Meeting the UK's climate change commitments ...

P. Smith et al.

1

Soil Use and Management (2000) 16, 1^11

Meeting the UK's climate change commitments: options for carbon mitigation on agricultural land P. Smith1, D.S. Powlson, J.U. Smith, P. Falloon & K. Coleman Abstract. Under the Kyoto Protocol, the European Union is committed to an 8% reduction in CO2 emissions, compared to baseline (1990) levels, during the first commitment period (2008^2012). However, within the overall EU agreement, the UK is committed to a 12.5% reduction. In this paper, we estimate the carbon mitigation potential of various agricultural land-management strategies (Kyoto Article 3.4) and examine the consequences of UK and European policy options on the potential for carbon mitigation. We show that integrated agricultural land management strategies have considerable potential for carbon mitigation. Our figures suggest the following potentials (Tg yrÿ1) for each scenario: animal manure, 3.7; sewage sludge, 0.3; cereal straw incorporation, 1.9; no -till farming, 3.5; agricultural extensification, 3.3; natural woodland regeneration, 3.2 and bioenergy crop production, 4.1. A realistic land-use scenario combining a number of these individual management options has a mitigation potential of 10.4 Tg C yrÿ1 (equivalent to about 6.6% of 1990 UK CO2- carbon emissions). An important resource for carbon mitigation in agriculture is the surplus arable land, but in order to fully exploit it, policies governing the use of surplus arable land would need to be changed. Of all options examined, bioenergy crops show the greatest potential. Bioenergy crop production also shows an indefinite mitigation potential compared to other options where the potential is finite. The UKwill not attempt to meet its climate change commitments solely through changes in agricultural landuse, but since all sources of carbon mitigation will be important in meeting these commitments, agricultural options should be taken very seriously. Keywords: Climatic change, land management, organic carbon, soil, carbon dioxide, emission, UK

A

I N T RO DUC T I O N

nnex B of the Kyoto Protocol (available at: http:// www.cop3.de/) lists the Quantified Emission Limitation or Reduction Commitments for 39 of the parties that ratified the United Nations Framework Convention on Climate Change (UNFCCC). The European Union is committed to an 8% reduction in CO2 emissions compared to baseline (1990) levels during the first commitment period (2008^ 2012). Since the Kyoto Protocol was signed, the EU member states have rearranged this commitment internally such that the sum of the EU commitments is still an 8% reduction, but the UK is now committed to a 12.5% reduction (DETR, 1998). The Kyoto Protocol allows carbon emissions to be offset by demonstrable removal of carbon from the atmosphere. Thus, land-use/land-management change and forestry activities that are shown to reduce atmospheric CO2 levels can be included in the Kyoto emission reduction targets. These activities include afforestation, reforestation and deforestation (Article 3.3 of the Kyoto Protocol) and may include the improved management of agricultural soils (Article 3.4). In this paper, we concentrate mainly on Kyoto Article 3.4 options, though afforestation of surplus arable land is also considered. Soil Science Department, IACR-Rothamsted, Harpenden, Herts. AL5 2JQ , UK. E-mail: [email protected] 1 Corresponding author.

In a recent series of papers, (Smith etal.1997a,b; 1998a) presented preliminary estimates of carbon mitigation potential in Europe by using long-term experiments to develop relationships between changes in soil organic carbon (SOC) content and various land-use=land-management changes. These relationships were then used to examine the potential for carbon mitigation in European agriculture by developing six scenarios whereby a single land-use=land-management change was implemented across Europe. Recently, Smith et al. (2000; Smith & Powlson 2000) assessed European carbon mitigation potential relative to a 1990 baseline condition and considered the competition for available land. In this paper we present a similar analysis of the potential for carbon mitigation on agricultural land in the UK. M ET H O DS (Smith et al. 1997a,b; 1998a) provided estimates for CO2carbon mitigation for the European Union and the wider Europe. In this paper, our estimates for agricultural CO2carbon mitigation are for the UK. Although some changes in SOC may occur below 30 cm, the great majority of the change will occur in the top 30 cm. For this reason (as in Smith et al., 1997a,b; 1998a), changes in SOC are assessed to a depth of 30 cm. The area and total SOC content of UK arable land The total SOC in UK arable soils (to 30 cm) was calculated using the figures of Howard et al. (1995; updated by Milne &

2

Agricultural carbon mitigation in the UK

Brown, 1997) and Eurostat (1995) as follows. Howard et al. (1995) divided all of the land area of Great Britain (England, Wales and Scotland) into 11 km and 1010 km squares. These squares were then categorized by one of seven landcover groups according to the dominant landcover type, one of which was arable agriculture. The SOC content to 1 m for all arable land in Great Britain can be calculated as 2500 Tg from the figures for England, Wales and Scotland in Howard et al. (1995). Assuming an exponential distribution of SOC from the surface, the top 30 cm of soil contains 87.3% of the carbon to 1 m. The figure to 30 cm is, therefore, 2184 Tg for Great Britain.The area of Great Britain reported in Howard et al. (1995) is 23 773 400 ha, whereas the total area of the UK (Great Britain plus Northern Ireland) is 24 414 000 ha (Eurostat, 1995). Assuming a distribution of SOC contents in Northern Ireland similar to that in the rest of the UK, the total UK SOC content of arable land can be derived as 2242 Tg. However, the area of grid squares where arable agriculture was the dominant landcover type in Howard et al. (1995) amounts to 12 964 900 ha. The actual arable area of the UK (from Eurostat, 1995) is 6 589 000 ha so the figure for 12 964 900 ha needs to be scaled by 0.51 to give a figure for the actual arable area (to 30 cm) of 1140 Tg. Changes in SOC resulting from changes in agricultural land management Only changes to arable agricultural land are considered since grassland is already under management conducive to SOC accumulation compared to arable land, and arable land shows the greatest potential for SOC increase.The ploughing under of grassland causes a rapid decrease in SOC that takes many years to recover (e.g. Smith et al., 1996a). As in previous papers (Smith et al., 1997a, b; 1998a), the logistics of resource redistribution are not considered.Similarly,although scenarios are chosen to minimise the risk of environmental damage (e.g. sewage sludge application rates of 1 t haÿ1 yrÿ1), potential environmental side- effects are not explicitly considered. In the papers of (Smith etal.1997a,b; 1998a), we derived statistical relationships between various agricultural land-management practices and changes in SOC using results derived from European long-term experiments contributing to the Global Change and Terrestrial Ecosystems Soil Organic Matter Network (GCTE-SOMNET; Smith et al., 1996b,c,d; 1998b). These changes in SOC were then applied to various soil carbon pools under different land-use, to examine scenarios for carbon mitigation. The scenarios examined were: the amendment of soil with organic additions (animal manure, sewage sludge and cereal straw; Smith et al., 1997a,b), conversion to no -till agriculture (Smith et al., 1998a), agricultural extensification (Smith et al., 1997a,b), and natural woodland regeneration (with varying degrees of bio fuel use of the wood; Smith et al., 1997a,b). In the papers of Smith et al. (1997a,b; 1998a) the land-management practices considered were assumed to be absent in present agricultural practice. For some scenarios this assumption was entirely safe (e.g. the permanent afforestation of arable set-aside land which, by definition, was not under woodland at the time) whilst for others (e.g. the addition of organic amendments to soils), the assumption was less safe, because some agricultural practices were in use to varying extent in the early 1990s. In this paper we estimate the extent of use of each land-management practice in the UK in 1990

