Calculating carbon savings from wind farms on Scottish peat lands - A New Approach

Listen

Appendix 1: Review of SNH Technical Guidance Note (2003) on Wind farms and Carbon Savings

A1.1. Executive Summary

The 2007 renewable energy policy of the Scottish Government ( SG) has a target of 50% of electricity demand in Scotland to come from renewable sources by 2020, with an interim target of 31% by 2011.

Scottish Natural Heritage ( SNH) has supported the Scottish Government's 2003 target of achieving 18% of electricity generation from renewable sources by 2010, to be increased to 40% by 2020. Large scale wind farm development proposals in Scotland have raised concerns about the reliability of methods used to calculate the carbon savings associated with wind farms, as compared to power derived from fossil-fuel and other more conventional sources of power generation. This is partly due to the siting of wind farms on peat lands as opposed to mineral soils and partly due to changes in the UK electricity generation mix.

Government policy is to deliver renewable energy in a way that "affords appropriate protection to the natural and historic environment without unreasonably restricting the potential for renewable energy development" ( SPP6). Other relevant policies seek to deliver biodiversity objectives, including the conservation of designated wildlife sites and important habitats including peat lands. These are considerations which the planning system takes into account.

Not all peat land sites are of equal environmental value. In general, less man modified peat land sites will have higher biodiversity value, are more likely to act as active carbon sinks and will probably have higher levels of stored carbon. It is therefore likely that development of any form will need to pass higher environmental standards on sites of higher conservation status. It is also worth noting that carbon stored in mineral soils will be significantly less than in peaty soils, with a consequent reduction in the need to address carbon balance issues.

Net carbon emissions from wind farm developments is not, of itself, an issue which the planning system should take into account, beyond a requirement on developers to ensure that projects are "designed to minimise soil disturbance when building and maintaining roads and tracks, turbine bases and other infrastructure to ensure that the carbon balance savings of the scheme are maximised" ( SPP6). Guidance on how to estimate carbon emissions is, however, a useful tool to allow all interested parties to come to a common view on net carbon dioxide benefits.

The purpose of this review is to analyse a wide range of literature and provide an assessment of the SNH guidance material highlighting the areas of uncertainty, and indicating areas where further research and technical guidance is required.

In the guidance note some factors such as CO 2 emission from the full life cycle of the wind farm and from backup power generation are not included. Emissions during material production, transportation, erection, operation and dismantling of wind turbines are significant and the payback time for this alone can add 2 - 24 months, when using the CO 2 emissions from conventional generation as the comparator. However, this comparator does not itself include full life cycle costs for the conventional plant. As a rule, it is expected that wind will compare very favourably with other means of power generation, such as nuclear, particularly when the full life cycle costs are taken into account.

The sources of uncertainty for calculation of CO 2 payback time include the capacity factor, emission factors for other power sectors, carbon fixation capacity of the peat land or forestry, losses from the carbon store in peat lands or forestry and the proportion of the site lost or affected by the wind farm development. These sources of uncertainty should be included as input values while calculating CO 2 payback time and should be site specific. A site specific capacity factor should be used wherever possible to provide a more realistic payback time for the site. Grid-mix emission factors should be updated annually.

Simulation modelling of soil erosion, soil hydrology and C cycling should be used to provide better estimates of the extent of the area affected by a specific wind farm development and the impact on C losses from the soil. Potential reductions in the emissions of greenhouse gases, methane and nitrous oxide, should also be included in the calculation of payback time.

Restoration of disturbed sites can significantly impact greenhouse gas emissions, potentially preventing further loss of and increasing the C stored in the restored habitat, but also possibly increasing the emissions of methane and nitrous oxide. These changes in greenhouse gas emissions should be accounted for in the calculation of payback time. Note that methane and nitrous oxide emissions on flooding are usually very small compared to CO 2 emissions on drainage.

The major pathways of C loss from soil are aerobic decomposition of organic matter, resulting in CO 2 losses; anaerobic decomposition of organic matter, resulting in CH 4 losses; leaching losses of dissolved organic C associated with movement of water through the soil; and losses of C associated with erosion or mass movement of soil. C losses associated with erosion and mass movement are highly site specific and can only be adequately estimated using dynamic simulation modelling driven by detailed site specific data. These are assumed to be negligible given adherence to Best Practice guidance. Aerobic and anaerobic decomposition losses can be estimated using IPCC defaults or more site specific relationships derived from dynamic C turnover simulations. These simulations will also provide estimates of losses as dissolved organic C.

A1.2. Abstract

This report presents a critical assessment of SNH Guidance Document (Windfarms and Carbon Savings, Technical Guidance Note; 2003). This article critically examines the approach used to estimate carbon savings or costs associated with wind farm development on peat lands. The SNH Guidance provides an excellent, easily applied methodology to explore whether a wind farm development is likely to represent a carbon saving or cost. However, in the guidance note, some major factors, such as carbon dioxide emission from the full life cycle of the wind farm and from backup power generation are not included. The present study identified sources of uncertainty in the calculation of carbon dioxide payback time which include the capacity factor, emission factors from other power sectors, carbon fixation capacity of the peat land, loss of carbon stored in peat lands and the proportion of the site lost in the wind farm development. This review provides recommendations for reducing uncertainties in the estimates.

A1.3. Background

The renewable energy policy of the Scottish Government ( SG) is set out in "Securing a Renewable Future: Scotland's Renewable Energy", published in March 2003. This states "Scotland should aspire to generate 40% of its electricity from renewable sources by 2020." The target has been revised with the Scottish Government Spending Review (Scottish Government, 2007, p. 54) noting that the national indicator for renewable energy is that 50% of electricity generated in Scotland should come from renewable sources by 2020, with an interim target of 31% by 2011. SG policy is to deliver renewable energy in a way that "affords appropriate protection to the natural and historic environment without unreasonably restricting the potential for renewable energy development" ( SPP6). Other relevant polices seek to deliver biodiversity objectives, including the conservation of designated wildlife sites and important habitats including peat lands.

Scottish Natural Heritage ( SNH) restate the value of renewable sources of electricity generation to tackling climate change but warn that a planned approach is needed to ensure that the right technologies are situated in the right places to minimise adverse impacts on the natural heritage (Scottish Natural Heritage, 2007, p. 36). Scottish Natural Heritage supported the original SG target of achieving 18% of electricity generation from renewable sources by 2010, to be increased to 40% by 2020. However, they noted that any attempt to meet a 40% target relying predominantly on onshore wind could result in increasingly difficult trade-offs with natural resources, if the economic environment and planning framework cause these developments to be associated with areas of semi-natural habitats. The SG has more recently highlighted the environmental and biodiversity considerations that need to be included when planning wind farm developments. They place strong emphasis on avoiding developments on important designated sites and priority habitats such as peat bogs - this technical note should therefore be read in conjunction with the National Planning Framework, SNH wind farm guidance and SPP14 and SPP6.

An issue when planning a wind farm development in forests or on blanket bogs is whether

there is potential for overall CO 2 savings when the changes in CO 2 emissions associated with the land use change are included. The SNH Guidance Document (Windfarms and Carbon Savings, Technical Guidance Note; 2003) provides a methodology to quantify the potential CO 2 savings for a wind farm development. However, as mentioned in the guidance note (page 2, para 7), the SNH guidance does not cover some issues associated with the wider environmental considerations in planning wind farm developments on important habitats. In addition, while there may be overall CO 2 savings associated with electricity generation from a wind farm on peat land, despite the loss of natural C stocks from the peats, the potential environmental benefits could have been greatly increased if the development had been sited on a mineral soil, where there would be a lower net loss, or even a net gain of C held in soil organic matter. Peat lands come in many forms and it is also important to recognise that changes in C stocks will differ, depending on the type of peat land, and factors such as its morphology, structure and the condition of the peat. Understanding the C characteristics of peat lands is still in its infancy.

A1.4. Factors not included in the current SNH Guidance

A1.4.1. Carbon dioxide emissions due to production, transportation, erection, operation, and dismantling of the wind farm

A full life cycle analysis of a wind farm includes CO 2 emissions during production, transportation, erection, operation, dismantling and removal of turbines, foundations and transmission grid from the existing electricity grid (Vestas, 2005). The average life time of the wind turbine and the internal cable is between 20 and 30 years (White, 2007). Emissions from material production are the dominant source of CO 2 (White, 2007). The source of energy used during material production (eg. coal or gas-fired) is a key determinant of the CO 2 emitted during construction. Against that must be set the productivity of a wind farm over its lifespan, giving a figure for the amount of CO 2 emitted (in tonnes) for every unit of electricity generated (in GWh).

An industry life cycle analysis of offshore and onshore wind farms (Vestas, 2005) shows an emission of 4.64 t CO 2 GWh -1 for onshore electricity generation in Denmark. This is a non-peer-reviewed industry value, and is probably lower than might be the case in the UK because 69% of electricity consumption for material production was from CO 2 neutral energy sources. However, to date the bulk of turbines erected in Scotland were manufactured in Denmark, so it could be argued that this value would be appropriate for the UK.

