow the spread of fire and reduce the intensity of the burn. Conversely, burning may only be possible when the moss layer is drier and more likely to burn, which could result in exposure of the soil/peat and potential drying out and erosion. The balance between these possible outcomes will depend on the site and weather conditions.
- A significant proportion of most moorland bird species will have commenced egg-laying within the currently legal dates for muirburn. However, population-level effects of this on bird breeding success are generally unknown.
- Most bird species whose nests are likely to be affected by spring muirburn have ‘unfavourable conservation status’ in the UK and include Hen Harrier, Merlin, Red Grouse, Golden Plover and Ring Ouzel.
- There is evidence that some moorland bird species are breeding earlier in the spring than previously so in areas where these birds are breeding, this does need to be considered in relation to legal muirburn dates in spring.
Impact of muirburn on ecosystem services
Ecosystem Services have been defined in a number of different ways but for the purposes of this short report, the definition adopted in the Millenium Ecosystem Assessment (MEA) in 2005 provides a reasonable overall context. In essence Ecosystem Services are the benefits that people can obtain from ecosystems. These include both tangible and intangible benefits which although sometimes separated into ‘goods’ and ‘services’ respectively (Daily 1997), the MEA refers collectively as Ecosystem Services.
The MEA classifies these services along functional lines using categories of provisioning, regulating, cultural and supporting services. In brief these are described as:
- Provisioning – the products obtained from ecosystems
- Regulating – the benefits obtained from the regulation of ecosystem processes
- Cultural – the nonmaterial benefits, such as spiritual enrichment, reflection and recreation that people obtain from ecosystems
- Supporting – the services that necessary for the production of all other ecosystem services.
Heather moorland and the management system of periodic burning contribute to all of these to some extent. In terms of supporting services, the growth per se of Calluna vulgaris (often referred to as ling orheather) and other plant species that make up heather moorland communities is primary production that can be exploited by livestock grazing and its existence provides the habitat for other ecosystem services. These other services rely on the presence of moorland for them to manifest themselves, for example grouse and some other moorland birds rely on the presence of heather for both food and cover.
Moorland indirectly provides food via grazing animals and other products such as wool are also derived from extensive grazing of ‘domestic’ animals. Wild animals such as red deer and grouse that also graze this habitat similarly provide a source of food for human consumption, albeit not as major component of the national diet. The activity of hunting wild animals may be seen more as a cultural service than a provisioning service. Muirburn aims to maintain and enhance the capacity of moorland to act as a grazing and habitat resource.
Heather also provides the necessary cover to allow the ecosystem to fulfil its role in climate regulation, water regulation, erosion control and water purification; all of these rely on a continuous, healthy and vigorous plant cover to minimise the loss of GHGs to the atmosphere, dissolved organic carbon (DOC) and particulate organic carbon (POC) to water courses and the implications that has for water quality and quantity. One of the objectives of muirburn is to maintain a vigorous sward of heather and thereby enable the ecosystem to provide these services in somewhat vulnerable environments.
Heather moorland provides a number of cultural services although it is here that the value of muirburn can be contested by different stakeholders. For example, it is vital for the maintenance of the moorland ecosystem in its current state, but it could be argued it is detrimental to woodland regeneration and the range of benefits that that could provide over time; or to the development of more natural blanket bog communities, diversity of habitat and associated fauna, increased carbon storage; and increased erosion control. Nevertheless muirburn does help maintain the habitat that provides a sense of place, inspiration, aesthetic values and somewhere that activities such as game sports, recreation and bird watching take place. All of these bring economic activity into places where other opportunities are relatively scarce.
This briefing note outlines the evidence of the impact that muirburn has on two of the ecosystem services that moorland provides; firstly on the carbon stock stored in the soil and secondly on biodiversity, particularly on the birds that utilise it for nesting. How these might change under a changing climate and/or management will be commented on and evidence and knowledge gaps identified.
Daily, G.C. 1997. Introduction: What are ecosystem services? In: Nature’s Services: Societal Dependence on Natural Ecosystems (ed. G.C. Daily), pp. 1-10. Island Press, Washington, D.C.
MEA [Millennium Ecosystem Assessment] 2003.Ecosystems and Human Well-being: A Framework for Assessment. Island Press, Washington, D.C.
Impact of muirburn on soil carbon
Is well managed muirburn beneficial, detrimental or neutral for soil carbon?
