General enquiries on this form should be made to: Defra, Procurements and Contracts Division (Science R&D Team) Telephone No. 0207 238 5734 E-mail: [email protected]
SID 5 Research Project Final Report
Note In line with the Freedom of Information Act 2000, Defra aims to place the results Project identification of its completed research projects in the public domain wherever possible. The NF0444 SID 5 (Research Project Final Report) is 1. Defra Project code designed to capture the information on the results and outputs of Defra-funded 2. Project title research in a format that is easily Assessment of the availability of marginal or idle land for publishable through the Defra website. A bioenergy crop production in England and Wales SID 5 must be completed for all projects. • This form is in Word format and the boxes may be expanded or reduced, as 3. Contractor appropriate. The Food and Environment Research organisation(s) Agency ACCESS TO INFORMATION The information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation 54. Total Defra project costs £ £110,912 acting as an agent authorised by Defra to (agreed fixed price) process final research reports on its
behalf. Defra intends to publish this form 5. Project: start date ...... 2 February 2009 on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental end date ...... 10 August 2010 Information Regulations or the Freedom of Information Act 2000. Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.
SID 5 (Rev. 07/10) Page 1 of 6 6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...... YES √ NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow. Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer. In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000. (b) If you have answered NO, please explain why the Final report should not be released into public domain
Executive Summary 7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work. A key aim of this study was to identify potential ‘idle’ and marginal land areas where expansion of biomass production is possible without incurring significant impacts on sustainability and competing with food production. This represents the maximum possible area suitable for biomass production. However, logistical implications, planning procedures, and other environmental considerations mean that area that is realistically available is likely to be much smaller. The key land areas of interest identified were: A) Existing land resources of agricultural value a. Land currently fallow or in voluntary set-aside b. Arable land where production of arable crops is of marginal profitability compared to production of energy crops c. Grassland where stocking rates have declined B) Land resources with no current agricultural value a. Hedgerows and lowland bracken b. Urban spaces, road and rail margins c. Brownfield sites Much of the work involved spatial analysis of resource availability, utilising spatial maps of the above land resources where available, or estimates of land areas associated with particular features. These were used to determine the scale of the available land resource in each category, along with estimates of yield potential for each resource. Where possible (arable crops and perennial energy crops) estimates of yield potential used spatially modelled assessments of yield. Areas of high biodiversity value were excluded from the analysis (e.g. SSSI’s and other protected habitats), along with urban, forest areas and lakes etc. Arable land of marginal profitability was determined by spatial analysis of yield potential and estimated profit, which were compared with spatial estimates of profitability for both willow short rotation coppice (SRC) and miscanthus production using a range of different market prices for each commodity. Land was assigned to the commodity delivering the highest profit to identify areas of arable land most likely to convert to perennial energy cropping under the scenarios evaluated. A different approach was used on grassland, due to difficulties in use of economic factors alone to predict potential land use changes. Changes in livestock numbers and the grassland area were examined to identify how utilisation of the grassland resource has changed in recent years with declining dairy and sheep numbers. Maximum stocking rates over the previous 5 years were compared with the most recent data available on current stocking rates to identify any under-utilisation of the available grass resources. Where stocking rates had been reduced, it was assumed that the under-utilised portion of the grass could be used for biomass production and the remaining grass returned to more intensive recent historic rates of stocking, thereby maintaining livestock production.
SID 5 (Rev. 07/10) Page 2 of 6 It is acknowledged that previous stocking rates were not necessarily sustainable in all areas and that intensification of livestock farming could lead to increased risk of diffuse pollution and soil erosion, potentially leading to eutrophication and other habitat damage. Higher nitrogen fertilizer requirements would also increase greenhouse gas emissions, both through nitrous oxide emissions at the field level and through energy use during fertilizer production. The local environmental impact of intensification will depend strongly on local ecological conditions and previous habitat damage. Therefore, although environmental impacts are likely, it is beyond the scope of this study to quantify the impacts of increased stocking densities at national or regional scales. Further study is recommended to inform any future decisions on intensification of livestock grazing to make room for energy crop cultivation. For each land category the potential for biofuel and/or biomass production was reviewed, taking into account suitability and impacts on soil carbon. Where there was potential for more than one use (i.e. use for biofuel and/or biomass production) then the potential contribution under each scenario was calculated. For each scenario, the area of each land resource was reported, along with the potential biomass production and its energy content (expressed as tonnes of oil equivalent (toe)). All potential contributions to UK energy demand are expressed as contributions to primary energy demand. A key issue affecting land use is the impact of land use change on soil carbon turnover and consequent Green House Gas (GHG) emissions. Changes in soil carbon turnover associated with changes in land use may offset some or all of the carbon savings associated with use of biomass feedstock in energy generation. Growth of annual crops on land previously under arable cultivation results in little or no additional GHG emissions. Growth of perennial crops on this type of land may even lead to net sequestration of carbon in soils, leading to further GHG benefits. However, GHG savings are much smaller on converted grassland and, in many cases, the overall GHG emissions can be greater than those of the fossil energy reference system when soil carbon losses are also accounted for. Information on soil carbon dynamics in perennial biomass systems is lacking and much more detailed research and evidence is required in this area. By utilising of all the ‘idle’ and marginal land resources identified in this study, the potential to generate up to 5 million toe of primary energy was estimated. This resource is capable of meeting 6.3% of UK energy demands for electricity generation or 7.6% of UK energy demands for primary heat energy. However, this represents an absolute maximum and does not take into account the logistical, environmental, or policy issues that will determine the availability of land in practice. The total area identified includes the following categories: • Uncropped arable land, including set-aside, fallow land, field margins, and field corners, which accounts for approximately 14% of the total area. This area may be constrained by environmental concerns, policy developments, and the impact of biomass crops on the appearance of the National Parks and Areas of Outstanding Natural Beauty. • Economically marginal arable and pasture land that is currently used for food production, which accounts for approximately 66% of the total area. This area may be constrained by concerns over competition with food production or over the impact of biomass crops on the appearance of protected environments. • ‘Idle’ land, including roadside verges, railway embankments, canal towpaths, golf courses, sports turf, hedgerows, and brownfield land, which accounts for approximately 20% of the total area. This area may be constrained by logistical difficulties in cultivation and harvesting; competition from other land uses (including recreation and urban development); and safety issues surrounding planting in close proximity to roads and railways. Clearly in some cases there would be strong opposition from the public to conversion for bioenergy crop production. The total area of non-agricultural ‘idle’ land that could potentially be used to produce biomass (from the existing vegetation resource) was estimated at 867,667 ha, capable of producing 2.7 million tonnes of biomass annually, with an energy content of 1,06 million toe. This is capable of supplying up to 1.3% of primary energy demand for electricity, or up to 1.6% of UK primary energy demand for heat. The largest contribution to this is Brownfield land (450,000 hectares), followed by roadside verges, lowland bracken and golf courses and hedgerows in descending order of importance. Sports turf, railway embankments and canal margins make very little contribution to the overall total from ‘idle’ land. Again, the true availability of this land for biomass production has not been assessed in this report and the total area that can practically be used is likely to be smaller. The utilisation of ‘idle’ and marginal land resources for biomass, at the scales identified in this study, could help to deliver a significant proportion of the UK targets for renewable energy as announced in the Renewable Energy Strategy (RES) for 2020, including up to 21% of the renewable electricity target (30% of electricity energy demand), or up to 63% of the renewable heat target (12% of heat energy demand). Alternately, if the priority is to produce biofuel feedstocks, then up to 18% of the RES renewable transport fuel energy target (10% by energy) could be met from biofuel feedstock production on fallow and un- cropped arable land and use of under-utilised temporary grassland. In addition, wheat straw production, plus biomass from other ‘idle’ and marginal land resources could potentially, and additionally, account for up to 17% of the renewable electricity target, or up to 52% of the renewable heat target. However, it remains unlikely that all the idle and marginal land identified in this study will be used for biomass
SID 5 (Rev. 07/10) Page 3 of 6 production. This analysis identifies the potential size of the ‘idle’ and marginal land resource that could be made available for biofuel and biomass production. This represents a maximum possible contribution without significantly affecting food production, soil carbon turnover or impinging on environmentally sensitive habitats. However, the proposed potential land changes will have some direct or indirect impacts on the environment. Loss of fallow and set-aside land will have impacts on farmland biodiversity that derives benefit from this resource, including farmland bird populations. But, conversion to perennial energy crops would have less of an impact than conversion to arable cropping for biofuel production. Encouraging intensification of livestock farming in order to free grassland for energy crops while retaining current livestock numbers is likely to lead to further, wider environmental impacts. The necessary increase in nitrogen fertilizer use may lead to higher nitrous oxide emissions from grassland systems and expose water courses to higher nitrogen inputs. Furthermore, the high energy requirement of nitrogen fertilizer production will lead to increased indirect greenhouse gas emissions. The balance of such impacts needs to be taken into consideration when contemplating the use of such land resources. ‘Idle’ land resources represent areas where there are no such concerns affecting use for biomass cropping. Further work is required to understand what proportion of the identified ‘idle’ and marginal land resource is likely to be utilised in practice, and what measures are likely to be required to encourage wider use of such land resources for biomass production.
Project Report to Defra 8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).
Full report appended.
SID 5 (Rev. 07/10) Page 4 of 6
References to published material 9. This section should be used to record links (hypertext links where possible) or references to other published material generated by, or relating to this project.
SID 5 (Rev. 07/10) Page 5 of 6
SID 5 (Rev. 07/10) Page 6 of 6 Appendix: Full Report Assessment of the availability of ‘marginal’ and ‘idle’ land for bioenergy crop production in England and Wales
Project NF0444
Project team
David Turley, Moray Taylor, Ruth Laybourn & John Hughes
Food and Environment Research Agency
John Kilpatrick, Chris Procter, Lucy Wilson & Phil Edgington
ADAS
Funded by the UK Department for Energy and Climate Change
Final Report
May 2010
Contents
1 Executive Summary ...... 3 2 Introduction ...... 6 2.1 Use of ‘marginal’ and ‘idle’ land terms in the study ...... 7 2.2 Historic and potential land use change ...... 8 2.3 Aims and objectives ...... 8 3 Approaches ...... 9 3.1 Resource mapping approaches ...... 10 3.2 Mapping areas unsuited for bioenergy production ...... 11 3.3 Identification of arable, grassland and fallow areas where biomass production is possible ...... 13 3.4 Mapping ‘idle’ land resources ...... 14 3.5 Mapping the yield potential and output of arable crops ...... 17 3.6 Mapping the yield potential of perennial energy crops ...... 18 3.7 Identification of arable land likely to convert to perennial energy cropping ...... 22 3.8 Identification of grassland likely to convert to arable or perennial energy cropping .23 3.9 Expression of biomass and biofuel energy contributions ...... 26 4 The agricultural land resource ...... 27 4.1 Fallow land resource ...... 27 4.2 Biomass supply from un-cropped arable land ...... 29 5 The economically marginal arable land resource for perennial bioenergy production ....32 6 The potential grassland resource for bioenergy production ...... 36 6.1 Use of temporary grassland for biofuel and biomass production ...... 38 6.2 Use of permanent grassland for biofuel and biomass production ...... 43 7 The ‘idle’ land resource available for bioenergy production ...... 44 7.1 Hedgerows ...... 44 7.2 Lowland bracken ...... 44 7.3 Roadside verges ...... 45 7.4 Railway embankments ...... 46 7.5 Canal margins ...... 47 7.6 Golf courses ...... 48 7.7 Sports turf ...... 49 7.8 Brownfield sites ...... 50 7.9 Contribution of ‘idle’ land resources to primary energy demands ...... 51 8 Impacts on soil carbon flux ...... 53 8.1 Land use change and carbon sequestration ...... 53 8.2 Soil carbon ...... 54 8.3 Net costs and benefits ...... 56 8.4 Carbon payback time ...... 61 8.5 Conclusions – soil carbon turnover ...... 62 9 Potential combinations of biomass supply from marginal and ‘idle’ land resources to meet UK renewable energy demands ...... 62 9.1 Potential supply of liquid biofuels...... 62 9.2 Potential supply of biomass ...... 64 10 Conclusions ...... 68 11 Further work ...... 70 12 References ...... 71 13 Annex 1 - Assumptions ...... 74 14 Annex 2 - NUTS 2 codes and regions ...... 76 15 Annex 3 Assumptions and estimated costs for derivation of base crop production costs ...... 77 16 Annex 4 – Regional data tables for un-cropped arable land resource ...... 80 17 Annex 5 – Regional data tables for economically marginal land resource ...... 82 18 Annex 6 – Regional data tables for grassland resource ...... 84
2 Assessment of the availability of marginal and ‘idle’ land for bioenergy crop production in England and Wales Project NF0444
1 Executive Summary
A key aim of this study was to identify potential ‘idle’ and marginal land areas where expansion of biomass production is possible without incurring significant impacts on sustainability and competing with food production. This represents the maximum possible area suitable for biomass production. However, logistical implications, planning procedures, and other environmental considerations mean that area that is realistically available is likely to be much smaller. The key land areas of interest identified were:
A) Existing land resources of agricultural value
a. Land currently fallow or in voluntary set-aside
b. Arable land where production of arable crops is of marginal profitability compared to production of energy crops
c. Grassland where stocking rates have declined
B) Land resources with no current agricultural value
a. Hedgerows and lowland bracken
b. Urban spaces, road and rail margins
c. Brownfield sites
Much of the work involved spatial analysis of resource availability, utilising spatial maps of the above land resources where available, or estimates of land areas associated with particular features. These were used to determine the scale of the available land resource in each category, along with estimates of yield potential for each resource. Where possible (arable crops and perennial energy crops) estimates of yield potential used spatially modelled assessments of yield. Areas of high biodiversity value were excluded from the analysis (e.g. SSSI’s and other protected habitats), along with urban, forest areas and lakes etc.
Arable land of marginal profitability was determined by spatial analysis of yield potential and estimated profit, which were compared with spatial estimates of profitability for both willow short rotation coppice (SRC) and miscanthus production using a range of different market prices for each commodity. Land was assigned to the commodity delivering the highest profit to identify areas of arable land most likely to convert to perennial energy cropping under the scenarios evaluated.
A different approach was used on grassland, due to difficulties in use of economic factors alone to predict potential land use changes. Changes in livestock numbers and the grassland area were examined to identify how utilisation of the grassland resource has changed in recent years with declining dairy and sheep numbers. Maximum stocking rates over the previous 5 years were compared with the most recent data available on current stocking rates to identify any under-utilisation of the available grass resources. Where stocking rates had been reduced, it was assumed that the under-utilised portion of the grass could be used
3 for biomass production and the remaining grass returned to more intensive recent historic rates of stocking, thereby maintaining livestock production.
It is acknowledged that previous stocking rates were not necessarily sustainable in all areas and that intensification of livestock farming could lead to increased risk of diffuse pollution and soil erosion, potentially leading to eutrophication and other habitat damage. Higher nitrogen fertilizer requirements would also increase greenhouse gas emissions, both through nitrous oxide emissions at the field level and through energy use during fertilizer production. The local environmental impact of intensification will depend strongly on local ecological conditions and previous habitat damage. Therefore, although environmental impacts are likely, it is beyond the scope of this study to quantify the impacts of increased stocking densities at national or regional scales. Further study is recommended to inform any future decisions on intensification of livestock grazing to make room for energy crop cultivation.
For each land category the potential for biofuel and/or biomass production was reviewed, taking into account suitability and impacts on soil carbon. Where there was potential for more than one use (i.e. use for biofuel and/or biomass production) then the potential contribution under each scenario was calculated. For each scenario, the area of each land resource was reported, along with the potential biomass production and its energy content (expressed as tonnes of oil equivalent (toe)). All potential contributions to UK energy demand are expressed as contributions to primary energy demand.
A key issue affecting land use is the impact of land use change on soil carbon turnover and consequent Green House Gas (GHG) emissions. Changes in soil carbon turnover associated with changes in land use may offset some or all of the carbon savings associated with use of biomass feedstock in energy generation. Growth of annual crops on land previously under arable cultivation results in little or no additional GHG emissions. Growth of perennial crops on this type of land may even lead to net sequestration of carbon in soils, leading to further GHG benefits. However, GHG savings are much smaller on converted grassland and, in many cases, the overall GHG emissions can be greater than those of the fossil energy reference system when soil carbon losses are also accounted for. Information on soil carbon dynamics in perennial biomass systems is lacking and much more detailed research and evidence is required in this area.
By utilising of all the ‘idle’ and marginal land resources identified in this study, the potential to generate up to 5 million toe of primary energy was estimated. This resource is capable of meeting 6.3% of UK energy demands for electricity generation or 7.6% of UK energy demands for primary heat energy. However, this represents an absolute maximum and does not take into account the logistical, environmental, or policy issues that will determine the availability of land in practice. The total area identified includes the following categories:
• Uncropped arable land, including set-aside, fallow land, field margins, and field corners, which accounts for approximately 14% of the total area. This area may be constrained by environmental concerns, policy developments, and the impact of biomass crops on the appearance of the National Parks and Areas of Outstanding Natural Beauty.
• Economically marginal arable and pasture land that is currently used for food production, which accounts for approximately 66% of the total area. This area may be constrained by concerns over competition with food production or over the impact of biomass crops on the appearance of protected environments.
• ‘Idle’ land, including roadside verges, railway embankments, canal towpaths, golf courses, sports turf, hedgerows, and brownfield land, which accounts for approximately 20% of the total area. This area may be constrained by logistical difficulties in cultivation and harvesting; competition from other land uses (including recreation and urban development); and safety issues surrounding planting in close
4 proximity to roads and railways. Clearly in some cases there would be strong opposition from the public to conversion for bioenergy crop production.
The total area of non-agricultural ‘idle’ land that could potentially be used to produce biomass (from the existing vegetation resource) was estimated at 867,667 ha, capable of producing 2.7 million tonnes of biomass annually, with an energy content of 1,06 million toe. This is capable of supplying up to 1.3% of primary energy demand for electricity, or up to 1.6% of UK primary energy demand for heat. The largest contribution to this is Brownfield land (450,000 hectares), followed by roadside verges, lowland bracken and golf courses and hedgerows in descending order of importance. Sports turf, railway embankments and canal margins make very little contribution to the overall total from ‘idle’ land. Again, the true availability of this land for biomass production has not been assessed in this report and the total area that can practically be used is likely to be smaller.
The utilisation of ‘idle’ and marginal land resources for biomass, at the scales identified in this study, could help to deliver a significant proportion of the UK targets for renewable energy as announced in the Renewable Energy Strategy (RES) for 2020, including up to 21% of the renewable electricity target (30% of electricity energy demand), or up to 63% of the renewable heat target (12% of heat energy demand). Alternately, if the priority is to produce biofuel feedstocks, then up to 18% of the RES renewable transport fuel energy target (10% by energy) could be met from biofuel feedstock production on fallow and un- cropped arable land and use of under-utilised temporary grassland. In addition, wheat straw production, plus biomass from other ‘idle’ and marginal land resources could potentially, and additionally, account for up to 17% of the renewable electricity target, or up to 52% of the renewable heat target. However, it remains unlikely that all the idle and marginal land identified in this study will be used for biomass production.
This analysis identifies the potential size of the ‘idle’ and marginal land resource that could be made available for biofuel and biomass production. This represents a maximum possible contribution without significantly affecting food production, soil carbon turnover or impinging on environmentally sensitive habitats. However, the proposed potential land changes will have some direct or indirect impacts on the environment. Loss of fallow and set-aside land will have impacts on farmland biodiversity that derives benefit from this resource, including farmland bird populations. But, conversion to perennial energy crops would have less of an impact than conversion to arable cropping for biofuel production. Encouraging intensification of livestock farming in order to free grassland for energy crops while retaining current livestock numbers is likely to lead to further, wider environmental impacts. The necessary increase in nitrogen fertilizer use may lead to higher nitrous oxide emissions from grassland systems and expose water courses to higher nitrogen inputs. Furthermore, the high energy requirement of nitrogen fertilizer production will lead to increased indirect greenhouse gas emissions. The balance of such impacts needs to be taken into consideration when contemplating the use of such land resources. ‘Idle’ land resources represent areas where there are no such concerns affecting use for biomass cropping.
Further work is required to understand what proportion of the identified ‘idle’ and marginal land resource is likely to be utilised in practice, and what measures are likely to be required to encourage wider use of such land resources for biomass production.
5 2 Introduction
In recognition of the significant contribution that energy consumption makes to global warming, a number of policies and strategies have been initiated in the EU and by member states to encourage increased use of renewable forms of energy in transport, electricity generation and the heating sector. In addition to permitting the use of fiscal and policy support measures to increase the uptake of renewable energy resources, targets for substitution have also been proposed. The EU Biofuels Directive proposed that 5.75% (by energy content) of transport fuels should be derived from renewable sources by 2010. The later EU Renewable Energy Directive (2008) (RED) proposed that the target should be increased to a binding 10% by 2020 1. In addition, the RED proposed that renewable energy should account for a binding minimum target of 20% of all energy consumption in the EU by 2020. The UK’s share of this target represents a requirement to deliver 15% of energy consumption from renewable sources by 2020. The recently published UK Renewable Energy Strategy (July 2009) (RES) outlined a feasible scenario to deliver this target, which includes delivery of individual energy sector targets for renewable energy amounting to 30% of electricity demand, 12% of heat demand and 10% of transport energy demand by 2020. The RES analysis indicates that around 30% of the UK’s renewable energy demand could be met from biomass sources, rising to 50% if transport biofuel targets are also included.
There are concerns regarding the ability to deliver a significant proportion of the renewable energy targets by using plant biomass resources due to concerns related to increasing competition for land with food production (and associated risk of food price increase). There are also concerns that expansion in biomass cropping that leads to conversion of land with high carbon stocks may exacerbate climate change through loss of biomass and soil carbon stocks.
In support of the introduction of the Renewable Transport Fuels Obligation in the UK, designed to increase biofuel uptake in the transport sector, a number of studies were commissioned to examine the impacts of meeting the proposed EU transport biofuel targets. The resulting Gallagher Review of the indirect effect of biofuels production (published July 2008) advised caution in supporting the further development of biofuels produced from biomass until a better understanding of the wider impacts on displaced food and feed production (land use change) was obtained, raising questions as to how much biofuel and biomass feedstock the UK can produce sustainably without affecting food production or causing significant displacement of less sustainable food and/or feedstock production to third countries.
The Gallagher Review advised that biofuel feedstock production should avoid use of agricultural land used for food production, proposing that biofuel production should instead focus on the use of ‘marginal’ or ‘idle’ land, or use appropriate wastes or agricultural residues. The review recognised that definitions of what constitutes ‘marginal’ and ‘idle’ land lack consensus across Europe and that this requires further work.
The Review also recognised the need to assess the sustainability criteria associated with any land use change. Changing land use to produce bioenergy feedstock crops should not result in an overall increase in greenhouse gas (GHG) emissions that cannot be relatively quickly recouped (within 10 years). In practice this can be difficult to achieve; figures in the Gallagher review suggest that converting grassland to wheat to provide feedstock for bioethanol production would result in a carbon payback period of between 72-123 years, depending on the type and management of the previous grassland. The potential impacts on carbon fluxes arising from potential land use change in UK situations need to be quantified in addition to other potential environmental impacts.
1 This target will be subject to review by the EC in 2014 to ascertain whether this can be achieved sustainably 6 The potential of forestry to produce biomass, and agricultural crops and residues to supply biomass from arable agricultural land has previously been reviewed (see UK Biomass Strategy, 2007). However, in the case of agricultural land, this has often been done without taking account of the direct or indirect impacts on land use change that may lead to increases in production elsewhere in the world to fill gaps in UK demand or left by removal of export surpluses.