A wider review turns up higher emissions. For three wind farms studied in USA (White, 2007), CO 2 emissions ranged from 14 to 34 t CO 2 GWh -1. Lenzen and Munksgaard (2002) carried out a survey of the environmental performance of wind farms all over the world and analysed ~70 studies including various sizes of wind turbines with power ratings ranging from 0.3 to 3000kW. The results showed that the CO 2 emissions varied from 7.9 to 123.7 t CO 2 GWh -1, the very high emissions occurring only in Japan, due to the national fuel mix and analysis methodology. A life cycle analysis of Italian wind farms showed CO 2 emission varied from 8.8 to 18.5 t CO 2 GWh -1 (Ardente et al., 2008). White and Kulcinski (2000) estimated birth to death CO 2 emissions from wind turbines and reported a value ranging from 6 to 15 t CO 2 GWh -1.

If we assume the range of CO 2 emissions to be in the more normal range, between 6 (White and Kulcinski, 2000) and 34 t CO 2 GWh -1 (White, 2007), then we can calculate what additional CO 2 payback time due to production, transportation, erection, operation and dismantling of the wind farm that this represents for the example given in the SNH Technical Guidance Note (2003). The additional CO 2 payback time for the best case scenario i.e. (6 t CO 2 GWh -1) would be 2 months assuming replacement of coal-fired power generation and 4 months assuming replacement of the grid-mix. For the worst case scenario ( i.e. 34 t CO 2 GWh -1), this would increase to 13 months and 24 months additional CO 2 payback time respectively. These increases are significant, and so it is essential that they are taken into account in the calculation of CO 2 payback time for a wind farm. However, it should also be noted that this may still compare very favourably with the life cycle analysis of other means of non fossil fuel based power generation, such as nuclear, particularly when the full energy costs of construction, operation, maintenance and decommissioning and long term waste management are taken into account in relation to the overall C budget.

Note that developments for other forms of power generation (eg. coal, oil, gas or nuclear) are not required to carry out a full life cycle analysis, and the Scottish Renewables Forum considers that requiring this from the wind farm developers would provide a significant competitive advantage to those industries ( 1J.Ormiston, pers.comm). However, because the Scottish Government's 50% target for renewable energy by 2020 is designed to reduce CO 2 emissions, a full life cycle analysis is essential to demonstrate reduced emissions. Full life cycle analysis is recommended for all forms of power generation, as this allows emissions from wind generated electricity to be compared directly to the emissions originating from fossil-fuel generation.
1 Jason Ormiston, Scottish Renewables Central Chambers, 93 Hope Street, Glasgow G2 6LD

A1.4.2. Carbon dioxide emission from backup power generation

Wind generated electricity is inherently variable, providing unique challenges to the electricity generating industry for provision of a supply to meet consumer demand (Netz, 2004). Backup power is required to accompany wind generation to stabilise the supply to the consumer. This backup power will usually be obtained from a fossil fuel source. At a high level of wind power penetration in the overall generating mix, and with current grid management techniques, the capacity for fossil fuel backup may become strained because it is being used to balance the fluctuating consumer demand with a variable and highly unpredictable output from wind turbines (White, 2004). The Carbon Trust (Carbon Trust/ DTI, 2004) concluded that increasing levels of intermittent generation do not present major technical issues at the percentages of renewables expected by 2010 and 2020, but the UK renewables target at the time of that report was only 20%. When national reliance on wind power is low (less than ~20%), the additional fossil fuel generated power requirement can be considered to be insignificant and may be obtained from within the spare generating capacity of other power sectors (Dale et al, 2004). However, as the national supply from wind power increases above 20%, without improvements in grid management techniques, emissions due to backup power generation may become more significant. The extra capacity needed for backup power generation is currently estimated to be 5% of the rated capacity of the wind plant if wind power contributes more than 20% to the national grid (Dale et al 2004). Moving towards the SG target of 50% electricity generation from renewable sources, more short-term capacity may be required in terms of pumped-storage hydro-generated power, or a better mix of offshore and onshore wind generating capacity. Grid management techniques are anticipated to reduce this extra capacity, with improved demand side management, smart meters, grid reinforcement and other developments. However, given current grid management techniques, it is suggested that 5% extra capacity should be assumed for backup power generation if wind power contributes more than 20% to the national grid. At lower contributions, the extra capacity required for backup should be assumed to be zero. These assumptions should be revisited as technology improves.

A1.4.3. Peat landslide hazard

The technical guidance does not include consideration of the risk of mass movements of peat and bog slides ( e.g. Warburton et al., 2004), such as seen at Derrybrien, Republic of Ireland, and their potential effects on carbon stocks. In reporting best practice for reducing risk of peat slides, which may be associated with wind farm development, the Scottish Executive (2006) has established a more comprehensive and rigorous procedure for identifying existing, potential and construction induced peat landslide hazards. This should lead to reduced peat landslide hazard and risk with development infrastructure being re-sited into areas of least hazard. Development consent may be declined due to the level of hazard identified, or where insufficient mitigation measures are proposed. It is stated that where complete avoidance of risk is not possible, the proposed design should be modified to incorporate engineering options for mitigation of risk. The report highlights the risk that any such measures may cause further damage to the peat, but lists a number of measures which may limit damage (Scottish Executive, 2006, pp. 34-35). These include catch ditches engineered into the peat substrate, and the installation of drainage measures to isolate areas of peat susceptible to risk. Such mitigation measures deliver on one requirement in terms of reducing the risk of mass movement of peat, but could potentially increase the level of disturbance, or drying out of the peat and release of carbon. Also, they may not be in the immediate vicinity of the proposed development, rather based upon geotechnical assessments of areas of risk, and as such could greatly increase the area of the peat where C losses may be likely to occur. Peat landslide is potentially the greatest source of CO 2 emissions from the soil. However, if it can be assumed that legislated measures to avoid peat landslides have been taken, so that the risk of peat landslide is minimal, then this potential source of loss can be emitted from the guidance.

A1.4.4. Restoration of the site

The SNH Technical Note does not include the impacts of restoration, or after-use of wind farm sites in areas of peat or peat land. The requirements for after-use planning include the provision of suitable refugia for peat forming vegetation, the removal of structures, or an assessment of the impact of leaving them in situ. If peat land is drained on construction of the wind farm, losses of CO 2 will continue until the soil organic matter reaches a new stable state. For peat soils, this may be close to zero ( IPCC, 1997). Restoration of the site could potentially halt these loss processes, allowing CO 2 emissions that occur up to the time of restoration to be included alone. This more realistic estimate of the CO 2 losses associated with the wind farm development could be very much lower than those calculated assuming 100% loss. Therefore, it is imperative that these practices be included in appraisals of potential C loss. Note that a peat land site is usually not in a steady state with respect to soil C even before the wind farm is constructed. This may be due to the impacts of previous land uses, long term adaptation or climate change. We are not suggesting that the peat land should be reinstated to be stable with respect to C stocks where that was not the previous state. However, if the site is reinstated so that any changes in CO 2 emissions directly attributable to the wind farm development can be assumed to have been halted, then the additional emissions of CO 2 from the soil due to the wind farm development can be limited to the period before restoration of the site.

A1.4.5. Change in grid-mix and payback time

When calculating the time required for a wind farm development to pay back the C emissions associated with its construction, operation and dismantling, it is important that the appropriate level of CO 2 saving be assigned to the electricity generated by wind energy. This depends on the type of energy generation displaced by wind energy. It can be argued ( e.g.BWEA, 2005) that increased energy generation from renewable energy sources in the UK represents a displacement of energy produced by coal-fired plants alone. Data published by The National Grid Company (National Grid Transco, 2004), describing the distribution of energy generation in the UK power system, demonstrate that in the day-to-day running of the national grid, coal plants tend to be taken off load when additional base load plants, such as nuclear or renewables, start to generate. It is also suggested ( BWEA, 2005) that, in the long-term, as older coal plants become uneconomic, these will tend to be replaced by the new energy plants. Clearly the proportion of the national grid generation displaced from the different energy sources depends on complex economic drivers, but it is the thermal plants (coal, gas and oil) that are currently used for grid balancing. The level of CO 2 saving will be somewhere between the higher level assuming displacement from coal-fired plants only, and a lower level obtained by comparison against the current fossil-fuel generated grid-mix. Therefore, in the absence of a detailed economic analysis, the technical guidance should allow calculation of payback time assuming both displacement of coal-fired plants and the fossil-fuel sourced grid-mix.

In the SNH Technical Guidance (2003), the payback time is calculated against the emissions from coal-fired plants and grid-mix, not fossil-fuel sourced grid-mix. The grid-mix emission factors will be misleadingly low, as they also include nuclear and renewable sources of power. Furthermore, between 2000 and 2005, the proportion of Scotland's electricity generated from coal fell from 33% to 25% (16624 to 12160 GWh y -1), with gas and oil reducing from 11283 to 11014 GWh y -1 over the same period. This will affect the pay-back time calculated for wind farms if comparison is made to the current grid-mix.

It is unclear whether the fossil-fuel sourced grid-mix used should be for Great Britain, or for the United Kingdom (given the exchange of electricity between Scotland and Northern Ireland, via the North Channel inter-connector). If Scotland is self-sufficient in electricity, as measured by the difference between generation and consumption, it could be argued that the fossil-fuel sourced grid-mix should be for Scotland alone. The technical guidance should use the current government / industry agreed value for fossil-fuel grid-mix.