Burning of heather moorlands has been practiced for a long time and typically occurs as patches with a return frequency of 10-20 years. There is a range of factors associated with burning that are likely to influence C loss, these may be considered as i) immediate and direct (loss of standing biomass C), ii) intermediate or indirect (subsequent erosion; recovery of vegetation after the fire) and/or iii) longer-term impacts on C cycling processes through influences on primary productivity (species or growth rate) or physical changes such as might be associated with modified water table depth, iv) long term effects associated with gross changes in vegetation type and land use, or long term degradation of peat (Legg et al 2009). Occasionally intense, uncontrolled burning can also remove a variable quantity of the surface organic-rich horizons with much longer implications for the C balance.
Effects on soil carbon dynamics
Studies of the effect of muirburn practices on soil carbon content, cation exchange capacity or peat accumulation rates are very sparse indeed and there appear to be no examples in the peer-reviewed literature that are specific to Scottish sites. Hence, most of the experimental studies presented here have been conducted in England and in fact there are only two studies that we are aware of where soil carbon has been measured directly, one in the north Pennines (Garnett et al., 2000) and the other in the Yorkshire dales (Farageet al., 2009). Additional data on impacts of fire on carbon come from 26 experimental management fires conducted in Scotland (Legg et al 2007; Davies et al 2009), with calculations using the Farange et al (2009) ratios of carbon:organic matter (Legg et al 2010).
The three major areas of concern with regards to rotational muirburn on moorland ecosystems are in the areas of 1) direct loss of carbon stocks due to ignition and consequent loss of peat 2) increased runoff from the site or increased soil erosion and 3) a change in the soil C dynamics with resulting decreases in peat and C accumulation rates alongside alterations in the emissions of gaseous end products of decomposition (CO2 and CH4). Much of the literature relevant to these themes has been reviewed in an English Nature document (Tucker, 2003)
The physical effects of fire on peat soilsare likely to be fundamentally rather similar in nature regardless of UK site location, albeit within the overall context that sites will tend to be wetter the further north and west they sit in the country. Fire intensity and depth of penetration of lethal temperatures into the litter/soil are critical factors in fire impact on carbon and on biodiversity. While the direct ignition or smouldering of peat can cause huge losses of carbon (Rein et al., 2009), the limited available literature suggests that low intensity, well managed, burns are unlikely to generate sufficient heat to cause the peat layers to ignite due to the insulating properties of peat as well as the high moisture content of peat. Ignition of peat can only occur when sufficient of the available water has been lost, and most studies have shown that few fires generate sufficient heat for long enough to elevate soil temperature to more than 100ºC for more than a few minutes (Lloyd, 1968; Hobbs et al., 1983). Laboratory tests suggest that sustained smouldering in peat can only occur where the moisture content is below 125% of oven-dry weight (Rein et al. 2008) and, in most cases, Scottish moorlands are sufficiently moist throughout the current muirburn season to withstand the short term heating without ignition (Legg & Davies 2009). Clearly a problem arises when muirburn is not “well-managed”, where there is potential for actual destruction of the underlying peat. The risk of this occurring, which can destroy thousands of years of peat accumulation in a very short time, has to be balanced against the benefits of muirburn. Indeed, one of the cited benefits of muirburn is the reduction of high fuel loads which may be prone to wildfire, though this argument may be less applicable to Scotland than to England where the risk of summer wildfire is very high in some places.
A muirburn will char or ash a substantial quantity of vegetation, and, in the case of higher severity fires, a fraction of the accumulated litter (e.g. Ward et al., 2007). While this is a short-term loss of carbon, it is likely that this only becomes an issue when the fire frequency is such that a repeated loss of this fraction of future soil C affects the carbon accumulation rate of the site. There are some reports that suggest that removal of thick layers of litter can reduce shading and thus can, in some particular circumstances, encourage regeneration of Sphagnum spp. (Gray, 2006), which has led to the proposed use of muirburn as a tool in peatland restoration in some situations (e.g. Rowell, 1988). However, most available evidence indicates that Sphagnum recovery only occurs where the burn return interval is of the order of hundreds of years, i.e. the frequency of wildfire (Yallopet al., 2009), although this work was carried out in the Pennines where the situation is complicated by high levels of pollution which also has strong detrimental effects on mosses. There are other circumstances when Sphagnum can withstand repeated burning if wet enough.