A key aim of this project is to identify potential land areas where expansion in biomass production is potentially viable in England and Wales, and the potential biomass resource that could be obtained, as well as examining areas of potential conflict with use for food production. Although this project aims to identify land that is suitable for growing and harvesting biomass crops, it does not assess the likelihood of that land being used for this purpose. Consequently, the area identified represents the maximum possible area; the actual area available will depend on policy developments, economic circumstances, social trends, ecological factors, and logistical limitations and is likely to be much smaller.
2.1 Use of ‘marginal’ and ‘idle’ land terms in the study
The definitions of idle and marginal or degraded land used in the Gallagher Review are:
Idle land: Former or current agricultural land that will not otherwise be used for food production and other unused land that is potentially suitable for agricultural production (this definition currently includes arable land set-aside).
Marginal or degraded land: Land unsuited to food production (e.g. on poor soils) or areas that have been degraded (making them unsuited to food production)
However, the above does not adequately recognise the significant flux in land use change that occurs as a result of natural responses to fluctuation in economic and political drivers, for example in relation to set-aside policy. Marginal land is more commonly defined as land where cost effective agricultural production is not possible under a given set of conditions.
In this study, idle and unused land is taken to include,
• Land temporarily or permanently taken out of production as set-aside or fallow • Land capable of agricultural production but currently unused – e.g. land in urban areas and associated with highways and other features • Degraded, contaminated, Brownfield and other land capable of producing biomass
The above represents land whose use is unlikely to conflict with food production.
Marginal land is taken to include
• Land of typically low agricultural productivity, which is likely to change use based on economic and social drivers
In this case the risk of conflict with food production is more contentious and policies introduced to support bioenergy production could tip the balance towards energy crop production at the margins of profitability.
It is recognised that there is potential for overlap in the margins of some of the above categories, particularly in the case of marginal and idle land where there is a continuum in terms of likelihood and timescale over which land will enter or be removed from commercial arable or livestock production.
7 2.2 Historic and potential land use change
The UK has seen very significant changes in land use over time. Before the Second World War, and as a result of the depressed agricultural markets of the 1930’s, one third of UK farmland was rough semi-natural vegetation used for extensive sheep grazing. At this time, arable land was reverting to rough pasture. The advent of the Second World War reversed this trend, resulting in an 8% growth per annum in the arable area during the war years, along with a significant decline in the forested area (by one fifth). This led to the current pattern of intensive arable production in the East and grassland dominance of the West. This included a significant change from permanent to temporary grassland to improve productivity and support an increase in livestock numbers. Around 15% (1 million hectares) of the nations rough grazing land was converted to improved arable or temporary pasture between the 1930’s and 1980’s. In part this was halted by reduction in incentives for cereal production under Common Agricultural Policy (CAP) regimes in the early 1980’s.
Clearly financial drivers play a significant role in land use change. It is the balance of enterprise profitability that in many cases determines land use and in financial terms affects the marginal profitability of enterprises. Less productive land is closer to the break-even economic margin and this is reflective of land where significant change in use is most likely to be observed.
Ongoing reform of the CAP, to reduce the cost burden on the EU, as well as recent changes in the set-aside rate (to increase cereal and oilseed production) will, and have led to changes in land use in the UK. In most cases, receipt of direct financial support from the EU is now linked to land management activities rather than to the growing of a specific crop, removing the link between crop or livestock production and the amount of subsidy received. Also, with reduced intervention support, agricultural production and prices are more reflective of changes in national and world market demand, leading to greater annual fluctuation in crop and livestock production and greater uncertainty for growers.
There is a demand for livestock products and with the exception of dairy production, few are in excess of current UK market requirements. However, livestock returns are typically lower than those of arable producers, and in many cases beef and sheep enterprises are unprofitable; reflected by declining animal numbers in recent years (Kilpatrick et al , 2008). Many current beef and sheep enterprises could consider conversion to biomass production as a more sustainable economic option. Intensification of livestock stocking rates within some areas to previous densities could also release grassland for bioenergy production. Such changes in land use may not be desirable in all cases, particularly where land has high organic soil carbon levels.
Other land has been abandoned due to contamination or other issues that make it unsuited to food production, or because it has become urbanised and isolated by incorporation into other infrastructure developments such as road and rail networks.
This work identifies abandoned or unused land that could be utilised for biomass production, as well as existing agricultural land that with appropriate economic drivers, could convert to bioenergy cropping. The likely sensitivities in the land areas affected are also examined.
2.3 Aims and objectives
The key objectives of this work are to
1. Identify areas of idle land or land of currently marginal economic production value in England and Wales that could be used to grow energy crops while minimising competition with existing food crop production
8 2. Ensure that any associated land use change does not have a significant impact on either the anticipated green house gas saving, or any other significant detrimental environmental impact, by identifying land types or protected land areas unsuited to bioenergy cropping 3. Identify appropriate bioenergy crops that could viably be grown on the land areas identified in 1) above, taking account of the yield potential and economic return that may accrue to growers as an alternative to the existing land use 4. Identify, on a spatial scale for England and Wales the areas of land that could potentially convert to energy crop production and the associated potential biomass yield in each land class examined, and the contributions that this could make towards meeting UK bioenergy targets
Using these objectives, this project will identify and quantify the maximum area of land suitable for biomass production, and the potential biomass yield that could be obtained from this area. The real availability of land is likely considerably smaller, but it will depend on various other factors and is outside the scope of this study.
3 Approaches
The key categories of land use identified for analysis of capacity for bioenergy production were
C) Existing land resources of agricultural value
a. Land currently fallow or in voluntary set-aside (conversion to annual or perennial energy copping)
b. Arable land where production of arable crops is of marginal profitability compared to production of energy crops (conversion to perennial energy cropping)
c. Grassland where stocking rates have declined (conversion to annual (temporary grassland only) or perennial energy cropping, or using under- utilised grass for bioenergy)
D) Land resources with no current agricultural value
a. Hedgerows and bracken
b. Urban spaces, road and rail margins
c. Brownfield land
The analysis relies on use of geographical information system (GIS) approaches, using spatial data and databases to map and quantify the land resource available for sustainable bioenergy production in defined land categories on a regional basis, as well as to identify areas where production would be constrained by environmental concerns and immovable pre-existing uses. Potential biomass production figures for the identified areas are derived using a mix of existing yield prediction functions and/or derived biomass or yield maps drawing on underlying mapping factors (described below). The economic returns for different land uses, drawing on a range of scenarios, are used to identify situations where land use change is most likely to occur on arable land.
Assessment of the impacts of land use change on soil carbon content are based on use of existing pertinent literature in the absence of a suitable spatial map of soil carbon content
9 and due to the difficulty in accurately modelling or describing changes in soil carbon turnover at a spatial scale. Example carbon payback times were calculated using published LCA studies for key energy conversion technologies in the liquid biofuel and electricity generation sectors.
3.1 Resource mapping approaches
The resource mapping work revolved around several steps that involved
• Identification of land unsuitable for energy crop production, or unlikely to be used for energy crop production (section 3.2) to exclude it from further GIS analysis
For the remaining area;
• Identification of fallow land (section 3.3) that could be used for bioenergy production and estimation of the potential biomass that could be generated from this land area using spatial estimates of yield potential for arable and perennial energy crops (sections 3.6.1 and 3.6.2)
• Identification of existing arable land that could potentially convert to perennial energy cropping given appropriate financial drivers (section 3.7) and estimation of the biomass yield that could be realised, utilising spatial estimates of yield potential.
• Identification of existing grassland that could potentially convert to annual or perennial energy cropping due to decline in livestock populations (section 3.8) and estimation of the biomass yield on any released land that could be realised, using spatial estimates of yield potential.
• Identification of additional non-agricultural land sources that could be better utilised for bioenergy production (section 3.4) and estimation of the potential biomass yields that could be obtained from various potential options.
The outputs of the above assessments are wherever possible expressed at the level of England and Wales, and broken down by the EU nomenclature of territorial units for statistics (NUTS) level 2 (see Figure 1 and Annex 1), except for current fallow land, where data was only available at NUTS 1 level. Data for grassland utilisation was also only analysed at a national level due to difficulties in clearly identifying how grassland is used in specific locations.
10
Figure 1. Regional boundaries (excluding Scotland) used for data aggregations (NUTS 2)
3.2 Mapping areas unsuited for bioenergy production
Drawing on similar approaches adopted by related studies funded by the UK Science Research Councils under the Rural Economy and Land Use Programme, (Lovett et al. , 2009 and Haughton et al ., 2009) land types deemed unsuitable for energy production were identified and discussed with Defra stakeholders (this also took account of land use considerations highlighted in the EU Directive on promotion of the use of energy from renewable sources) to agree a primary list of land areas that should be excluded (masked) from the analysis. This included a mix of habitat-related features, such as protected areas and landscapes of high environmental value many of which are well designated and mapped (Table 1). In addition, other features were masked including, slopes of greater than 15% (unsuited to mechanised planting and harvesting), urban areas, roads, lakes. Areas of woodland and forestry were also excluded as these represent complimentary sources of biomass and are unlikely to be converted to perennial or annual biomass crop production. These features were combined to produce a single mask (primary mask) to exclude these areas of land from the GIS assessment of land suitability for biomass production.
A secondary map of additional land class categories was drawn up to identify areas where there may be sensitivity to planting, but where there are no specific or absolute restrictions on planting currently. These included Environmentally Sensitive Areas, National Parks and Areas of Outstanding Natural Beauty. These areas were in addition to, those of the primary mask and were combined with the primary mask to produce a secondary mask. The impact of this is shown in the later results tables where data are presented based on the exclusion of both primary and secondary masked land areas. The difference between the two values help to identify the proportion of the potentially available land area where there may be wider social concerns over planting.
11 GIS datasets were rationalised from a range of spatial scales to a 1ha grid resolution for suitability mapping purposes using ArcGIS software version 9.3.1 (ESRI, Redlands, California, USA). To standardise the spatial scale for reporting, all the boundaries of excluded landscape features were converted to a 1 ha resolution. This was used to produce a mask to exclude the above land class features from the land and biomass resource assessment.
Table 1. Data sources associated with land areas and land features excluded from the spatial analysis
Feature Data source
1. Primary exclusion features Slopes steeper than 15% Ordinance survey profile from Digimap Existing woodland MAGIC (Multi-Agency Geographic Information for the Countryside, 2008 dataset) (ancient woodland) and National Inventory of Woodland and Trees, Forestry Commission (Community Forest areas) Urban areas SPIRE – 2001 Census of urban area – settlement data cover Lakes Ordnance Survey Mastermap Special areas of conservation, Special MAGIC Protection areas, Sites of Special Scientific Interest, National Nature Reserves, local nature reserves, RAMSAR sites, country parks, historic battlefields, Areas of Outstanding Natural Beauty (AONB), Environmentally Sensitive Areas, Heritage Coastline, common land World Heritage Sites, battlefields, Natural England GIS digital boundary datasets biodiversity action plan priority habitats, local nature reserves and country parks Registered parks and gardens, English Heritage scheduled monuments Salt marsh, littoral zones and montane CEH Landcover map 2000 (Classes 15.1, 18.1, habitats 19.1, 20.1, 21.1, 21.2)
2. Secondary exclusion features Environmentally Sensitive Areas, MAGIC Areas of outstanding natural Beauty, National Parks
12
Figure 2. Impacts of primary mask on available Figure 3. Impacts of combined primary and land resource (shaded). The primary mask secondary mask on available land resource removed 32.1% of the land area from (shaded). The secondary mask removed 47.0% consideration. of the land area from consideration.
3.3 Identification of arable, grassland and fallow areas where biomass production is possible
Following similar approaches adopted by other studies (Haughton et al .,2009) it was assumed that perennial energy crops would not be grown on the best agricultural land classification (ALC) of grade 1 land, as most energy crop planting agreements to date are on land in ALC grades 3 or 4 (Natural England, cited by Haughton et al .,2009).
Spatial information on areas and locations of temporary and permanent grassland and arable cropping were drawn from Edina datasets based on 2004 census data (the most recently available GIS dataset).
Reference to fallow in this study is taken to mean all forms of fallow land, including land under the now voluntary set-aside scheme. There may be limitations affecting future conversion of fallow land to bioenergy cropping, as Defra seeks options to compensate for the loss of environmental benefits associated with the decline in set-aside, which may further affect future availability of the fallow land resource for energy crop production.
Spatial information on set-aside fallow land is available from the Edina database. However, as this draws on the 2004 census as the most recent publicly available sources of information it is significantly out of date following the decision in 2007 to reduce the set-aside requirement to zero for the 2008 cropping year. More recent information is available via reports from the Defra Observatory project (Defra, 2009). Data from this report was used to identify the areas of fallow land in each Government Office region in England and in Wales as a whole.
13 3.3.1 Assessment of the biomass cropping potential of fallow land
It was assumed that the fallow land resource available for annual arable biomass crop production would be cropped in rotations of similar type and length to those currently in place in each Government Office region and for Wales. The proportion of the arable land area currently growing wheat and oilseed rape in each English Government Office region and in Wales was used to estimate the proportion of the arable land resource that would be cropped with wheat and oilseed rape each year. Regional average yields for wheat and rape were derived from the spatial wheat and oilseed rape yield maps (see section 3.5.1) to determine local production figures, and ensure compatibility in approaches in different aspects of the study. The anticipated ethanol and biodiesel yield from the additional wheat and oilseed rape area was estimated, as well as the potential additional straw biomass resource that could be exploited (see assumptions in section 14)
In addition, and in contrast to the above, it was also assumed that the whole un-cropped land resource could be used for perennial energy cropping. Average perennial energy crop yields for each government office region were calculated drawing on the spatial potential yield maps (see section 3.6). Along with appropriate net calorific values (see assumptions in section 14), these were used to calculate the potential bioenergy resource that could be realised by utilising the currently un-cropped arable land resource in each region.
3.4 Mapping ‘idle’ land resources
Table 2. Currently non-cropped areas that may be suited to bioenergy production and associated sources of spatial information
Feature Data source Roadside verges OS Master Map Due to the difficulties extracting large AA Roads data amounts of Master Map data, a sample of 1km tiles was used within each region to obtain an average verge area per length of road of each type. Regional totals were calculated by multiplying total road length by average area. Railway embankments OS Master Map Calculated by totalling embankment area in OSMM by region. Canal towpaths AA Canals data Used standard width of 5m for canal towpaths to buffer lines. Golf courses AA GB data Point data so had to assume an average size for golf courses from ‘A Guide to Golf Course Planning’ (European Institute of Golf Course Architects). 2 Sports turf England & Wales Areas per Local Authority Sports Councils Hedgerows Countryside Survey Statistical estimate of hedgerow length per 2007 country from sampled squares. Lowland Bracken CEH Land Cover Total area of bracken below moorland line Map 2000 (and therefore potentially harvestable) Brownfield Land National Land Use Point dataset of previously developed sites, Database (NLUD) with information on the type and area of the site. Only covers England and is likely to be incomplete.
2 The point data for golf courses simply show the location of the clubhouse and the LCM2000 data do not include a golf course category. Therefore, the actual area could only be obtained by manually digitising each course. As there are more than 2000 courses in England and Wales (English Golf Courses 2010), this was deemed impractical during this study. 14 There are a number of currently under-utilised sources of land that with the appropriate drivers could be used for bioenergy production. Previous work by ADAS has identified some of these (Kilpatrick et al , 2008). Further work was undertaken to look at the scale of the available source and the biomass resource capable of being generated by each. The key land uses examined were as detailed in Table 2. Forestry was not included in this analysis as the potential of forestry for biomass production has been studied in detail previously, including assessment of the currently under-utilised resource that could be used for biomass without compromising other end uses (see Kilpatrick et al , 2008).
The biomass production potential for these non-agricultural areas was estimated based on utilisation of either the current vegetation cover for biomass and assumptions regarding the yield potential, or by estimation of yields that could be obtained from planting the area with biomass crops (using spatial estimates of yields for biomass crops (primarily miscanthus) derived as detailed in Section 3.6)
Biomass yields in all cases are expressed as oven dry tonnes (odt) on an annual basis. Frequency of harvesting will vary depending on land use. Sports turf and golf courses will be cut frequently throughout the grass growing season. Hard wood trees on railway embankments will be cut back on long-term cycles of 5+ years.
In the majority of cases, current vegetation on the land resources detailed in Table 2 is a mixture of grass and trees. It has been assumed that annual yields will be either ‘low’, in the range 2-3 odt/ha or ‘medium’ in the range 4-6 odt/ha. Typical grassland yields were obtained from Frame (1992) whilst forestry yields were obtained from Tubby (pers.comm.). Where typical yields are known (e.g. Bracken) these are used.
For many categories, data is not available at NUTS2 level, although NUTS2 overlays could be applied to the maps to provide the appearance of a spurious precision. Data is therefore, presented at the NUTS1 level.
The viability of producing biomass on non-agricultural land will depend on a number of factors. The most important of these will be market demand for biomass within a reasonable distance of the point of production. In many cases, the biomass currently produced at these sites is actively managed, at not insignificant cost. Without seeking to minimise the challenges involved in harvesting, capturing and transporting biomass, this is likely to result in only marginal increases in the existing cost of managing such sites.
3.4.1 Hedgerows
Statistical estimates of the length of hedgerow in England and in Wales were obtained from Countryside Survey 2007 summary data (Countryside Survey data owned by NERC’s Centre for Ecology & Hydrology). The Field Survey is a very detailed study of a sample of 1km squares, located all over England, Scotland and Wales. The individual squares are chosen so that they represent all major habitat types in the UK. Estimates of the total quantity of habitat features (including hedgerow length) at a country level are made based on the area of each major habitat type in the country, and the density of the feature within sample squares for that habitat type.
3.4.2 Lowland bracken
The CEH Land Cover Map (2000) was used to define areas of bracken below the moorland line. This extracted data was overlain with regional boundaries to give areas of bracken below the moorland line by region. Checks were undertaken to confirm that there was no chance of double accounting with grassland areas. The two types of land use were found to be distinct.
15 3.4.3 Roadside verges
OS Mastermap data was used. The complexity of roadside verges in the Mastermap dataset makes the task of processing this at a national scale difficult. Instead, 10 random 1km sample tiles per region and road classification were used to calculate an average road verge area per kilometre of road. For each region the length of each road class was calculated, and the total verge area for the region is the sum of these road lengths multiplied by the appropriate average verge area per kilometre of road.
3.4.4 Railway embankments
OS Mastermap data was used. Railway embankment areas were extracted from the Mastermap dataset for the country. These were then overlain with the regional boundaries, giving areas of railway embankment per region.
3.4.5 Canal towpaths
No dataset containing areas classified as canal towpaths exists. However, using OS Mastermap and AA Canals data, an estimation of an average width of canal towpath is 5m. The AA Canals data was overlain with the regional boundaries, and the length of canal per region was calculated. This was then multiplied by the estimated average width of canal towpath (5m) to give a regional total canal towpath area.
3.4.6 Golf courses
AA data giving point location of golf courses was used. An average area for golf courses that are 9, 18, and 36 holes was taken from “A Guide To Gold Course Planning” (European Institute of Golf Course Architects). The point locations of golf courses were overlaid with regional boundaries to provide regional classification. Area of golf course was calculated using the appropriate average. The lack of readily available golf course boundary data and the large number of courses in England and Wales meant that it was impractical to derive real areas from the available point data (see Table 2 above).
3.4.7 Sports turf
Local Authority data on sports facilities was used, giving areas of turfed sports grounds. A lookup was used to sum these areas to provide regional totals.
3.4.8 Brownfield land
Four sources of data were used to make an estimate of the quantity of Brownfield land per region in England and Wales.
(i) National Land Use Database (NLUD), created to provide evidence to English Partnerships and Central Government on the extent of Previously Developed Land (PDL) in England. Site data is provided through annual returns by local planning authorities (LPAs). The size and location (X,Y coordinate) of the site is provided, along with attribute data such as the category of PDL, previous land use and general site information. The type of PDL recorded in NLUD considered suitable to locate biomass is category C – ‘Derelict Land and Buildings’. Site locations were buffered so that the buffer area matched the size of the site recorded in the database.
16 (ii) BRITPITS database of mineral workings is a list of mines and quarries in the UK including information on operational status (i.e. current or historic), purchased from the British Geological Survey (BGS). The dataset is supplied as point locations in GIS format. Historic sites were selected from the database and buffered to a distance that covered an area of 1 hectare, which is the median area of a mine/quarry.
(iii) Historic Landfill Dataset (HLD) is available free of charge from the Environment Agency for Defra funded projects, and details the boundaries of closed landfill sites in England and Wales as GIS polygons.
(iv) The phase 1 habitat map for Wales was provided by the Countryside Council for Wales. This dataset holds comprehensive habitat cover data for the whole of terrestrial Wales derived from a programme of field recording that was begun by the Wales Field Unit (WFU) of the Nature Conservancy Council (NCC) in 1979 and continued by the Countryside Council for Wales (CCW). Brownfield land classification included mine, quarry, spoil and tips, GIS polygons were selected for these land types.
Once all four datasets were represented as polygons in a GIS (i.e. the point data buffered), the polygons from all four were merged to avoid double counting of sites that were represented in more than one database. The resulting layer was intersected with regional boundaries to give a total per region, and with 5km cells to illustrate the spatial variation in the resource. However, this approach means that it is not possible to identify the type of Brownfield land resource available individually.
3.5 Mapping the yield potential and output of arable crops
3.5.1 Cereals and oilseed rape yields
There are no simple spatial yield prediction models for wheat or oilseed rape. Existing yield prediction models rely on specific data inputs related to both agronomic and environmental parameters that while effective for single site assessments are difficult to effectively model spatially. To derive an indication of agricultural productivity for each grid mapped square, data on cereal and oilseed rape yields from the Defra 2008 Cereal Production and Oilseed Rape Surveys were plotted spatially to identify the relationship with data on underlying soil class and natural fertility factors.
Winter wheat yield figures from 2374 fields in England were mapped using the geographical information system ArcGIS 9.3.1 (Environmental Systems Research Institute, Inc. (ESRI), Redlands, California). Information on soil fertility classes were obtained from NATMAP Soilscapes (National Soil Resources Institute (NSRI), Cranfield University, UK) and the Agricultural Land Classification (Natural England, Sheffield, UK). Wheat yield results located on land defined as land class 1, Urban and Non-Agricultural were excluded leaving 2144 fields. The Soilscapes data classifies the natural fertility of all soils in England and Wales into 12 categories of which 11 were coincident with wheat yield locations. The mean yield found in each combination of land class and fertility category was used to estimate the wheat yield for all unmasked land in England and Wales (amounting to 10.2 million ha after accounting for the primary mask and 8.0 million ha after accounting for the secondary mask). For land class and fertility category combinations where no yield data was available (typically representing areas where little or no wheat is grown) then the mean wheat yield for the respective parent land class was used. The land areas affected by this (after primary masking) by land class are shown below (Table 4).
17 Table 3. Data sources associated with derivation of arable crop yield maps
Feature Data source Agricultural land class NSRI NATMAP vector and Defra Soil fertility category NSRI NATMAP vector and Defra 2008 July census data, by farm holding, supplied Wheat and oilseed rape yields directly by Defra and mapped by Fera
Much of the land in class 5 is characterised as grassland rather than arable so the higher level of impact in this category is not unexpected. The areas of affected land in classes 2 to 4 are small relative to the total land areas available in each land class.
Table 4. Areas in each land class (after primary masking) where there was no reported field yield for wheat for particular ALC and fertility class combination
Agricultural Land class Area affected (hectares) 2 2,799 3 552 4 48,818 5 173,034
For oilseed rape, 622 oilseed rape field yields for 2008 were obtained. As the number of data points was limited in this case, mean field yields were only calculated for each soil fertility class, rather than for each ALC and fertility class combination. Mean yields for each fertility class were used as the yield estimate for all soils in the same fertility category. There were two soil fertility categories where no oilseed rape yields were represented. In these cases, land areas in such situations were allocated the mean oilseed rape yield associated with the underlying soil land class. As with wheat, the areas of land affected by this are relatively small (Table 5).