A1.4.6. Carbon Implications of Forestry

Average yield class for forestry

In their Environmental Statement for the proposed Kyle wind farm ( AMEC, 2004), AMEC note that a major source of uncertainty is the average yield class for forestry. In the SNH technical guidance, the amount of biomass produced is assumed to be 16 m 3 ha -1 y -1, compared to the value of 14 m 3 ha -1 y -1 provided by the Forestry Commission. AMEC comment that the values used by SNH overstate the calculations of C sequestered in woodlands, thus exaggerating the negative impacts of the wind farm. The technical guidance should require site specific entry of yield class, so allowing a more accurate assessment to be made of the annual C sequestration.

Other factors associated with forestry

European forest soils contain approximately the same amount of C as is found in tree biomass (Smith et al, 2006). The presence of extensive areas of forestry on, and in the vicinity of, the wind farm site can significantly reduce the yield of wind energy so it has often been the practice to clear existing forestry from the area surrounding the site prior to wind farm development. The cleared land has then often been left as open ground. However, to reduce the long term loss of woodland alternative approaches should always be considered - such as 'key-holing' combined with replacing felled trees with short rotation coppice/ short rotation forestry or low-height native woodland. The losses of C from the tree biomass depend on the fate of the wood product following felling. For example, if wood is removed from the site and used in furniture production, C is stored in the wood for longer than if wood is used in biomass heating. Forestry may be felled earlier than planned due to the wind farm development, so limiting the nature and longevity of wood products. However, over time, most of the sequestered C will be emitted from the wood products, irrespective of the development of the wind farm, so this issue is of less importance than the fate of the considerable stocks of C in the soil. Managed arable and grassland soils store significantly less C than soils under forestry or semi-natural management (Smith et al, 2005; Smith et al, 2006). The change in soil C stocks on tree removal will depend on the subsequent land management, especially the drainage regime. In addition, if a forestry plantation is due to be felled with no plan to replant, the effect of the land use change is not attributable to the wind farm development. These factors accounted for in the technical guidance.

A1.5. Uncertainties in the current SNH Guidance

A1.5.1. Capacity factor or utilisation rate

The capacity factor is the ratio of actual energy produced in a given period of time to the hypothetical maximum amount of energy that could be produced if the turbine was running full time at the rated power. The rated power of the turbine is the energy the turbine will produce per hour of operation, when running at its maximum performance ( i.e. at high wind speeds above ~15 m s -1). For wind generated power, the utilisation rate is usually limited by the capacity factor of the wind farm rather than by user demand. For this reason, capacity factor and utilisation rate are assumed here to be synonymous. Globally, the average capacity factor of a windpower plant varies from 16% (Windstats News Letter, 2006) to 30% ( DTI, 2006). Global capacity factor statistics provide an average capacity factor for different countries; in Spain the average capacity factor is 24.6%, in Denmark 24%, in Germany 16%, in Sweden 19% and in the UK 28.2%.The average capacity factor between 1998 and 2004 for Scotland was 30% ( DTI, 2006). The UK average load factor for wind in 2006 was 27.4% (Annual tables: Digest of UK Energy Statistics ( DUKES)). However, the capacity factor used in the SNH Guidance is even higher, at 40%. The capacity factor is a key determinant of the payback time for a wind farm. In the SNH guidance, the payback time for an example 10 MW wind turbine, assuming a capacity factor of 40%, was 1.45 and 2.63 years if the alternative to wind generation is electricity generated from coal-fired capacity and grid-mix respectively. However, for the same wind turbine, the payback time would be more than doubled at the lowest capacity factor in the range, increasing to 3.63 and 5.69 years if a capacity factor of 16% is assumed. The average value for Scotland ( i.e. 30%) would give a payback time that is 33% higher than that given in SNH guidance, increasing to 1.93 and 3.51 years for the same 10 MW turbine. The high capacity factor used in the SNH guidance is likely to result in the payback time being significantly underestimated. In practice, this figure is seldom used in environmental impact statements. The capacity factor is usually measured before the decision to develop a wind farm at a particular site is taken. Use of this site specific capacity factor is recommended as it will greatly improve the calculation of payback time.

A1.5.2. Emission factors for other power sectors

The SNH guidance uses CO 2 emission factors for energy related emissions from the guidelines for the measurement and reporting of emissions by direct participants in the UK Emissions Trading Scheme ( DEFRA, 2002). Over the short term, this is a good approximation because energy related emissions do not vary significantly from year to year. However, potential improvements in future technology to reduce emissions from other power sectors may reduce the emission factors, so increasing the calculated payback time associated with wind generation of electricity. A DTI study suggests that by 2015 the national emission factor across all power sectors could drop to 0.3 tCO 2 MWh -1 as electricity generation from renewable sources increases and old coal fired plants are phased out under EU Directive ( IPA, 2005). A reduction from the current emission factor of 0.43 tCO 2 MWh -1 to 0.3 tCO 2 MWh -1 for grid-mix would increase the CO 2 payback time for the 10 MW turbine cited in the SNH guidance from 2.63 to 3.77 years. This is equivalent to a 43% increase in the calculated payback time.

This is, however, an overly pessimistic view as it can be assumed that the contribution from renewable sources would not be displaced by other renewables and should not therefore be included in the reduction in the emission factor. Furthermore, grid-mix figures include the contribution from nuclear power which, for technical reasons, is not affected by renewable energy generation. Depending upon the results of discussions between the BWEA and the ASA, the current BWEA proposals should be considered, using either the average fossil fuel mix, revised on an annual basis (2006 figure) or a range calculated from a 5 year average. The 2006 emission factor based on estimated CO 2 emissions and electricity generation from each type of supply for 2006 from the Digest of the United Kingdom Energy Statistics (2007) is 0.627 tCO 2 MWh -1 (627 tCO 2 GWh -1 in table A1.5.1). The 5 year average emission factor calculated using estimated CO 2 emissions for 2002 and 2003 from the National Atmospheric Emission Inventory (Baggott et al, 2007) and for 2004 to 2006 from the Digest of the United Kingdom Energy Statistics (2007) is 0.607 tCO 2 MWh -1 (607 tCO 2 GWh -1 in table 1). An arbitrary proposed range of ï'± 10% gives a range from 0.546 - 0.668 tCO 2 MWh -1. An increase in the emission factor from the grid-mix value used in the SNH Technical Guidance (0.43 tCO 2 MWh -1) to the fossil-fuel mix of 0.607 tCO 2 MWh -1 would decrease the CO 2 payback time for the 10 MW turbine cited in the SNH guidance from 2.63 to 1.87 years. This is equivalent to a 28.8% decrease in the calculated payback time.

1Emission
factor
(t C GWh -1)
Emission
factor
(t CO 2 GWh -1)
2Electricity generated by each supply (GWh)3Emission factor accounting f or the share of the total fossil fuel supply (t CO 2 GWh -1)
200620062002200320042005200620022003200420052006Ave.
Coal239876118475131760125689128600142681382411388399438404
Oil16159042174171409444664271999999
Gas101370148870145134153734149280138253203191201196180194
All fossil fuels
17246312715622810652835172823462852055956115986044627607

Notes
1 DUKE Stats (2007) Table 5C, p.120.
2 DUKE Stats (2007) Table 5.6, p. 136.
3 Assumes emission factors as f or 2006
4 Note the dif ference in emission factor calculated due to rounding errors

Table A1.5.1. Calculation of emission factors from fossil fuels from 2002-2006

A1.5.3. Carbon- fixation capacity of peat land

The major factors controlling the C cycle in peat lands are plant community, temperature, water table depth and the chemistry of the peat. The estimated global average for apparent C accumulation rate in peat land ranges from 0.12 to 0.31 tC ha -1 yr -1 (Turunen et al., 2001; Botch et al., 1995). The SNH guidance has taken an average value of 0.25 tC ha -1 yr -1, which is within the range of the literature. However, the accumulation of C in peat is highly site specific. This factor is of importance in considering the improbable but worst-case scenario for loss of C fixing potential where C fixation in the entire peat area is lost due to the wind farm development. In the SNH guidance for calculation of payback time, the loss of C fixing potential of the peat land is not included because it is considered to be insignificant. However, if the capacity factor of the wind farm is low, then the amount of C loss through loss of C-fixing capacity can have a significant impact on the payback time. For example in the SNH guidance with an assumed 40% capacity factor, to pay for the loss of C fixation capacity of the peat land for 25 years ( i.e. the average life time of the wind farm), a 10 MW wind farm will take 2-4 months. This increases to 3-6 months with the 30% capacity factor estimated for Scotland between 1998 and 2004 ( DTI, 2006). These calculations assume that the C-fixing capacity of the peat bog would have remained constant without the development of the wind farm and that it returns to the initial state of the peat land immediately after decommissioning of the wind farm. In some cases, regeneration of vegetation could require significantly more time than this, whereas in others, regeneration of vegetation will have occurred during the operation of the site. Indeed, in some cases, there may be an opportunity, through habitat management plans, to improve the status of the peat bog compared to the baseline scenario. These factors are not insignificant and should be accounted for in the planning and habitat management for the site.

A1.5.4. Proportion of site lost in wind farm development

For a peat land habitat, most wind farm environmental statements would describe two potential types of habitat loss arising from the wind farm development:

1. direct loss of habitat at the site of construction, whether this be roads, turbines or construction infrastructure;
2. indirect loss or degradation of habitat caused by drainage effects arising from the sites of constructed features.