The literature on changes in soil erosion potential and site runoff has not yet generated a clear consensus. Some studies suggest that the physical changes to the peat after severe wildfire (not managed burning), such as increased hydrophobicity (water repellence), formation of crusts, generation of bare areas of peat, etc. may lead to higher rates of erosion through both increased wind dispersal of crust material and a lower structural resistance to peat slumping in gullies (Maltbyet al., 1980; 1990). There are, however, also reports to the contrary where peat shrinkage due to fire causes higher peat stability. The promotion of Calluna by muirburn as the dominant species in the vegetation may also cause some long term changes in soil physical structure that may make burned sites more prone to desiccation and oxidation (Yallopet al., 2006) and hence lead to higher erosion risk. Similarly, some literature reports the potential for higher runoff from, and/or reduced infiltration on, burned sites due to the higher hydrophobicity of the peat (e.g. Malliket al., 1984; Shaw et al., 1996; Glaves & Haycock, 2005). Others, in contrast, have found no conclusive evidence for long-term changes in losses of C due to runoff (Worrall et al., 2007; Clay et al., 2009). There is some evidence that there may be short-term increases in dissolved organic carbon concentrations in soil water (Clay et al., 2009) but these appear to be short lived.Farage et al (2009) suggested that carbon loss associated with subsequent erosion might be in the order of 5-21 g C m-2, but this based on estimates of reduction in peat surface level following burning that do not take soil compaction into account and may therefore be overestimates (Legg et al 2010).
As suggested, continuous removal of vegetation and/or litter inputs to a moorland may result in lower peat accumulation rates. Ward et al. (2007) and Garnett et al. (2000) showed this at a site (Hard Hill, Moorhouse NNR) in the North Pennines where both carbon accumulation and soil C content in the upper layers of the site was reduced due to burning. Ward et al., (2007) also found evidence of increased soil carbon dynamics in that both gross photosynthesis and soil respiration were higher in burned plots. This is entirely feasible, given that burning returns mineral nutrients to an ecosystem where most plant-available nutrients are generally bound in organic form and thus less accessible. Burning, and specifically the ashing of vegetation, thus returns some of these nutrients in a more readily accessible form to the vegetation. Indeed, plant nutrient status in such burned sites is elevated compared to unburned controls (Ward et al., 2007) and therefore may contribute to a higher photosynthetic capacity in burned plots during the regrowth cycle between muirburn applications. Similarly, increased mineral nutrient availability could conceivably elevate microbial activity thus contributing to increased soil respiration. Increased heather vigour during the regeneration stage will no doubt increase root exudation or carbohydrates into the rhizosphere (the part of the soil adjacent to roots). Therefore, some of the direct losses of vegetation and litter carbon during the burn appear to be alleviated by a higher net C sequestration capacity during the regeneration between burns. At a moorland site in the Yorkshire Dales, Farageet al. (2009) reported minimal losses of carbon due to burning over a 15-20 year burning cycle (100-200 g C m-2 during two seasons of burning) and concluded that, on average, their burned site was still accumulating carbon. However, Legg et al. (2010) have recently commented on this report and conclude that the extent of the burns reported were atypically low and additionally question the appropriateness of the control unburned sites in this comparison. Legg et al (2010) suggest that typical muirburn would result in losses of the order of 400-500 g m-2.
There are some anecdotal reports of an increase following burning in the occurrence of sedge species such as Eriophorum spp., which have aerenchymatous tissue that is known to promote direct methane losses to the atmosphere (Gray, 2006) and Gray (2006) also reported high levels of methane emission from a blanket bog site shortly after a fire. However, Ward et al., (2007) reported a slight reduction in net methane emissions in burned plots nine years after the fire. These limited data thereforeimply that post-fire methane fluxes are likely to varysignificantly through time. Similarly, changes in soil physical structure leading to increased periods of desiccation would logically favour increased potential for methane oxidation, though Chen et al. (2008) found no conclusive evidence for this.
It would therefore seem logical to investigate whether there is a ‘best practice’ muirburn frequency that could be adopted for a given site, based on the likely losses of C during the burn and the carbon sequestration capacity, so that total losses are minimised. Some reports advocate the mixed burning strategy that would allow for the development of a mixture of normal burn cycle as well as long term unburned stands. Such a strategy may favour the maximisation of C sequestration potential as long as fuel loads in the long term unburned stands do not lead to a high risk of a more severe fire.