and estimate the effects of changes in the extent of each practice relative to this baseline. In addition to a different focus area (the UK instead of Europe) in this study, various other improvements to the relationships used in (Smith et al. 1997a,b; 1998a) have been made as detailed below. Animal manure The total amount of animal manure produced each year (based on a 1990 baseline) was calculated in the manner employed for Europe by Smith et al. (1997a). Animal manure is derived from housed cattle and pigs. In the UK in 1990 there were 1 184 300 cattle and 7 380 000 pigs (Eurostat, 1995). Using figures derived from MAFF (1994) for the amount of manure produced per animal per year (mean for cattle [bullock and dairy] 6.35 t yrÿ1; mean for pigs [sow and piglets] 2.4 t yrÿ1), the total animal manure production for the UK for 1990 was calculated as 92.9106 t yrÿ1. The arable and non-arable agricultural areas of the UK are 6 589 000 ha and 11 858 000 ha respectively (Eurostat, 1995), giving a fraction of non-arable land (of total agricultural land) of 0.643. Assuming that animal manure was spread proportionately upon arable and non-arable agricultural land, in 1990, 33.2106 t would have been spread on arable land (an average application rate of 5 t haÿ1 yrÿ1), and 59.7106 t spread on non-arable land (which will provide a conservative estimate of the amount not spread on arable land since a disproportionately high amount is probably applied to grassland). The 59.7106 t yrÿ1 spread on nonarable land would be better employed for the purposes of increasing SOC by incorporating into arable land because the manure is incorporated by cultivation so less carbon is oxidized, and arable soils generally have a greater potential to increase in SOC because they are more carbon depleted (Poulton, 1996). This quantity represents the `surplus' animal manure. In this paper we use a variable application rate of animal manure (those commonly used in the UK; between 5 and 20 t dry matter haÿ1 yrÿ1) to allow different areas to be used in combined land management scenarios. The statistically significant (r2 0.49, t15 3.76, P 0.0017) relationship between manure application rate and SOC increase of Smith et al. (1997a) was used here following: y 0:038x ÿ 0:0538 where: y % change in SOC yrÿ1, and x animal manure added (t haÿ1 yrÿ1). This gives potential SOC accumulation rates ranging from 0.14% yrÿ1 (for 5 t haÿ1 yrÿ1) to 0.71% yrÿ1 (for 20 t haÿ1 yrÿ1). Sewage sludge The relationship between SOC increase and sewage sludge application at a rate of 1 t dry matter haÿ1 yrÿ1 used by Smith et al. (1997a; 1998a) was based on results from long-term experiments where rates of 6.5 t haÿ1 yrÿ1 and above were used. The relationship below 6.5 t haÿ1 yrÿ1 was extrapolated to derive SOC changes at 1 t haÿ1 yrÿ1 of 1.96% yrÿ1. This is an erroneously large change in SOC. In this paper we have assumed a linear relationship between 0 and 6.5 t haÿ1 yrÿ1 to give an SOC accumulation rate of 0.49% yrÿ1 at 1 t haÿ1 yrÿ1. The application rate of 1 t haÿ1 yrÿ1 of sewage sludge was

P. Smith et al. chosen by Smith et al. (1997a) because it was considered to be an environmentally safe level (based on the lowest recommended maximum rate in Europe, Williams, 1988). Because of this, sewage sludge, unlike straw and animal manure, was always applied at a fixed rate in this study. In the UK, 45% of sewage sludge was applied to agricultural land in the late 1980s (including horticulture and gardens; Webber et al., 1986) but this may have increased recently after the 1998 ban on dumping of sewage sludge at sea. The amount of sewage sludge produced in the UK in 1990 is assumed (based on late 1980s figures) to be 1 210 000 t (dry matter) yrÿ1. Using these figures, 544 500 t yrÿ1, of sewage sludge was applied to agricultural land, of which, 194 500 t yrÿ1 was applied to arable land (assuming the sludge was spread pro portionately to arable and grassland according to area). Assuming an application rate of 1 t haÿ1 yrÿ1, a maximum arable area of 194 500 ha would have been treated with sewage sludge which represents about 3% of UK arable land.We use these figures for the 1990 baseline. Given that a calculated 194 500 t yrÿ1 was applied to arable land in the baseline year (and this will continue to be applied), an extra 350 000 t yrÿ1 is available for use on arable land each year. If all the sewage sludge produced in the UK were spread on arable land at an application rate of 1 t haÿ1 yrÿ1, the maximum area covered would be 544 500 ha. This represents 8.3% of all arable land, or an extra 5.3% of the arable land in addition to that receiving sewage sludge in 1990. Straw incorporation The same relationship between increase in SOC and straw incorporation was used in this study as presented in Smith et al. (1997a,b). Before the introduction of strict regulations governing straw burning in1983, very little (2%) straw incorporation took place in the UK (Christian & Ball, 1994). However, straw burning regulations increased the incidence of straw incorporation during the 1980s such that in 1988, 18% of straw in the UK was ploughed-in (MAFF, 1989; Christian & Ball, 1994). We use these figures for the 1990 baseline, which means that 18% of cereal land, i.e. 545580 ha, were subject to straw incorporation in 1990. This is equivalent to 8.3% of arable land. The amount of surplus straw (i.e. after the quantities required for other purposes had been removed) produced in the UK has been estimated at between 5 and 7106 t yrÿ1 (Staniforth, 1982; Prew et al., 1995) with the figure 6.5106 t yrÿ1 most commonly used (D.G. Christian, pers. comm.; MAFF, 1984). The 8.3% of arable land that was subject to straw incorporation in the baseline year received 18% of the surplus straw, i.e. 1170 000 t. This leaves an annual straw surplus of 5 330 000 t yrÿ1 to be incorporated in scenarios of land-management change. As with animal manure, we used variable incorporation rates for straw (those commonly used for low to high yielding cereal crops in the UK; between 2 and 10 t dry matter haÿ1 yrÿ1) to allow different areas to be used in combined land management scenarios. The statistically significant (r2 0.41, t8 2.38, P 0.04) relationship between straw incorporation rate and SOC increase of Smith et al. (1997a) was used here following: y 0:1115x 0:192:

3

where y % change in SOC yrÿ1, and x straw incorporated (t haÿ1 yrÿ1). This gives SOC accumulation rates ranging between 0.42% yrÿ1 (for 2 t haÿ1 yrÿ1) and 1.31% yrÿ1 (for 10 t haÿ1 yrÿ1). No-till The same relationship between SOC accumulation rate and no -till (0.73% yrÿ1) as presented in Smith et al. (1998a) was used in this study, and again, changes were applied only to the 0^25 cm layer (with the 25^30 cm layer assumed to be unaffected; see Smith et al., 1998a). As well as increasing SOC accumulation, no -till farming also reduces fossil fuel carbon emissions, even after accounting for the extra herbicides needed (Frye, 1984). This saving of fossil fuel carbon is estimated to be 23.8 kg C haÿ1 yrÿ1. (Kern & Johnson, 1993; Frye, 1984). The total carbon mitigation potential of no -till was calculated as the yearly SOC accumulation plus the fossil fuel carbon saving. No -till/reduced-till agriculture is far less prevalent in the UK than in North America (Christian & Ball, 1994; Lal et al., 1998). Unlike in North America (CTIC, 1997), accurate yearly figures for the extent of no -till farming in the UK are not available. In the UK in 1973 it was estimated that 3% of arable crops in England and 0.2% in Scotland were direct drilled (Christian & Ball, 1994). By 1990 (after stricter straw burning regulations were introduced in 1983), these proportions were likely to have been even lower (D. Christian, pers. comm.). The maximum area of arable land under no -till in the baseline year (1990) is therefore set at 3%. The maximum area on which no -till could be practised is 80% of the cereal area (based on soil suitability either for winter plus spring cereals or for winter cereals alone; Cannell et al., 1978).The area suitable for no -till farming is, therefore, 2 424 800 ha, which represents 36.8% of arable land. Agricultural extensification The agricultural extensification scenario reported in Smith et al. (1997a,b) relies upon a 20^30% surplus of arable land by 2010 as projected by Flaig & Mohr (1994). Given recent trends in the areas in agricultural set-aside (which has remained relatively constant in the UK over the last 10 years), 20^30% surplus now seems very unlikely. A figure of 10% for surplus arable land in the UK by 2010 was regarded as more likely and was used here. A two in six period in grass is required for a viable ley-arable rotation similar to that described in Smith et al. (1997a,b). Using the 10% surplus arable land, 1/3 of the intensive arable area of the UK could be extensified. The same relationship between extensification and change in SOC as reported in Smith et al. (1997a) is used, i.e. an SOC accumulation of1.02% yrÿ1 under extensive farming. Because extensification takes place on land that was under conventional arable cultivation in 1990, the area under extensive farming on this land in 1990 is, by definition, zero. Natural woodland regeneration A recent spatial application of the dynamic SOM models RothC (Falloon et al., 1998) and CENTURY (Falloon et al., 1999) showed estimates for the increase in SOC during woodland regeneration reported by Smith et al. (1997a) to be too high (Falloon et al., 1999). This is due to an atypically high

4


SOC accumulation in one of the experiments used to derive the mean. For this reason, the lower estimate of SOC accumulation during natural afforestation (1.17% yrÿ1) was used here. For aboveground biomass C in trees, we used the mean of the range limits used by IPCC (1996, p782; Nabuurs & Mohren, 1993) for the average net annual rate of carbon accumulation in broadleaf forests on former agricultural land (range: 2.2^3.4; mean of these figures: 2.8 t haÿ1 yÿ1). This replaced the relationship between SOC and above-ground carbon used by Smith et al. (1997a, b). In this paper we assumed that the trees were grown to maturity without harvest. As for the extensification scenario, surplus arable land in 2010 was assumed to be 10%. The area of arable land under this land-use in 1990 is, by definition, zero. Bioenergy crop production In this study we assume the same accumulation of SOC under short-rotation woody bioenergy crops as that seen under natural woodland regeneration (1.17% yrÿ1). We assume an annual bioenergy crop production of 12 t haÿ1 yrÿ1 (Hall et al., 1981; all production figures are for oven dry tonnes) which is the average production target for 2000 of the EU's Biomass Development Programme (Hall et al., 1997). This figure is realistic; in the UK, short rotation woody crops such as willow (Salix spp.) can yield between 13.5 and 30 t haÿ1 yrÿ1 (Kofman & Spinelli, 1997; calculated from their Figure 1). The target yield of 12 t haÿ1 yrÿ1 by 2008 is, therefore, considered attainable (Hall et al., 1997). In this study we assumed that all dry matter was used in bioenergy production. Carbon was assumed to constitute 50% of dry matter. The energy utilization figure of 100%

Fig. 1. Maximum yearly carbon mitigation potential, and maximum percentage offset of UK 1990 CO2 ^ carbon emissions (156.5 Tg C yrÿ1; Marland et al., 1994), of each land-management practice in isolation. Animal manure was applied at 20 t haÿ1 yrÿ1 to 45.3% of arable land. Sewage sludge was applied at 1 t haÿ1 yrÿ1 to an extra 5.3% of arable land (compared to 1990 levels). Straw was incorporated at a rate of 10 t haÿ1 yrÿ1 to an extra 40.4% of arable land (compared to 1990 levels). No -till was applied to all arable land suitable for no -till farming (i.e. 36.8% of arable land). Extensification was applied to 1/3 of all arable land not under the same land-management as 1990 (i.e. 28.6% of arable land). Woodland regeneration and bioenergy pro duction were each applied to surplus (i.e. 10%) arable land.

(for dedicated temperate energy crops; IPCC, 1996; p755) was used, as was the mean of the lower (0.65) and higher (0.75) energy substitution factors for dedicated temperate energy crops (i.e. 0.7) of Sampson et al. (1993). As for the extensification scenario, surplus arable land in 2010 was assumed to be 10%. The area of arable land under this land-use in 1990 is, by definition, zero. Summary of land-management changes Table 1 summarizes the yearly SOC accumulation rates for each land-management practice as well as the baseline (1990) occurrence of these practices, and the maximum potential area over which these scenarios can be implemented. For the organic amendments, the estimated yearly surplus of each resource is also provided. Using the figures derived in this section, we applied the land management changes described to the areas and carbon stocks of the UK to examine the impact of each land-management change in isolation, and in combination with others. Combined land-management scenarios For the combined land management scenarios, no management practice takes place on the same area of land as another (e.g. an area under no -till cannot also undergo straw incorporation or manure addition). The combined scenarios do not, therefore, combine different practices on the same portion of land; instead, they utilize different management practices on different portions of the available land. Furthermore, some land-management practices require the use of the same portion of land as others, and so are mutually exclusive in a given combined scenario. An example of this is the use of surplus arable land: agricultural extensification utilizes all of the 10% surplus arable land and so cannot be used in combination with natural woodland regeneration or bio energy production, since they rely upon exploitation of surplus arable land. The combined scenarios were formulated on the basis of potential policy decisions in the UK and Europe. The first policy decision considered is how to utilize surplus agricultural land.This land can either be used for agricultural extensification which would mean that the area was no longer surplus to requirements, or it can be used for bioenergy production or woodland regeneration.This first policy decision divides the combined scenarios into three groups that we term `bioenergy' (B),`woodland' (W) and `extensification' (E).The second policy decision considered relates to how the remaining agricultural land is managed. Potential policies include those encouraging no -till farming, straw incorporation, or the more efficient use of organic amendments (such as animal manure and sewage sludge). On the basis of these different policy decisions, we examined 12 combined landmanagement scenarios, and one other which is considered to be an optimal realistic scenario. In all scenarios, the area under a practice favourable to carbon accumulation in the baseline year remained unaltered.The policy aims and resulting combined land-management scenarios are shown in Table 2. R E SU LTS The carbon mitigation potential of each land management change Relationships were derived between the carbon mitigation potential of each land management practice, and the propor-