Direct Losses

The direct effects of site construction include loss of C from the excavated peat, the amount of C lost depending on the fate of the removed peat and the characteristics of the soil. Direct loss of habitat at the site of construction is a very small proportion of the site. For a turbine spacing of 400 m between turbines (6.5 turbines km -2), with a turbine base area of 20 m x 40 m, in the worst case scenario where all turbines are sited on peat, the area directly affected by the turbine bases would be only 0.5% of the peat land habitat, with a similar additional proportion lost to roadways. However, in practice a regular close grid distribution of turbines is rarely achieved. At the range of sites considered in table A1.5.2, the average density of turbines was only 2.5 turbines km -2, which would result in only 0.2% of the site being lost assuming the same turbine base area.

Wind Farm

No. of turbines

Height to tip (m)

Site Area (ha)

Density of turbines (turbines km -2)

Hare Hill extension

39

44 - 91

1279

3.0

Achany

23

105

1279

1.8

Invercassley

23

100

1669

1.4

Kildrummy

8

93

286

2.8

Drumderg

16

107

~440

3.6

Dunbeath

23

125

~1300

1.8

Fallago Rig

48

110 & 125

1050

4.6

Lewis

181

140

~25000

0.7

Average

2.5


Table A1.5.2. Density of turbines at selected wind farms in Scotland ( 1A. Coupar, pers.comm.)

1A. Coupar, pers.comm. Policy & Advice Manager, Uplands & Peat lands, Scottish Natural Heritage, Great Glen House, Leachkin Road, Inverness, IV3 8NW

Indirect Losses

Indirect loss of habitat mainly occurs due to drainage of the site during construction and operation. Recent best practice guidance from the Scottish Executive (2006) draws attention to the high conservation value of peat lands and their susceptibility to changes in hydrology, and under the terms of Section 1 of the Nature Conservation (Scotland) Act, 2004, the Scottish Government and planning authorities are required, in all their decisions, to have regard for the Scottish Biodiversity Strategy and UK Habitat Action Plan targets, which include no loss of the current habitat extent and an improvement in the condition of what remains. Hence, even where there is a direct loss of peat land associated with any development, considerable input is required to identify the environmental impact and mitigate during and after development to ensure minimum deterioration of the hydrological conditions of the site, so reducing potential indirect effects.

Drainage can result in a reduction in the level of the water table, loss of habitat structure and subsidence of the peat surface (Lukkala 1949). The reduction in the level of the water table increases the aerobic volume of the surface peat and leads to increased CO 2 emissions through enhanced aerobic organic matter decomposition (Silvola 1986; Moore and Dalva 1993; Silvola et al. 1996). Hogg et al. (1992) suggested that C losses from peat following a reduction in the level of the water table depend on the quality of the peat. Peats that have previously been exposed to long periods of aerobic decomposition may be resistant to further decomposition (Bridgham & Richardson, 1992). A peat land with a water table 20 cm or more below the peat surface for most of the summer might not exhibit a significant increase in C release when areas of the peat are drained during wind farm construction. Many forested bogs and fens would have such a water regime. By contrast, in a peat land where organic strata near the surface have been continuously inundated, the peat containing highly labile organic matter is likely to decompose at increased rates when the surface is drained.

Extent of Drainage

The efficacy of drainage is related to the depth of ditching, distance between ditches, and the hydraulic conductivity of the peat (Boelter, 1972; Armstrong, 2000). The reduction in the water level is greatest close to the ditch, and diminishes rapidly with distance, depending on hydraulic conductivity. For example, Prevost et al. (1997) found that drainage was most effective within 15 m of ditches in a drained tree-covered bog near Riviere-du-Loup, Québec, whereas Boelter (1972) found ditches were effective at up to 50 m in fibric peat in Minnesota, but ineffective beyond 5 m in more decomposed peat where the hydraulic conductivity was lower. A review of the available literature shows the extent of drainage effects are reported as being anything from 2m (Burke,1961) to (in some unusual conditions) as much as 200m (Trettin et al., 1991) around the site of disturbance. The amount of information decision makers will require depends on the conservation importance of the site with strong legal obligations on internationally and nationally important sites, requiring rigorous examination of the potential impacts. Ideally, estimation of the extent drainage should use this information together with a dynamic simulation model to account for the impact of hydraulic conductivity and the structure and slope of the site. Where there is less imperative for such detailed measurements, it is advisable to at least account for the hydraulic conductivity of the soil in the estimation of the extent of the drainage. The extent of drainage will be an important determinant of the indirect effects due to the wind farm development.

The SNH Guidance suggests that if the site infrastructure can be floated, disturbance to hydrology will be minimal. A review of Environmental Statements for wind farms shows that developers often attempt to construct floating roads to reduce the requirement for drainage and minimise disturbance of the peat land ( e.g. Entec UK Limited, 2002). However, according to Lindsay (2005), floating roads often suffer from subsidence and so may require drainage or some amelioration after a few years. It is the industry experience over multiple sites that such drains are temporary during the construction period and will be back filled following construction ( 2M.Mathers, pers.comm.). There have been situations, however, where the backfilled drains have continued to operate as drainage channels with the level of impact depending on the nature of the back filled material ( 3C.Bain, pers.comm.). Back filling of drains is desirable for many reasons, including the impact on runoff water quality. Whether indirect effects should be considered around floating roads depends on whether the roads can be assumed to remain intact without further drainage and the choice of practice of back-filling drains.

2M. Mathers, pers.comm. Scottish Renewables, Central Chambers, 93 Hope Street, Glasgow G2 6LD. Information obtained through personal communication with construction teams at ScottishPower, Airtricity, Amec, Morrisons, Novera, SSE, Natural Power. This group covers > 80% of all wind farm developments in the UK.
3C. Bain, pers.comm. Royal Society for Protection of Birds.

Impacts of Drainage

Carbon Dioxide Emissions

The impacts of drainage, peat removal and site disturbance on peat land hydrology and CO 2 emission are complex. Undisturbed peat bogs comprise an upper active 'acrotelm' peat layer with a high hydraulic conductivity and ßuctuating water table, and a more inert lower 'catotelm' layer, which corresponds to the permanently saturated main body of the peat (Ivanov, 1948). Peat excavation typically removes the acrotelm to expose the more decomposed peat of the catotelm which usually has a lower hydraulic conductivity than the acrotelm. This profoundly affects the balance between water storage and runoff (Schouwenaars and Vink, 1992), the nature and magnitude of evapotranspiration losses (Price, 1996) and so also the soil processes controlling soil C dynamics and CO 2 emissions (Waddington and Warner, 2001). Drainage and subsequent lowering of the water table has been shown to change peat lands from C sinks into sources of C emissions to the atmosphere as a result of increased oxidation of organic matter (Holden et al., 2007).

Methane Emissions

In addition to acting as a C sink, saturated organic soils can emit methane (CH 4), a significantly more potent greenhouse gas than CO 2, due to anaerobic decomposition of organic matter. Methane is formed under anaerobic conditions at the end of the reduction chain when all other electron acceptors such as, for example nitrate and sulphate, have been used. Methane emissions from freely drained organic soils are usually negligible, aerobic soils tending to oxidise rather than emit CH 4 (Goulding et al., 1995; Willison et al., 1995). However, when the soil becomes saturated, it will tend to emit CH 4, as well as fixing C due to the reduced rate of aerobic decomposition. In order to assess the overall impact of greenhouse gas emissions, CH 4 emissions are expressed in terms of CO 2 equivalents. This is done by integrating the radiative forcing of CH 4 and CO 2 over different timescales and comparing the result to the radiative forcing of CO 2. This measure is called the global warming potential, and one unit by weight of CH 4 has the same global warming potential as 23 units by weight of CO 2 ( IPCC, 2001). Smith et al (2007a) used IPCC default values ( IPCC, 1997) to estimate a mean CO 2 sink of 36.67 t CO 2 ha -1 y -1 (3.67 to 69.67 t CO 2 ha -1 y -1) on restoration of organic soils in cool moist regions. Based on figures from Le Mer and Roger (2001), the same land use change resulted in CH 4 emissions of 3.32 t CO 2 eq. ha -1 y -1 (0.05 to 15.30 t CO 2 eq. ha -1 y -1). If it can be assumed that drainage of peat results in the opposite response, but of the same magnitude, the estimated CO 2 equivalent emissions of the drained site are ~9% lower if the reductions in CH 4 emissions on drainage are accounted for than if they are neglected (~1% to ~22% assuming low C accumulation and low CH 4 emissions tend to occur at the same site). However the figures could vary greatly depending on the nature of the peat land (eg. whether a bog or a fen). Some peat lands emit very little methane in their natural condition, and so ideally, simulation modelling should be used to get a more site specific estimate of losses of CH 4. Reductions in CH 4 emissions should be included in the calculation of payback time, but more specific data will be required about the type and condition of the habitat in order to accurately determine CH 4 emissions prior to drainage. This will tend to reduce the payback time calculated for a highly organic wetland site. Similarly, increased CH 4 emissions that occur on restoration of a site to a high water table should be included.