Table 5. Land areas in each soil fertility classes (after primary masking) where there was no reported field yield for oilseed rape Agricultural Land class Area affected (hectares) 2 659 3 8143 4 10054 5 443
All spatial yield maps were converted to a grid cell format with the same dimensions and a land cell size of 50m x 50m (0.25 ha). Masked areas were removed from the yield maps (see wheat yield map Figure 4).
3.6 Mapping the yield potential of perennial energy crops
3.6.1 Willow short rotation coppice yields
Willow short-rotation coppice yield potential was mapped using models derived by Forest Research. The yield potential map is based on yield estimates obtained from a network of 49 field experiments established across the UK during the mid to late 1990s. These experiments were part of the research project; “Yield Models for Energy Coppice of Poplar and Willow” funded by Defra, the former DTI and the Forestry Commission. Further details of this project are available from Forest Research (contact Ian Tubby tel: 01420 22255). The yield model, used to transform site-specific yield estimates for five willow varieties into a national map, takes into account the following variables:
18 • Annual rainfall • Seasonal rainfall (March to October) • Growing degree days • Frost days • Soil pH • Soil texture (sand, clay or loam).
Environmental data sets based on 5 x 5 km grid squares representing 30-year averages for each climatic variable are used along with soil data from the National Soil Resources Institute. The potential productivity of the following five willow varieties were considered (parent species in parenthesis). These varieties have either been planted commercially or are similar genetically to those planted commercially:
• Tora (Salix viminalis x S. schwerinnii) • Bjorn (Salix viminalis x S. schwerinnii) • Stott 10 (Salix burjatica x S. viminalis) • Jorunn (Salix viminalis x S. viminalis) • Jorr (Salix viminalis x S. viminalis).
Average yield estimates for these five varieties grown for two three-year cutting cycles in 5 x 5 km grid squares were calculated. The model includes a constraint that classifies sites at altitudes greater than 300 m above sea level as ‘unsuitable’.
All spatial yield maps were converted to a grid cell format with the same dimensions and a land cell size of 50m x 50m (0.25 ha). Masked areas were removed from the yield maps (see Figure 5).
3.6.2 Miscanthus yields
The potential yield of miscanthus was estimated using a yield map developed by ADAS. This map is based on the following dry matter accumulation equation, applied on a daily basis for the length of the growing season, with the cumulative yield calculated at the end of the growing season
n = ε ε YP()() k∑ S t k i ()k c d f() k k=1
-2 Where, ( Yp ) is the total above-ground dry matter yield at final harvest (g m ). The argument ()k denotes the value of the associated variable on the kth day during the growing season; -2 ε St is the daily incident solar radiation over the growing season (MJ m PAR); i is the ε efficiency with which the crop intercepts that radiation (dimensionless); c is the efficiency with which the intercepted radiation is converted into above-ground biomass (g d.m. MJ -1 PAR
intercepted); d f is the drought reduction factor (dimensionless) and n is the length of the
growing season. For potential yield d f is unity.
In order to estimate the potential yield across England and Wales, the model was applied to a 30-year time series of daily weather data generated by a stochastic weather generator including incident radiation, maximum and minimum temperature and rain days. Available water capacity was calculated using the SEISMIC soils database obtained from the National
19 Soil Research Institute (NSRI) which contained a statistical summary of the soil series present in each 1 × 1 km unit, and their moisture holding capacities. Land cover in each 1 x 1 km cell was derived from the ADAS Land Cover Database. For cells that contained more than 20% agricultural land, the dominant soil was extracted and the AWC calculated. The AWC values were then averaged over the 5x 5 km grid unit. The water-limited yield was calculated on a grid basis modified via d f to account for water availability acting as a limiting factor on yield. The model calculates total above ground dry matter production. However, significant leaf shedding and dry matter loss occurs after maturity. Clifton-Brown and Jones, (2001) reported in-field harvestable dry matter losses of 30% or more between maturity and harvest, while more recent investigations (Gezan and Riche,2008) suggest field losses in dry matter of around 20%. For this study, it was assumed that 25% of the biomass produced was left in the field as stubble or shed leaf litter.
All spatial yield maps were converted to a grid cell format with the same dimensions and a land cell size of 50m x 50m (0.25 ha). Masked areas were removed from the yield maps (see Figure 6).
3.6.3 Apportioning the available arable land resource to biomass and agricultural production
A further step in the process is accounting for the different land uses within each 50 x 50 grid square to ensure that production of biomass is apportioned only to the specific land types of interest, particularly in the case of the arable land resource.
The process for masking land as ‘unavailable’ is detailed in Section 3.2. The productivity of land is estimated by the spatial yield mapping approaches detailed above. This provides a grid (50 x 50m squares) of land area in England and Wales and an estimate of productivity for each square – assuming the whole grid area could be utilised. However in most cases only a proportion of this can be utilised. The EDINA Agcensus GIS database was used to derive spatial estimates of arable land availability within 2km x 2km grid cells. Overlaying the two grid sets provides an estimate of A) The land area that is available for cropping after masking, and B) the area of arable land that was present in the grid square in 2004.
In situations where the estimate from A is greater, not all the potentially available land is suited to, or used for arable production and so the proportion that is suitable was calculated (area in B as a proportion of area in A) the output from each 2km x 2km grid square (yield or financial return) is then reduced proportionately. In situations where the figure from B exceeds that of A, then it is assumed that the whole area identified in A can be utilised without adjustment.
20
Figure 4. Wheat yield potential (after primary Figure 5. Willow SRC yield potential (after Figure 6. Miscanthus yield potential (after masking) primary masking) primary masking)
21 3.7 Identification of arable land likely to convert to perennial energy cropping
To identify areas of arable land that could potentially convert from arable to perennial biomass crop production (under a given set of economic incentives), the most profitable land use option for all identified (and suitable (see section 3.2)) arable land was calculated.
The potential wheat, miscanthus and willow SRC yield that each hectare of arable land in England and Wales could support was derived (as described above). These were converted to potential revenues, by allocating prices to outputs for each crop. This was done according to a number of derived scenarios representing current, low and high market prices (see Table 6)
Table 6. Market prices for wheat and biomass used in scenario generation. Figures in bold represent typical values currently in the market place wheat willow SRC miscanthus Scenario £/t (85% dm) £/odt £/odt High £120 £90 £90 £100 £75 £75 £80 £60 £60 Low £60 £50 £50
The prices for wheat used in the above scenarios reflect the typical average grain prices observed in recent years (Table 7).
Table 7. Average annual grain price calculated from Defra's monthly wheat trade and price figures for the last 14 years Average Year price/tonne 2009 113.56 2008 130.91 2007 116.90 2006 79.17 2005 67.00 2004 76.00 2003 75.00 2002 65.00 2001 77.00 2000 68.00 1999 75.00 1998 77.00 1997 90.00 1996 122.00
The prices for bioenergy reflect figures reported by industry and collated in bioenergy market statistics, such as monthly market reports by Enagri; £60 per oven dried tonne is widely reported as a typical current market value for bulk-traded UK biomass crops.
Base costs of production/hectare for each crop were derived from default values used in earlier work undertaken by Fera to develop costing worksheets for farmers considering
22 energy crop production 3, which have been adopted and used by the biomass supply industry. The raw data used are shown in Annex 3. These include all planting and input costs and account for machinery, depreciation and labour costs. Costs related to output per tonne of product were stripped out (e.g. haulage and drying costs). All costs for perennial energy crops were annualised over a 13-year period. Planting grants were estimated at 40% of total incurred planting costs (the current repayment rate under the Energy Crop Scheme), but no EU energy aid payment was assumed to be available (following closure of the scheme). The base costs of production were calculated as £482/ha for wheat, £187/ha for willow SRC and £159/ha for miscanthus
As air dried SRC typically only achieves 65% dry matter and air dried miscanthus 85% dry matter, a drying cost of £24.50 per tonne was allocated to SRC and £10.50 to miscanthus to reflect the lower calorific value of typical air dried feedstocks and the current deductions observed in the biomass market. Haulage costs of £10 per tonne were allocated to biomass crops and £5.53/tonne to wheat to reflect the different bulk densities of each. A further baling cost of £10.60/tonne harvested was applied to miscanthus. Total enterprise margins were then calculated as follows for each crop type, at a spatial scale using the GIS yields, derived as described earlier.
SRC enterprise margin = market price (£/t) x Yield (t/ha) – (production cost (£/ha) + yield t/ha) x £34.50)
Miscanthus enterprise margin = market price (£/t) x Yield (t/ha) – (production cost (£/ha) + yield (t/ha) x £31.10)
Wheat enterprise margin = market price (£/t) x Yield (t/ha) – (production cost (£/t) + yield (t/ha) x £5.53)
As the dominant crop in arable rotations driving economic performance, the enterprise margin for wheat production was used as an indicator of current land use and profitability. Where enterprise margins for the perennial crops were greater than those estimated for arable production then it was assumed that conversion to energy crop production was feasible and use of the land was allocated to the most profitable of the two biomass crop options, drawing on underlying productivity estimates.
3.8 Identification of grassland likely to convert to arable or perennial energy cropping
It is very difficult to clearly identify at a spatial scale how any specific piece of grassland is being utilised in the UK. In addition, livestock moves through different types of grassland during production, from upland breeding to lowland finishing operations, which means there is an important dependent relationship between these operations affecting the demand for the grassland resource. Trying to utilise enterprise margins associated with grassland enterprises to identify where livestock production may not be the most profitable option is also fraught with difficulties, as found in previous studies (Kilpatrick et al , 2008). Kilpatrick et al (2008) and discussion with upland land managers reveals that many beef and sheep enterprises and smaller dairy enterprises are not profitable per-se and rely on underlying single farm payments to maintain a living. In such cases, profitability is not the only issue influencing how land is managed.
There are a number of different factors affecting trends in livestock populations. Different livestock sectors show different trends (Figure 7). Dairy cattle numbers have been in decline
3 Available to view via the NNFCC website at www.nnfcc.co.uk (search for ‘energy crops calculator’)
23 over decades, reflecting falling profitability in this sector, particularly for smaller herds. In contrast the beef herd had been gradually growing until early 2000, before declining. Currently the beef cattle population is showing small levels of growth. Sheep numbers peaked during the 1990s’ but the foot and mouth epidemic of 2001 significantly reduced sheep numbers that have not yet recovered to former levels. Under reform of the CAP, since 2005 ‘hedage payments’ (subsidies paid per head of livestock) have been replaced with a production-decoupled area-based single farm payment, and this will also be a factor affecting stocking rates.
In response to these changes in livestock populations, there has been a decline in temporary grass (Figure 7), by around 13% over the last decade alone. The area of permanent grass (grass >5 years old) had been in a slow gradual decline, but has shown significant growth since 2004, in part this may be due to EU requirements to protect and enhance the area of permanent grass in the EU.
12.00 4000.00
3500.00 10.00
3000.00 Dairy herd 8.00 total 2500.00 Beef herd total Total sheep and lambs one year old and over 6.00 2000.00 Grasses < five thou ha thou years old millionhead Grasses >five 1500.00 years old 4.00 Sole right rough grazing 1000.00
2.00 500.00
0.00 0.00
year
Figure 7. Changes in livestock numbers (millions) and the grassland resource (thousand hectares) in England between 1983 and 2009 (Defra June Survey data)
The temporary grassland area, which by its nature can more readily be converted to arable production, tends to track changes in livestock production. However, an examination of stocking rates over time indicates how efficiently livestock is using the available grass resource. Using the same methodology proposed by Kilpatrick et al , (2008), as stock rearing becomes more extensive as animal numbers decline, the grass resource is utilised less efficiently. Assuming that the remaining stock can be managed at former higher stocking rates, it can be estimated what proportion of the grass resource could potentially be released for other uses, including biomass production (given the appropriate drivers) and without affecting food production.
Beef, Sheep and Dairy cattle numbers were obtained for England and Wales, along with areas of temporary grass, permanent grass and rough grazing (Defra Statistics and Welsh Assembly Statistics). Unfortunately it was only possible to source grassland statistics that differentiated between temporary and permanent grassland in Wales back to 2000. It is
24 possible to trace such statistics back further for England, but this would then result in the use of stocking rates that are generally lower than those typical for Wales. To maintain consistency between data sets, changes in stocking rates between 2000 and 2004 (where spatial information is available on grassland locations) were used to identify changes in utilisation of the grassland resource in England and Wales over this period.
Different enterprises utilise grass resources to different degrees. Proportionate use figures derived by ADAS (Kilpatrick et al , 2008 (see Table 8)) were used to provide estimates of the total areas of grass resource available to each livestock type (grassland area/head) in each year from 2000 to 2004. To derive these, national average stocking rates were calculated for each livestock type and each grass resource type for each year (using the associated livestock numbers and grassland resources in each year).
These figures provide an indication of changes in the efficiency of grassland use by each livestock type over time. The highest levels of grassland utilisation/head (hectares/head) for each livestock type on each grass type was identified and used as an indication of maximum grassland stocking rate for each livestock and grassland type combination.
Table 8. % Utilisation of the grassland resource types in England and Wales by different livestock types (Kilpatrick et al , 2008) Temporary grass Permanent grass Rough grazing Sheep 15 25 85 Dairy 54 35 0 Beef 30 40 15
For each NUTS2 region, areas of permanent, temporary and rough grazing, and livestock numbers were derived from the June census data. The area of each grassland type available to each livestock category was allocated according to the ADAS proportionate use figures given above (Table 8). The maximum (most intensive) grassland use rate for each livestock type (Table 9) was used to calculate the minimum area of grassland resource required for the livestock numbers in each region. The difference between this figure and the grassland areas identified as available via the spatial 2004 EDINA database was used to identify where there was potentially any grassland resource that was not being utilised to its full potential.
Table 9. Maximum grassland utilisation rates by each livestock type (hectares/head) between 2000 and 2004 Livestock type Temporary Permanent Rough grazing Wales Sheep 0.002 0.021 0.017 Dairy 0.153 0.836 0 Beef 0.112 1.327 0.105 England Sheep 0.006 0.039 0.033 Dairy 0.216 0.639 0 Beef 0.242 1.473 0.120
However, livestock utilise many areas of land that would otherwise be unavailable for biomass production due to protected status or inability to cultivate etc. These are areas that are in many cases masked out from the spatial analysis (see section 3.2). It is assumed that the grassland encompassed by any masking would be used first to meet the grazing needs of livestock in the region.
The potential livestock carrying capacity of masked land was calculated using the derived maximum stocking rate. The grassland requirement for the remaining livestock that could not
25 be accommodated on the masked areas was calculated in similar fashion. The additional demand for each type of grassland was then compared with the actual availability in each region (after masking) and any surplus was deemed to be potentially available for alternative uses. In some cases a specific grassland resource may be indicated as being in deficit in a region. This is reflective of the inter-relationships between upland and lowland units, particularly in relation to sheep enterprises. However, summing to provide a national picture takes account of such regional discrepancies to provide an overall estimation of availability in England and Wales.
Recognising that different grass types typically reflect different underlying soil or climatic factors affecting productivity, estimates drawn from the spatial yield maps for arable and perennial energy crops were used to derive estimates of the average estimated yields that would be associated with production on land associated with temporary and permanent grassland (Table 10)
Table 10. Estimated average yields for arable and energy crop production on grassland sites Temporary grass Permanent grass Willow SRC 9.91 9.83 Miscanthus 12.14 11.68 Wheat 7.72 7.51 Oilseed rape 3.26 3.26
It is conceivable that retaining grassland in grass production and utilising any spare grass resource for bioenergy uses is also a potential option. While there is no spatial analysis of grass yield potential, for the grassland resource aggregated at a national level, the following production potentials were used; temporary grass 8-10 odt/ha, permanent grass, 4-6 odt/ha and rough grazing 1-3 odt/ha, following the values used by Kilpatrick et al , (2008).
3.9 Expression of biomass and biofuel energy contributions
As there is a clearly defined route of conversion from wheat and oilseed rape to liquid biofuels (for transport use) it is assumed that all wheat and oilseed rape is converted to bioethanol and biodiesel respectively. Estimated production potential for arable crops are converted to equivalent volumes of fuel (see assumptions (section 13)) for comparison against UK Renewable Transport Fuel Obligation (RTFO) targets, expressed by volume. In addition, the energy content of the fuel is also calculated for comparison against the EU RED targets. Therefore in the case of liquid biofuels, contributions to renewable energy targets are therefore expressed in terms of volumes of fuel and energy actually delivered .
Biomass can be used for heat or power generation or a combination of the two. In addition the efficiency with which energy in the feedstock is delivered as power varies between conversion processes. Currently, in most cases, generation of electricity from biomass occurs in large scale plants dedicated to biomass or co-firing biomass with fossil fuels or other biomass wastes. In such plants, energy conversion efficiency can range between 30- 40% and even less in some cases. By contrast, use in combined heat and power or heat- only applications can raise this to 80%+. Clearly the route by which biomass is exploited for energy generation will have a significant impacts on its contribution to delivered energy at point of use. As there is no clear policy to influence which routes biomass follows, all of the biomass produced on ‘idle’ or marginal land resources is expressed in terms of energy content in the feedstock. Contributions to renewable energy targets are therefore expressed in terms of primary energy. Actual delivered energy will be lower, and considerably so where used for electricity generation . However, as all fuels are
26 subject to the same general conversion efficiencies, it would be fair to equate the contribution from biomass energy to primary energy demand to that supplied as delivered energy.
4 The agricultural land resource
According to June census figures (Table 11) just under 5 million hectares of land on agricultural holdings in England and Wales is currently in arable cropping, or is bare fallow or temporary grassland that is capable of being cropped for both food and non-food uses There is a further 4-5 million hectares of land in long-term grass that may to a lesser extent be suited to bioenergy cropping.
Table 11. Cropped and grassland areas (thousand hectares) in England and Wales according to 2008 June Census results (provisional as at 3/9/2009)
England Wales Total Arable cropping 4,031 66 4,097 Bare fallow 4 159 8 167 Grass < 5yrs old 636 87 723 Grass > 5yrs old 3,429 1,017 4,446 Rough grazing 578 380 958 Total 8,834 1,558 10,391 Total excluding 4,826 161 4,987 permanent grass and rough grazing
In the spatial assessment of land suitability undertaken in this study, taking account of non- cropped areas (urban), slope restrictions, unsuitable areas and areas of environmental sensitivity as detailed earlier (primary mask), 10.25 million hectares of land was identified in England and Wales as being potentially suitable for bioenergy cropping (which compares well with June Census data (Table 11)). This included the exclusion of agricultural land in class 1, the most fertile soils in the UK, where it was deemed that due to horticultural and other high value use of such soils they are highly unlikely to be used for non-food cropping. When secondary features such as AONB’s and ESA’s are discounted, this fell to 8 million hectares of land (primary and secondary masks combined).
4.1 Fallow land resource
Since 1993, significant areas of land have been taken out of agricultural production under the EU set-aside scheme, designed to control the supply of home-produced cereals in the EU. In recent years, between 0.5 and 0.6 million hectares have been taken out of production in the UK annually. It is permitted to use such designated land for industrial crop production, including biofuel and biomass crop production. In the face of tightening cereal supplies, the compulsory set-aside rate was reduced to zero for the 2008 cropping year, allowing farmers to crop as much of their land as they wished. However, farmers are presently being strongly encouraged to maintain uncropped land for environmental reasons under the Campaign for the Farmed Environment. If voluntary targets are not met, mandatory areas of uncropped land may be reintroduced and so in practice not all former set-aside land is available for energy crop cultivation, at least in the short term.
4 Note figures derived from the June census data for uncultivated land in England are higher than those derived using single payment scheme data reported later in this section, as in the latter, areas in un-cropped agri-environment scheme options (e.g. wild bird or game cover) are excluded.
27 In addition to set-aside, other land may be left fallow between crops because of cultivation problems (late harvesting, poor soil or weather conditions during critical periods) or due to low productivity in years of anticipated low financial return. Between 165 and 185 thousand hectares of land have been reported as fallow or un-cropped in England and Wales in recent years, and as yet, this has not been significantly influenced by the removal of set-aside. The provisional June Survey figures for 2009 (17 Sept 2009) suggest that the area of un-cropped land in England has increased to 221 thousand hectares, but this estimate now also includes some 50 thousand hectares of un-cropped land in agri-environment schemes.
Clearly, removing the direct compulsion to include set-aside in the farmed area has resulted in a significant fall in the overall area of un-cropped land in England and Wales. Data from Defra’s June Survey estimates show that this has fallen by 63% in England between 2007 and 2008. It is difficult to identify comparable data for Wales due to aggregation in Welsh Assembly statistical datasets, but a fall of 7% was recorded in ‘other crops and bare fallow’. This represents a fall of over 260 thousand hectares in un-cropped land (set-aside and fallow) in England and Wales.
A detailed analysis of this change in areas of un-cropped land was undertaken for Defra’s Environmental Observatory project (Defra, 2009). This study estimated that, according to Single Payment Scheme (SPS) figures, there were 144 thousand hectares of arable land out of production in England (as GAEC12 5 land) in 2008. It is difficult to derive comparable data for Wales, but typically between 700 and 1000 hectares of land are typically left as un- cropped bare fallow annually, in addition to any set-aside requirement.
The Defra Observatory project also identifies from SPS returns how much of this un-cropped land is represented by field margins and corners (17.4%) and how much is non rotational (63.3%) and rotational in nature (19.3%- (Table 12). The removal of the compulsive element of set-aside has lead to an increasing proportion of the ‘un-cropped’ land resource being either located in field margins and corners (typically lower yielding or awkward to manage) or as land taken out of production on a more permanent or semi-permanent basis. The latter most probably represents land associated with difficulties in management or land of lower productivity.
Table 12. Un-cropped land in England, according to SPS data collated for Defra’s Environmental Observatory (Defra 2009) and its division into field margins/corners and by rotational/non-rotational history (thousand hectares) (Welsh data from Welsh Agricultural Statistics, Welsh Assembly) Government Total area of Field margins Non rotational Rotational Office Region un-cropped and corners (NUTS1) land NE 5.0 0.7 2.9 1.4 NW 3.4 0.4 2.5 0.5 Y&H 14.6 2.9 8.5 3.2 E Mids 26.0 4.7 15.3 5.9 W Mids 9.6 2.0 5.9 1.7 Eastern 42.5 7.4 28.0 7.1 SE 28.5 4.3 19.4 4.8 SW 14.4 2.7 8.7 3.0 England total 144.1 25.1 91.2 27.7 Wales 0.9 NA NA NA
The majority of the current un-cropped land resource is located in Eastern Counties of England (Table 12). Similarly detailed data was not available for Wales.
5 GAEC – Land reported in single payment system returns as in good agricultural and economic condition and managed as fallow or voluntary set-aside
28 A further investigation 6 of the length of time that non-rotational land had been taken out of arable production indicated that around 30% of non-rotational fallow or set-aside fields had been left un-cropped for ten or more years. Again the highest proportion of ‘older’ rotational fallow/set-aside can be found in the Eastern and South East counties of England.
4.2 Biomass supply from un-cropped arable land
The ability of the above identified land area to supply biomass for a range of end uses was evaluated, firstly its capacity to supply 1 st generation biofuels 7 and secondly to supply biomass for heat or power. It should be noted that there is some risk of double counting outputs from un-cropped arable land with those from arable land that may convert to bioenergy cropping. This arises because of the limits on access to up to date GIS datasets on the location of un-cropped or fallow arable land following the reduction of the set-aside rate to zero, which would help to separate production from these two potential resources.