P. Smith et al.

5

Table 1. Summary for each land-management practice of SOC accumulation rates, extent of occurrence in the UK during the baseline year (1990), and the maximum area to which each land-management practice could be applied. Land-management practice Animal manure1 (5 ^ 20 t haÿ1 yrÿ1) Sewage sludge (1 t haÿ1 yrÿ1) Cereal straw (2 ^ 10 t haÿ1 yrÿ1) No -till farming Agricultural extensification Natural woodland regeneration Bioenergy crop production

SOC accumulation rate (% yrÿ1) 0.14 ^ 0.71 0.49

Area under this practice in 1990 (103 ha)

% of arable area exclusive to this practice in 19903

Maximum area to which the practice could be applied (103 ha)4

% of arable area to which the practice could be applied

Surplus organic resource available each year (103 t yrÿ1)10

0

0

5647

85.73

59727

195.3

3

545

8.35

350

6

5330

3

0.42 ^ 1.31

546.

8.3

2662

40.4

0.73 (plus fossil fuel C saving)2 1.02

198.3

3

2425

36.87

-

8

1.17 (plus aboveground C)2 1.17 (plus aboveground C)2

0

0

1884

28.6

-

0

0

659

10.9

-

659

9

-

0

0

10.

1 Lowest rate of animal manure used is 10.6 t haÿ1 yrÿ1 since this is the rate resulting from applying 59.7106 t yrÿ1 to cover all available arable land (i.e. 85.7% of arable area; see footnote 3); 2 The main impact of woodland regeneration and bioenergy crop production lies in the production of aboveground carbon. For no -till farming, fossil fuel C savings also need to be added (see text); 3 In this study, areas already under a practice favourable to SOC accumulation in 1990 (see text) were not manipulated. Hence, 14.3% of the UK's arable area remains in the same land-use as in 1990, which leaves a maximum of 85.7% of the arable area for manipulation; 4 Calculated from % in column 6; 5 Given the total surplus of sewage sludge produced in the UK each year, this is the maximum area covered at 1 t haÿ1 yrÿ1; 6 Given the surplus of straw produced in the UK each year, this is the maximum area covered at the highest incorporation rate (10 t haÿ1 yrÿ1); 7 Defined by soil suitability (see text for details). 8 The 10% surplus arable land allows 1/3 of current conventional agriculture to be extensified (see text), i.e. 1/3 of the 85.7% arable area available for manipulation; 9 Since these land-management changes are applied exclusively on surplus arable land, the maximum area available for these changes is 10% of arable land; 10 See text for details of how these figures were derived.

tion of arable area to which it was applied.These relationships are given in Table 3 These relationships were used, with estimates of the maximum area to which the practices could be applied, to derive the maximum yearly carbon mitigation potential for each land-management practice in isolation. Figure 1 shows these values and compares them to the 1990 CO2- carbon emissions from the UK.

The carbon mitigation potential of combined land-use=land-management scenarios Using the values derived above, it was also possible to calculate the carbon mitigation potential ofvarious combined land management scenarios (Table 2). In all scenarios, the area under a practice favourable to carbon accumulation in the baseline year remained unaltered. Therefore, all scenarios have 14.3% of arable land remaining in1990 land-use.Tables 4, 5, 6 and 7 show

Table 2. Potential policy decisions and resulting combined land-management scenarios. Policy aim for 10% surplus arable land Policy aim for remaining arable land

Resulting combined land management scenario Scenario Abbreviation

Encourage Bioenergy (B) Encourage Bioenergy (B) Encourage Bioenergy (B) Encourage Bioenergy (B)

Bioenergy plus no -till Bioenergy plus straw incorporation Bioenergy plus organic amendments Bioenergy plus organic amendments at highest allowed rates and put remaining area into no -till Woodland plus no -till Woodland plus straw incorporation Woodland plus organic amendments Woodland plus organic amendments at highest allowed rates and put remaining area into no -till Extensification plus no -till Extensification plus straw incorporation Extensification plus organic amendments Extensification plus organic amendments at highest allowed rates and put remaining area into no -till Use 50% of surplus arable land for bioenergy production and the other 50% for woodland, use organic amendments at the highest rates allowed and put the remaining area into no -till

Encourage Woodland (W) Encourage Woodland (W) Encourage Woodland (W) Encourage Woodland (W) Encourage Extensification (E) Encourage Extensification (E) Encourage Extensification (E) Encourage Extensification (E) Optimal Realistic Scenario (Opt)

Encourage no -till (NT) Encourage straw incorporation (S) Encourage organic amendments (O) Encourage organic amendments at highest allowed rates and put remaining area into no -till (O NT) Encourage no -till (NT) Encourage straw incorporation (S) Encourage organic amendments (O) Encourage organic amendments at highest allowed rates and put remaining area into no -till (O NT) Encourage no -till (NT) Encourage straw incorporation (S) Encourage organic amendments (O) Encourage organic amendments at highest allowed rates and put remaining area into no -till (O NT) Optimal realistic scenario (Opt)

B NT B S B O B O NT W NT WS WO W O NT E NT E S E O E O NT Opt

6


Fig. 2. Maximum yearly carbon mitigation potential, and maximum percentage offset of UK 1990 CO2 emissions (156.5 Tg C yrÿ1; Marland et al., 1994) of each combined land management scenario. SeeTable 3 for scenario abbreviation codes. The lines at 8% (solid) and 12.5% (dashed) of UK 1990 CO2 ^ carbon emissions represents the Quantified Emission Limitation or Reduction Commitments for the first commitment period (2008 ^2012) of the Kyoto Protocol for the EU and the UK, respectively. Table 3. Relationship between the carbon mitigation potential of each land management practice and the proportion of UK arable area to which it is applied. Land management practice

Animal manure (10.6 t haÿ1 yrÿ1) Animal manure (20 t haÿ1 yrÿ1) Sewage sludge (1 t haÿ1 yrÿ1) Cereal straw (2 t haÿ1 yrÿ1) Cereal straw (10 t haÿ1 yrÿ1) No -till farming Agricultural extensification Natural woodland regeneration Bioenergy crop production

Maximum Maximum yearly Trend-line5 extra % carbon mitigation area of potential for the arable land1 UK (Tg yrÿ1) 85.7 45.3{ 5.3{ 40.4{ 8.0{ 36.8 28.6 10. 10.