Erosion and Mass Movement

Other problems associated with drainage and peat excavation are erosion and mass movement of peat, usually reported as bog bursts or peat slides. Drains in blanket peats have been observed to incise rapidly (Mayfield and Pearson, 1972) and drains cut to 50 cm depth may erode to several metres. Peat land erosion in the UK has been shown to be a major problem as it removes particulate organic C (Holden, 2005). Peat landslides represent one end of a spectrum of natural processes of peat degradation. Longer term processes of degradation include incision and upslope extension of gully networks by water action (Evans and Warburton, 2005), development of subsurface piping creating extensive sub-surface voids (Holden, 2004), and desiccation cracking and wind erosion (deflation) of the top surface of peat deposits (Evans and Warburton, 2007). Again, recent Best Practice guidance from the Scottish Executive (2006) draws attention to and outlines methods for identifying, mitigating and managing peat slide hazards and their associated risks. However, the potential C loss through the above processes should also be considered while calculating the C saving of the wind farm. This can only be done with any accuracy using detailed measurements at the site and rigorous simulation of erosion risks because erosion losses are highly site specific, depending on topography, formation of gullies and many other factors. For the current purposes, it is assumed that erosion management given in the Best Practice guidance is adhered to and is sufficient to avoid large losses due to runoff of particulate organic matter. Note that if Best Practice guidance is not adhered to, losses of C by erosion may be high, and greatest losses can occur at already degraded peat lands where erosion structures have already formed.

Dissolved Organic Carbon

Another process by which organic material may be lost from the soil is by leaching as dissolved organic carbon ( DOC). The export from temperate and boreal peat lands ranges between 10 and 500 kg DOC ha -1 yr -1(Dillon and Molot 1997), which typically represents around 10% of the total C release. DOC is important in peat lands because any change in the flux of DOC will result in a significant regional redistribution of terrestrial C. There is increasing evidence of large increases in DOC concentrations in rivers, lakes and reservoirs across Northern Europe (Worrall et al., 2004), North America (Driscoll et al., 2003) and in Central Europe (Hejzlar et al., 2003). High DOC concentrations and water discoloration in rivers are particularly associated with catchments where there is extensive peat cover (Aitkenhead et al., 1999). Increases in DOC concentrations in the channels of a peat catchment could indicate shifts in the C budget, suggesting either a decrease in soil storage of C, or an increase in the turnover rates of the soil organic matter. Possible reasons for increases in DOC are: increasing air temperature (Freeman et al., 2001a); changes in land management (Worrall et al., 2003); changes in pH (both increases and decreases - Bouchard, 1997); change in the amount and nature of water ßow (Tranvik and Jansson, 2002); eutrophication (Harriman et al., 1998); and increasing atmospheric CO 2 (Freeman et al., 2004). Drainage of the soil during construction and operation of a wind farm will impact the quantities of DOC leached. For Minnesota peat soils, Clausen (1980) found that DOC concentrations increased upon drainage, and Mitchell and McDonald (1995) have shown that at a catchment scale the areas of the highest drainage density are the largest sources of DOC. For drained catchments, i.e. with no remedial works, the increase in DOC is forecasted to be between 15% and 33% over the next decade (Worrall et al., 2007).

Timeframe of Drainage Impacts

Many of the loss processes potentially initiated by development on peat lands are long term processes. The decomposition processes of the more recalcitrant soil organic matter pools have half lives of more than 50 years (Coleman and Jenkinson, 1996). Processes resulting in C loss, greenhouse gas emissions and potential further erosion and mass movement of peats, could continue well beyond the lifetime of any development. It could justifiably be argued that the developer has no control over the management of the land beyond decommissioning and so the calculation of C loss should be restricted to the lifetime of a development. However, if they remain unchecked by remediation, processes that are directly attributable to the decisions taken in the development will continue beyond its lifetime. The counterargument would assert that the full extent of the losses should be included by calculating a new stable C content of the soil after the wind farm development (this may be close to zero for many peat lands). This viewpoint would suggest that all losses of C occurring before the soil reaches its new stable C content should be included in the calculation of the C balance of the wind farm. If, however, restoration of the site included in the decommissioning plan prevents further C loss, it would seem justifiable to only include losses occurring over the lifetime of the wind farm. A peat land site may not have been in a steady state with respect to soil C even before the wind farm was constructed, due to the impacts of previous land uses, long term adaptation or climate change. However, if the site is restored so that any changes in soil C stocks directly attributable to the wind farm development can be assumed to have been halted, then the additional emissions of CO 2 from the soil due to the wind farm development can be limited to the period before restoration of the site. Equally, any increases in soil C content, associated with restoration that has occurred as part of the habitat management plan at the site should be included in the overall C balance. Resolution of the issue of the timeframe to be considered requires agreement between Scottish Government, the power companies, environmental groups and other stake-holders. If the full extent of C losses is included in the planning process, careful planning of wind farm structures, habitat management and restoration of the site has the potential to change the developed site into a C sink rather than a source, especially at damagedintactintact sites where there is more potential for habitat restoration. The implications for wider biodiversity would nonetheless have to be considered as well, particularly since the favourable status of a habitat depends on it supporting healthy populations of typical species, many of which could be adversely affected by the operation of a wind farm.

Previous Drainage Events

In the SNH guidance, loss of C fixation capacity is calculated assuming a fixed value for the annual C fixation by the peat land ( i.e. 0.25 tC ha -1 yr -1) and that the peat land is stable or active. However, Usher et al (2000) reported that of around 1100000 ha of blanket peat in Scotland, in excess of 25%, no longer supports peat bog habitat due to agricultural reclamation or forestry, with further unquantified losses due to agricultural changes such as drainage, heavy sheep grazing and burning. More localised threats include peat extraction and erosion. The above processes impact the C content of the peat as well as primary productivity of peat land. Site specific data for the C fixation capacity (primary productivity) of the peat land should be used while calculating the loss of C fixation capacity. Previous drainage conditions should also be accounted for when determining the extent of the site likely to be affected by the wind farm development.

A1.6. Recommendations for reducing uncertainties in technical guidance

A1.6.1. Capacity factor

The capacity factor is a key determinant of CO 2 payback time, and so wherever possible, the value used in technical guidance should be obtained from the information provided for the environmental impact assessment at the specific site. In the absence of a value measured at the site, the lowest capacity factor for the region within Scotland should be assumed to calculate the worst case scenario payback time. If the mean capacity factor was to be used as an estimate of the likely payback time, the high uncertainty in the estimate should be expressed by also calculating the worst and best-cases using the lowest and highest capacity factors as given for Scotland ( e.g.DTI, 2006).

A1.6.2. Grid-mix emission factor

The calculation of emission factors should be updated annually, based upon the latest available figures reported by DEFRA (or its equivalent into future years). The emission factors used should relate to the type of power generation that is most likely to be replaced by wind generated power, i.e. the fossil fuel mix. For consistency across the UK, the approach that will shortly be agreed between BWEA and ASA should be adopted.

A1.6.3. Extent of the site affected by development

The assessment of extent of site affected by the wind farm development should consider both the direct effect of disturbance at the site due to construction and the indirect effects due to drainage and lowering of the water table. Ideally, a detailed survey of the hydrological features of the site should be used to calculate the extent of site affected by the wind farm development. Mapping the depth and hydrological characteristics of the peat land, and simulation modelling of the anticipated changes in the soil hydrology are recommended to improve assessment of the extent of the site impacted by the wind farm development. In the absence of such detailed information, the hydraulic conductivity of the soil could be used to provide a more site specific estimate of the extent of indirect effects than is currently used in the SNH Guidance.

A1.6.4. Estimates of percentage C loss

In the SNH Guidance, the C loss from the site is calculated assuming an improbable worst-case scenario of 100% loss of the C stored in the area impacted by the development. A much lower percentage of C may be lost from sites with well-designed hydrology and habitat management plans. An improved estimate of the percentage C loss due to the wind farm development could be obtained using evaluated simulations from a dynamic model of C turnover in highly organic soils, such as ECOSSE (Smith et al, 2007b). The calculated uncertainty in the simulations can then be included as a range of values for the rate of C loss. In order to reduce the impact of wind farm development on loss of habitat and disturbance, and C losses associated with erosion and peatslides, it is recommended in the Best Practice Guide for Proposed Electricity Generation Developments (Scottish Executive, 2006) that mitigation measures are taken. Although these measures are not outlined in any detail, there is much background information on the principles of management and restoration of peat lands in Brooks and Stoneman (1997) and Wheeler and Shaw (1995). Site-specific opportunities may also arise and be adopted by a particular wind farm developer, such as the restoration of an open cast mine site to recreate wetland habitat (Black Law Windfarm, Central Scotland). Carbon losses associated with practices such as restoration of borrow pits by filling them with surplus peat require further consideration. Government Forestry Policy (Scottish Forestry Strategy 2006) and the Forestry Commission guidelines (Forests and Peat land habitats, 2000) support the removal of plantation forestry for bog restoration where there are high net environmental benefits to be obtained from permanent removal of trees, but not for the development of wind farms alone. Such restoration and mitigation practices are usually outlined in the Habitat Management Plan for the site and should also be accounted for in the C savings for the wind farm. It should be recognised, however, that in some cases the early removal of trees will facilitate better bog restoration success and therefore prevent long term deterioration and loss of the C store, as well as allowing restored C fixation.

A1.7. Detailed Comments on the SNH Technical Guidance Document

A1.7.1. Introduction

Paragraph 2 - SNH has supported the Scottish Executive's target of achieving 18% of electricity generation from renewable sources by 2010, and for that target to be increased to 40% by 2020. Note that the target for electricity generation from renewable sources has changed since the publication of the 2003 Technical Note to 50% of electricity generated in Scotland to come from renewable sources by 2020, with an interim target of 31% by 2011 (Scottish Government, 2007, p. 54).

