Introduction
The direct emissions of CO2, CH4 and N2O from agriculture account for approximately one-fifth of the annual increase in radiative forcing leading to climate changeReference Houghton, Ding, Griggs, Noguer, van der Linden, Dai, Maskell and Johnson1. This proportion is even higher if indirect emissions due to the energy use by agriculture in fertilizers, machinery, etc. are included in the estimatesReference Olesen, Schelde, Weiske, Weisbjerg, Asman and Djurhuus2. Agriculture accounts for approximately 50% of the anthropogenic CH4 emissionsReference Mosier, Duxbury, Freney, Heinemeyer, Minami and Johnson3. Emissions from farms with a stock of ruminant animals are particularly high. Agriculture also accounts for 80% of the global anthropogenic N2O emissions and about 40% of these originate from manure managementReference Mosier, Kroeze, Nevison, Oenema, Seitzinger and van Cleemput4. In addition, soils are also important sources for N2O emissionsReference Mosier5. Furthermore, volatilization of NH3 from animal wastes can occur after application to soil, during housing and storage, or grazingReference Bussink and Oenema6. The volatilized NH3 is deposited partly locally and partly transported long distances from the source before being depositedReference Asman, Sutton and Schørromg7. Deposited NH3 causes eutrophication of N-limited aquatic and terrestrial ecosystems and contributes to the increased acidification of sensitive ecosystemsReference Schulze, de Vries, Hauhs, Rosen, Rasmussen, Tamm and Nilsson8.
Concern about energy self-sufficiency and global warming in recent decades has stimulated interest in using biomass for energy production. Biogas digesters are able to produce energy from dedicated energy crops like silage maize and cereal grains, as well as from organic residues like stable wastes, crop residues, etc. The emergence of commercial-scale technologies for producing methane from wastes and dedicated energy crops is expected to create important economic opportunities for farmers, while reducing fossil fuel dependenceReference Greene, Celik, Dale, Jackson, Jayawardhana and Jin9, Reference Farrell, Plevin, Turner, Jones, O'Hare and Kammen10, GHG emissionsReference Möller and Stinner11, Reference Möller, Stinner and Leithold12, nitrate leaching riskReference Möller, Stinner and Leithold12 and enhancing crop productivity by using digested effluents (=digestate) as manuresReference Stinner, Möller and Leithold13, Reference Möller, Stinner, Deuker and Leithold14. Biogas digestion encompasses the aim of minimizing emissions of odors, NH3, N2O and CH4 during storage of manure, if the effluents are stored in enclosed tanks after digestionReference Clemens, Trimborn, Weiland and Amon15, Reference Weiske, Vabitsch, Olesen, Schelde, Michel, Friedrich and Kaltschmitt16. Therefore, biogas production by anaerobic digestion is a promising means of achieving multiple environmental benefits such as substitution of fossil fuels, reduction of nitrate leaching and GHG emissions. The use of digested materials on farmland can improve the nutrient use efficiencyReference Möller17.
The implementation and operation of a biogas digester are related to additional inputs and environmental burdens, e.g., for handling the substrates and returning the effluents, as manure, to the fields, for construction and dismantling the digester, etc. The big differences among diverse biogas systems make them complex to study from an environmental point of view; the environmental impact of each system is more or less uniqueReference Börjesson and Berglund18. Furthermore, the total environmental impact of the implementation of a biogas digestion system depends mainly on the reference system replaced, concerning energy consumption, animal waste handling and farming system. There is a lack of broad environmental systems analyses, which include the entire biogas fuel-chain, and the indirect environmental impact of replacing different reference systemsReference Börjesson and Berglund18.