3.40 3.65 0.30 1.91 1.20 3.452 3.34 3.183 4.104

0.0397 0.0805 0.0560 0.0473 0.1506 0.0938 0.1176 0.3178 0.4101

1 This is the area in addition to that already under this land management practice in 1990. Those figures marked with { show the maximum areas covered applying all of the surplus organic material; 2 This figure includes SOC accumulation and fossil fuel carbon savings; 3This figure includes SOC accumulation and increases in woody biomass carbon; 4This figure includes SOC accumulation and carbon mitigation through the substitution of fossil fuel carbon with bioenergy crop carbon; 5These trend-lines show the carbon mitigation potential per unit area, and are the constants in the equations describing the increase in carbon mitigation potential with increasing area to which the land-management change is applied. All equations have the form: Carbon Mitigation Potential for UK [Tg yrÿ1] [trend-line]% arable area to which the land-management change is applied.

the carbon mitigation potential for each of the combined landmanagement scenarios outlined inTable 2. The total carbon mitigation potential for each scenario compared to UK 1990 CO2- carbon emissions is summarized in Figure 2. D I S C USS I O N Single land-management options This study demonstrates that bioenergy production has the greatest potential for carbon mitigation of all land-manage-

ment options examined. It relies upon a 10% surplus of arable land. The mitigation potential of the bioenergy scenario (on surplus arable land) is a little higher than that of the best case mitigation potential on the remaining arable land, highlighting that surplus arable land is an important resource for carbon mitigation in agriculture. The extra advantage of bioenergy production over all other scenarios is that the fossil fuel carbon substitution component (about 67% of the total carbon mitigation potential) continues indefinitely. In all other scenarios (with the exception of the small fossil fuel carbon saving from the no -till scenario) the systems tend toward new equilibria after about 50^100 years, whereby carbon accumulation slows and eventually stops. From Figure 1, it is clear that no single land-management option can offset all of the UK's 1990 CO2- carbon emissions. However, it is not appropriate to dismiss mitigation options on the basis that they contribute only small proportions towards the mitigation target (Paustian et al., 1997; Royal Society, 1999). The finding that no one land-management change in isolation can deliver the UK's climate change commitments highlights the importance of using integrated land management for carbon mitigation. In attempting to quantify the maximum potential for carbon mitigation through agriculture, little account was taken in this study of the suitability of soils or climate in a given region of the UK or of the local availability (or surplus) of resources used (such as animal manure), nor of the potential problems of moving resources from one locality to another. In combined land-management scenarios, however, it is easier to select appropriate strategies for given regions since many options are available, each occupying a relatively small percentage of the available arable land area. For the scenarios using organic amendments, it is important to weigh the benefits against potential undesirable side effects such as increased risk of nitrate leaching, trace gas fluxes from the soil, increased use of fuels to apply the amendments, and increased heavy metal and organic pollutant concentrations in the environment. The rates of organic amendment suggested here are low, however, and as such are unlikely to lead to serious environmental side effects, but we have not attempted to quantify these. As well as potential problems associated with the use of organic materials in agriculture, it is important also to consider the additional benefits (e.g. improved soil fertility and structure, and increased agricultural productivity and sustainability; Arden-Clarke & Hodges, 1988; Paustian et al., 1997; Lal et al., 1998) leading, in many cases, to `win-win' land-management strategies (Lal et al., 1998). Combined land-management options The use of the surplus arable land is an important main factor influencing the efficacy of a combined carbon mitigation strategy. Where the surplus arable land is used to extensify 1/3 of current intensive agricultural production (ENT, ES, EO and EONT scenarios), mitigation potentials of combined scenarios are less effective than those allocating surplus arable land for bioenergy production.There are, however, other potential benefits associated with extensification. The area under grass each year in the new rotational arable/ grass systems would be 732 000 ha yrÿ1. Given stocking densities of 13 and 1000 animals haÿ1 for pigs (MAFF, 1983) and

P. Smith et al.

7

Table 4. Carbon mitigation potential for UK combined land-management scenarios, with surplus agricultural land used for bioenergy production (B). Scenario1

Land-management during commitment period

B NT

% of arable land used

Carbon mitigation potential

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Bioenergy crop C mitigation2 Extra No -till3 Remainder unchanged Total

3.0 8.3 3.0 10.0 36.8 38.9

0.00 0.00 0.00 4.10 3.45 0.00 7.55

B S

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Bioenergy crop C mitigation2 Extra Straw incorporation (2 t haÿ1) Remainder unchanged Total

3.0 8.3 3.0 10.0 40.4 35.3

0.00 0.00 0.00 4.10 1.91 0.00 6.01

B O

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Bioenergy crop C mitigation2 Extra sewage sludge Animal manure (12.9 t haÿ1) Total

3.0 8.3 3.0 10.0 5.3 70.4

0.00 0.00 0.00 4.10 0.30 3.49 7.89

B O NT

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Bioenergy crop C mitigation2 Extra sewage sludge Animal manure (20 t haÿ1) Extra Straw incorporation (10 t haÿ1) Extra No -till3 Total

3.0 8.3 3.0 10.0 5.3 45.3 8.0 17.1

0.00 0.00 0.00 4.10 0.30 3.65 1.20 1.60 10.85

1 3

See Table 3 for abbreviation code; 2 Includes 2.77 Tg yrÿ1 from bioenergy crop C mitigation and 1.33 Tg yrÿ1 from SOC accumulation under woody crops; The total carbon mitigation potential for extra no -till of which 97.8% is from SOC accumulation and 2.2% is from fossil fuel carbon savings.

poultry (Lampkin, 1990), respectively, either all pigs or all poultry currently reared in intensive indoor units could be raised outdoors. Using the new grassland, 56% of pigs plus poultry could be reared in outdoor units. If reared in outdoor units, pigs would add about 24 t haÿ1 of animal manure to the land during the grass phase which is about 8 t haÿ1 yrÿ1. The extensification scenario does, however, preclude the use of any arable land for the more beneficial (in terms of carbon mitigation) purpose of bioenergy crop production. The figures for scenarios where surplus arable land is used for woodland regeneration are similar to those for extensification, but unlike extensification, woodland regeneration can be used in combination with bioenergy crop production. When considering the remainder of arable land, scenarios relying upon straw incorporation (B S, W S and E S) perform least well when compared to scenarios with the same use of surplus arable land, followed by those where no till is used (B NT, W NTand E NT). The poorer performance of the straw incorporation and no -till scenarios occurs because more than one third of the arable land (in B and W ) scenarios remain unchanged in management. The application of low rates of manure (B O, W O and E O), in which all arable land is used, show slightly higher potential. A greater carbon mitigation potential, and a more realistic mix of land-management options, arises from the use of all available organic amendments at the highest allowed rates with the remainder of arable land (17% for B and W

scenarios) used for no -till farming (B O NT, W O NTand E O NT). In these scenarios, the application rates for organic amendments are attainable, and the extent of extra no -till farming is not unreasonable representing less than half of the land suitable for no -till farming (Cannell et al., 1978). Of the scenarios with attainable combinations of land-management options, the B O NT scenario shows the greatest mitigation potential (Figure 2). No combined scenario is able, by itself, to offset more CO2carbon than the level the Europe Union is committed to (8%) in the Kyoto Protocol or the UK target of 12.5% (see Figure 2). It may not be possible to use all surplus arable land for bioenergy production since it must be sited in close proximity to the bioenergy processing and production plant (Hall et al., 1997). If we assume that 50% of surplus arable land could be sited close to bioenergy production plants, the remainder of surplus arable land could be used to grow woodland to maturity. This combined use of surplus arable land is the basis of the optimal realistic scenario (Opt; Table 8). This scenario uses levels of bioenergy production and woodland growth, rates and areas of organic amendment, and an area for no -till farming that are all attainable by the beginning of the first commitment period (2008). This combined landmanagement scenario comes close to the European Union's commitments for CO2- carbon mitigation (Fig. 2). The realisation of Opt (or similar scenario) would entail changes in European land-management/agricultural policy