A1.7.2. Background

Paragraph 4 - In 1999 it was estimated that 20% of Scotland's annual CO 2 emissions were from land use change and forestry (Key Scottish Environment Statistics, 1999, p.14). The estimate of annual CO 2 emissions from land use change and forestry is highly uncertain. The latest issue of 'Key Scottish Environment Statistics' (2007) notes that the net change in land use and forestry had acted as a net C sink of 1.26 Mt C eq. in 2004, and comments that, "Estimates of emissions and removals from this sector are particularly uncertain since they depend critically on assumptions made on the rate of loss or gain of carbon in Scotland's carbon rich soils."

Paragraph 6 - Because of the uncertainties involved in estimating the relevant C budgets, the 2003 Technical Note suggests it will only be possible to give an indication of payback time to within 5 years. It is suggested that this estimate of uncertainty may be too high. Revised technical guidance should include a rigorous calculation of uncertainty.

Paragraph 7- The note does not include the need to maintain electricity supply from other sources (including fossil fuel sources) on still days. The note assumes that the emissions per unit from back-up sources before and after the wind farm development are unaffected by the wind farm. If the national dependence on wind generated electricity is less than ~20%, it is likely that there will be sufficient slack in the overall grid supply to allow the back-up mix of fuel sources to remain unchanged. However, as the demand for wind generated sources increases to meet the 50% SG target for electricity generated from renewable sources, this assumption may no longer hold. It is commented that the back-up sources are allowed for in the assumption of a 40% capacity factor (which limits the utilisation rate in paragraph 13). However, the capacity factor used is higher than the 30% average rate for Scotland observed between 1998 and 2004 ( DTI, 2006). To account for the higher C emissions due to back-up sources of electricity, the capacity factor should have been lower than the average. The calculations also exclude off-site C costs, for example, associated with the erection of transmission lines. The exclusion of off-site C costs is a potentially significant weakness. Two issues arise (i) the development of infrastructure associated with the wind farm development can be expected to increase the area of disturbance of peat lands; (ii) the wider infrastructural requirements of wind farm developments are not borne by individual developments. Careful design and siting of wind farms may obviate the need for these factors to be considered, but disturbance due to increasing over-, or under-ground development of electricity transmission capability is likely to impact CO 2 emissions. Note that these increase emissions can equally be associated with other forms of electricity generation.

A1.7.3. Carbon emissions from electricity generation

Paragraph 10 - Energy-related emission factors are given, but no allowance is made for improvements in technology. The assumption made is that the emission factors used will remain constant for the period of the development. The same assumption is made in the calculation of emissions from backup (paragraph 7). However, the likely change in emission factors is difficult to predict. Assuming the baseline value (as done here) is the norm in all other sectors. Therefore, the approach used is the pragmatic approach, but it should be noted that it is likely to provide an over-optimistic estimation of pay-back time as the efficiency of technology is improving in all sectors.

Paragraph 11 - The average electricity emission factor, taken across the mix of electricity sources supplying the UK grid as a whole is taken to be 0.43 t CO 2 MWh -1. The value of the grid-mix emission factor should be replaced with the fossil-fuel sourced grid-mix and updated to allow for the change in production mix since publication of the 2003 Technical Note.

A1.7.4. Wind farm carbon emission savings

Paragraph 13 - A utilisation rate of 0.4 is assumed ( i.e. the use of energy from the turbines is 40% of the rated generation capacity). This was based on the best available figures at the time. Figures for the utilisation rate ranged from 0.7 (The Carbon Trust, 2007) to 0.35-0.45 (wind industry), with opponents of wind energy using more pessimistic figures. The figure of 0.4 was chosen because it was the median of the available estimates ( 1Mitchell, pers.comm). This utilisation rate is high compared to the figures published on the British Wind Energy website ( www.bwea.org), and by the Department for Trade and Industry ( DTI, 2006). The average value used should be reduced to 30% to reflect current estimates. The median value used in the 2003 Technical Note may have been biased upwards by the over-optimistic utilisation rate reported by the Carbon Trust. This assumed that the utilisation rate was limited by the user-demand, whereas it will normally be limited by the capacity factor of the turbine (the ratio of the actual power output over a period of time and the output if it had operated a full capacity of that time period). The capacity factor may change over time as turbine technology improves and as wind speeds vary due to climate change. However, the value of 40% is over-optimistic. Ideally a site specific capacity factor should be used. However, if this is not available, the capacity factor used could be refined to draw on values published on the BWEA website, or to use averages provided on a sub-national basis. The examples used in this paragraph should cover the range of developments found in Scotland, as well as potential larger developments of the future.

1C. Mitchell, pers.comm. Strategy & Communications Manager, Strategic Direction, Scottish Natural Heritage, Battleby, Redgorton, Perth, Scotland PH1 3EW

A1.7.5. Loss of carbon stored on bogs

Paragraph 20 - The 2003 Technical Note uses a worst-case scenario of the entire bog area being lost due to wind farm development. There are now sufficient wind farm developments on peat and peat lands to allow more refined approaches to the estimation of the proportion of the area impacted upon by the development. The entire bog area could be considered to have some meaning in an enclosed peat deposit ( e.g. a small basin bog). However, in most cases for which information can be found ( e.g. Achkeepster, Caithness), the development does not cover the entire bog area, or even a single 'sub-area' as defined in resource surveys (Macaulay Institute for Soil Research, 1985). Therefore, discussion about the area, depth and volume of the peat should be based upon more reliable data than that considered to be available by SNH (2003).

Paragraph 21 - The calculations presented in the 2003 Technical Note assume the hydrology of the bog is affected in an area double that required for the infrastructure of the wind farm. The use of a factor of 2 over the area required for the infrastructure should be reconsidered. Hydrological modelling enables the buffering of the infrastructure proposed for a proposed development to be assigned more precise distances. As the spatial layout of turbines, access roads, and links to off-site infrastructure are often quite complex ( i.e. follow hill crests, contours or areas of shallow peat), the proportions of the area which potentially cover different depths of peat can be calculated. Data for this can be obtained from the developer, because it will be gathered as part of the assessment of peat slide risk, as required for best practice in such environments (Scottish Executive, 2006). This information should be used in assessing the area where hydrology of the bog is affected.

A1.7.6. Loss of carbon-fixing potential as a result of woodland clearance

Paragraph 27 in the 2003 Technical Note states that land most likely to be favoured for wind farm development is likely to be occupied by commercial plantations of poor quality (in terms of productivity and carbon sequestration potential or biodiversity interest), which tend to be found at the higher altitudes favoured by wind farm developers. Historically this was the case but current and future wind farm proposals, based on lower windspeeds, are likely to involve better quality forest land, the potential extent of which is significant on upland areas with peat soil. Whilst forestry policy no longer supports planting on deep peat soils, there are some historic sites where such planting did occur and some of these are areas where tree removal would be beneficial for priority habitat restoration.

The consequences on C stocks of removing forestry from peat soils are still not clearly understood. A study in Wales in 2000 concluded that planting on peat soils results in very substantial emissions of C, which exceed the level of C storage in livewood (Bateman and Love, 2002). Further work by Hargreaves et al (2003) has shown that the C losses from peat begin to exceed sequestration from new forestry within three rotations. This, however, takes no account of additional soil C loss due to forest harvesting and planting disturbance to the peat soil. The other aspect which needs to be considered when assessing the consequences in C stocks of tree removal from peat is the planned fate of the crop, which in most cases would lead to very rapid C loss to the atmosphere. If the crop was destined for paper pulp where the C loss would be rapid after harvesting then tree removal as part of the wind farm would have little additional C implications compared to the planned harvesting of the crop.

Tree removal from peat soils can have considerable biodiversity benefits and contributes to the restoration of EU priority blanket bog and raised bog habitat. Furthermore the early removal of trees may facilitate more successful peat land restoration and hence retention of greater C stocks compared to leaving the trees longer which may make restoration less successful.

A1.8. Suggested approach for calculating carbon dioxide savings associated with wind farms

CO 2 emission from the life cycle analysis of wind turbines and the backup source should be included in the calculations of CO 2 savings of wind farms. This will require generic figures to be provided in guidance.

A site specific capacity factor should be used wherever practicable to provide a more realistic payback time for the site. Failing that, the best (34%) and worst case capacity factors for Scotland (27%) should be used ( e.g.DTI, 2006).

Fossil-fuel sourced grid-mix emission factors should be updated annually.

Calculation of C loss should include both direct loss of C from the excavated peat and indirect loss due to drainage and other site disturbance.

A combined study on the effect of the wind farm development on hydrology and the C cycle will give a better understanding of C loss.

The major pathways of C loss from soil are aerobic decomposition of organic matter, resulting in CO 2 losses; anaerobic decomposition of organic matter, resulting in CH 4 losses; leaching losses of dissolved organic C associated with movement of water through the soil; and losses of C associated with erosion or mass movement of soil. C losses associated with erosion and mass movement are highly site specific and can only be adequately estimated using dynamic simulation modelling driven by detailed site specific data. These are assumed to be negligible given adherence to Best Practice guidance. Aerobic and anaerobic decomposition losses can be estimated using IPCC defaults or more site specific relationships derived from dynamic C turnover simulations. These simulations will also provide estimates of losses as dissolved organic C.