4.2.1 Use of un-cropped land resource for biofuel production
4.2.1.1 Cropped with annual energy crops
Taking account of the available fallow land resource (un-cropped area in Table 12), and the frequency with which the available arable land resource is cropped with wheat and oilseed rape in each region (results for each individual region are tabulated in Annex 3), the potential wheat grain, wheat straw and oilseed rape production that this could support in England and Wales was derived (Table 13). It was also estimated what contribution these could make to current UK fuel demand, and in the case of wheat straw to UK electricity or heat demand. In total it was calculated that in England and Wales around 169,000 tonnes of ethanol and 29,000 tonnes of biodiesel could be derived from the full utilisation of the currently available fallow land resource, under current rotational practices, equating to 0.1% of UK diesel demand and 0.95% of UK gasoline demand (by volume), equating to just over 0.4% of UK total fuel use by volume (0.26% by energy content). In addition, a straw resource of around 333,000 tonnes could be generated, capable of meeting 0.14% of the primary energy demand consumed in UK electricity generation OR 0.17% of the primary energy consumed in UK heating applications. These represent upper limit potentials that would not impinge on current food production areas, and would fit with current land use. However conversion of this land to arable production would result in increased use of crop inputs, as much of this land resource in currently non-rotational in nature
The biomass potentials are based on utilisation of all of the current fallow land resource, which is unlikely to be realised in full in practice as land is left un-cropped for a number of agronomic, economic, environmental or social reasons.
4.2.1.2 Cropped with perennial crops
The impact of cropping the above fallow land resource with perennial energy crops is summarised in Table 13, with a fuller regional analysis in Annex 3.
6 Cropping records back to 1995 were available for 9,000 fields covering a total of 33,000 hectares. 7 1st generation biofuels refers to biofuels derived from simple sugar or starch crop feedstocks (bioethanol) or crop seed oils (biodiesel) utilising the conventional outputs from agricultural crops (e.g. wheat starch, corn starch, sugar beet or sugar cane). This represents the current technology route for most commercial biofuel production.
29 The biomass production potential in England and Wales was estimated at between 1.4m tonnes of biomass if cropped with miscanthus, rising to 1.7m tonnes if cropped with short rotation coppice, capable of supplying the equivalent of between 0.7-0.9% of the energy consumed in UK electricity generation OR 0.9-1.1% of energy consumed in UK heat applications. While this production would take place in arable-dominated landscapes, there may be some concerns related to visual impacts of any block planting. However, environmental impacts associated with use of crop inputs would be lower than if converted to arable cropping, due to low pesticide and fertiliser requirements of perennial energy crops. As mentioned above, these represent maximum use of suitable land; the actual area available may be smaller due to other economic, social, and ecological factors.
4.2.1.3 Comparison of total energy output
When cropped with annual energy crops a total of 252 toe of biomass energy could potentially be generated from the current fallow land resource, compared to 646-721 toe if cropped with perennial energy crops. This difference reflects on the intensity with which the land is currently likely to be cropped with cereal and oilseed crops. Increasing the intensity of wheat and oilseed cultivation is possible through shortening of rotations, but at the cost of increasing inputs to maintain yield levels as weed, pest and disease pressures will increase and both crops are relatively nitrogen demanding.
30 Table 13. Potential bienergy and biofuel resource obtained from planting un-cropped arable land in England and Wales with either arable or perennial biomass crops (note use for electricity or heat are mutually exclusive options) Engla nd Wales Total % of UK % of UK % (by volume) of UK Alternative land use options primary energy primary fuel use in transport consumption energy in electricity consumption generation in heating applications If cropped with annual arable biomass crops: wheat area (,000 ha) 72.5 0.2 72.7 wheat production (,000 tonnes) 574.5 1.2 575.7 wheat ethanol production (,000 tonnes) 168.9 0.4 169.3 0.95% UK petrol 0.40% total UK fuel use (0.25% total fuel PLUS energy) oilseed rape area (,000 ha) 21.3 <0.1 21.3 oilseed rape production (,000 tonnes) 71.0 0.1 71.1 biodiesel (FAME) production (,000 tonnes) 29.1 <0.1 29.1 0.10% UK Diesel = 0.06% total fuel use (0.06% total fuel PLUS energy) Wheat straw production ((,000 tonnes) 333.2 0.7 333.9 Wheat straw biomass energy (,000 toe) 119.0 0.2 119.2 0.14% 0.17% OR, If all cropped with perennial biomass crops (EITHER miscanthus OR willow SRC): miscanthus production (,000 tonnes) 1,750.5 10.8 1,761.4 miscanthus bioenergy (,000 toe) 721.1 4.4 725.5 0.88% 1.06% SRC production (,000 tonnes) 1,460.0 8.0 1,468.0 SRC biomass energy (,000 toe) 646.5 3.5 650.1 0.79% 0.95%
31 5 The economically marginal arable land resource for perennial bioenergy production
The financial return obtained from different agricultural enterprises is an important driver for change in land use, though it is recognised that it is not the only factor influencing change. To examine situations where arable production on current arable land results in returns that are lower than those that could be obtained from cropping with perennial energy crops on the same land, a range of output price scenarios were used (Table 6) to identify situations where significant areas of land could potentially realise greater returns by changing from arable to perennial energy cropping (as described in section 3.7).
The results of the analysis (Table 14 to Table 17) indicate that at current typical grain and biomass prices in the UK market place (c £100/t for wheat and £60/odt for biomass), perennial energy cropping could potentially be a more profitable option on 128,000 ha of suitable land in England and Wales, falling to 87,000 ha if excluding National Parks and AONB’s where there may be additional landscape concerns affecting use of such land. This land would be capable of delivering between 1.0 million tonnes (excluding National Parks and AONBs) and 1.7 million tonnes of biomass, which would equate to between 0.63 and 1.0% of total UK primary heat energy demand or 0.53-0.83% of total UK electricity primary energy demand (Table 19 and Table 20).
Figure 8 identifies the areas capable of providing biomass contributions towards these totals. In relation to the actual land area available in these areas to deliver biomass contributions, East and West Wales, Devon and East Anglia are amongst the largest contributors (a regional breakdown of biomass production potential is provided in Annex 5). Any move to restrict biomass crop planting in National Parks or AONB’s would have the greatest impacts in East Anglia, Devon, Surrey, East and West Sussex and Hampshire.
The results of the analysis from all the scenarios (Table 14 to Table 17) identifies that relatively small changes in the value of crop outputs has a significant impact on the comparative profitability of biomass crops. At current wheat prices and even at wheat prices of up to £120/t, and at a biomass price of £60/odt, biomass should be a profitable option on much more land than it is currently being grown on. Above this price, biomass crops would face no financial barrier to wider adoption. A return to recent low grain prices of around £80/t could encourage a significant shift to biomass crop production, and has the potential to deliver more than twice the biomass estimated as feasible under current pricing scenarios even at a lower biomass price of £50/odt. In contrast, a return to biomass prices of around £50/tonne at current wheat prices would be a significant disincentive. In this situation bioenergy crops would only potentially be the most profitable option on a mere 4,600 ha of land in England and Wales, not dissimilar to the current biomass planted area.
Uptake of biomass crops by UK farmers will, however, be affected by a number of other factors. Competition with readily available and inexpensive imported biomass could lead to lower prices being offered by power stations. Transport costs can also have an impact on price, with some generators reducing prices offered to growers to cover the cost and others requiring that farmers pay for transport themselves. Biomass crops are also seen as a risky investment by farmers. Costs of harvesting may be higher than anticipated and, even with support payments, it is a matter of years before costs can be recovered through the sale of biomass. Uncertainty over the long term support for perennial energy crops in the face of unforeseen changes in the economy and in government policy could also discourage adoption. This is in contrast with established and conventional food crops, which provide an immediate financial return and for which there is a clear long term demand. The perennial
32 nature of energy crops therefore represents a loss of flexibility in the view of some growers compared to annual conventional crops.
Clearly in all the above cases, conversion of land to biomass is at the expense of food crops, and so would have an impact on food production. If however, as expected, food prices rise in relation to any tightening supply, this would then shift the balance back to favour conversion back to food cropping.
Figure 8. Identification of areas where wheat, willow SRC or miscanthus production is estimated to be the most profitable enterprise for a specific location (wheat @ £100/t, biomass crops @ £60/odt)
33 Table 14. Land areas (,000 ha) associated with situations where perennial energy cropping could be more profitable than arable cropping for a range of different wheat and perennial energy crop prices (after primary masking) (totals may not match due to rounding errors). Shading indicates combinations where production of energy crops would be maximised based on estimated profitability. Bi omass price (£/odt) Wheat £50 £60 £75 £90 price SRC Misc Total SRC Misc Total SRC Misc Total SRC Misc Total £60/t Eng 16.6 3621.1 3637.5 183.6 3458.0 3641.6 382.9 3258.7 3641.6 466.1 3175.5 3641.6 Wales 0.1 62.8 63.0 0.2 62.7 63.0 0.6 62.3 63.0 0.7 62.2 63.0 Total 16.5 3684.0 3700.4 183.8 3520.7 3704.5 383.6 3321.0 3704.5 466.9 3237.6 3704.6 £80/t Eng 0.4 297.7 298.2 165.3 3318.0 3483.4 382.8 3258.2 3641.1 466.1 3.174.0 3641.1 Wales 0 36.6 36.8 0.2 62.4 62.6 0.6 62.3 62.9 0.7 62.2 62.9 Tota l 0.5 334.4 335.0 165.5 3380.4 3545.9 383.5 3320.5 3704.0 466.9 3237.1 3704.0 £100/t Eng 0 4.2 4.2 3.1 105.7 108.8 343.1 3102.1 3445.1 466.1 3174.7 3.640.9 Wales 0 0.3 0.3 0 19.5 19.5 0.6 61.6 62.1 0.6 62.1 62.8 Total 0 4.6 4.6 3.1 125.2 128.3 343.6 3163.6 3507.2 466.8 3236.9 3703.7 £120/t Eng 0 0.5 0.5 0 4.9 4.9 28.8 573.0 601.8 405.6 3018.2 3423.8 Wales 0 0.1 0.1 0 0.4 0.4 0.2 39.7 39.9 0.6 61.1 61.7 Total 0 0.6 0.6 0 5.3 5.3 29.0 612.6 641.6 406.3 3079.2 3485.5
Table 15. Biomass production (,000 tonnes) associated with situations where perennial energy cropping could be more profitable than arable cropping for a range of different wheat and perennial energy crop prices (after primary masking) (totals may not match due to rounding errors). Shading indicates combinations where biomass production by energy crops would be maximised. Biomass price (£/odt) Wheat £50 £60 £75 £90 price SRC Misc Total SRC Misc Total SRC Misc Total SRC Misc Total £60/t Eng 211.5 44216.0 44427.5 2482.3 42459.0 44941.3 5149.5 40199.5 45349.0 6199.9 39253.3 45543.1 Wales 1.2 757.8 758.9 2.1 757.0 759.1 8.3 751.7 760.0 9.6 750.4 760.1 Total 212.6 44973.8 45186.4 2484.4 43216.0 45100.3 5157.8 40951.1 46109.0 6209.5 40003.6 46213.2 £80/t Eng 6.8 3857.8 3864.6 2297.9 41146.8 43444.6 5149.5 40197.8 45347.3 6199.9 39251.5 45451.4 Wales 0.6 448.0 448.6 1.9 754.0 755.9 8.1 751.7 759.8 9.5 750.4 759.9 Total 7.4 4305.8 4313.2 2299.8 41900.7 44200.5 5157.6 40949.4 46107.0 6209.3 40001.9 46122.2 £100/t Eng 0 53.4 35.4 43.5 1374.7 1418.2 4748.4 38685.3 43433.7 6199.5 39250.0 45449.6 Wales 0 4.2 4.2 0.1 242.4 242.5 7.6 745.2 752.8 8.9 750.1 759.1 Total 0 57.5 57.5 43.6 1617.1 1660.7 4756.0 39430.4 44186 6208.5 40000.1 46208.6 £120/t Eng 0 6.3 6.3 0.5 61.4 61.9 401.4 7527.3 7928.7 5591.4 37705.5 43296.9 Wales 0 0.6 0.6 0 5.5 5.5 2.7 491.0 493.7 8.9 740.2 749.1 Total 0 6.9 6.9 0.5 66.9 67.4 404.2 8018.3 8422.5 5600.3 38445.7 44046.0
34 Table 16. Land areas (,000 ha) associated with situations where perennial energy cropping could be more profitable than arable cropping for a range of different wheat and perennial energy crop prices (after secondary masking) (totals may not match due to rounding errors). Shading indicates combinations where production of energy crops would be maximised based on estimated profitability. Biomass price (£/odt) Wheat £50 £60 £75 £90 price SRC Misc Total SRC Misc Total SRC Misc Total SRC Misc Total £60/t Eng 13.4 3042.5 3055.9 164.5 2895.5 3060.0 343.3 2716.8 3060.0 413.8 2646.3 3060.1 Wales 0.1 50.6 50.7 0.1 50.6 50.7 0.6 50.1 50.7 0.7 50.0 50.7 Total 13.5 3093.2 3106.6 164.6 2946.1 3110.7 343.9 2766.9 3110.7 414.5 2696.3 3110.8 £80/t Eng 0.2 216.3 216.5 153.6 2801.5 2655.1 343.3 2716.3 3059.6 413.8 2645.7 3059.5 Wales 0.0 30.4 30.4 0.1 50.4 50.5 0.5 50.1 50.7 0.6 50.0 50.7 Total 0.2 246.7 246.9 153.7 2851.9 3005.5 343.9 2766.4 3110.2 414.5 2695.7 3110.2 £100/t Eng 0.0 2.8 2.8 1.8 68.6 70.4 321.5 2612.6 2934.1 413.8 2645.7 3059.4 Wales 0.0 0.3 0.3 0 16.5 16.5 0.5 49.7 50.2 0.7 50.0 50.7 Total 0.0 3.1 3.1 1.9 85.1 87.0 322.0 2662.3 2984.3 414.4 2695.7 3110.1 £120/t Eng 0.0 0.5 0.5 0.0 3.1 3.1 23.2 442.4 465.6 380.9 2542.2 2923.1 Wales 0.0 0.0 0.0 0.0 0.4 0.4 0.2 32.6 32.8 0.7 49.3 50.0 Total 0.0 0.5 0.5 0.0 3.5 3.5 23.4 475.0 498.4 381.5 2591.5 2973.0 Table 17. Biomass production (,000 tonnes) associated with situations where perennial energy cropping could be more profitable than arable cropping for a range of different wheat and perennial energy crop prices (after secondary masking) (totals may not match due to rounding errors). Shading indicates combinations where biomass production by energy crops would be maximised. Biom ass price (£/odt) Wheat £50 £60 £75 £90 price SRC Misc Total SRC Misc Total SRC Misc Total SRC Misc Total £60/t Eng 171.1 36433.0 36604.1 2191.2 34870.8 37062.0 4567.0 32856.8 37423.7 5456.8 32053.8 37510.5 Wales 0.6 583.6 584.2 1.5 582.9 584.4 7.2 577.9 585.1 8.4 576.8 585.2 Total 171.7 37016.6 37188.3 2192.7 35453.6 37646.3 4574.1 33434.7 38008.8 5465.2 32630.5 38095.7 £80/t Eng 3.1 2661.6 2664.7 2082.6 33997.5 36080.1 4567.0 32855.3 37422.2 5456.8 32052.0 37508.8 Wales 0.1 352.1 352.2 1.3 581.2 582.5 7.1 577.9 585.0 8.3 576.8 585.1 Total 3.2 3013.6 3016.9 2083.9 34578.7 36662.6 4574.0 33433.2 38007.2 5465.1 32628.8 38093.8 £100/t Eng 0.0 30.6 30.6 24.7 833.5 858.3 4351.0 31866.3 36217.2 5456.5 32051.0 37507.5 Wales 0.0 3.4 3.4 0.1 193.4 193.5 7.1 574.1 581.2 8.3 576.7 585.0 Total 0.0 34.0 34.0 24.8 1027.0 1051.7 4358.0 32440.4 36798.4 5464.8 32627.7 38092.5 £120/t Eng 0.0 5.2 5.2 0.2 34.8 35.0 311.9 5563.7 5875.6 5133.8 31047.3 36181.1 Wales 0.0 0.2 0.2 0.0 4.5 4.5 2.6 383.3 385.8 8.2 570.8 579.0 Total 0.0 5.4 5.4 0.2 39.3 39.4 314.4 5947.0 6291.4 5142.1 31618.0 36760.1
35 Table 18. Percentage of UK primary energy demand for either heat or electricity generation that could be met from the potential production of biomass (see table Table 15) identified under a range of financial scenarios (after primary masking) Biomass price Wheat price £50 £60 £75 £90 1. % UK heat primary energy demand £60 27.19% £80 2.59% 26.69% £100 0.03% 1.00% 26.79% £120 <0.01 0.04% 5.08% 26.75% 2. % UK electricity primary energy demand £60 22.64% £80 2.16% 22.22% £100 0.03% 0.83% 22.31% £120 <0.01% 0.03% 4.23% 22.27%
Table 19. Percentage of UK primary energy demand for either heat or electricity generation that could be met from the potential production of biomass (see Table 17) identified under a range of financial scenarios (after secondary masking)
Biomass price Wheat price £50 £60 £75 £90 1. % UK heat primary energy demand £60 22.38% £80 1.81% 22.15% £100 0.02% 0.63% 22.33% £120 <0.01% 0.02% 3.78% 22.34% 2. % UK electricity primary energy demand £60 18.63% £80 1.51% 18.44% £100 0.02% 0.53% 18.59% £120 <0.01% 0.02% 3.15% 18.61%
6 The potential grassland resource for bioenergy production
Historic stocking rates (2000-2004) were used to identify areas of temporary, permanent and rough grazing land currently being less intensively utilised and where a return to previous rates of utilisation could release ‘underutilised’ grassland for other uses (see section 3.8). After accounting for livestock use of sensitive areas of grassland, which would be unsuitable for biomass production, the remaining ‘underutilised’ grass resource was estimated. The results for individual NUTS2 regions were very variable across grass types (see Annex 6) as might be expected, with a range of deficits and surpluses reflecting the different availability of each type of grassland in each region in relation to dominant livestock type and type of livestock operation (breeding or finishing) in each region. This reflects the inter-connected nature of grassland use, with livestock moving between regions to exploit grassland resources most effectively. A national overview was taken of the grassland resource to overcome the limitations posed by regional analysis. While this indicates the scale of the resource available, it is difficult to clearly identify spatially where such resources may become available. However the results of the regional analyses were included (see Annex 6) to assist identification of regions in England and Wales where relatively large areas of grassland resource are indicated as being potentially ‘underutilised’.
36 At a national level, just under 90,000 ha of temporary grassland (12.1% of the temporary grassland in England and Wales) was identified as potentially ‘underutilised’, though no spare capacity was found in Wales (Table 20). Exclusion of areas of potential landscape concern made little difference to the potential area available. This represents land that has been in grass for 5 years or less, and can be relatively easily returned to arable cropping in most circumstances. The relatively small scale of land identified reflects that the temporary grass resource tends to follow trends in livestock populations relatively closely compared to permanent grassland, where other options for use are typically more limited.
Conversion of temporary grassland to biomass cropping would have no direct impact on food production, but may require inputs to the remaining grassland to be increased to maintain productivity/livestock carrying capacity. As this land resource has been in grass for less than 5 years, it is assumed that there would be little impact on soil carbon from conversion to either arable or perennial energy cropping.
At a national level a significant area of permanent grass, around 0.5m hectares (11.3% of the permanent grassland in England and Wales) was identified as potentially ‘underutilised’: As with temporary grassland, exclusion of areas of potential landscape concern made little difference to the estimated resource area available (Table 20). This resource represents land that has been in grass for 5 years or more. This land resource is more difficult to convert to arable cropping due to legislative and agronomic constraints.
As with temporary grass, conversion to biomass cropping would have no direct impacts on food production, though inputs to the remaining grassland may need to be increased to support the higher stocking rates. However, as a high soil carbon stock, impacts of land use change need to be considered. The impacts of land use change on soil carbon are detailed in Section 8. Conversion of permanent grassland to arable cropping for biofuel production would result in a soil carbon loss that would take several decades to recoup through the carbon gains derived from use of the biofuels produced, much more than the 10-year maximum payback period proposed by the Gallagher Review of the RTFO. For this reason it is concluded that such land is unsuited to so called ‘1st generation’ ( see footnote 7) biofuel feedstock production from wheat and oilseed rape.
The impacts of converting permanent grassland to perennial energy cropping on soil carbon are less well understood (see Section 8). IPCC default values suggest that conversion of grassland to perennial crop systems could result in a carbon payback time of less than 20 years using some technologies (e.g. miscanthus for electricity and gassification of SRC for electricity, although this has yet to be commercialised in the UK), although this would not meet the RTFO sustainability requirement of 10 years (where biomass is intended for use in transport fuel production). Other sources suggest that conversion to miscanthus or SRC could have a minimal effect on soil carbon, or even lead to net carbon sequestration. Therefore, the existing data are inadequate and definitive conclusions about the effect of perennial crops on grassland soil carbon stocks cannot be drawn. In this study it is assumed that permanent grassland could be converted to perennial energy crops for uses where GHG balances were well studied, take into account any land use change impacts and by doing so meet minimum specified GHG saving requirements (as specified in the EU Renewable Energy Directive). The impact on soil carbon flux of grassland conversion to perennial energy cropping is an area that needs further detailed study.
For the reasons outlined in section 3.8, due to limited availability of data for Wales, changes in livestock stocking rate were only considered between 2000 and 2004. However, it is possible to examine stocking rates over a longer period for England, taking into account the higher stocking densities seen since the early 1980’s. Using the same approach, the ‘under- utilised’ grassland resource available could increase by up to 60% for temporary grassland and by up to 84% for permanent grassland. However, permanent pasture is protected under
37 current Single Farm Payment ‘cross compliance’ measures. Under these measures, the area of permanent pasture cannot be reduced by more than 5% of the total agricultural area (as recorded in 2003). This represents a maximum permitted change of around 0.8m ha. The 641 thousand hectares of grassland identified as ‘under-utilised; by use of more recent changes in stocking rate therefore represents a more realistic and contemporary assessment of potential resource availability.
At a national level, there was found to be very little spare capacity in rough grazing land, and when additional possible landscape concerns are taken into consideration by exclusion of areas in National Parks or AONB’s, then no rough grazing land was identified as under- utilised (Table 20). Rough grazing represents areas of land that are of biodiversity and habitat value and in some case would be difficult to bring back into arable or energy crop production. A small resource ‘surplus’ was identified in England, with Cumbria, Northumberland and Tyne and Wear being the largest contributors to this. However, as a relatively small resource and in view of other potential habitat concerns, this was excluded from the analysis of areas for potential biomass production.
Table 20. Grassland areas (ha) identified as ‘under-utilised’ after primary and secondary masking Temporary grass Permanent Rough grazing Grass Primary mask England 89,078 413,994 31,539 Wales Nil 89,205 1,360 Net Total 87,516 503,199 30,179 Secondary mask England 83,478 383,996 Nil Wales Nil 89,205 Nil Net Total 81,916 473,200 Nil
6.1 Use of temporary grassland for biofuel and biomass production
Based on the estimates of ‘under-utilised’ temporary grassland in England and Wales, it was estimated what this land resource could potentially support if converted to arable production. It was assumed that the released land would be cropped in the same proportions as current arable areas nationally, and appropriate national yields were used based on estimates obtained for soils and land areas represented by temporary grassland (Table 10).
As this represents additional land converting to arable production and without affecting current livestock use, it is assumed that this change of use represents very little competition with use of land for food use. In fact as part of a rotation, it would increase the output of other food crops grown as part of the rotation.
As an alternative and for comparison, it was also estimated what the existing grass resource itself could deliver as a potential biomass resource if not fully utilised for livestock production.