8


Table 5. Carbon mitigation potential for UK combined land-management scenarios, with surplus agricultural land used for woodland (W). Scenario1




W NT

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Woodland C accumulation2 Extra No -till3 Remainder unchanged Total

3.0 8.3 3.0 10.0 36.8 38.9

0.00 0.00 0.00 3.18 3.45 0.00 6.63

WS

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Woodland C accumulation2 Extra Straw incorporation (2 t haÿ1) Remainder unchanged Total

3.0 8.3 3.0 10.0 40.4 35.3

0.00 0.00 0.00 3.18 1.91 0.00 5.09

WO

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Woodland C accumulation2 Extra sewage sludge Animal manure (12.9 t haÿ1) Total

3.0 8.3 3.0 10.0 5.3 70.4

0.00 0.00 0.00 3.18 0.30 3.49 6.97

W O NT

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Woodland C accumulation2 Extra sewage sludge Animal manure (20 t haÿ1) Extra Straw incorporation (10 t haÿ1) Extra No -till3 Total

3.0 8.3 3.0 10.0 5.3 45.3 8.0 17.1

0.00 0.00 0.00 3.18 0.30 3.65 1.2 1.6 9.93

1

SeeTable 3 for abbreviation code; 2 Includes1.84 Tg yrÿ1 from aboveground carbon accumulation and1.33 Tg yrÿ1 from SOC accumulation under woodland; 3 The total carbon mitigation potential for extra no -till of which 97.8% is from SOC accumulation and 2.2% is from fossil fuel carbon savings.

such that surplus arable land could be put into alternative long-term use, instead of short-term rotational set-aside. Additional policies would be required to encourage (a) bio energy crop production on surplus arable land where feasible, (b) woodland regeneration on surplus arable land where bio energy crop production is not feasible, (c) greater adoption of conservation tillage practices in areas where the land is suitable, and (d) application of and the majority of organic amendments to arable land, with inorganic fertilization replacing current organic fertilization of grassland and nonarable crops. The mechanisms to implement such policies include the use of tax benefits, subsidies, joint implementation projects, and improved extension and information dissemination (Paustian et al., 1997). The combined scenarios show that the carbon mitigation potential of agricultural land is significant in the context of the UK's Kyoto commitments, accounting for over 50% of the target reduction. These options should be used in addition to the main mechanism for carbon mitigation which is emission reduction, which many argue provide the only long-term solution to the problems posed by high atmospheric carbon dioxide concentrations. The remainder of the target would need to be met by forestry activities (under Article 3.3 of the Kyoto Protocol) such as afforestation on non-arable land and improvement of land already under forestry (Cannell, 1999). Sources of land for forestry-related carbon mitigation other than surplus arable land (e.g. under

grassland, natural plant cover, or land already under some form of forestry; Nabuurs et al., 1999) are beyond the scope of this paper but it is likely that these activities will also make a significant contribution to UK carbon mitigation (Cannell, 1999; Nabuurs et al., 1999). In terms of a reduction of CO2- carbon emissions, agricultural land accounts for about 7% of the 12.5% emission reduction target, so a further 5.5% reduction in emissions is required. In 1995, this reduction had almost been achieved in the UK, with 1995 CO2- carbon emissions 5.2% lower than those in 1990. However, by the following year emissions increased again and the 1996 figure was only 1.1% lower than in 1990. The best UK combined scenario is able to offset just under 1% of Europe's CO2- carbon emissions reported in Smith et al. (1999). However, per unit area, the UK has a greater carbon mitigation potential than Europe as a whole (0.44 compared to 0.23 t haÿ1 for the UK and Europe, respectively). This difference appears to be due to the higher SOC content of arable soils in the UK compared to Europe (Smith et al., 2000). The estimates for carbon mitigation potential presented in this study are based upon a large number of European longterm experiments (Smith et al., 1997a; 1998a). Some of the estimates for SOC accumulation rate are based upon statistically robust relationships (e.g. animal manure, sewage sludge, straw incorporation) whilst others are based on fewer

P. Smith et al.

9

Table 6. Carbon mitigation potential for UK combined land-management scenarios, with surplus agricultural land used for agricultural extensification of1/3 of available arable land (E). Scenario1


E NT



1990 Sewage sludge 1990 Straw incorporation 1990 No -till Extensification Extra No -till2 Remainder unchanged Total

3.0 8.3 3.0 28.6 36.8 20.3

0.00 0.00 0.00 3.36 3.45 0.00 6.81

E S

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Extensification Extra Straw incorporation (2 t haÿ1) Remainder unchanged Total

3.0 8.3 3.0 28.6 40.4 16.7

0.00 0.00 0.00 3.36 1.91 0.00 5.27

E O

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Extensification Extra sewage sludge Animal manure (17.5 t haÿ1) Total

3.0 8.3 3.0 28.6 5.3 51.8

0.00 0.00 0.00 3.36 0.30 3.61 7.26

E O NT

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Extensification Extra sewage sludge Animal manure (20 t haÿ1) Extra Straw incorporation (10 t haÿ1) Total

3.0 8.3 3.0 28.9 3.5 45.3 8.0

0.00 0.00 0.00 3.36 0.20 3.65 1.2 8.41

1

SeeTable 3 for abbreviation code; 2 The total carbon mitigation potential for extra no -till of which 97.8% is from SOC accumulation and 2.2% is from fossil fuel carbon savings.

experiments and have greater associated uncertainty (e.g. extensification, woodland regeneration, bioenergy production and no -till). Where uncertainty has been estimated for the latter category, the 95% confidence intervals are about 50% of the mean value (Smith et al., 1998a). The estimates in this study take no account of differences in soil type and local climate. To improve these estimates, more explicit account needs to be taken of the impact of soil characteristics and local climate on SOC accumulation rates, and on the suitability of different portions of the UK's agricultural land for different land-management options. Dynamic simulation

models, coupled to high-quality GIS databases are the ideal tools for such spatially- explicit analyses. C O N C LUS I O N S (1) An important resource for carbon mitigation in UK agriculture is the surplus arable land. In order to fully exploit the potential of arable land for carbon mitigation, policies will need to be implemented that allow surplus arable land to be put into alternative long-term land-use, instead of shortterm rotational set-aside.