Simulation modelling of soil hydrology, soil erosion and C cycling should be used wherever sufficient information is available to provide better estimates of the extent of the area affected by the wind farm development and the impact on C losses from the soil.

Potential effects of mitigation measures to increase C storage should be included in calculations of CO 2 savings of wind farms.

The impacts of restoration, or after-use of wind farm sites in areas of peat or peat land should be included in appraisals of potential C loss.

A1.9. Suggested best practice to improve carbon savings associated with wind farms

Below we provide some recommendations for improving C savings associated with wind farm developments. Note that these best practice guidelines aim to improve C savings only; biodiodiversity issues should be considered separately, although many of the suggested practices will also benefit habitat and biodiversity. Although, wind farms sited on active peat lands do save C compared to fossil-fuel power generation, a key factor in further improving C savings would be to site the wind farm on a mineral soil. In an example calculation, 36% of the estimated C losses were associated with loss from soil organic matter. If it can be assumed that C losses from soil organic matter held in a mineral soil would be negligible, the C savings at an equivalent site on a mineral soil would be significantly higher than on an active peat land. This assumption is likely to be applicable at most sites, indeed it is likely that a mineral site would gain soil organic C if the management moved to more natural vegetation (Smith et al., 2005). However, because mineral soils tend to be found on less exposed, lowland sites, the capacity factor might be expected to be lower, so reducing C savings by another route. This issue is complex, because increasing C savings by one route can reduce C savings by another. These factors should be considered further by reviewing the payback time of existing wind farms, sited on different soils across Scotland, but this is beyond the scope of this report.

A1.9.1. Best Practise

  • When excavating areas of peat, excavated C-layer turfs should be as intact as possible, often it is easiest to achieve this by excavating in large turfs or clumps. An intact excavated block will be less prone to drying out.
  • Excavations should be prevented from drying out or desiccating as far as possible. This can be achieved by minimizing disturbance or movement of the excavated peat once excavated. Consideration should also be given to spraying the peat to keep it moist in appropriate circumstances.
  • Stockpiling of peat should be in large amounts, taking due regard to potential loading effects for peat slide risk. Piles should be bladed off at the side to minimise the available drying surface area.
  • The peat should be restored as soon as possible after disturbance. When constructing tracks, this requires restoration as track construction progresses. However for borrow pits and crane pads it may be more difficult to reinstate before construction is complete.
  • Floating roads should be used where peat is deeper than 1 m to avoid cutting into peat and disturbing it leading to drying out. Cut roads should only be used in areas where the peat is less than 1 m deep.
  • Submerged foundations should be employed on deeper areas of peat, to maintain hydrology around the turbine base and avoid draining the peat area. Submerged foundations are designed to be larger than normal drained foundations so can they withstand water pressure lifting up the foundation where they are submerged.
  • The design of tracks should be such that they do not act as a conduit or channel for water or a dam or barrier to water flow. This is a highly site specific consideration, and requires the consideration of track design at the construction stage, when all geotechnical investigations are proceeding, rather than deciding on a final track design at the planning stage.
  • Good track design should be employed with appropriate cross drains, minimizing the collection of water and ensuring overall catchment characteristics are maintained.
  • Developers should take ancillary opportunities to improve habitats, by including simple practices such as drain blocking and re wetting of areas. These practices can be included as mitigation.