6.1.1 Cropped with annual biofuel crops
It was assumed that wheat and oilseed rape would be grown in rotation providing feedstocks for biofuel production and wheat straw as a bioenergy resource. Due to lack of available grassland resource no production was assumed in Wales. It was estimated that 341.4 thousand tonnes of cereals, and 40.5 thousand tonnes of oilseed rape could be produced on the under-utilised temporary grass resource (Table 21). This assumes that there would be no particular landscape impact concerns as much of this land would be in a mixed farming landscape.
38 This could make a contribution of 0.56% to current Ultra Low Sulphur Petrol (ULSP) demand and 0.06% to current Ultra Low Sulphur Diesel (ULSD) demand (by volume) and 0.27% to UK fuel use overall (by volume)
In addition around 200 tonnes of wheat straw would be generated capable of meeting 0.09% of UK primary energy demand for electricity and 0.1% of UK primary energy demand for heating applications.
This amounts to a total primary bioenergy output of 149 toe for this option.
6.1.2 Cropped with perennial energy crops
It was estimated that 867 thousand tonnes of willow SRC or 1.06 m tonnes of miscanthus could be produced on the under-utilised temporary grass resource (Table 21), without impacting on food production. If landscape impacts are a concern in some more sensitive areas (National Parks and AONB’s) this could fall to 811 and 994 thousand tonnes respectively. This is capable of supplying the primary energy requirements of around 0.5- 0.6% of UK primary heat OR electricity demand, falling slightly where landscape concerns are a potential issue (Table 21).
In total this option could deliver a total of up to 384-437 toe in biomass primary energy production (359-410 toe where possible landscape concerns constrain production)
6.1.3 Use of the resource for bioenergy
There is no reason why the grass resource itself cannot be utilised for biomass generation, either as a dried biomass resource or as a ‘wet’ feedstock for anaerobic digestion operations. Based on typical grassland production rates, it was estimated that the ‘under-utilised’ temporary grass resources could deliver between 713-890 odt with no landscape restrictions on use. However this level of productivity is dependent on maintaining fertiliser and other inputs to grass to optimise yield, and rotational re-sowing to maintain high yields. This resource is capable of supplying 0.2-0.3% of UK primary heat OR electricity demand.
In total, this option could deliver 171-214 toe in primary energy, which is more than could be delivered through conversion to arable production, but significantly less than the potential if cropped with perennial energy crops.
39 Table 21. Potential total liquid biofuel and biomass contributions that could be derived from arable cropping of all of the available under-utilised temporary grass resource in England and Wales (% contributions based on net total biomass potential) England Wales Net % of UK % of UK % by volume of UK Alternative land use options Total primary energy primary fuel use in transport consumption energy in electricity consumption generation in heating applications A) PRIMARY MASK wheat area (,000 ha) 44.5 No net 44.2 0.56% ULSP wheat production (,000 tonnes) 343.8 resource 341.4 0.235% total volume wheat ethanol production (,000 tonnes) 101.1 100.4 (0.149 %by total PLUS energy) oilseed rape area (,000 ha) 12.4 12.4 oilseed rape production (,000 tonnes) 40.6 40.5 0.06% ULSD biodiesel (FAME) production (,000 tonnes) 16,7 16.6 0.033% total volume PLUS (0.03% by total energy) Wheat straw production ((,000 tonnes) 199.4 198.0 Wheat straw biomass energy (,000 toe) 71.2 70.7 0.086 0.103 B) SECONDARY MASK wheat area (,000 ha) 41.7 41.4 wheat production (,000 tonnes) 322.2 No net 319.8 0.53% ULSP wheat ethanol production (,000 tonnes) 94.8 resource 94.1 0.220% total volume PLUS (0.140% by total energy) oilseed rape area (,000 ha) 11.7 11.6 oilseed rape production (,000 tonnes) 38.1 37.9 0.05% ULSD biodiesel (FAME) production (,000 tonnes) 15.6 15.5 0.031% total volume PLUS (0.03% by total energy) Wheat straw production ((,000 tonnes) 186.9 185.5 Wheat straw biomass energy (,000 toe) 66.7 66.3 0.080 0.096
40 Table 22. Potential biomass contributions that could be derived from cropping of all of the available under-utilised temporary grass resource in England and Wales with either willow SRC or miscanthus (% contributions based on total biomass potential) England Wales Total % of UK % of UK primary Alternative land use options primary energy energy consumption consumption in (Willow SRC or Miscanthus) in electricity heating generation applications A) PRIMARY MASK SRC Production (,000 odt) 882.7 No net 867.3 SRC energy content (,000 toe) 390.9 resource 384.1 0.466 0.559 OR Miscanthus production (,000 odt) 1,081.1 1,062.4 Miscanthus energy content (,000 toe) 445.4 437.6 0.530 0.637 B) SECONDARY MASK SRC Production (,000 odt) 827.3 No net 811.8 SRC energy content (,000 toe) 366.4 resource 359.5 0.436 0.523 OR Miscanthus production (,000 odt) 1,013.4 994.5 Miscanthus energy content (,000 toe) 417.4 409.6 0.497 0.596
Table 23. Potential biomass contributions that could be derived from harvesting grass from the available under-utilised temporary grass resource in England and Wales (% contributions based on total biomass potential). Ranges indicate production from lower (8 odt/ha) to upper (10 odt/ha) grass yield potentials England Wales Total % of UK % of UK prim ary Alternative land use options primary energy energy consumption consumption in Grass for bioenergy uses in electricity heating generation applications A) PRIMARY MASK Grass production (,000 odt) 712.6-890.8 No net 700.1-875.2 Energy content (,000 toe) 174.7-218.4 resource 171.7-214.6 0.208-0.260% 0.250-0.312% B) SECONDARY MASK Grass production (,000 odt) 667.8-834.8 No net 655.3-819.2 Energy content (,000 toe) 163.8-204.7 resource 160.7-200.9 0.195-0.244% 0.234-0.292%
41 Table 24. Potential biomass contributions that could be derived from cropping of all of the available under-utilised permanent grass resource in England and Wales with either willow SRC or miscanthus (% contributions based on total biomass potential) England Wales Total % of UK % of UK primary Alternative land use options primary energy energy consumption consumption in (Willow SRC or Miscanthus) in electricity heating generation applications A) PRIMARY MASK SRC Production (,000 odt) 4,070 877 4,946 SRC energy content (,000 toe) 1,802 388 2,190 2.655% 3.189% OR Miscanthus production (,000 odt) 4,835 1,042 5,877 Miscanthus energy content (,000 toe) 1,992 429 2,421 2.934% 3.542% B) SECONDARY MASK SRC Production (,000 odt) 3,775 877 4,652 SRC energy content (,000 toe) 1,671 388 2,060 2.497% 2.999% OR Miscanthus production (,000 odt) 4,485 1,041 5,257 Miscanthus energy content (,000 toe) 1,847 429 2,277 2.760% 3.314%
Table 25. Potential biomass contributions that could be derived from harvesting grass from the available under-utilised permanent grass resource in England and Wales (% contributions based on total biomass potential). Ranges indicate production from lower (4 odt/ha) to upper (6 odt/ha) grass yield potentials England Wales Total % of UK % of UK primary Alternative land use options primary energy energy consumption consumption in Grass for bioenergy uses in electricity heating generation applications A) PRIMARY MASK Grass production (,000 odt) 1655-2484 357-535 2012-3019 Energy content (,000 toe) 406-609 88-131 493-740 0.598-0.897% 0.719-1.078% B) SECONDARY MASK Grass production (,000 odt) 1536-2304 357-535 1892-2839 Energy content (,000 toe) 377-565 88-131 464-696 0.563-0.844% 0.676-1.014%
42 6.2 Use of permanent grassland for biofuel and biomass production
Based on the estimates of ‘under-utilised’ permanent grassland in England and Wales (Table 20), it was estimated what this land resource could support if converted to perennial energy crop production or if retained in grass and used as a biomass resource. As with the temporary grassland resource it is assumed that change of use to biomass production represents very little competition with use of land for food use, though it is recognised that there could be some sensitivities in certain locations with regard to landscape issues where perennial energy crops are planted.
As indicated earlier, the amount of land that can be converted to perennial energy crops could be limited by SFP Cross Compliance regulations designed to protect the permanent grass habitat, though the available resource highlighted by this exercise could be accommodated within the area tolerance allowed for change, though it is accepted that concerns might be raised if the upper end of the potential was actually exploited.
6.2.1 Cropped with perennial energy crops
It was estimated that 4.9 million tonnes of willow SRC OR 5.9 m tonnes of miscanthus could be produced on the under-utilised temporary grass resource (Table 21), without impacting on food production. If landscape impacts are a concern in some more sensitive areas (National Parks and AONB’s) this could fall to 4.7 and 5.3 million tonnes respectively. This is capable of supplying the primary energy requirements of up to 2.7-2.9% of UK primary electrical energy demand OR 3.3-3.5% of UK primary heat energy demand if used to the full.
Where production in landscape sensitive areas were excluded, the available land resource is capable of supplying around 2.5-2.8% of UK primary electrical energy demand OR 3.0-3.3% of UK primary heat energy demand.
In total this option could deliver a total of up to 2.2-2.4 million toe in biomass primary energy production (2.1-2.3 million toe where possible landscape concerns constrain production).
6.2.2 Use of the perennial grass resource for bioenergy
Based on typical grassland production rates, it was estimated that the ‘under-utilised’ permanent grass resources could deliver between 2-3 million odt with no landscape restrictions on use. However this level of productivity is dependent on maintaining inputs to grass to optimise yield. If realised, this resource is capable of supplying 0.7-1.1% of UK heat primary energy OR 0.6-0.9% of UK electrical primary energy demand.
In total, this option could deliver 493-740 million toe in primary energy, which although a fraction of that delivered through conversion of the same area to perennial energy cropping, would face fewer restrictions in terms of landscape impacts and soil carbon depletion. In contrast inputs, particularly of fertiliser would be higher to grass production. This is an area where again more work is required to compare the full carbon balance of such options to encourage the most sustainable use of the available resource.
43 7 The ‘idle’ land resource suitable for bioenergy production
The following is an assessment of the idle land resource suitable for bioenergy production in the UK. This represents the maximum area that could be used and the actual area available is likely to be smaller due to the logistical, economic, social, and policy implications associated with use of these land resources.
7.1 Hedgerows
Only very limited data for hedgerows is available which shows a total linear length of 455,720km comprising 401,900 in England and 53,820 in Wales. If a mean width of 1.5 metres is assumed, this equates to a total area of 68,358ha. An annual broadleaf tree yield of 3 odt/ha may be assumed, giving a potential annual yield of 205,074odt (91,105 toe). This refers to the annual incremental growth, additional to the hedge’s trimmed size and shape. As now, hedges could continue to be trimmed at various intervals, annually or less frequently. As an agricultural operation it is quite easy to imagine that the material could be readily collected by simple adaptation of existing equipment. Taken together with a local requirement for woodchip, this could make a valuable contribution to local biomass production.
7.2 Lowland bracken
It is assumed that all lowland bracken, by definition, is harvestable. An area of 30,836 ha of bracken grows below the moorland line. Annual yields of bracken have been established by Lawson et al. (1986) at either 7 odt/ha (low) or 12 odt/ha (high). Potential annual yield is calculated to be 215,855 odt (108,267 toe) or 370,037odt/ha (185,602 toe) for `low` and `high ` estimates respectively. Whilst this is very significant in both area and yield potential, effective utilisation for heat on a local basis, is probably the only potential use which might be considered viable when compared to other biomass crops; if only because of the potential logistical challenges likely to be encountered in trying to harvest this resource. Although harvesting bracken for fuel may not constitute a change in land use, it is not clear how exploitation of this natural resource would be viewed under the terms of the EU Renewable Energy Directive (RED) if by removal this altered the biodiversity of the natural habitat, or significantly reduced carbon returns to the soil. Taking account of sensitive environmental habitats would reduce the available bracken resource to 23,000 ha (primary mask).
Further work would be required to evaluate any impacts of bracken removal. However, as bracken is viewed as a ‘problem` weed on agricultural land, and is cut or sprayed off, it is assumed that harvesting for biomass would be viewed favourably. Areas of bracken and spatial assessment of the density of potential annual biomass production is shown in Figure 9 and Figure 10 respectively.
44
Figure 9. Density of land areas associated with Figure 10. Density (t/ha) of potential annual lowland bracken bracken biomass production (low yield estimate)
7.3 Roadside verges
The total area of land occupied by roadside verges is calculated to be 126,706 ha. Assuming a low annual yield of 3 odt/ha, this represents a biomass potential of 380,118 odt (131,191 toe equivalent) or 633,530 odt (218,652 toe) at a medium annual yield of 5 odt/ha. The spatial assessment of estimated annual yield potential (assuming low yield potential) is shown in Figure 11.
Roadside verges represent a good example where the costs of managing grass and tree growth is already borne by the road operators and the additional cost of capturing and transporting the biomass would be fairly marginal. The logistical feasibility of such an operation has already been demonstrated. In the 1970’s grass was harvested and made into hay across a significant proportion of the UK’s major roads.
Whilst a key requirement of roadside verges is that they maximise driver-vision, and therefore only grass can be grown, there are parts of the system where trees are grown. In this instance, it could be possible to plant fast growing SRC or miscanthus in order to maximise biomass production. This could be possible on up to 10% of the system. If miscanthus were to be grown on this area of 12,700 ha it could produce a net increase of 114,300 odt per annum (47,229 toe) over and above the total of 152,047 odt referred to above.
45 Due to high moisture content of grass, and difficulties in ensuring suitable drying and storage, it is anticipated that the bulk of biomass from existing roadside verges would be best suited to utilisation in local anaerobic digestion (AD) plants. If more specialist dedicated energy crops such as miscanthus were to be grown these could be used for heat and/or power applications.
Figure 11. Density (t/ha) of potential annual biomass production from roadside verges (low yield estimate)
7.4 Railway embankments
Railway embankments occupy some 17,472 ha. They are similar to roadside verges in many ways although they tend to have a greater proportion of trees than grass. The same assumptions for annual growth of 3 odt/ha (low) and 5 odt/ha (medium) yield are used. The low annual yield potential is calculated at 52,418 odt (23,287 toe) and medium yield is calculated at 87,363 odt (38,811 toe). The area of railway embankment is shown in Figure 12 and the variation in density of annual yield potential (Low yield) is shown in Figure 13.
Railway embankments are actively managed at significant cost to ensure safety and efficient scheduling on the rail network. The logistical advantage of railways suggests that the biomass could, for example, be readily delivered to power stations for co-firing to offset the marginal additional costs of capturing this valuable biomass resource.
46
Figure 12. Density of land areas associated Figure 13. Density (t/ha) of potential annual with railway embankments biomass production from railway embankments (low yield estimate)
7.5 Canal margins
The canal system is relatively small compared to roads and railways at a total of 1,123 ha but again considerable costs are expended to maintain the towpaths and bank-sides. Canal side vegetation is similar to roadsides and annual yields of 3 odt/ha (low) and 5 odt/ha (medium) have been used to estimate annual yields of 3,368 odt (1,162 toe) and 5,613 odt (2,937 toe) respectively.
Again, the costs of capturing and transporting biomass would only be marginally greater than the current costs of managing the vegetation. In the case of canals, towpath mowing and tree cutting are done at different times and it is envisaged that grass could be used in local AD plants and wood could be sold to waterways users.
The area of land available on the canal system is shown in Figure 6 and the density of estimated annual biomass production at low yield is shown in Figure 7.
47
Figure 14. Density of areas associated with canal Figure 15. Density (t/ha) of biomass production side verge from canal side verges (low yield estimate)
7.6 Golf courses
Golf courses make up a very significant area nationally, in rural and semi-rural locations and in proportional densities to population centres. At 130,200 ha they occupy a similar area to the national potato crop. Grass production on golf courses is of critical importance and the subject of considerable nurturing by green keepers, who make effective use of herbicides and fertilisers. However, grass yield is not a key objective and the annual low and medium yield level assumptions made in each of the cases above remain valid at 3 or 5 odt/ha. In this case the Medium level is the more likely to be achieved.
Annual yield estimates are 390, 600 odt (96,092 toe) and 651,000 odt (160,153 toe) respectively. Grass from golf courses, being close to centres of population, producing a consistent high fresh/wet, are likely to be particularly suited to utilisation in AD plants. Assuming close proximity of suitable AD plants, this could be done at minimal additional cost beyond current management costs for the golf course. Densities of areas occupied by golf courses in shown in Figure 16 and the density of annual biomass production at the low yield estimate is shown in Figure 17.
48
Figure 16. Density of land area associated with Figure 17. Density (t/ha) of potential annual golf courses biomass production from golf courses
7.7 Sports turf
Sports turf sites are very similar to golf courses in the way that they are managed, though rather less intensively in terms of herbicide and fertiliser use. Again the same annual yield assumptions of 3 odt and 5 odt are made respectively for low and medium yield estimations.
Sports turf sites are similar to golf courses, having national coverage but following centres of population. They occupy a total area of 40,988 ha. This equates to a potential low annual yield estimation of 122,964 odt (30,250 toe) and a medium yield estimation of 204,940 odt (50,417 toe). Again, this would be highly suited to utilisation in AD plants at minimal added cost. Densities of areas of sports turf are shown by Local Authority in Figure 18 and potential annual biomass yields are shown in Figure 19.
49
Figure 18. Densities of sports turf areas Figure 19. Density (t/ha) of biomass production from sports turf areas (low yield estimate)
7.8 Brownfield sites
The combined total area derived from the Brownfield datasets (Section 3.4.8) totals 451,984 ha. The yield potential of these sites will be extremely variable. Again, the annual low and medium yield assumptions of 3 odt and 5 odt/ha are considered reliable estimates.
For the purpose of this study, it was assumed that all of the identified land has the potential to be utilised for biomass production. However, it is likely that large proportions might not be capable of high yields. If the whole were cropped at the low annual yield level this would equate to 1,355,955 odt (582,956 toe) and at the medium yield level, 2,259,921 odt (917,614 toe). Maps of these two potential yield levels are shown in Figure 20 and Figure 21.
It is not clear what proportion of Brownfield land is capable of currently being brought into production as a number of different land uses fall into this category, including derelict land and buildings, mines, spoil heaps quarries and closed landfill sites. The factors that could limiting use of sites are not clearly identified (i.e. presence of specific contaminants) much further detailed work is required to clearly define what proportion of the Brownfield resource could be exploited, but it is clearly a potentially significant land resource.
Whilst it is unlikely that large amounts of Brownfield land might be brought into production, it is conceivable that a significant area of Brownfield land might be returned to agriculture. If for example only 10% of the current area (45,198ha) were suited to miscanthus production at
50 a relatively low annual yield of 8 odt/ha, a total of 361,584odt (149,408 toe) could be produced.
Figure 20. Density (t/ha) of potential annual Figure 21. Density (t/ha) of potential annual biomass production on Brownfield sites (low biomass production on Brownfield sites yield estimate) (medium yield estimate)
7.9 Contribution of ‘idle’ land resources to primary energy demands
The potential biomass production for each non-agricultural and `idle` land use, and the potential contribution that this could make towards UK primary energy demands is shown in Table 26 below. The total area of non-agricultural land that could be used to produce biomass (from the existing vegetation resource) is calculated to be 867,667 ha. Assuming the lowest yield potentials described above, this land could produce 2,7 million odt annually, with an energy content of 1,06 million toe. This is capable of supplying up to 1.3% of primary energy demand for electricity, or up to 1.6% of UK primary energy demand for heat.
The largest contributions to this come from Brownfield land, followed by roadside verges, lowland bracken and golf courses and hedgerows in descending order of importance. Sports turf, railway embankments and canal margins make very little contribution to the overall total from ‘idle’ land. None of these resources conflict with each other, and with the exception of bracken harvesting, none are likely to be constrained by environmental or landscape concerns, as most represent use of the existing vegetation as a bioenergy resource.
51 Table 26. Potential contribution of biomass from ‘idle’ land to UK primary energy demands England Wales Total Hedgerows Area (ha) 60,285 8,073 68,358 Potential biomass production (tonnes) 180,855 24,219 205,074 Energy content (toe) (18.6 GJ/t) 80,346 10,759 91,105 Potential contribution to energy use in UK electricity production 0.10% 0.01% 0.11% Potential contribution to energy use in UK heat production 0.12% 0.01% 0.13% Lowland bracken Area (ha) 19,400 10,986 30,836 Potential biomass production (tonnes) 138,947 76,908 215,855 Energy content (toe) (21 GJ/t) 69,692 38,575 108,267 Potential contribution to energy use in UK electricity production 0.08% 0.05% 0.13% Potential contribution to energy use in UK heat production 0.10% 0.06% 0.16% Roadside verges Area of roadside (ha) 112,864 138412 126,706 Potential biomass production on roadside (tonnes) 338,593 41,525 380,118 Energy content (toe) (14.5 GJ/t) 116,860 14,331 131,191 Potential contribution to energy use in UK electricity production 0.14% 0.02% 0.16% Potential contribution to energy use in UK heat production 0.17% 0.02% 0.19% Railway embankments Area(ha) 16,003 1,469 17,472 Potential biomass production (tonnes) 48,010 4,408 52,418 Energy content (toe) (18.6 GJ/t) 21,328 1,958 23,286 Potential contribution to energy use in UK electricity production 0.02% 0.002% 0.03% Potential contribution to energy use in UK heat production 0.02% 0.002% 0.03% Canal margins Area(ha) 1,062 61 1,123 Potential biomass production (tonnes) 3,186 182 3,368 Energy content (toe) (14.5 GJ/t) 1,099 63 1,162 Potential contribution to energy use in UK electricity production 0.001% Negligible 0.001% Potential contribution to energy use in UK heat production 0.001% Negligible 0.001% Golf courses Area (ha) 118,950 11,250 130,200 Potential biomass production(tonnes) 356,850 33,750 390,600 Energy content (toe) (10.3 GJ/t) 87,789 8,303 96,092 Potential contribution to energy use in UK electricity production 0.11% 0.01% 0.12% Potential contribution to energy use in UK heat production 0.13% 0.01% 0.14% Sports turf Area (ha) 39,502 1,486 40,988 Potential biomass production(tonnes) 118,506 4,458 122,964 Energy content (toe) (10.3 GJ/t) 29,153 1,097 30,250 Potential contribution to energy use in UK electricity production 0.04% 0.001% 0.04% Potential contribution to energy use in UK heat production 0.04% 0.001% 0.04% Brownfie ld land Area (ha) 394,546 57,438 451,984 Potential biomass production (tonnes) 1,183,641 172,314 1,355,955 Energy content (toe) (18 GJ/t) 508,874 74,082 582,956 Potential contribution to energy use in UK electricity production 0.62% 0.09% 0.71% Potential contribution to energy use in UK heat production 0.74% 0.11% 0.85% Totals Area (ha) 638,492 229,175 867,667 Potential biomass production (tonnes) 2,368,588 357,764 2,726,352 Energy content (toe) 915,141 149,168 1,064,309 Potential contribution to energy use in UK electricity production 1.11% 0.18% 1.29% Potential contribution to energy use in UK heat production 1.33% 0.22% 1.55%
52 8 Impacts on soil carbon flux
The efficacy of biofuels and bioenergy crops as sources of renewable, carbon-neutral energy, relies on the GHG emissions associated with their production not exceeding the GHG savings that result from their use. The Renewable Fuels Association, as part of its efforts to develop sustainability criteria for biofuels, stipulates that any land use change associated with biofuel production should not have a carbon payback time of longer than 10 years, i.e. additional emissions resulting from the production of the fuel must be offset by the carbon savings associated with the use of the fuel within 10 years (RFA, 2008). The Gallagher Review investigated the indirect environmental impacts of transport biofuels and concluded that they “can only contribute GHG savings from transport if significant emissions from land use change are avoided” (RFA, 2008). However, the data presented in the review mostly refer to imported transport fuel and cover only a small range of possible land use change scenarios. Here, the effects of land use change associated with biofuel and bioenergy crop growth in the UK are investigated. A range of crops and land use changes will be discussed in the context of life cycle assessment (LCA) in order to evaluate the overall carbon dioxide emissions of these fuels.