Table 7. Carbon mitigation potential for the optimum realistic scenario (Opt) in the UK. Scenario1


Opt

1990 Sewage sludge 1990 Straw incorporation 1990 No -till Bioenergy crop C mitigation2 Woodland regeneration3 Extra sewage sludge Animal manure (20 t haÿ1) Extra Straw incorporation (10 t haÿ1) Extra No -till4 Total

1



3.0 8.3 3.0 5.0 5.0 5.3 45.3 8.0 17.1

0.00 0.00 0.00 2.05 1.59 0.30 3.65 1.20 1.60 10.39

See Table 3 for abbreviation code; 2 Includes 1.38 Tg yrÿ1 from bioenergy crop C mitigation and 0.67 Tg yrÿ1 from SOC accumulation under woody crops; Includes 0.92 Tg yrÿ1 from aboveground carbon accumulation and 0.67 Tg yrÿ1 from SOC accumulation under woodland; 4 The total carbon mitigation potential for extra no -till of which 97.8% is from SOC accumulation and 2.2% is from fossil fuel carbon savings.

3

10


(2) No single land-management change or combined landmanagement change can mitigate all of the carbon needed to meet the UK's climate change commitments. However, integrated combinations of land-management options show considerable potential. (3) Bioenergy crops show the greatest potential for carbon mitigation and unlike other options provide an indefinite potential. To fully exploit the bioenergy option, surplus arable land needs to be made available for longer-term land use changes and the infrastructure for bioenergy production needs to be significantly enhanced before the beginning of the first Kyoto commitment period in 2008 (Mangan, 1997). (4) It is not expected that the UKwill attempt to meet its climate change commitments solely through changes in agricultural land-use. A reduction in CO2- carbon emissions will be key to meeting Europe's Kyoto targets, and forestry activities (Kyoto Article 3.3) will play a major role. In this study, however, we demonstrate the considerable potential of changes in agricultural land-use and management (Kyoto Article 3.4) for carbon mitigation and highlight the policies thatwould need to be implemented to promote these agricultural activities. AC K N OW L E D G E M E N TS We are very grateful to Dudley Christian, Margaret Glendining, and David Glen for advice, and for help in locating some of the many values used in this study. This work contributes to the following projects: the UK-Biotechnology and Biological Sciences Research Council Project ``Modelling SOM dynamics using SOMNET'' (Grant: 206/A 06371), and the EU projects Modelling Agroecosystems under Global Environmental Change (MAGEC - ENV4 -CT97-0693) and Euro SOMNET (ENV4 -CT97-0434). IACR receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom. R E FE R E N C E S ARDEN-C LARKE, C. & HODGES, R.D. 1988. The environmental effects of conventional and organic/biological farming systems. II. Soil ecology, soil fertility and nutrient cycles. Biological Agriculture and Horticulture 5, 223 ^ 287. C HRISTIAN, D.G. & BALL, B.C. 1994. Reduced cultivation and direct drilling for cereals in Great Britain. In: Conservation tillage in temperate agroecosystems (ed. M.R. Carter), Lewis Publishers, Boca Raton, Florida, pp. 117 ^ 140. CANNELL, M.G.R.1999. Forests, Kyoto and Climate. Outlook on Agriculture 28, 171 ^177. CANNELL, R.Q.D., DAVIES, D.B., MACKNEY, D. & PIDGEON, J.D. 1978. The suitability of soils for sequential direct drilling of combine-harvested crops in Great Britain: A provisional classification. Outlook on Agriculture 9, 303 ^ 316. CTIC 1997. National survey of conservation tillage practices. Conservation Tillage Information Center,West Lafayette, Indiana. DETR 1998. UK Climate Change Programme ^ Department of the Environment, Transport and the Regions Consultation Paper. Department of the Environment, Transport and the Regions, HMSO, London. EUROSTAT 1995. Agriculture StatisticsYearbook, 1995. Theme 5, Series A. Commission of the European Communities, Luxembourg. FALLOON, P., SMITH, P., SMITH, J.U., SZABO¨, J., COLEMAN, K. & MARSHALL, S. 1998. Regional estimates of carbon sequestration potential: linking the Rothamsted carbon model to GIS databases. Biology and Fertility of Soils 27, 236 ^ 241. FALLOON, P., SMITH, P. et al. 1999. Linking GIS and dynamic SOM models: estimating the regional C sequestration potential of agricultural management options (Abstract). Journal of Agricultural Science 133, 341 ^ 342. FLAIG, H. & MOHR, H. (eds) 1994. Energie aus Biomasse - eine Chance fuel die Landwirtschaft. Springer, Berlin-Heidelberg-New York.