A1.10. References

Aitkenhead, J.A., Hope, D. and Billet, M.F. (1999). The relationship between dissolved organic carbon in stream water and soil organic carbon pools at different spatial scales. Hydrol. Process 13, 1289-1302.
AMEC (2004). Kyle Windfarm Proposal, Environmental Statement, October 2004. pp. 374.
Ardente, F., Beccali, M., Cellura, M. and Brano V.L. (2008). Energy performances and life cycle assessment of an Italian wind farm. Renewable and Sustainable Energy Reviews 12, 200-217.
Armstrong, A.C. (2000). DITCH: A model to simulate field conditions in response to ditch levels managed for environmental aims. Agric. Ecosyst. & Envir. 77, 179-192.
Baggott, S.L., Cardenas, L., Garnett, E., Jackson, J., Mobbs, D.C., Murrells, T., Passant, N., Thomson, A. and Watterson, J.D. (2007). UK Greenhouse Gas Inventory, 1990 to 2005. Annual Report for submission under the Framework Convention on Climate Change. London, HSO. ISBN 0-9554823-1-3. http://www.naei.org.uk/reports.php. Report AEAT/ENV/R/2429 13/04/2007
Botch, M.S., Kobak, K.I., Vinson, T.S. and Kolchugina, T.P. (1995). Carbon pools and accumulation in peat lands of the Former Soviet Union. Global Biogeochemical Cycles, 9, 37-46.
Bouchard, A. (1997). Recent lake acidification and recovery trends in Southern Quebec, Canada. Water Air and Soil Pollution 94, 225-245.
Brooks, S. and Stoneman, R. (1997). Conserving Bogs: The Management Handbook. The Stationary Office, Edinburgh.
BWEA, 2005. Blowing away the myths: a critique of the renewable energy foundation's report: reduction in carbon dioxide emissions: estimating the potential contribution from wind power. February, 2005. www.bwea.com.
Carbon Trust/Dti (2004), Renewables Network Impacts Study.
Clausen, J.C. (1980). The quality of runoff from natural and disturbed Minnesota peat lands. The role of peat lands in a world of limited resources, energy, food, fibre and natural areas. In: Proceedings 6 th International Peat Congress, Duluth, Minnesota 1980, International Peat Society, pp. 523-532.
Coleman, K. and Jenkinson, D.S. (1996). RothC-26.3- A Model for the turnover of carbon in soil. In: Evaluation of Soil Organic Matter Models Using Existing, Long-Term Datasets, NATOASI Series I, Vol.38 (eds Powlson DS, Smith P, Smith JU), pp. 237-246. Springer-Verlag, Heidelberg, Germany.
Council of European Energy Regulators ( CEER) (2004). Current Experience With Renewable Support Schemes in Europe, pp. 57-59.
Dale, L, Millborrow, D, Slark, R and Strbac, G (2004) Total Cost Estimates for Large-Scale Wind Scenarios in UK, Energy Policy, 32, 1949-56.
Department of Environment, Food and Rural Affairs ( DEFRA) (2002). Guidelines for the measurement and reporting of emissions by Direct Participants in UK Emissions Trading Scheme ( DEFRA,Oct 2002)
Department of Environment, Food and Rural Affairs ( DEFRA) (2003). Guidelines for the measurement and reporting of emissions by Direct Participants in UK Emissions Trading Scheme, UKETS(01)05rev2(June 2003).
Department of Trade and Industry ( DTI) (2006). Energy Trends, March 2006.
Digest of the United Kingdom Energy Statistics (2007). Department for Business, Enterprise and Regulatory Reform. A National Statistics Publication, London, HSO. ISBN 9780115155208. http://www.berr.gov.uk/energy/statistics/source/electricity/page18527.html
Dillon, P.J. and Molot, L.A. (1997) Effect of landscape form on export of dissolved organic carbon, iron, and phosphorus from forested stream catchments. Water Resources Research 33, 2591-2600.
Douglas, E. (2006). The hidden cost of wind turbines, New Scientist ( http://environment.newscientist.com/channel/earth/mg19125591.600-the-hidden-cost-of-wind -turbines.html
Driscoll, C.T., Driscoll, K.M., Roy, K.M. and Mitchell, M.L. (2003). Chemical response of lakes in the Adirondack region of New York to declines in acidic deposition. Environmental Science Technoly 37, 2036-2042.
Entec UK Limited (2002). Black Law Windfarm Environmental Statement, Volume 1. pp.357
Evans, M.G. and Warburton, J. (2005). Sediment budget for an eroding peat moorland catchment in northern England. Earth Surface Processes and Landforms 30, 557-577.
Evans, M.G. and Warburton, J. (2007). Geomorphology of upland peat: erosion, form and landscape change. Blackwell Publishers Limited, Oxford.
Forestry commission (2000). Forest and Peat land Habitat Guidelines.
Freeman, C., Fenner, N., Ostle, N.J., Kang, H., Dowrick, D.J., Reynolds, B., Lock, M.A., Sleep, D., Hughes, S., Hudson, J. (2004). Export of dissolved organic carbon from peat lands under elevated carbon dioxide levels. Nature 430, 195-198.
Freeman, C., Evans, C.D., Montieth, D.T., Reynolds, B., Fenner, N. (2001). Export of organic carbon from peat soils. Nature 412, 785.
Goulding, K.W.T., Hutsch, B.W., Webster, C.P., Willison, T.W. and Powlson, D.S. (1995). The effect of agriculture on methane oxidation in soil. Philosophical Transactions of the Royal Society of London A 351, 313-325.
Hall, M. (2006) Peat, carbon dioxide payback and wind farms, 2006 - A guide to calculating the carbon dioxide debt and payback time for windfarms, Renewable Energy Foundation.
Harriman, R., Curtis, C., Edwards, A.C. (1998). An empirical approach for assessing the relationship between nitrogen deposition and nitrate leaching from upland catchments in the United Kingdom using runoff chemistry. Water Air and Soil Pollution 105, 193-203.
Hejzlar, J., Dubrovsky, M., Buchtele, J., Ruzicka, M. (2003). The apparent and potential effects of climate change on the inferred concentration of dissolved organic matter in a temperate stream (the Malse River, South Bohemia). Science of the Total Environment 310, 143-152.
Hogg, E.H., Lieffers, V.J. and Wein, R.W. (1992). Potential carbon losses from peat profiles: effects of temperature, drought cycles, and fire. Ecological Applications 2, 298-306.
Holden, J.H. (2004). Hydrological connectivity of soil pipes determined by ground-penetrating radar tracer detection. Earth Surface Processes and Landforms 29, 437-442.
Holden, J., Chapman, P.J. and Labadz, J.C. (2004). Artificial drainage of peat lands: hydrological and hydrochemical process and wetland restoration. Progress in Physical Geography 28, 95-123.
Holden, J. (2005). Peat land hydrology and carbon cycling: why small-scale process matters. Philosophical Transactions of the Royal Society, Series A 363, 2891-2913.
IPA Energy Consulting (2005). Implications of the EU Emissions Trading Scheme for the UK Power Generation Sector.
International Panel on Climate Change, IPCC (1997). Revised 1996 IPCC guidelines for national greenhouse gas inventories workbook, vol. 2. Cambridge, UK. Cambridge University Press.
International Panel on Climate Change, IPCC (2001). Climate change 2001: The scientific basis (Contribution of Working Group I to the third assessment report of the Intergovernmental Panel on Climate Change). Cambridge, UK. Cambridge University Press.
Ivanov, K.E. (1948). Filtration in the top layer of convex mire massifs. Meteorol. Gidrol. 2, 46-59.[In Russian.]
Le Mer, J. and Roger, P. (2001). Production, oxidation, emission and consumption of methane by soils: a review. Eur. J. Soil Biol. 53, 343-357.
Lenzen, M., Munksgaard, J. (2002). Energy and CO2 life-cycle analyses of wind turbines Review and applications. Renew. Energy. 26, 339-362.
Lindsay, R. (2005). Lewis Wind Farm Proposals: Observations on the Environmental Impact Statement, RSPB.
Lukkala, O.J. (1949). Soiden turvekerroksen painuminen ojituksen vaikutuksesta. (Referat: Über die Setzung des Moortorfes als Folge der Entwässerung). Commun. Inst. For. Fenn. 37, 1-67.
Macaulay Institute for Soil Research (1985), Peat resources of Causeymire, Caithness: topographic survey. Macaulay Institute for Soil Research, Craigiebuckler, Aberdeen.
Mayfield, B. and Pearson, M.C. (1972). Human interference with the north Derbyshire blanket peat. East Midland Geographer 12, 245-251.
Mitchell, G., McDonald, A.T. (1995). Catchment characterisization as a tool for upland water quality management. Journal of Environmental Management 44, 83-95.
Moore, T.R. and Dalva, M. (1993). The inßuence of temperature and water table position on carbon dioxide and methane emissions from laboratory columns of peat land soils. Journal of Soil Science 44, 651-664.
National Grid Transco (2004). Seven Year Statement.
Netz, E.O. (2004). Wind Report 2004, p.6.
Prevost, M., Belleau, P. and Plamondon, A.P. (1997). Substrate conditions in a treed peat land: Responses to drainage. Ecoscience 4, 543-544.
Price, J.S. (1996). Hydrology and microclimate of a partly restored cutover bog, Québec. Hydrol. Proc. 10, 1263-1272.
Schouwenaars, J.M. and Vink, J.P.M. (1992). Hydrophysical properties of peat relics in a former bog and perspectives (sic) for Sphagnum regrowth. Int. Peat J. 5, 15-28.
Scottish Executive (2003). Securing a renewable energy future - Scotland's renewable energy policy paper. pp.11.
Scottish Executive (2006). Peat Landslide Hazard and Risk Assessments. Best Practice Guide for Proposed Electricity Generation Developments. Scottish Executive, Edinburgh.
Scottish Natural Heritage (2003). Wind Farms and Carbon Savings, SNH Technical Guidance Note. pp. 11.
Scottish Natural Heritage (2007). Annual Review 2006/07. Scottish Natural Heritage, Battleby. pp 36.
Silvola, J. (1986). Carbon dioxide dynamics in mires reclaimed for forestry in eastern Finland. Annales Botanici Fennici 23, 59-67.
Silvola, J., Alm, J., Ahlholm, U., Nyka¨nen, H., Martikainen, P.J. (1996). CO 2 ßuxes from peat in boreal mires under varying temperature and moisture conditions. Journal of Ecology 84, 219-228.
Smith, J.U., Smith, P., Wattenbach, M., Zaehle, S., Hiederer, R., Jones, R.J.A., Montanarella, L., Rounsevell, M.D.A., Reginster, I., Ewert, F., 2005. Projected changes in mineral soil carbon of European croplands and grasslands, 1990-2080. Global Change Biology 11, 2141-2152. (doi: 10.1111/j.1365-2486.2005.01075.x).
Smith, P., Smith, J.U., Wattenbach, M., Meyer, J., Lindner, M., Zaehle, S., Hiederer, R., Jones, R., Montanarella, L., Rounsevell, M., Reginster, I., Kankaanpää, S. 2006. Projected changes in mineral soil carbon of European forests, 1990-2100. Canadian Journal of Soil Science 86, 159-169.
Smith, P., Martino, D., Cai, Z., Gwary, D., Janzen, H.H., Kumar, P., McCarl, B., Ogle, S., O'Mara, F., Rice, C., Scholes, R.J., Sirotenko, O., Howden, M., McAllister, T., Pan, G., Romanenkov, V., Schneider, U., Towprayoon, S., Wattenbach, M. and Smith, J.U. 2007a. Greenhouse gas mitigation in agriculture. Philosophical Transactions of the Royal Society, B. doi:10.1098/rstb.2007.2184
Smith, P., Smith, J.U., Flynn, H., Killham, K., Rangel-Castro, I., Foereid, B., Aitkenhead, M., Chapman, S., Towers, W., Bell, J., Lumsdon, D., Milne, R., Thomson, A., Simmons, I., Skiba, U., Reynolds, B., Evans, C., Frogbrook, Z., Bradley, I., Whitmore, A., Falloon, P. 2007b. ECOSSE: Estimating Carbon in Organic Soils - Sequestration and Emissions. Final Report. SEERAD Report. ISBN 978 0 7559 1498 2. 166pp.
The Carbon Trust, (2007). http://www.carbontrust.co.uk/resource/measuring_co2/ Measuring_CO2_Examples.htm
Tranvik, L.J. and Jansson, M. (2002). Terrestrial export of organic carbon. Nature 415, 861-862.
Trettin, C.C, Johnson, J.R., and Misiak, R.D. (1991). Hydrologic effect of a prescriptive drainage system on a forested wetland in northern Michigan pp. 175-183. In J.K. Jeglum and R.P. Overend (eds). Proc. Symposium '89 Vol. 1 - Peat land Forestry, Quebec City, Quebec 6-10 August,1989. Can. Soc. Peat and Peat lands, Dartmouth, Nova Scotia.
Turunen, J., Pitkänen, A., Tahvanainen, T. & K. Tolonen, (2001). Carbon accumulation in West Siberian mires, Russia. Global Biogeochemical Cycles, 15, 285-296.
Usher, MB, Bain, C, and Kerr, S, (2000) Action for Scotland's Biodiversity, Scottish Biodiversity Group
Vestas (2005). Life cycle assessment of offshore and onshore wind power plants based on Vestas V90-3.0 MW turbines. Vestas Wind Systems A/S Alsvej 21, 8900 Randers, Denmark, www.vestas.com. pp.59.
Warburton, J., Higgit, D., and Mills, A.J. (2004) Anatomy of a Pennine peat slide, Northern England. Earth Surface Processes and Landforms, 28, pp457-473.
Wheeler, B.D. and Shaw, S.C. (1995). Restoration of Damaged Peat lands. HMSO, London.
White, S.W. and Kulcinski, G.L. (2000). Birth to death analysis of the energy payback ratio and CO 2 gas emission rates from coal, fission, wind, and DT fusion electrical power plants. Fusion Eng. Des. 48, 473-481.
White, W.S. (2007). Net energy payback and CO 2 emissions from three Midwestern wind farms: an update. Natural Resources Research. 15, 271 - 281.
White. D. (2004). Reduction in carbon dioxide emissions : estimating the potential contribution from wind-power. Renewable Energy Foundation, December.
Willison, T.W., Goulding, K.W.T. and Powlson, D.S. (1995). Effect of land use change and methane mixing ratio on methane uptake from United Kingdom soil. Global Change Biology 1, 209-212.
Windstats News Letter (2006). Vol:19,No:2 (Spring 2006) http://www.windstats.com/
Worrall, F. and Burt, T.P. (2004). Time series analysis of long term river DOC records. Hydrological Processes 18, 893-911.
Worrall, F., Burt, T.P. and Shedden, R. (2003). Long terms records of riverine carbon ßux. Biogeochemistry 64, 165-178.
Worrall, F., Shedden, R. and Burt, T.P. (2007). Modelling the impact of drainage and drain-blocking on dissolved organic carbon release from peat lands. Journal of Hydrology 338, 15-27.