8.1 Land use change and carbon sequestration
The terrestrial biosphere contains various major pools of carbon, the sizes of which are strongly dependent on vegetation type and land management. Biomass is an important store of carbon and is most obviously affected by land use change. Forests and woodland have much larger amounts of both above- and below-ground biomass than grassland or cropland and so deforestation leads to an immediate loss of carbon. Decomposition of any remaining biomass results in a net input of carbon dioxide to the atmosphere. Dead organic matter, consisting of wood and litter, can also be a major pool of carbon in forests and woodland areas. Lack of oxygen can prevent decomposition in waterlogged areas, leading to a build up of belowground dead organic matter. Draining of land for agricultural purposes can lead to increased decomposition and, therefore, increased net carbon dioxide emissions. This is in contrast to the impact of waterlogging on methane emissions. In anaerobic conditions, methanogenic organisms cause soils to act as a net source of methane; in drier, aerobic conditions, methanotrophs cause soils to act as a net sink of methane. However, a recent review conducted by Natural England (2010) concluded that, in general, the impact of soil water content on methane production is outweighed by the impact on carbon dioxide production in UK peat systems. Actual net greenhouse gas emissions or sequestration will depend on the production and consumption of both these gases and will likely vary according to local conditions.
The single largest terrestrial pool of organic carbon is soil (Schlesinger, 1997). Carbon enters the soil through deposition of dead organic matter. It leaves the soil through erosion and through decomposition and respiration of organic matter. Gain and loss of carbon is influenced by a number of biotic and abiotic factors. Vegetation type determines the structure and depth of the root system and also influences inputs of organic carbon through changes in above-ground litter fall and below-ground root deposition. Soil disturbance improves oxygen supply, kills roots, and makes dead organic matter more available to soil organisms, leading to increased respiration of soil organic carbon (SOC). Importantly, the amount of SOC in the soil influences the rate of soil respiration and so there is feedback between these processes, leading to a long-term equilibrium carbon content. Agricultural practices affect all these factors and so changes in land use associated with biofuel or bioenergy crop cultivation should be expected to cause changes in the carbon equilibrium, and thus the carbon dioxide emissions, of soils.
53 8.2 Soil carbon
Soil is a three-dimensional medium, but it is often discussed in terms of area in the context of agricultural and environmental research. Unless stated otherwise, the soil carbon data given in this report refer to areas of soil to a depth of 30 cm. The SOC contents of arable land, grassland, and woodland are of the order of 80, 100, and 130 t C ha −1 respectively, although it is important to note that there is considerable regional variation (Bradley et al ., 2005). Here, use of annual and perennial crops for energy production are considered. Wheat (for bioethanol) and oilseed rape (for biodiesel) are already an important part of conventional arable land crop rotations. Short rotation coppice willow and miscanthus, by contrast, are perennial crops, harvested regularly over the course of 15–25 years, which makes them quite different from existing food crops. The following outlines the potential consequences for soil carbon content of converting existing land uses to these crops.
8.2.1 Annual arable bioenergy crops
Growing arable crops for bioenergy does not currently involve any changes to existing growing practices. It may be assumed, therefore, that the conversion of land to cultivation of annual crops for biofuels results in the same soil carbon changes that would result from the establishment of any arable crop. Soil carbon in arable systems is generally low (Bradley et al ., 2005; Smith et al ., 2000a) and conversion of arable to grassland, woodland, or bioenergy crop plantation is usually thought to increase the SOC content of soil. Therefore, although conversion from another arable crop would be expected to have little impact on SOC, conversion from any other land use is likely to result in a net loss of carbon from the soil. Using IPCC methods, the rate of change in soil carbon following conversion from grass or woodland has been calculated as −0.634 t C ha−1 y−1 (equivalent to emissions of −1 −1 2.325 t CO 2 ha y ) (St Clair et al ., 2008). This is supported by a meta-analysis of international data (Guo and Gifford, 2002). This paper gives 59 and 42% losses of soil carbon following conversion of grassland and woodland to conventional arable land respectively.
An important factor contributing to the low SOC content of arable soils is the disturbance that occurs during ploughing. It has been estimated that switching an existing arable system from conventional to reduced tillage could lead to an increase in soil carbon of 0.040 t C ha −1 y−1 (King et al ., 2004). Using the methods of Lal (2004), St. Clair et al . (2008) calculated an increase of 0.115 t C ha −1 y−1 following adoption of reduced tillage. Adopting zero tillage practices could increase soil carbon by between 0.145 t C ha −1 y−1 and 0.235 t C ha −1 y−1 (King et al ., 2004). Reduced tillage can lead to increased emissions of nitrous oxide in some circumstances (Snyder et al . 2009), although controlled traffic farming, where satellite guidance is used to ensure that machinery only follows certain paths, has been shown to reduce nitrous oxide emissions by between 20 and 50% in reduced tillage systems (Vermeulen and Mosquera 2009).On balance, although these practices should improve the soil carbon content of existing arable soils, they do not fully offset the predicted carbon losses associated with a change from grassland or woodland.
8.2.2 Short rotation coppice willow
Willow SRC is expected to remain viable for approximately 25 years, assuming that it is harvested every three years. Due to this long lifespan of individual plantations and the relatively recent advent of this form of cultivation, there are no long-term data available for soil carbon content under SRC systems. In the absence of direct evidence, it might be assumed that SRC interacts with soil carbon in the same way as native deciduous woodland, giving an SOC content of around 130 t C ha −1 at equilibrium. Using this assumption, various
54 authors have estimated the rate at which carbon accumulates in the soil following conversion of land from arable cultivation. A database of results from long-term experiments gave a −1 −1 −1 −1 range of 0.552–0.828 t C ha y (net savings of 2.024–3.036 t CO 2 ha y ) based on this assumption (King et al ., 2004). Another review of existing data gave a figure equivalent to −1 −1 −1 −1 0.983 t C ha y (saving 3.604 t CO 2 ha y ) (Smith et al ., 2000b). A further study assumed perennial crops such as SRC to be equivalent to reduced tillage systems, with −1 −1 −1 −1 SOC accumulating at 0.115 t C ha y (saving 0.422 t CO 2 ha y ) (St Clair et al ., 2008). A more detailed model, based on data from the long-term regeneration of Geescroft Wilderness and adapted to the biological characteristics of SRC willow, predicts a carbon −1 −1 −1 −1 accumulation rate of 0.41 t C ha y (saving 1.503 t CO 2 ha y ) to a depth of 23 cm following conversion from arable land (Grogan and Matthews, 2002). A meta-analysis also provides strong evidence for the direction of this change in soil carbon following conversion from an arable crop, although the lack of European data makes it difficult to quantify the effect from this source (Guo and Gifford, 2002).
Evidence concerning the conversion of grassland or woodland to SRC willow plantations is less readily available. The meta-analysis mentioned above gives 10 and 13% declines in soil carbon content following conversion from grassland and woodland respectively (Guo and Gifford, 2002). However, the studies cited for these changes were primarily paired site analyses conducted in Australia and New Zealand and so may be a poor match to UK soils. Modelling studies that have focused explicitly on soil carbon in the UK have assumed no change in SOC following the conversion of grassland (King et al. 2004) or woodland (Smith et al. 2000b) to SRC willow, based on the results of long term studies of analogous soil systems. Although direct evidence is lacking, data from similar environments do not give reason to expect major additional losses of carbon from soils following conversion to SRC willow. Conversion of arable land is likely to result in a net accumulation of soil carbon over time. The Defra research project “Planting Biomass Crops: Assessment of Options to Reduce Soil Carbon Loss (NF0441)”, which had yet to publish results at time of writing, is expected to provide direct data on the impact of SRC willow on soil carbon.
8.2.3 Miscanthus
Due to the similar methods of cultivation, the effect of miscanthus on SOC may be assumed to be the same as that of SRC willow (St Clair et al ., 2008). Evidence from one model suggests that carbon accumulation following conversion of arable might be lower than with SRC, though still positive between 0.490 t C ha −1 y−1 and 0.734 t C ha −1 y−1 (saving between −1 −1 −1 −1 1.797 t CO 2 ha y and 2.691 t CO 2 ha y ) (King et al ., 2004). Another model, also calibrated against the Geescroft Wilderness data, indicates that carbon accumulation −1 −1 −1 −1 following arable conversion could be as high as 0.93 t C ha y (saving 3.41 t CO 2 ha y ) to a depth of 23 cm (Matthews and Grogan, 2001). Experimental data from a site in Denmark suggests that miscanthus has little impact on soil carbon during its first nine years of growth, but that it causes an increase in soil carbon content after 16 years (Hansen et al ., 2004). However, this experiment did not take measurements of soil carbon prior to miscanthus cultivation and so its data must be treated with some caution.
Some experimental data are also available for the conversion of grassland to miscanthus plantations. These data, collected from a site in southern Ireland, showed soil carbon content to be 64.0 t C ha −1 under 15-year-old miscanthus, compared to 59.7 t C ha −1 under adjacent grassland, although there was considerable variability within each dataset and no significant differences were found (Clifton-Brown et al ., 2007). Experimental evidence regarding the conversion of woodland to miscanthus in northern Europe is lacking. The similarity to SRC cultivation has led some to make the assumption that miscanthus and deciduous woodland have similar SOC contents (St Clair et al ., 2008).
55 8.3 Net costs and benefits
8.3.1 Emissions from land use change
The relative importance of biomass, dead organic matter, or soil carbon to the overall carbon mitigation effect of land use change depends on the land use prior to that change. For example, woodland would be expected to have a greater store of carbon in biomass than grassland. In addition, the full effects of land use change on some carbon pools may take many years to complete. It is often assumed, for example, that between 50 and 100 years are required for soil carbon to reach a new equilibrium following land use change (Falloon, et al . 2004; King et al ., 2005). It is important that calculations of the carbon mitigation potential of biofuels and bioenergy crops take these long-term effects into account. As mentioned above, the RTFO requires that any greenhouse gas emissions resulting from the production of liquid biofuels from biomass must be offset by the greenhouse gas savings associated with the use of the fuel within 10 years (RFA, 2008). A further consideration is that the effects of land use change on soil carbon are reversible. If the cultivation of a bioenergy crop leads to an increase in below-ground carbon storage, then a reversion to the previous land use and management regime will almost certainly result in carbon dioxide being released back into the atmosphere.
The IPCC publish guidelines for calculating the net GHG emissions associated with land use change (IPCC, 2006). The guidelines are divided into three tiers, which differ in the amount of data that must be supplied. Tier 1 does not require any data beyond the IPCC default values, which are derived from an extensive literature review. The Tier 2 approach uses some data specific to the nation or region being investigated. Tier 3 requires further information on the changes in carbon stocks through time. The default values supplied for the Tier 1 approach assume that it takes 20 years for soil carbon stocks to reach equilibrium following land use change. Equilibrium values and stock change factors are selected for a region based on clearly defined climate zones, ecological zones, soil types, management regimes, and land uses. Land use is divided into broad categories, such as forest, grassland, and cropland. These are then further divided into more specific subcategories, for example cropland is divided into annual and perennial crops. The default values take into account changes in biomass, dead organic matter, and soil carbon. They also take into account emissions of carbon dioxide, nitrous oxide, and methane and so are expressed as carbon dioxide equivalent (CO 2e), which takes into account the individual warming effects of the gases and their atmospheric lifetime.
The Tier 1 approach has been used to calculate the default GHG emissions associated with various land use change scenarios in the UK given in the RFA technical guidance documents (E4tech, 2007; RFA, 2008). The climate zone for the UK was assumed to be cool, temperate, and moist (the mean of the values for wet and dry) and the ecological zone was assumed to be temperate oceanic forest. The results of this analysis are shown in Table 27. It is important to note that the Tier 1 calculations used to obtain these data assume little or no difference in soil carbon between annual and perennial crops, which may not reflect the European data presented above (see Soil carbon). The methods are also not well suited to estimating the GHG emissions resulting from land use change within a major land category, such as conversion from annual to perennial crops, and so these potential changes were not examined. Therefore, this report will use the soil carbon changes reported by Matthews and Grogan (2001) for changes from arable cultivation to SRC and miscanthus (see above). These figures are derived from a model that takes into account the biological characteristics of the individual crops and so may be considered more accurate than the assumption that perennial crops have the same effect on soil carbon as woodland. It will also be assumed that the change from annual to perennial crops has no effect on biomass or dead organic matter as these carbon stocks would be harvested in either form of cultivation.
56 Table 27: Increase or decrease in GHG emissions caused by changes in biomass, dead organic matter (DOM), and soil carbon (to a depth of 30 cm) following land use change in the UK. Data were calculated by E4tech (P. Watson, pers. comm.) using the IPCC methods (IPCC, 2006) and published in the RFA guidance documents (E4tech, 2007; RFA, 2008). Data for arable to perennial conversions draw on the work by Matthews and Grogan (2001). Changes in carbon stocks are assumed to take place at a constant rate and to take 20 years to reach equilibrium. Impact of land use change (t CO eq ha −1 y−1 ) Previous land use New land use 2 Biomass DOM Soil C Total Forest Annual crop 16.5 3.9 6.3 26.7 Forest Perennial crop 16.1 3.9 0.0 20.0 Grassland Annual crop 0.7 0.0 6.3 7.0 Grassland Perennial crop 0.4 0.0 6.3 6.7 Arable SRC 0 0 -1.5 -1.5 Arable Miscanthus 0 0 -3.4 -3.4
8.3.2 Energy and fuel applications
Life cycle assessment (LCA) is a standardised means of calculating the overall environmental impact of a product or process. In the case of biofuel and bioenergy crops, LCA covers the period from the growth of the crop to the delivery of the fuel. This should include the GHG emissions associated with fertiliser and pesticide use, harvesting, transport of raw harvested material, processing and refining of fuel, and distribution of the end product. Different assumptions can lead to contrasting results being obtained for the same product. This report examined LCA data for the following energy application case studies.
Diesel from oilseed rape Rapeseed oil is refined to produce diesel. Current commercial practice for any biodiesel produced from UK feedstock. Data were obtained from the RFA (2008) and Woods and Bauen (2003).
Diesel from SRC (Fischer–Tropsch) SRC willow is converted to “syngas” (gasification), which is converted to liquid hydrocarbons and refined to diesel (2 nd generation). Future potential 2 nd generation technology, currently only at demonstration scale in Europe. Data were obtained from Woods and Bauen (2003).
Electricity from miscanthus combustion Miscanthus is burned to produce energy for electricity generation. Current commercial practice. Data were obtained from Elsayed et al . (2003).
Electricity from SRC combustion SRC willow is chipped and burned to produce energy for electricity generation. Current commercial practice. Data were obtained from Elsayed et al . (2003).
Electricity from SRC gasification SRC willow is converted to gas, which is burned to produce energy for electricity generation. Potential future practice, no commercial scale production in UK currently. Data were obtained from Elsayed et al . (2003).
57 Electricity from SRC pyrolysis SRC willow undergoes pyrolysis and the resulting gases are burned to produce energy for electricity generation. Pilot scale only in Europe currently. Data were obtained from Elsayed et al . (2003).
Electricity from wheat straw combustion Wheat straw is burned directly to produce energy for electricity generation. Current commercial practice. Data were obtained from Elsayed et al . (2003).
Ethanol from SRC (lignocellulosic) SRC willow lignin and cellulose are digested with enzymes to produce monomer sugars, which are fermented and refined to produce ethanol as a substitute for petrol. Potential future practice, currently at early demonstration scale. Data were obtained from Woods and Bauen (2003).
Ethanol from wheat (lignocellulosic) SRC willow lignin and cellulose are digested with enzymes to produce monomer sugars, which are fermented and refined to produce ethanol as a substitute for petrol. State of commercialisation as above. Data were obtained from Elsayed et al . (2003) and Woods and Bauen (2003).
Ethanol from wheat (grain) Wheat grain is fermented and refined to produce ethanol as a substitute for petrol. Current commercial practice. Data were obtained from Elsayed et al . (2003).
The LCA data for these fuels and energy crops were compared with fossil fuel reference systems to determine the carbon saving of substituting a fossil energy source with a bioenergy source, utilising existing and potential near-future energy conversion technologies. Biodiesel technologies were compared with fossil diesel (Woods and Bauen, 2003); bioethanol technologies were compared with petrol (Woods and Bauen, 2003); and electricity generated from bioenergy crops was compared with grid electricity in 1996 (Elsayed et al ., 2003). The results of these comparisons are shown in Figure 22. It can be seen that most biofuels and bioenergy crops provide a carbon saving relative to the fossil fuel reference systems, although in some cases, particularly biodiesel from oilseed rape, the savings are relatively small. Therefore, when land use is not taken into account, biofuels and bioenergy crops appear to provide clear GHG benefits.
8.3.3 Energy and land use change
When the GHG emissions resulting from land use change shown in Table 27 are combined with the LCA data shown in Figure 22, it becomes apparent that the benefits of biofuels and bioenergy crops strongly depend on previous land use. Arable land is often low in SOC. Therefore; conversion of existing arable land to energy crop cultivation does not lead to increased GHG emissions (Figure 23). Indeed, conversion from annual crops to perennial crops such as miscanthus and SRC willow leads to increased carbon sequestration in soils and so the carbon savings increase when this land use change occurs.
Conversion from other types of land, however, leads to large increases in GHG emissions that are not offset by the carbon savings associated with fossil fuel substitution. In the case of grassland, clear carbon savings are only shown for miscanthus combustion and SRC willow gasification for electricity (Figure 24). Other processes only provide carbon savings under certain circumstances.
58 • Fischer–Tropsch biodiesel produced using SRC willow may or may not provide GHG benefits when land is converted from grassland. The smallest carbon savings were achieved when low crop yield and low conversion efficiencies were assumed (Woods and Bauen, 2003). Commercialisation of this technology is still in its early stages and so higher efficiencies may be more likely in future. • The GHG emissions from electricity generation using pyrolysis of SRC are influenced by the harvesting method. Combined harvesting and chipping results in higher emissions than harvesting and baling followed by chipping at point of use (Elsayed et al ., 2003). In the case of conversion from grassland, GHG benefits are marginal and so the choice of method could be very important. • The net GHG savings of lignocellulosic digestion to produce ethanol will also be influenced by SRC harvesting method. In addition, uncertainties surrounding this emerging technology lead to a wide range of estimates (Woods et al ., 2003).
Carbon savings without land use change
15 ) −1
y 10 −1
eq ha 5 2
0
-5 Carbon savingCarbon (t CO -10
Figure 22: Ranges of carbon savings from biofuel and bioenergy crop fuel chains without accounting for land use change (based on LCA data in Elsayed et al ., 2003; RFA, 2008; Woods and Bauen, 2003).
The overall impact of the perennial crops SRC willow and miscanthus shown in these graphs may appear to contrast with the model and experimental evidence mentioned above (see section 8.2). Some studies have assumed no change in soil carbon following conversion of grassland (King et al ., 2004) or woodland (Smith et al ., 2000b) to SRC. At least one experiment has shown no significant difference in SOC between grassland and a miscanthus crop (Clifton-Brown et al ., 2007). However, these are isolated studies and primarily relate to soil carbon and so do not take into account biomass or dead organic matter returns, or emissions of nitrous oxide. For these reasons, and due to its wider use in other studies, the IPCC methods have been preferred over the models, experimental evidence, or assumptions from the literature discussed above for calculation of carbon payback periods. However, even use of the IPCC approach is not without problems. In the latest 2006 IPCC guidelines
59 for National Greenhouse Gas Inventories, the example perennial crops listed represent a wide range of different default croplands typified by woody crops – which include orchards and plantation crops and agro-forestry, but there is no specific mention or inclusion of grass or woody bioenergy crops. This reflects on the need for more reliable data on the impacts of using grassland for perennial energy crops before firm conclusions can be drawn on the impacts of such cropping changes on overall carbon balances.
Land converted from arable
15 ) −1
y 10 −1
eq ha 5 2
0
-5 Carbon savingCarbon (t CO -10
Figure 23: Ranges of carbon savings from biofuel and bioenergy crop chains when converting from arable land (biomass crops only) (based on LCA data in Elsayed et al ., 2003; RFA, 2008 and Woods and Bauen, 2003).
60 Land converted from grassland
15 ) −1
y 10 −1
eq ha 5 2
0
-5 Carbon savingCarbon (t CO -10
Figure 24: Ranges of carbon savings from biofuel and bioenergy crop chains when converting from grassland (based on LCA data in Elsayed et al ., 2003; RFA, 2008 and Woods and Bauen, 2003).
8.4 Carbon payback time
Where land use change resulted in additional GHG emissions, carbon payback time was calculated. The methods published in the RFA technical guidance documents were employed (RFA, 2008), which define the carbon payback time of a fuel as the total carbon loss as a result of land use change divided by the annual carbon saving resulting from the use of fuel. The carbon payback times are shown in Table 28. It can be seen that biofuel and bioenergy crops grown on arable land do not result in GHG emissions that exceed the GHG savings associated with their use.
Carbon payback times for crops grown on former grassland or forest land show wide variation. None of the fuel or energy options examined in this study met the RFA requirement for a carbon payback time of less than 10 years. However, several perennial crop technologies did show payback times less than 15 years, such as combustion of miscanthus for electricity generation, or less than 25 years, such as SRC willow gasification and pyrolysis. Miscanthus and SRC willow are typically grown for 15 and 25 years respectively. Consequently, these technologies could be seen as sustainable over the lifetime of a single crop. Furthermore, when it is considered that soil carbon can take many decades to reach equilibrium, other fuel options with relatively short payback times could also be seen as sustainable in the long term even if they do not meet the sustainability criteria for liquid fuel production set out by the RFA.
Table 28: Maximum and minimum carbon payback times of the biofuel and bioenergy technologies examined in this report. A value of zero shows that GHG emissions from land use change do not exceed the GHG savings associated with using the fuel.
61 Carbon payback time by previous land use class (years) Technology Arable Grassland Forest Max. Min. Max. Min. Max. Min. Diesel (oilseed rape) 0 0 174.16 46.96 664.31 179.11 Diesel (SRC Fischer–Tropsch) 0 0 63.16 14.42 188.52 43.06 Electricity (miscanthus burning) 0 0 14.89 14.67 44.44 43.80 Electricity (SRC burning) 0 0 39.56 31.49 118.08 94.01 Electricity (SRC gasification) 0 0 17.34 14.24 51.77 42.52 Electricity (SRC pyrolysis) 0 0 24.15 19.55 72.09 58.34 Electricity (wheat straw burning) 0 0 123.26 113.40 470.15 432.54 Ethanol (SRC, lignocellulosic) 0 0 82.86 13.13 247.33 39.20 Ethanol (wheat, lignocellulosic) 0 0 430.61 50.45 1642.47 192.44 Ethanol (wheat, grain) 0 0 392.75 34.31 1498.06 130.86
Conversion of grassland to perennial energy cropping results in significant payback periods using IPCC default values and approaches. However, limited data relevant to UK conditions questions whether there is any significant long-term impact on soil carbon associated with conversion of grassland to perennial energy cropping. Clearly this is an area of contention where more reliable UK data sets are required if conversion of long-term grassland is contemplated, and at present the net impact on soil carbon is unclear. The ongoing Defra research project “Planting Biomass Crops: Assessment of Options to Reduce Soil Carbon Loss” (NF0441), which had yet to publish its results at time of writing, is intended to provide data on the impact of energy crops on soil carbon.