FRYE, W.W. 1984. Energy requirement in no -tillage. In: No-Tillage Agriculture. Principles and Practices (eds. R.E. Phillips & S.H. Phillips), Van Nostrand Reinhold, New York, pp. 127 ^151. HALL, D.O., MYNICK, H.E. & WILLIAMS, R.H.1981. Cooling the greenhouse with bioenergy. Nature. 353, 11 ^ 12. HALL, D.O., SCRASE, J.I. & ROSILLO -CALLE, F. 1997. Biomass energy: The global context now and in the future. Aspects of Applied Biology 49, 1 ^10. HOWARD, P.J.A., LOVELAND, P.J., BRADLEY, R.I., DRY, F.T,. HOWARD, D.M. & HOWARD, D.C.1995.The carbon content of soil and its geographical distribution in Great Britain. Soil Use and Management 11, 9 ^ 15. IPCC 1996. Climate Change 1995. Impacts, Adaptations and mitigation of climate change: Scientific-Technical Analyses. Cambridge University Press, New York. KERN, J.S. & JOHNSON, M.G. 1993. Conservation tillage impacts on national soil and atmospheric carbon levels. Soil Science Society of America Journal 57, 200 ^ 210. KOFMAN, P.D. & SPINELLI, R. 1997. Recommendations for the establishment of Short Rotation Coppice (SRC) based on practical experience of harvesting trials in Denmark and Italy. Aspects of Applied Biology 49, 61 ^70. LAL, R., KIMBLE, J.M., FOLLET, R.F. & COLE, C.V. 1998. The potential of U.S. cropland to sequester carbon and mitigate the greenhouse effect. Ann Arbor Press, Chelsea, Michigan. LAMPKIN, N. 1990. Organic farming. Farming Press, Ipswich. MAFF 1983. Pigs: the outdoor breeding herd. Booklet 2431. HMSO, London. MAFF 1984. Straw use and disposal. Booklet 2419, HMSO, London. MAFF1989.Strawsurvey1988EnglandandWales.MAFFStats 25/89,Guildford. MAFF 1994. Fertilizer recommendations for agricultural and horticultural crops. RB209, 6th edition, HMSO, London. MANGAN, C. 1997. Overview of EU energy crop policy. Aspects of Applied Biology 49, 11 ^ 15. MARLAND, G., ANDRES, R.J. & BODEN, T.A. 1994. Global, regional, and national CO2 emissions. In: Trends '93: A Compendium of Data on Global Change (eds T.A. Boden D.P. Kaiser R.J. Sepanski & F.W. Stoss) ORNL/ CDIAC-65. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, Oak Ridge, Tennessee, pp. 505 ^ 584. M ILNE, R. & BROWN, T.A. 1997. Carbon in the vegetation and soils of Great Britain. Journal of Environmental Management 49, 413 ^433. NABUURS, G.J. & MOHREN, G.M.J. 1993. Carbon fixation through forestation activities. IBN Research Report 93/4, Institute for Forestry and Nature Research (IBN-DLO),Wageningen. NABUURS, G.J., DOLMAN, A.J. et al. 1999. Resolving issues on terrestrial biospheric sinks in the Kyoto Protocol. Dutch National Programme on Global Air Pollution and Climate Change. Report no. 410 200 030 (1999) PAUSTIAN, K., ANDRE¨N, O. et al. 1997. Agricultural soils as a sink to mitigate CO2 emissions. Soil Use and Management 13, 229 ^ 244. POULTON, P.R.1996.The Park Grass Experiment,1856 ^ 1995. In: Evaluation of soil organic matter models using existing, long-term datasets (eds D.S. Powlson P. Smith & J.U. Smith) NATO ASI 138, Springer-Verlag, Heidelberg, pp. 376 ^ 384. PREW, R.D., ASHBY, J.E. et al. 1995. Effects of incorporating or burning straw, and of different cultivation systems, on winter wheat grown on two soil types, 1985-91. Journal of Agricultural Science 124, 173 ^ 194. ROYAL SOCIETY 1999. Nuclear energy ^ the future climate. Summary Report. Royal Society document 11/99 (June 1999), The Royal Society, London. SAMPSON, R.N., WRIGHT, L.N. et al. 1993. Biomass management and energy. Water, Air, and Soil Pollution 70, 139 ^159. SMITH, P., SMITH, J.U. & POWLSON, D.S. 1996a. Moving the British cattle herd. Nature 381, 15. SMITH, P., SMITH, J.U. & POWLSON, D.S. (eds.) 1996b. Soil Organic Matter Network (SOMNET): 1996 Model and Experimental Metadata. GCTE Report 7, GCTE Focus 3 Office,Wallingford. SMITH, P., POWLSON, D.S., SMITH, J.U. & GLENDINING, M.J. 1996c. The GCTE SOMNET. Abstract. A global network and database of soil organic matter models and long-term datasets. Soil Use and Management 12, 104 SMITH, P., POWLSON, D.S. & GLENDINING, M.J. 1996d. Establishing a European soil organic matter network (SOMNET). In: Evaluation of soil organic matter models using existing, long-term datasets (eds D.S. Powlson P. Smith & J.U. Smith) NATO ASI 138, Springer-Verlag, Heidelberg, pp. 81 ^ 98. SMITH, P., POWLSON, D.S., GLENDINING, M.J. & SMITH, J.U.1997a. Potential for carbon sequestration in European soils: preliminary estimates for five scenarios using results from long-term experiments. Global Change Biology 3, 67 ^ 79. SMITH, P., POWLSON, D.S., GLENDINING, M.J. & SMITH, J.U.1997b. Opportunities and limitations for C sequestration in European agricultural soils through changes in management. In: Management of carbon sequestration in soil (eds R. Lal J.M. Kimble R.F. Follett & B.A. Stewart) Advances in Soil Science CRC Press, Boca Raton, Florida, pp. 143 ^ 152.

P. Smith et al. SMITH, P., POWLSON, D.S., GLENDINING, M.J. & SMITH, J.U. 1998a. Preliminary estimates of the potential for carbon mitigation in European soils through no -till farming. Global Change Biology 4, 679 ^685. SMITH, P., POWLSON, D.S., FALLOON, P. & SMITH, J.U. 1998b. Soil organic matter and global environmental change: the contribution of GCTESOMNET to recent research. Proceedings of the 16th World Congress of Soils, Montpellier, France, August 1998 (on CD - presentation no. 708). SMITH, P., POWLSON, D.S., SMITH, J.U., FALLOON, P. & COLEMAN, K. 2000. Meeting Europe's climate change commitments: Quantitative estimates of the potential for carbon mitigation by agriculture. Global Change Biology (in press) SMITH, P.B. & POWLSON, D.S. 2000. Considering manure and carbon sequestion. Science 287, 428 ^ 429.

11

STANIFORTH, A.R. 1982. Straw for fuel, feed and fertilizer. Farming Press, Ipswich. WEBBER , M.D., DUVOORT-VAN ENGERS, L.E. & BERGLUND, S. 1986. Future developments in sludge disposal strategies. In: Factors influencing sludge utilisation practices in Europe (eds R.D. Davis H. Haeni & P. L'Hermite). Elsevier Applied Science Publishers, London, pp. 103 ^116. WILLIAMS, J.H. 1988. Guidelines, Recommendation, Rules and Regulations for Spreading Manures, Slurries and Sludges on Arable and Grassland. Commission of the European Communities, Luxembourg.

Received August 1999, accepted after revision December 1999. # British Society of Soil Science 2000

BOOK REVIEW Urban soilsöApplications and Practices By P. J. Craul, Published by John Wiley & Sons, inc. 1999. 366 pp. »51.95 Hardback (ISBN 0-471-18903-0). How often have we seen over the years in the urban environment, dead or dying trees and shrubs, sick looking vegetation or areas of bare and waterlogged soils. Whilst ever larger amounts of money are being spent on `greening' our urban environment with landscaping projects, how much consideration is necessarily given to the sustainability of these areas? Deterioration can occur relatively quickly because insufficient attention has been paid to the whole of the plant environment, but especially the soils. This very helpful book aims to address this situation and endeavours to show that for successful landscaping projects, it is imperative that the landscape professional understands the interrelationship between the plant, soil and the biophysical environment. The book stresses the need to understand the importance of soil design in the urban situation where the existing soils have been severely disturbed or `new' soils have to be created. The authorstarts byexaminingthe importance and methodology involved in any site assessment to help determine the potential opportunities and problems associated with an area. This leads on to a chapter dealing with the urban environment which is very often not `plant friendly', due to extremes in local climatic conditions. Although temperature, humidity, wind and radiation can all be measured at weather stations, such information is generally not available to the specialist for a specific site. This chapter explains how a biophysical analysis of a site can be undertaken pro viding valuable information that needs to be considered before the design of the soil profile can properly be determined.

The following two chapters deal with the design and construction of sustainable soil profiles and preparing soil specifications. The author provides the reader with an understanding of natural soil properties before embarking on the methodology of designing sustainable soil profiles and pro viding helpful, measurable parameters and characteristics for achieving this. Chapters on drainage and irrigation techniques follow, firstly dealing with the principles and basic concepts before going into detail on the different design and techniques necessary for successful installation. The book continues with chapters on tree planting techniques, erosion and sedimentation control before concluding with a number of case studies. The author says that the book has been written for the landscape architect, horticulturist and urban forester, to give them an insight into the importance of soil for successful landscaping works. Whilst it would certainly be of value to them and expresses the need to become knowledgeable in soil science, a book of this nature requires a fundamental understanding of soils and will therefore be of greater value to the practical soil scientist who is engaged in reclamation works.The book has drawn on the author's considerable consultancy experience in the USA and therefore inevitably has a strong American flavour, but the principles discussed are of value in a much wider context. Whilst it is not necessarily an easy book to read from cover to cover, it is a useful reference book to delve into, for anyone involved in landscaping works in the urban environment.

Nick Duncan