The life-cycle assessment (LCA) methodology is a holistic tool used to assess the environmental impacts of a system or product from ‘cradle to grave’. It was designed to study all environmental impactsReference Consoli, Allen, Boustead, Fava, Franklin, Jensen, de Oude, Parrish, Perriman, Postlethwaite, Quay, Séguin and Vignon19. The present paper relates the studies presented by Stinner et al.Reference Stinner, Möller and Leithold13 and Möller et al.Reference Möller, Stinner, Deuker and Leithold14 in order to aggregate the fossil fuel consumption and the environmental impact of various biogas systems and the replacement of different reference systems common in organic farming. The aims of the present study are: (1) to compare the environmental performance of organic farming dairy systems with the currently prevalent animal housing systems [solid farmyard manure (FYM) versus liquid slurry] as the main reference systems; (2) to analyze the effect of the implementation of a biogas digestion system on the consumption of fossil fuels and production of electrical energy; (3) to quantify the effects of the implementation of a biogas digestion system on the environment; and (4) to compare the obtained net energy yields with other means of obtaining energy by using the farmland area. The chosen environmental impact categories are the greenhouse gas (GHG) balances, the acidification potential, the eutrophication potential and the potential groundwater pollution, which represent the most important impacts connected with agricultural activities.
Materials and Methods
LCA
To analyze the environmental impact of different manuring systems and of biogas digestion in organic farming systems, an LCA was carried out. The LCA method is internationally standardized. According to ISO 14040 ff, LCA is divided into four steps: (1) goal and scope definition; (2) life-cycle inventory (LCI) analysis; (3) life-cycle impact assessment (LCIA); and (d) interpretation20–23. To carry out a full LCA, it is usual to use a number of impact categories ranging from energy consumption to acidification.
System boundaries and delimitations
The present analysis deals with the entire life-cycle flow of the described cropping systems, including crop production, animal husbandry and power generation. In the case of biogas combustion, electricity is generated from the produced gas. The time frame is the milk, crop and electrical energy production over 1 year and follows the results obtained in field experiments in the time period 2002–2005Reference Stinner, Möller and Leithold13, Reference Möller, Stinner, Deuker and Leithold14.
Description of the systems
The present publication is based on studies published by Stinner et al.Reference Stinner, Möller and Leithold13 and Möller et al.Reference Möller, Stinner, Deuker and Leithold14. The reference cropping systems studied were designed according to the most common systems in organic farming practice in Germany, representing best management practices. Five cropping systems (two reference systems and three biogas systems), simultaneously representing five different manuring treatments, were implemented in trial series I, representing a mixed organic farming system with dairy husbandry at 0.8 livestock units ha−1 held loose in cubicles (Table 1). The designed arable crop rotation in trial series I included: year 1–2, clover/grass-ley (2 years); year 3, winter wheat (Triticum aestivum)+cover crops; year 4, potatoes (Solanum tuberosum) on 20% of the field and silage maize (Zea mays) on 80% of the field; year 5, winter rye (Secale cereale)+cover crops; year 6, spring peas (Pisum sativum)+cover crops; year 7, spelt (T. aestivum ssp. spelta)+cover crops; year 8, spring wheat (T. aestivum) with undersown clover/grass-ley. The following five manuring treatments were implemented in trial series I (in parentheses: used abbreviations).
Table 1. Description of cropping systems assessed in the present study.
Abbrev., abbreviation.
FYM
Wastes from stables (urine and feces) with high additions of straw from animal bedding results in FYM. The straw for animal bedding was harvested from all cereal crops. Cover crop biomass was mulched and ploughed in.
Undigested liquid slurry (US)
Liquid cattle slurry. Bedding consisted of a mixture of daily use of 0.6 kg straw and 2 kg of lime (DM content: 94%; CaCO3: 60%; MgCO3: 25%). Straw of winter and spring wheat was removed (harvested) for animal bedding in the stable. The remaining crop residues and the cover crops were mulched and ploughed in.
Digested liquid slurry (DS)
Liquid cattle slurry that was digested. After digestion effluents were stored in closed boxes. The animal bedding and the management of crop residues and cover crops matches the description of the US treatment.
Digestion of liquid slurry and field residues (DS+FR)
Liquid cattle slurry was collected and digested as described in the DS system. Straw not needed for animal bedding (straw of peas, rye and spelt) and the cover crop biomass were also harvested for digestion, in comparison with, e.g., DS, where the residues not needed for animal bedding were left on the field and incorporated directly. The effluents of the digester were reallocated mainly in spring within the same crop rotation as manures for the nonlegume crops.