8.5 Conclusions – soil carbon turnover
Biofuels and bioenergy crops represent a renewable and sustainable alternative to fossil energy resources. However, they will only provide a real environmental benefit if their production does not lead to greater GHG emissions than would be saved by their use. LCA calculates the environmental impact of these crops from cultivation to delivery and shows clear GHG savings during this phase. This report has shown that land use change associated with stimulation of biofuel and bioenergy crop cultivation may offset some or all of these savings. Growth of annual crops on land previously under arable cultivation results in little or no additional GHG emissions. Growth of perennial crops on this type of land may even lead to net sequestration of carbon in soils, leading to further GHG benefits. However, GHG savings are much smaller on converted grassland and, in many cases, the overall GHG emissions can be greater than those of the fossil energy reference system when soil carbon losses are also accounted for, though more evidence is required in this area. Therefore, conversion from grassland to bioenergy production should only take place when the crop and the production system are assessed thoroughly for their overall GHG impact.
9 Potential combinations of biomass supply from marginal and ‘idle’ land resources to meet UK renewable energy demands
9.1 Potential supply of liquid biofuels
The un-cropped arable land resource (set-aside and bare fallow) and potential land derived from conversion of temporary grassland to arable production, were identified as key resources capable of supplying wheat and oilseed rape to meet the needs of liquid biofuel producers, but without conflicting with food demands.
62 Assumptions;
• Set-aside and bare-fallow land is utilised predominantly for biofuel production, using land that would otherwise not be in food production (in fact some food production would arise as a consequence (from non-wheat and oilseed phases of the arable rotation)).
• Temporary grassland resource is released without affecting food production (due to increase in stocking rates on the remaining temporary grass resource).
• As the temporary grassland resource is less than 5 years old, there are no significant impacts on soil carbon turnover rates from conversion to arable production.
However, there would be some environmental consequences arising including
• Loss of bare fallow and rotational set-aside habitat that, for example, provides a valuable resource for farmland birds and arable biodiversity.
• Some increase in inputs may be required to the remaining temporary grassland supporting livestock, to increase livestock carrying capacity.
The total biofuel and biomass output of this combined resource is shown in Table 29. Up to 1.8% of UK fuel sales (by volume) could be met from biofuels derived from fallow arable land and conversion of under-utilised temporary grassland to biofuel crop production. The majority of this resource would be derived from former temporary grassland. There may be some landscape concerns over use of some of this grassland resource (National Parks and AONBs), which could cut the overall contribution to 0.7%. In addition, a small contribution to heat and electricity power demand (0.27% or 0.22-0.23% respectively) could be made from use of straw from the wheat crops grown for biofuel production.
The above assumes that only 1 st generation biofuels, derived from arable crops, would be produced on the available ‘idle’ and marginal land resource. As the technology for lingocellulosic and thermo-chemical routes of biomass conversion to liquid biofuels matures, then the range of biomass resources that can be exploited for biofuel production and associated ‘idle’ and marginal land resources that could be used, would increase. This would increase the competition between biofuel and biomass market outlets for the limited biomass resource available (only a proportion of the RES targets for renewable energy generation in the UK can be met by utilisation of all of the resource identified in this study (see Conclusions)). While there are well-developed alternative renewable power options for electricity generation (wind, wave, hydro, solar), the same cannot be said for alternatives to transport fuels, though thermodynamically, conversion to transport biofuels may not represent the most energy efficient use of biomass.
Table 29. Potential contribution from ‘idle’ and marginal land resources to UK biofuels Excluding a reas of Excluding additional environmental concern areas of landscape concern Land 1.% UK 2.% UK 3.% UK 4.% UK Notes resource fuel use by primary fuel use by primary volume energy volume energy Biofuel un-cropped 0.46 0.31 as 1 as 2 No landscape feedstock arable concerns, but (wheat and biodiversity OSR) consequences
63 Under- 1.35 0.18 0.25 0.17 Inputs to utilised remaining temporary grassland grassland may increase ? Total 1.81 0.49 0.70 0.48 Biomass energy 5. Energy 6. % UK 7. Energy 8. % UK produced as a by- in biomass Primary in biomass Primary product energy energy un-cropped 119,200 toe 0.14 elec as 5 as 6 Assume no Biomass arable 0.17 heat landscape feedstock restrictions (wheat Under- 70,700 toe 0.09 elec 66,300 toe 0.08 elec electricity OR straw) utilised 0.10 heat 0.10 heat heat only temporary grassland Total 189,900 toe 0.23 elec 185,500 toe 0.22 elec electricity OR 0.27 heat 0.27 heat heat only
9.2 Potential supply of biomass
Assumptions
• Set-aside and bare-fallow land is utilised predominantly for biomass production, using land that would otherwise not be in food production.
• Conversion of arable land to perennial energy cropping would have some impact on food cropping, however this would be countered by impacts on grain and oilseed commodity prices. Any increase in value of these would dissuade further growers from converting to perennial energy crop production.
• Where there is choice in selection of biomass outputs from the preceding analysis,(i.e. SRC v miscanthus) results from the most productive biomass resource are used. (Results for retaining grass as the main biomass resource (to minimise environmental, soil carbon change and or landscape impacts) are included in the table, but are not included in the sub-totals and overall totals).
• Biomass for ‘marginal’ arable land scenario taken as situation where wheat @ £100/tonne (85% DM) and biomass @ £60/odt (any change to this will significantly alter the size and scale of the resource available)
• No environmental constraints or landscape impacts affecting conversion from arable to perennial biomass production, or use of ‘idle’ land resources.
• Conversion of temporary grassland to perennial energy cropping would have no detrimental impact on soil carbon levels.
• Conversion of permanent grassland to bioenergy cropping is deemed feasible and acceptable in terms of any change in soil carbon, although the RFA requirement of a carbon payback time of less than 10 years (for liquid biofuels) will not be met using any of the crops examined (based on current IPCC data and approaches).
There would be some environmental consequences arising including
64 • Loss of bare fallow and rotational set-aside habitat that, for example, provides a valuable resource for farmland birds and arable biodiversity.
• Some increase in inputs may be required to the remaining temporary and permanent grassland to support increased livestock carrying capacity.
• Soil carbon stocks will be affected by the conversion of permanent grassland to perennial energy cropping, although further research is required to determine the direction and scale of that change. Information on the impact of grassland conversion on emissions of other greenhouse gases, including methane and nitrous oxide, is also lacking.
It should also be noted that there is the possibility of some double accounting between resource derived from fallow land and arable land converting to perennial energy cropping.
The contribution of agricultural and ‘idle’ land resources to biomass energy production, and the potential contributions to primary energy demands are shown in Table 30. A total of just over 5 million toe of primary energy could be generated from full utilisation of all the ‘idle’ and marginal land resources highlighted in this study. This represents the total area of land suitable for energy crop production. The actual area available depends on other economic, social, and policy drivers outside the scope of this study. This resource is capable of meeting 6.3% of UK energy demands for electricity generation and 7.6% of UK energy demands for primary heat energy. Around 82% of this would be derived from agricultural land resources, and the bulk of this is derived from conversion of perennial grass to perennial energy cropping. Initial evidence suggests that the resulting greenhouse gas emissions would prevent these crops meeting the RFA sustainability criteria. However, direct evidence is lacking and further research is required to fully understand the impact of the land use change on soil carbon dynamics. If landscape concerns are a significant barrier to uptake of biomass cropping in National Parks and AONBs, then this resource could be reduced by around 10%, a relatively small impact.
65
Table 30. Potential total contribution from ‘idle’ and marginal land resources to UK heat and electricity energy demands Excluding areas of Also excluding environmental concern additional areas of landscape concern Option Land 1. Energy 2.% UK 3. Energy 4.% UK Notes resource in primary in biomass primary biomass energy (toe) energy (toe) Arable to Un-cropped 725,500 0.88 elec as 1 as 2 No miscanthus arable land 1.06 heat landscape concerns, but possible biodiversity impacts Arable to marginal 687,572 0.83 elec 435,385 0.53 elec see note* SRC and arable land 1.00 heat 0.63 heat no other miscanthus concerns Temp underutilised 437,600 0.53 elec 409,600 0.50 elec Inputs to grass to temporary 0.64 heat 0.60 heat remaining miscanthus grassland grass may increase ? Retain as underutilised 171,700- 0.21-0.26 160,700- 0.20-0.24 Alternative grass temporary 214,600 elec 200,900 elec to above but (no grassland 0.25-0.31 0.23-0.29 requires change) heat heat higher inputs Perm grass underutilised 2,421,000 2.93 elec 2,277,000 2.76 elec Poss. to permanent 3.54 heat 3.31 heat Increase in miscanthus grass soil carbon flux. Inputs to remaining grass increase ? Retain as underutilised 493,000- 0.60-0.90 464,000- 0.56-0.84 Alternative grass (no permanent 740,000 elec 696,000 elec to above but change) grass 0.72-1.08 0.68-1.01 requires heat heat higher inputs Sub Total All (excluding 4,271,672 5.17 elec 3,847,485 4.76 elec see note* retained as 6.24 heat 5.60 heat grass)
‘Idle’ land All 1,064,309 1.29 elec 956,042 1.16 elec 3 and 4 resources 1.55 heat 1.39 heat exclude bracken Total 5,335,981 6.33 elec 4,803,527 5.96 elec see note* 7.63 heat 7.15 heat *Potential for some double counting of the marginal arable land with the un-cropped arable land resource
Another area of concern is the loss of habitat, and its dependent biodiversity, that could occur if all existing fallow land (including voluntary set-aside) is converted to energy cropping. However, the loss of set-aside since 2007 has already raised a number of concerns with regard to loss of a valuable habitat resource in the farmland landscape. Recognising this, even in the absence of any significant drive to use this resource for bioenergy production, Defra initiated a consultation exercise to identify means of re-capturing the biodiversity benefits of losing set-aside. As a result, the Campaign for the Farmed
66 Environment was established, which aims to double the uptake of agri-environment Entry Level Stewardship applications and other measures to encourage and support farmland biodiversity within the existing arable landscape. However one of the other aims is to increase the amount of un-cropped land to 20,000 hectares more than existed in January 2008. This will be monitored to ensure targets are achieved. This could conflict with use of this resource for biofuel production. However, there may be benefits from converting un- cropped arable land to perennial energy production, particularly SRC where the grassland rides around plantations and the edge of plantations provide important habitats for a number of farmland and woodland birds of interest (Rowe et al , 2009). Therefore permitting planting of energy crops on this resource may be justified where they are managed sympathetically with environmental aims.
The above analysis shows a contribution of around 1% to UK primary energy demands by biomass derived from the conversion of unprofitable arable land to perennial energy crop production. The scenario of typical current prices for wheat and perennial energy crops suggests that between 87,000 and 128,000 hectares of arable land in England and Wales would current be more profitably engaged in energy crop production. However, currently only around 6000 ha of miscanthus and SRC can be accounted for under Energy Crop Payment or Energy Crop Planting schemes. Clearly there are other barriers to uptake of perennial energy crop production. Recent studies (Sherrington et al., 2008) have identified some of the issues affecting wider uptake of biomass cropping:
• Uncertainty over yield potential (and therefore economic return)
• Lack of confidence in long-term sustainability of bioenergy enterprises and guarantee of market outlets
• Consideration of energy crops as a ‘diversification option’, i.e. only a proportion of the farm is likely to convert to energy production in most cases (no more than 20-30%)
• Uncertainty over costs of removal at end of useful life
• Unattractive to managers of tenanted land, where long-term contracts may be untenable
• Farmers unused to working with long-term contracts
• Delayed cash-flow until first harvest returns – can take several years to recoup initial high set-up costs
• Transport/isolation may make use of some upland and other opportunities difficult
However some positive features were also noted:
• The advent of long term contracts offer stability from volatile arable crop commodity markets.
• Introduction of a Renewable Heat Obligation could offer new incentives to supply more diverse market place and develop more local supply contracts.
• Potential to reduce labour on farm
This analysis identifies the potential size of the ‘idle’ and marginal land resource that could potentially be made available for biofuel and biomass production, which represents a maximum possible contribution without significantly affecting food production, soil carbon
67 turnover or causing other significant environmental impacts. However, further work is required to understand what proportion of the resource can actually be utilised in practice, and what measures are likely to be required to encourage wider use of such land resources.
10 Conclusions
This review has identified the significant resource of up to 5 million toe of biomass energy that could potentially be generated from full utilisation of the identified ‘idle’ and marginal land sources for biomass production without,
A) Impacting significantly on food production (although any use of existing farmland will reduce the total food production capacity of the UK and may increase the environmental impact of existing production through intensification).
B) Causing significant losses of soil carbon or other greenhouse gas emissions associated with land use change (although the available evidence is incomplete and further research is required to determine the impact of conversion of all land types to perennial energy cropping), or
C) Resulting in any other significant impacts on environmentally sensitive habitats (although there is an increased risk of soil erosion and nitrogen pollution affecting water courses, greenhouse gas emissions associated with fertilizer use are likely to increase, and the impact of energy crop plantations on biodiversity is unlikely to be analogous to the impacts of previous agricultural or environmental regimes).
This represents an absolute maximum that could be produced using all suitable land; the actual land area available will be smaller due to competing demands on land and other economic, social, and policy factors. Around 82% of this is resource is derived from agricultural land resources (fallow land, economically marginal arable land converting to perennial energy cropping and by more efficient utilisation of the available grassland resources in England and Wales). The most important contribution to the total biomass resource (46%) derives from conversion of under-utilised perennial grassland to perennial energy cropping. The initial evidence suggests that conversion of grassland will lead to net loss of soil carbon and carbon payback times exceeding the RFA maximum (specified for liquid biofuels) of 10 years, although specific data are lacking. However, further research is necessary to fully understand the impact of this change in land use on greenhouse gas emissions from soils.
Utilisation of currently un-cropped or fallow arable land represents a significant land resource, capable of supplying biomass (accounting for 14% of the biomass total from ‘idle’ and marginal land). Similarly, a significant proportion of current arable land, at current market prices for crops and biomass, could more profitably produce perennial energy crops (accounting for up to 13% of the biomass total from ‘idle’ and marginal land).
A key factor affecting exploitation of the un-cropped/fallow land resource will be the impacts of the Campaign for the Farmed Environment (CFE), and within this the current Government aim to increase the amount of un-cropped land in England and Wales, which would conflict with its use for biomass production, especially for liquid biofuel production. The aim of the CFE is to recapture or preserve the biodiversity benefits associated with setting-aside land from arable production (following reduction of the compulsory set aside rate to zero after 2007). The impacts of perennial bioenergy crops on biodiversity have only been studied to a limited extent, and comparisons have been drawn with arable crops rather than fallow land.
68 There is some evidence to suggest that perennial biomass crops do have some biodiversity benefits, particularly at field margins and along grassy access rides in crops (Boatman 2009). However, they are unlikely to be beneficial to the same range of species and so the impact of energy crops may not be directly comparable to previous land use regimes (e.g. Sage et al ., 2006; 2010). Therefore, it is questionable whether this can completely compensate for loss of biodiversity benefits associated with un-cropped habitats. More detailed study is required in this area. If significant tracts of un-cropped land are converted to biomass production, in order to deliver on both biomass and biodiversity targets, there is likely to be a requirement to provide environmental enhancements either in the biomass crop itself or in other areas in the vicinity.
Of the ‘idle’ land resources, the most significant land resource is the range of land uses covered by ‘Brownfield’ designation, representing around 450 thousand hectares in England and Wales. The area identified in this study is substantially larger than that identified in previous reports, such as the report by Bardos (2009), which put the available area of brownfield land at around 62,100 ha. However, our assessment does not attempt to evaluate the potential for lower yields on brownfield land, the logistical and practical implications of cultivating brownfield land, or competing demands on the brownfield land resource (including house building). Again further work is required; including more detailed regional analysis of the land resources in the Brownfield category, to identify how much Brownfield land can be effectively brought into energy crop cultivation.
The work in this study on the marginal arable land resource, demonstrates that despite the potential economic advantages that could currently accrue on some land from converting to perennial energy crop production, the actual land area planted with energy crops is far below that where biomass production is indicated as a potentially viable option at current market prices. Economic factors are therefore not the only issues influencing the willingness of farmers on marginal arable to change to perennial energy cropping. Confidence in the crop needs to be built, effective supply chains and localised end uses developed to help reduce issues associated with isolation from end point of use and high cost of biomass transport.
This study identified that up to 150,000 hectares of arable crops could be produced as biofuel feedstocks on currently un-cropped arable land and through better utilisation of the existing temporary grassland resource. These are the most likely sustainable sources of biofuel production from the identified ‘idle’ and marginal land resources that would not conflict with food production. The overall contribution from this land resource to current UK fuel use (1.18% by volume or 0.49% by energy) is relatively small, and so meeting the proposed biofuel targets sustainably, without impinging significantly on food production represents a significant challenge in the UK.
The utilisation of the potentially available ‘idle’ and marginal land resources for biomass, if used at the potentials identified in this study, could help to deliver a significant proportion of the UK targets for renewable energy as announced in the Renewable Energy Strategy (RES), including up to 21% of the renewable electricity target (RES target of 30% of electricity energy demand), or up to 63% of the renewable heat target (RES target of 12% of heat energy demand).
Alternately, if the priority is to produce biofuel feedstocks, then up to 18% of the RES renewable transport fuel energy target (10% by energy) could be met from wheat and oilseed rape biofuel feedstock production on fallow and un-cropped arable land and use of under- utilised temporary grassland. In addition, further biomass would be generated from wheat straw production, plus other ‘idle’ and marginal land resources could be used for biomass production that could potentially account for up to 17% of the RES renewable electricity target, or up to 52% of the RES renewable heat target.
69 11 Further work
There are some specific areas that require further research and development. Conversion of grassland to perennial energy cropping results in significant carbon payback periods for biomass and biofuel technologies when using IPCC default values and approaches. Although there is some indication that conversion of grassland to perennial energy crops does not lead to loss of soil carbon stocks (Clifton-Brown et al. 2007; Hansen et al. 2004), direct evidence that takes into account dead organic matter, biomass stocks, methane production, and nitrous oxide emissions is lacking. There is a particular need for thorough research on methane and nitrous oxide emissions during agricultural land use changes. More reliable and consistent data from direct measurements and monitoring of UK systems are required to understand the impact of grassland conversion on agricultural greenhouse gas emissions and sequestration. It is also important to ensure that net agricultural emissions are combined with relevant LCA data on processing technologies in order to measure the full life cycle impact of the crop and its use.
This report has examined the maximum possible area of land suitable for biomass crops, but has not made any assessment of the practical implications of growing and harvesting crops on the areas identified. Logistical and safety concerns and competition for resources are likely to restrict significantly the amount of land available. For example, the use of road verges would be impractical in many cases because of problems in carrying out planting and harvesting on such narrow linear areas, and because of the safety issues inherent in growing tall perennial crops adjacent to roads, where they would impact upon visibility for road users. Similarly, the conversion of a significant area of golf courses to bioenergy production seems unlikely at the present time. Further work could be undertaken to provide some quantification of the amounts of land in different categories of potential availability, and the constraints operating.
Further work is also required to identify what proportion of the brownfield land resource is capable of being utilised for biomass, and what if any restrictions may be posed by certain sites, for example, how ash is categorised/treated when biomass is derived from contaminated sites. A more detailed regional analysis is required of the brownfield land resource to help Local Authorities and other stakeholders identify possible opportunities for use of biomass, particularly in association with other developments such as biomass-fuelled combined heat and power installation in council house holdings etc. The potential productivity of brownfield sites, the logistical implications of using them, and the impact of competing demands for previously developed land also require further evaluation to allow a more accurate assessment of the ability of brownfield land to produce biomass.
Use of the proposed areas of uncropped and economically marginal land for biomass production will involve changes to habitats and intensification of agriculture. Further research is required to understand the environmental implications of these changes, which may have significant consequences for biodiversity, greenhouse gas emissions, soil erosion, and pollution of water courses. These factors could influence decisions over the total extent of biomass cropping and the need to retain uncropped land for environmental reasons.Further work is required to understand the economic and market factors that currently prevent wider uptake of biomass to identify possible means and approaches to encourage greater utilisation of the land resources identified in this study help make a significant contribution towards the renewable energy targets proposed in the RES. Any study of farmer decisions regarding biomass crop adoption should also investigate the importance of non-financial barriers to uptake, including loss of flexibility, uncertainty over long-term support, and perceptions of future demand.
70 12 References
Bardos, P. (2009). Crop Based Systems for Sustainable Risk Based Land Management for Economically Marginal Degraded Land. Rejuvenate Project SN-01/20. http://www.snowman- era.net/downloads/REJUVENATE_final_report.pdf (accessed 20 May 2010).
Boatman, N. (2009) Campaign for the Farmed Environment: Assessment of Voluntary Measures and Monitoring Proposals. Final report to Defra. http://www.defra.gov.uk/corporate/consult/gaec/farmed-environment-report090612%20.pdf (accessed 6 July 2010).
Bradley RI, Milne R, Bell J, Lilly A, Jordan C, Higgins A (2005) A soil carbon and land use database for the United Kingdom. Soil Use and Management, 21, 363-369.
Clifton-Brown JC, Breuer J, Jones MB (2007) Carbon mitigation by the energy crop, Miscanthus. Global Change Biology, 13, 2296-2307
Clifton-Brown JC and Jones MB, (2001) Yield performance of M.x giganteus during a 10-year field trial in Ireland. Aspects of Applied Biology 65 , Biomass and energy crops II , 153-160.
Defra (2007) UK Biomass Strategy, 2007 pub: Department for the Environment, Food and Rural Affairs, London, UK. 48 pages.
Defra (2009) Change in the area and distribution of un-cropped land in England: January 2009 Update. Defra Agricultural and Environment Observatory Research Report No 13. 29 pages
E4tech (2007) Carbon Reporting within the Renewable Transport Fuel Obligation— Methodology
Elsayed MA, Matthews R, Mortimer ND (2003) Carbon and energy balances for a range of biofuels options
English Golf Courses (2010) English Golf Courses: Your Complete Guide to Golfing in England. http://www.englishgolf-courses.co.uk/ (accessed 5 July 2010).
Falloon P, Powlson D, Smith P (2004) Managing field margins for biodiversity and carbon sequestration: a Great Britain case study. Soil Use and Management, 20, 240-247
Frame, J. (1992). Improved Grassland Management. Farming Press Books, UK. 351 pages
Gallagher, E. (2008) The Gallagher review of the indirect effect of biofuels production. Pub: Renewable Fuels Agency, UK. 90 pages
Gezan SA and Riche AB (2008). Over-winter yield decline in Switchgrass and Miscanthus. Aspects of Applied Biology 65 , Biomass and energy crops II , 219-223.
Grogan P, Matthews R (2002) A modelling analysis of the potential for soil carbon sequestration under short rotation coppice willow bioenergy plantations. Soil Use and Management, 18, 175-183.
Guo LB, Gifford RM (2002) Soil carbon stocks and land use change: a meta analysis. Global Change Biology, 8, 345-360.
71 Hansen EM, Christensen BT, Jensen LS, Kristensen K (2004) Carbon sequestration in soil beneath long-term Miscanthus plantations as determined by C-13 abundance. Biomass & Bioenergy, 26, 97-105.
Haughton A.J. Bond A.J..Lovett A.A., Dockerty T. Sünnenberg G., Clark S.J., Bohan D.A., Sage R.B., Mallott, M.D., Mallot, V.E. Cunningham, M.D., Riche, A.B. Shield I.F., Finch J.W., Turler M.M. and Karp, A. (2009). A novel, integrated approach to assessing social, economic and environmental implications of changing rural land use: a case study of perennial biomass crops. Journal of Applied Ecology 56 , 315-322.
IPCC (2006) Guidelines for National Greenhouse Gas Inventories.
Kilpatrick J, Haywood C, Smith C, Wilson L, Proctor C, Spink J, TubbyI, Matthews R, Mackie E, Hogan G and Randle T. (2008) Addressing the land use issues for non-food crops in response to increasing fuel and energy generation opportunities. Report for NNFCC (NNFCC Project 08-004) 97 pages. (available to view via NNFCC web-site)
King JA, Bradley RI, Harrison R, Carter AD (2004) Carbon sequestration and saving potential associated with changes to the management of agricultural soils in England. Soil Use and Management, 20, 394-402.
Lal R (2004) Carbon emission from farm operations. Environment International, 30, 981- 990.
Lawson, G.J.; Callaghan, T.V. & Scott, R. (1986). Bracken as an energy resource. In: Bracken: ecology, land use and control technology, edited by R.T. Smith & JA. Taylor, 239- 248. Carnforth: Parthenon Press.
Lovett AA, Sünnenberg GM, Richter GM, Dailey AG, Riche AB, Karp A. (2009) Land use implications of increased biomass production identified by GIS-based suitability and yield mapping for miscanthus in England. Bioenrgy Research, 2, No1-2.
Matthews R, Grogan P (2001) Potential C-sequestration rates under short-rotation coppiced willow and Miscanthus biomass crops: a modelling study. Aspects of Applied Biology, 65, 303-312.
Natural England (2010) England’s Peatlands: Carbon Storage and Greenhouse Gases. http://naturalengland.etraderstores.com/NaturalEnglandShop/NE257 (accessed 5 July 2010).
RFA (2008) Carbon and Sustainability Reporting Within the Renewable Transport Fuel Obligation: Technical Guidance Part One.
Rowe, R, Street NR, and Taylor G (2009) Identifying potential environmental impacts of large-scale deployment of dedicated bioenergy crop in the UK. Renewable and Sustainable Energy Reviews 13, 271-290.
Sage, R.B., Cunningham, M., & Boatman, N.D. (2006) Birds in willow short-rotation coppice compared to other arable crops in central England and a review of bird census data from energy crops in the UK. Ibis , 148 , 184-197.
Sage, R., Cunnigham, M., Haughton, A. J., Mallott, M. D., Bohan, D. A., Richie, A., and Karp, A. (2010) The environmental impacts of biomass crops: use by birds of miscanthus in summer and winter in southwestern England. International Journal of Avian Science 152 : 487-499.
72 Schlesinger WH (1997) Biogeochemistry: An Analysis of Global Change . Academic Press, London, UK.
Sherrington C., Bartley J. and Moran D. (2008) Farm-level constraints on the domestic supply of perennial energy crops in the UK. Energy Policy 36, 2504-2512.
Smith P, Milne R, Powlson DS, Smith JU, Falloon P, Coleman K (2000a) Revised estimates of the carbon mitigation potential of UK agricultural land. Soil Use and Management, 16, 293-295.
Smith P, Powlson DS, Smith JU, Falloon P, Coleman K (2000b) Meeting Europe's climate change commitments: quantitative estimates of the potential for carbon mitigation by agriculture. Global Change Biology, 6, 525-539.
Snyder, C. S., Bruulsema, T. W., Jensen, T. L., and Fixen, P. E. (2009) Review of greenhouse gas emissions from crop production systems and fertilizer management effects. Agriculture, Ecosystems and Environment 113: 247-266.
St Clair S, Hillier J, Smith P (2008) Estimating the pre-harvest greenhouse gas costs of energy crop production. Biomass & Bioenergy, 32, 442-452.
Vermeulen, G. D. and Mosquera, J. (2009) Soil, crop, and emission responses to seasonal- controlled traffic in organic vegetable farming on loam soil. Soil & Tillage Research 102: 126- 134.
Woods J, Bauen A (2003) Technology status review and carbon abatement potential of renewable transport fuels in the UK
73 13 Annex 1 - Assumptions
UK Energy statistics
1. UK road transport fuel use (2008) (UK Digest of Energy Statistics, 2009)
Total energy use by sector = 42.8 million toe
Demand by sector Motor Spirit 16.67 million tonnes (22.7 billion litres) Diesel 26.58 million tonnes (31.8 billion litres)
2. UK heat energy consumption
Total energy consumption (all fuel sources, for heating, drying and cooking operations for domestic, services and industry) = 68.7 million toe
3. UK electricity generation energy consumption
Total energy consumption (all fuel sources, for domestic, services and industry) = 82.5 million toe
Feedstock to biofuel conversion rates
2.44 tonne of oilseed rape (at 9% MC) required to produce 1 tonne of biodiesel
3.4 tonnes of wheat (at 15% MC) required to produce 1 tonne of ethanol
Harvestable straw yields
Wheat straw yield = 0.58% of wheat yield (accounts for both harvest index and amount that can be physically removed)
OSR straw yield = 1.44% of oilseed rape yield (accounts for both harvest index and amount that can be physically removed)
Calorific values used
Miscanthus 17.3 GJ/odt Willow SRC 18.6 GJ/odt Wheat straw 15 GJ/odt Ryegrass 10.3 GJ/odt Bracken 21 GJ/odt Gasoline/motor spirt 44.7 GJ/t (Net) Diesel oil (DERV) 42.8 GJ/t (Net) Bioethanol 26.7 GJ/t (Net) Biodiesel (FAME) 37.3 GJ/t (Net)
74 Conversion factors
1 tonne of oil equivalent (toe) = 41.868 GJ
Density conversions
1 tonne ethanol = 1274 litres 1 tonne biodiesel (FAME) = 1087 litres 1 tonne ULSP = 1361 litres 1 tonne ULSD (DERV) = 1198 litres
75 14 Annex 2 - NUTS 2 codes and regions
UKC1 Tees Valley and Durham UKC2 Northumberland and Tyne and Wear UKD1 Cumbria UKD2 Cheshire UKD3 Greater Manchester UKD4 Lancashire UKD5 Merseyside UKE1 East Riding and North Lincolnshire UKE2 North Yorkshire UKE3 South Yorkshire UKE4 West Yorkshire UKF1 Derbyshire and Nottinghamshire UKF2 Leicestershire, Rutland and Northamptonshire UKF3 Lincolnshire UKG1 Herefordshire, Worcestershire and Warwickshire UKG2 Shropshire and Staffordshire UKG3 West Midlands UKH1 East Anglia UKH2 Bedfordshire and Hertfordshire UKH3 Essex UKI1 Inner London UKI2 Outer London UKJ1 Berkshire, Buckinghamshire and Oxfordshire UKJ2 Surrey, East and West Sussex UKJ3 Hampshire and Isle of Wight UKJ4 Kent UKK1 Gloucestershire, Wiltshire and North Somerset UKK2 Dorset and Somerset UKK4 Devon UKL1 West Wales and The Valleys UKL2 East Wales
76 15 Annex 3 Assumptions and estimated costs for derivation of base crop production costs
Willow SRC Revenue (other than from biomass yield) Establishment grant (£/ha) 544.96 Pre -planting costs Vegetation removal/cutting (£/ha) 10.00 Ploughing (£/ha) 40.50 Sub-soiling (£/ha) 46.60 Number of times herbicides are applied 2 Herbicide farmer cost per application (£/ha) 9.30 Herbicide cost (£/ha) 10.00 Total cost of herbicide (£) 38.60 Number of cuttings/ha 15000 Price per cutting (£) 0.05 Cost of cuttings (£/ha) 750.00 Establishment costs Planting contractor (£/ha) 300.00 Power harrowing (£/ha) 32.00 Herbicide farmer cost per application (£/ha) 9.30 Herbicide cost (£/ha) 10.00 Total cost of herbicide (£/ha) 19.30 Fetiliser farmer application cost (£/ha) 6.40 Fertiliser cost (£/ha) 25.00 Total fertiliser cost (£/ha) 31.40 Cut back (£/ha) 35.00 Gapping up (£/ha) 39.00 Fencing (£/ha) 0.00 Operational Costs Harvesting/carting (£/ha) 311.00 Fertiliser application cost (£/ha) 6.40 Fertilisers (£/ha) 25.00 Total fertiliser cost (£/ha) 31.40 Miscellaneous (£/ha) 20.00
1. Crop is harvested after every three years
2. Fertiliser input: 85kg/N/ha at establishment and 40kg/N/ha every harvest year
3. Establishment grant = 40% of actual planting costs
77
MISCANTHUS Revenue (other than from biomass yield) Establishment grant (£/ha) 2 710.40
Pre -planting costs Vegetation removal/cutting (£/ha) 20.00 Ploughing (£/ha) 40.50 Subsoiling (£/ha) 46.90 Number of rhizomes/ha 15000.00 Price per rhizome (£) 0.08 Cost of rhizomes (£/ha) 1200.00 Number of times herbicides are applied 1.00 Herbicide farmer application cost (£/ha) 9.30 Herbicide cost (£/ha) 24.00 Total cost of herbicide (£/ha) 33.30 Fertiliser costs (£/ha) 3 30.00 Establishment costs Power harrowing (£/ha) 32.00 Planting contractor (£/ha) 300.00 Farmer cost per herbicide application (£/ha) 9.30 Herbicide cost (£/ha) 24.00 Total cost of herbicides (£/ha) 33.30 Topping (£/ha) 20.00 Operational Costs Mowing (£/ha) 21.00 Baling costs (£/bale) 3.00 Total baling costs (£/ha) 120.00 carting (£/ha) 36.00 Fertilisers (£/ha) 3 30.00 Miscellaneous (£/year) 20.00
1. Crop is harvested annually after establishment
2. Fertiliser inputs: 85kg/N/ha at establishment and 40kg/N/ha as maintenance application every three years
3. Establishment grant = 40% of actual planting costs
78
WHEAT Revenue (other than from grain yield) Straw yield (tonnes/ha) 3.50 Price of straw (£/tonne) 30.00 Variable costs Seeds 47.00 Fertilisers (£/ha) 130.00 Herbicides (£/ha) 30.00 Fungicides (£/ha) 75.00 Other sprays e.g. foliar feeds, aphicides (£/ha) 15.00 Rolling (£/ha) 13.00 Drilling (£/ha) 25.30 Combine harvest (£/ha) 73.35 Disc harrowing (£/ha) 22.20 Power harrowing (£/ha) 32.00 Miscellaneous (£/ha) 20.00
79 16 Annex 4 – Regional data tables for un-cropped arable land resource
Table 31. Potential for ethanol, biodiesel (FAME) and wheat straw biomass energy production on the 2008 un-cropped arable land resource - regional analysis Region Potential Potential Potential Potential OSR Potential OSR Potential Potential Potential (NUTS1) wheat area wheat ethanol area production biodiesel wheat straw straw (ha) production production (ha) (tonnes) (FAME) yield energy (tonnes) (tonnes) production (tonnes) resource (tonnes) (toe) NE 2,439 19,413 5,710 863 2,762 1,132 11,260 25,021 NW 1,051 5674 1,669 119 423 173 3,291 16,412 Y&H 7,366 59,374 17,463 2,212 7,697 3,155 34,437 64,011 E Mids 14,174 114,808 33,767 4,985 16,250 6,660 66,589 125,506 W Mids 4,777 35,158 10,340 1,158 3,890 1,594 20,391 45,065 Eastern 22,567 183,698 54,029 5,494 18,790 7,701 106,545 188,214 SE 14,441 114,662 33,724 4,952 16,044 6,576 66,504 127,476 SW 5,748 41,732 12,274 1,500 5,160 2,115 24,205 54,843 England total 72,563 574,519 168,976 21,283 71,016 29,105 333,221 646,549 Wales 184 1,240 365 39 129 53 719 3,535 Overall total 72,747 575,759 169,341 21,322 71,145 29,158 333,940 650,084
80
Table 32. Potential perennial biomass energy production on the 2008 un-cropped arable land resource - regional analysis Region Potential Potential Potential Potential (NUTS1) miscanthus miscanthus willow SRC willow SRC production biomass Production biomass (tonnes) energy (toe) (tonnes) energy (toe) NE 51,000 21,007 56,500 25,021 NW 38,080 15,685 37,060 16,412 Y&H 164,980 67,956 144,540 64,011 E Mids 306,800 126,372 283,400 125,506 W Mids 115,200 47,451 101,760 45,065 Eastern 548,250 225,827 425,000 188,214 SE 347,700 143,219 287,850 127,476 SW 178,560 73,550 123,840 54,843 England total 1,750,570 721,068 1,459,950 646,549 Wales 10,799 4,448 7,982 3,535 Overall total 1,761,369 725,516 1,467,932 650,084
81 17 Annex 5 – Regional data tables for economically marginal land resource
Table 33. Potential crop area (ha) (and associated potential biomass production (tonnes)) where biomass production would be a potentially more profitable option on arable land (Primary Mask) under the scenario of wheat at £100/tonne (85% DM) and biomass at £60/odt Area (ha) Biomass (tonnes) NUTS2 Region SRC Miscan. Total SRC Miscan. Total Derbyshire and Nottinghamshire 73 4,203 4,275 1,054 50,246 51,300 Bedfordshire and Hertfordshire - 271 271 - 3,530 3,530 Shropshire and Staffordshire 673 1,202 1,875 9,435 13,653 23,088 Lincolnshire - 1,783 1,783 - 19,899 19,899 Greater Manchester - 480 480 - 6,272 6,272 Leicestershire, Rutland and Northamptonshire - 224 224 - 2,609 2,609 East Anglia - 31,867 31,867 - 417,315 417,315 Tees Valley and Durham 146 138 284 2,026 1,718 3,744 West Yorkshire 3 691 693 36 7,939 7,975 West Midlands - 338 338 - 4,121 4,121 South Yorkshire - 790 790 - 9,221 9,221 Merseyside - 562 562 - 6,951 6,951 Cheshire - 754 754 - 9,775 9,775 Herefordshire, Worcestershire and Warwickshire - 3,051 3,051 - 39,666 39,666 Cumbria 700 471 1,170 9,786 5,402 15,188 North Yorkshire 1,019 1,205 2,223 14,094 14,086 28,179 Lancashire 38 1,171 1,209 537 14,660 15,197 Northumberland and Tyne and Wear 469 81 550 6,522 872 7,394 East Yorkshire and Northern Lincolnshire - 678 678 - 8,622 8,622 Devon - 15,007 15,007 - 194,705 194,705 Outer London - 0.2 0.2 - 7 7 Surrey, East and West Sussex - 7,292 7,292 - 100,024 100,024 Kent - 4,821 4,821 - 65,718 65,718 Hampshire and Isle of Wight - 8,134 8,134 - 110,316 110,316 Dorset and Somerset - 4,894 4,894 - 62,570 62,570 Cornwall and Isles of Scilly - 8,425 8,425 - 109,216 109,216 Berkshire, Buckinghamshire and Oxfordshire - 1,894 1,894 - 24,676 24,676 Essex - 3,916 3,916 - 52,254 52,254 Gloucestershire, Wiltshire and Bristol/Bath area - 1,997 1,997 - 26,793 26,793 East Wales - 8,601 8,601 - 107,572 107,572 West Wales and The Valleys 7 11,334 11,341 99 140,237 140,336
82
Table 34. Potential crop area (ha) (and associated potential biomass production (tonnes)) where biomass production would be a potentially more profitable option on arable land (Secondary Mask) under the scenario of wheat at £100/tonne (85% DM) and biomass at £60/odt Area (ha) Biomass (tonnes) NUTS2 Region SRC Miscan. Total SRC Miscan. Total Derbyshire and Nottinghamshire 73 4,172 4,245 949 47,427 48,376 Bedfordshire and Hertfordshire - 271 271 - 3,155 3,155 Shropshire and Staffordshire 643 1,120 1,763 8,612 12,210 20,823 Lincolnshire - 1,783 1,783 - 18,910 18,910 Greater Manchester - 480 480 - 5,617 5,61 7 Leicestershire, Rutland and - 224 224 - 2,421 2,421 Northamptonshire East Anglia - 15,677 15,677 - 195,914 195,914 Tees Valley and Durham 145 131 276 1,957 1,561 3,518 West Yorkshire 3 691 693 32 7,691 7,723 West Midlands - 338 338 - 3,836 3,836 South Yorkshire - 781 781 - 8,535 8,535 Merseyside - 562 562 - 6,504 6,504 Cheshire - 747 747 - 9,450 9,450 Herefordshire, Worcestershire and - 2,285 2,285 - 28,059 28,059 Warwickshire Cumbria 499 167 666 6,724 1,668 8,392 North Yorkshire 76 975 1,051 1,025 11,504 12,529 Lancashire 19 1,100 1,119 253 12,902 13,154 Northumberland and Tyne and Wear 391 81 472 5,183 842 6,026 East Yorkshire and Northern Lincolnshire - 678 678 - 8,268 8,268 Devon - 11,431 11,431 - 137,609 137,609 Outer London - 0.2 0.2 - 3 3 Surrey, East and West Sussex - 3,815 3,815 - 49,346 49,346 Kent - 3,215 3,215 - 40,570 40,570 Hampshire and Isle of Wight - 4,021 4,021 - 49,543 49,543 Dorset and Somerset - 2,144 2,144 - 24,564 24,564 Cornwall and Isles of Scilly - 7,619 7,619 - 94,101 94,101 Berkshire, Buckinghamshire and - 1,143 1,143 - 14,052 14,052 Oxfordshire Essex - 1,380 1,380 - 16,884 16,884 Gloucestershire, Wiltshire and Bristol/Bath - 1,799 1,799 - 23,362 23,362 area East Wales - 7,392 7,392 - 86,813 86,813 West Wales and The Valleys 7 9,372 9,379 92 109,631 109,722
83 18 Annex 6 – Regional data tables for grassland resource
Table 35. Temporary (temp), permanent (perm) and rough grassland in each NUTS2 region and resource available after primary and secondary masking (1 of 2) Unmasked After primary masking After secondary masking NUTS 2 Region Temp Perm Rough Temp Perm Rough Temp Perm Rough Derbyshire and Nottinghamshire 21,136 115,709 27,752 16,393 84,761 13,582 14,010 59,725 5,658 Bedfordshire and Hertfordshire 6,288 27,768 2,700 4,830 20,936 1,984 4,514 19,127 1,694 Shropshire and Staffordshire 56,333 212,752 13,143 48,278 178,170 10,352 41,266 137,659 6,443 Lincolnshire 10,882 44,257 3,979 9,350 37,385 2,939 8,319 32,323 2,637 Greater Manchester 3,685 21,327 7,885 2,076 11,622 2,797 2,033 11,274 2,092 Leicestershire, Rutland and Northamptonshire 20,323 110,493 3,574 18,210 98,162 3,117 18,155 97,816 3,110 East Anglia 17,518 101,061 21,530 13,509 78,262 13,955 11,533 66,392 9,677 Tees Valley and Durham 11,643 76,091 31,020 8,678 50,244 16,056 7,634 38,412 8,371 West Yorkshire 7,461 46,154 12,722 4,567 27,135 5,803 4,522 26,764 5,486 West Midlands 1,482 6,477 328 1,041 3,984 153 1,041 3,984 153 South Yorkshire 5,792 19,789 12,189 3,783 12,198 5,638 3,525 11,110 4,032 Merseyside 1,864 3,984 443 1,109 1,754 113 1,109 1,754 113 Cheshire 30,582 79,933 5,652 26,569 67,737 4,405 25,998 64,410 3,751 Herefordshire, Worcestershire and Warwickshire 39,954 162,594 6,808 33,332 134,419 5,505 31,077 126,990 5,198 Cumbria 44,784 250,690 114,455 30,303 159,903 67,208 22,477 100,539 31,062 North Yorkshire 37,239 212,154 96,302 27,531 147,085 59,873 19,263 83,196 18,723 Lancashire 18,703 122,677 39,139 13,885 93,126 26,305 10,747 67,016 15,812 Northumberland and Tyne and Wear 19,072 135,616 110,707 13,600 88,921 65,094 11,422 69,053 41,984 East Yorkshire and Northern Lincolnshire 7,719 25,070 2,760 6,871 21,964 2,377 6,721 21,741 2,366 Devon 63,927 292,109 32,632 48,029 216,972 21,978 36,045 157,865 8,945 Outer London 1,028 4,065 1,027 528 1,476 425 523 1,433 423 Surrey, East and West Sussex 25,187 120,106 11,943 16,187 76,829 7,511 7,762 37,506 3,421 Continued overleaf
84 (Table 35 continued) Temporary (temp), permanent (perm) and rough grassland in each NUTS2 region and resource available after primary and secondary masking (2 of 2)
Unmasked After primary masking After secondary masking NUTS 2 Region Temp Perm Rough Temp Perm Rough Temp Perm Rough Kent 13,083 61,776 5,501 8,975 40,691 3,595 4,959 22,017 2,082 Hampshire and Isle of Wight 17,917 57,052 6,674 12,294 37,239 4,059 7,363 20,867 2,146 Dorset and Somerset 62,853 221,601 21,098 48,615 167,317 12,725 31,987 102,969 4,304 Cornwall and Isles of Scilly 42,817 144,193 16,029 28,413 97,204 9,117 23,907 80,383 6,389 Berkshire, Buckinghamshire and Oxfordshire 25,856 107,323 6,127 21,059 87,113 4,519 14,642 64,045 2,566 Essex 6,956 28,474 3,967 5,679 21,970 2,744 5,017 19,514 2,052 Gloucestershire, Wiltshire and Bristol/Bath area 51,389 197,102 23,992 39,877 151,102 14,393 25,480 92,394 7,150 East Wales 38,353 383,993 51,708 25,447 246,079 27,716 23,168 221,192 25,321 West Wales and The Valleys 68,860 622,702 149,946 47,279 407,922 72,904 39,671 340,359 44,185
85 Table 36. Potential underutilised grassland resource (temporary (temp), permanent (perm) and rough grassland (rough)) (hectares) in each NUTS2 region identified after primary and secondary masking After primary masking After secondary masking NUTS 2 Region Temp Perm Rough Temp Perm Rough Derbyshire and Nottinghamshire -1719 17137 8755 -1719 17137 5498 Bedfordshire and Hertfordshire 3262 12673 -409 3246 12439 -409 Shropshire and Staffordshire 3983 1727 -27111 3983 1727 -27111 Lincolnshire 3246 4462 -3599 3246 4462 -3599 Greater Manchester 480 6952 2797 480 6952 2092 Leicestershire, Rutland and 1296 21386 -21072 1296 21386 -21072 Northamptonshire East Anglia 3983 33676 10844 3983 33459 7718 Tees Valley and Durham -1279 5288 7925 -1279 6474 7584 West Yorkshire -513 8140 3934 -513 8140 3934 West Midlands 510 1938 -432 510 1938 -432 South Yorkshire 1704 1706 5638 1704 1706 4032 Merseyside 977 560 56 977 560 56 Cheshire 3470 -9986 -1687 3470 -9986 -1687 Herefordshire, Worcestershire and 11580 24948 -38881 11580 24948 -38881 Warwickshire Cumbria -6643 13888 41245 -6643 13888 31062 North Yorkshire -4479 13870 31394 -4479 13870 18723 Lancashire -6705 20458 14340 -6705 20458 12974 Northumberland and Tyne and Wear -5562 -11445 60035 -6738 - 41984 18399 East Yorkshire and Northern 2455 -1991 -1902 2455 -1991 -1902 Lincolnshire Devon 7189 43894 -23848 7189 43894 -23848 Outer London 460 1216 409 457 1190 408 Surrey, East and West Sussex 9183 47246 -4707 7136 36339 -4707 Kent 5219 22588 -8477 3346 16118 -8477 Hampshire and Isle of Wight 6974 10359 -941 6560 8411 -941 Dorset and Somerset 11574 25785 -6691 11504 23800 -6691 Cornwall and Isles of Scilly 11435 8882 -6369 11435 8882 -6369 Berkshire, Buckinghamshire and 9974 32336 -10636 9974 32336 -10636 Oxfordshire Essex 3935 14061 1344 3935 14061 1344 Gloucestershire, Wiltshire and 13090 42241 -423 13090 39796 -423 Bristol/Bath area East Wales 1987 53276 -36582 1987 53276 -36582 West Wales and The Valleys -3549 35929 35222 -3549 35929 35222
86