Digestion of liquid slurry, field residues and external substrates (DS+FER)
As in DS+FR, stable wastes and crop residues were collected, digested and reallocated within the same crop rotation. Off-farm purchased substrates (silage maize) were digested in amounts according to EC guidelines for organic farming, resulting in an additional N input (40 kg N ha−1 yr−1). The effluents of the digester were reallocated mainly in spring within the same crop rotation as manures for the nonlegume crops.
Trial series II represented a crop rotation for a stockless organic farming system, with three manuring treatments implemented (Table 1). The cropping system consisted of a six-field crop rotation, which included six crops: year 1, clover/grass-ley; year 2, potatoes; year 3, winter wheat+cover crops; year 4, spring peas+cover crops; year 5, winter wheat+cover crops; year 6, spring wheat with undersown clover/grass-ley. The implemented treatments were as follows (in parentheses: used abbreviations).
Stockless system without livestock with the common mulching practice (wL)
The plant biomass of the clover/grass-ley, cover crops and the crop residues remained on the field, evenly spread over the surface and ploughed into the soil. No mobile manure was available. The clover/grass-ley was cut with a mulching machine, which chops the plant material into small pieces.
Stockless system without livestock with digestion of field residues (wL-FR)
The clover/grass-ley, cover crops and crop residues (straw of wheat and pea crops) were harvested, removed and digested. The effluents of the digester were reallocated mainly in spring within the same crop rotation as manures for the nonlegume crops.
Stockless system without livestock with digestion of field residues and external substrates (wL-FER)
The clover/grass-ley, cover crops and crop residues were harvested, removed and digested, as described for wL-FR. External substrates (silage maize) were digested in amounts according to EC guidelines for organic farming, resulting in an additional N input (40 kg N ha−1 yr−1). The effluents of the digester were reallocated mainly in spring within the same crop rotation as manures for the nonlegume crops.
The biogas systems with the use of off-farm purchased substrates (DS+FER and wL-FER) were modeled on the basis of silage maize derived from dedicated energy crops. As reference yields (Table 3), the results of the field trials reported by Möller et al.Reference Möller, Stinner, Deuker and Leithold14 for series I and by Stinner et al.Reference Stinner, Möller and Leithold13 for series II were used.
LCI analysis
The inventory analysis consists of the recompilation of data concerning energy consumption, emissions and products resulting from each activity in the production system. Data of each process were collected and allocation, to the farm compartments animal production, plant production and energy production, were performed. The present analysis included the extraction of raw materials (e.g., fossil fuels and fertilizers), the production and transportation of farming resources (e.g., fertilizers, seeds and fuels) and all agricultural operations in the field (e.g., manure application, tillage and harvest) and related emissions as recompiled in Table 2. The related consumption of fossil fuels was assessed according to data in the literature24, 25. All calculations were performed for a reference farm of 200 ha, and refer mainly to German conditions and the state-of-the-art technologies. The use of farm machinery was calculated for a mean field area of 5 ha and fitted according to our own records and supplemented by data available in the literature24, 25. Total production of plant products (grains+tubers), milk and electrical energy are presented in Table 3. It was calculated that mineral nutrient outputs (phosphorus and potassium) by sold products were replaced by an equivalent amount of purchased mineral nutrients, e.g., by spreading rock dusts. Furthermore, the application of 200 kg lime ha−1 yr−1 to keep soil at optimal pH was assumed. For the modeled cropping systems, it was assumed that pesticides were only applied to potatoes (250 g active ingredient ha−1). The environmental impact related to the preparation of the mineral fertilizers, pesticides and for the production of the required seeds and seed tubers was considered according to data reported in literature26. With respect to LCI of electricity, the German mixture was used. A primary energy input of 2.654 MJ per MJ electricity, and specific gaseous emissions of 172 g CO2, 0.348 g CH4, 5.04 mg N2O and 2.78 mg NH3 per MJ electricity, or 0.0502 kg CO2-equiv. kWh−1 electrical energy were assumed26.
Table 2. Inventory of the agricultural operations in the field and frequency of operations.
1 Only in the treatment wL.
2 Only in the treatments wL, FYM, US and DS.
3 Only in the treatments DS+FR, DS+FER, wL-FR and wL-FER.
4 Only in the treatments DS+FR and DS+FER.
Table 3. Total production of plant products (grains+tubers), milk and electric power.
1 Without fodder.
The requirement of electrical energy for dairy husbandry was calculated according to data provided by ClausenReference Clausen27. The requirement for fossil fuels (diesel) was set at 22.5 liters of livestock unit−1 yr−1.28 The environmental impact for production and disposal of the used electrical energy and fossil fuels was estimated according to data available in the literature26. The feeding plan was compiled according to available data29, 30. The milk yield per cow was set at 7000 kg milk yr−1.
It was assumed that digestion and storage of slurry and effluents were carried out in tanks made from reinforced concrete. The digestion period depends on the kind of substrate and its biodegradability: (1) 30 days for digesters with high inputs of liquid animal slurries and low additions of other kind of residues; (2) 35 days for digesters with digestion of slurry and high amounts of other kind of substrates; and (3) 45 days for digesters without slurry as substrate. It was assumed that effluents were stored in covered tanks. The assumed capacity of stores allowed the storage of slurry and effluents for a period of 210 days. Electrical energy production was achieved by the use of a combined block heat and power plant (CHP). The amounts of substrates digested are summarized in Table 4. It was assumed that the storage of stable wastes was performed in open stores, if no digestion was carried out. Further, it was assumed that milk cows spend the entire year in the stable and the adjacent outdoor run.
Table 4. Amounts of biomass for digestion in the biogas plant (t fresh matter yr−1).
DM, dry matter.
The analyses of biogas systems include inputs and emissions from energy input in the entire biogas production chain. For example, a biogas digester has a need for electrical energy, for example, to power pumps, bubbler and control technology. It was assumed that the biogas digester has a requirement for electrical energy equivalent to 10% of the produced electricity. The heat used in the biogas plants was assumed to be produced from biogas in the CHP unit. The transport of the purchased substrates (silage maize) was assumed to be carried out in trucks. The average transport distance was assumed to be 20 km.
The environmental impact of providing and consuming fossil fuels and imported maize silage as substrate for biogas digestion was calculated according to Ecoinvent26. Methane oxidation by soils was estimated according to the data published by Boeckx and van CleemputReference Boeckx and van Cleemput31. The N2O background emissions of soils were calculated according to Neufeldt (Deutsches Biomasse-Forschungs-Zentrum gGmbH, Leipzig, pers. comm. 2005). Furthermore, N2O emissions related to the application of organic manures (stable wastes, effluents of the biogas digester, crop residues and green manures incorporated into the soil) were included in the GHG inventory. Data published by IPCC32, 33 were used. According to these data, 1.00% of the applied manure-N was emitted as N2O after soil application. For series II, the data regarding the differences in clover/grass-ley management published by Möller and StinnerReference Möller and Stinner11 were used. Ammonia emissions related to spreading liquid manures were considered according to data published by Søgaard et al.Reference Søgaard, Sommer, Hutchings, Huijsmans, Bussink and Nicholson34. Emissions after the application of FYMs were calculated according to data by Hutchings et al.Reference Hutchings, Sommer, Andersen and Asman35.
The assessment of methane emissions derived from animal rumen and excrements was carried out according to data by IPCC33, based on the gross energy intake. Emissions of N2O from excrements in the stable were assumed to be negligible. Emissions of ammonia from the stable were considered according to data provided by Poulsen et al.Reference Poulsen, Børsting, Rom and Sommer36 and Olesen et al.Reference Olesen, Weiske, Asman, Weisbjerg, Djufhuus and Schelde37. Calculations of N2O and CH4 emissions from manure storage (liquid and solid manures) were performed according to the data of IPCCReference Poulsen, Børsting, Rom and Sommer36. However, for the inventory of ammonia emissions, data by Olesen et al.Reference Olesen, Weiske, Asman, Weisbjerg, Djufhuus and Schelde37 were used, because no data were recorded in IPCC33.
The environmental impact of construction and dismantling of the biogas digester was calculated according to Ecoinvent26. According to the available dataReference Olesen, Weiske, Asman, Weisbjerg, Djufhuus and Schelde37, it was assumed that 1.8% of the methane produced in the digester was lost to the atmosphere due to leakages. The transformation of biogas to electrical energy in the CHP unit was connected to further emissions, which were estimated according to published dataReference Edelmann, Schleiss, Engeli and Baier38. Savings on emissions from consumption of fossil fuels due to the produced renewable electrical energy were accounted as credits. Only electrical energy was balanced, assuming that there is no use for the heat energy.
The inventory of the impact category groundwater quality of the LCA was assessed by the N farmgate balance, based on a method developed by the German Soil Science Association39. The components of the farmgate balance for N were the N input by biological N2 fixation, by the purchase of seeds and biogas substrates, the N output via sold animal and plant products, and the gaseous N losses in the entire field–stable–field chain. Biological N2 fixation and gaseous N losses were assessed as described by MöllerReference Möller17.
LCIA
In a third step, an LCIA is required in order to evaluate the inventory data. Within the LCIA, the different inputs and outputs are summarized into environmental effects, the so-called impact categories. During this weighing step, each normalized indicator value is multiplied by a weighing factor, which represents the potential of the respective impact category to harm resources, natural ecosystems and human health. The data sources for resource consumption and emissions related to the different sub-systems are presented in Table 5. Calculations of the environmental impact of the use of fossil fuels were also carried out using published data26. For the impact category ‘climate change’, CO2, N2O, CH4 and NH4 emissions are aggregated (in CO2-equivalents) using data published by IPCC32 (for CO2, 1; CH4, 21; NH4, 4.01; N2O, 310; period of time, 100 years). To calculate the acidification potential of the different trace gases emitted by the modeled cropping systems the SO2-equivalent factors (for SO2, 1; NOx, 0.7; NH3, 1.88; HCl, 0.88; HF, 1.6; H2F, 1.88) and for calculating the eutrophication potential the PO4-equivalent factors (for NOx, 0.13; NH3, 0.35) were usedReference Heijungs43.
Table 5. Data sources for resource consumption and emissions related to the different sub-systems.
1 Emissions due to energy consumption and direct emissions in all sub-systems, including direct emissions from stable and farmland.
In order to compare the systems, during the second step in LCIA, a functional unit was introduced, which describes the primary function fulfilled by a product system and enables different systems to be treated as functionally equivalentReference Guinée, Gorrée, Heijungs, Huppes, Kleijn, de Koning, van Oers, Wegener Sleeswijk, Suh, Udo de Haes, de Bruijn, van Duin, Huijbregts, Lindeijer, Roorda, van der Ven and Weidema44. This study relates all resource consumption and emissions to 1 ha farmland. We chose the impact categories energy use, climate change, acidification, eutrophication and groundwater pollution (Table 6).
Table 6. Environmental impact categories and indicators considered in the LCA.
Results
Consumption of fossil fuels
In series I, the main requirements per ha of farmland are related to production of plant biomass (Fig. 1). The digestion of slurry (DS) was related to additional energy consumption due to the resources required for digestion and for construction and removal of the digester. However, the credits for the offsets of fossil fuel emissions by the produced electrical energy led to a balanced result, indicating nearly energy self-sufficiency of the modeled farming system.
Figure 1. Inventory of the impact category primary energy consumption of the LCA (GJ ha−1 yr−1).
The harvesting of organic residues like crop residues and cover crops for digestion, in addition to the stable wastes (DS+FR), increased the total energy inputs in comparison with the reference system US, due to higher energy consumption related to the operations of the engine to produce electrical energy, as well as due to additional transport of substrates and the spreading of the effluents back on the farmland. However, it nearly triples the energy yields of the cropping system (from 8.2 to 21.4 GJ ha−1 yr−1). The total balance of the system indicated that more fossil energy was produced than consumed in such a system. The comparison of the additional energy needs for the biogas system and the amount of energy obtained by digestion indicated that a net energy yield (i.e., gross energy output minus energy input) of approximately 15 GJ ha−1 yr−1 of electrical energy was obtained. Approximately one-half of this energy was needed to meet the total energy demand for production of milk and plant products. The overall energy net surplus was 8.0 GJ ha−1 yr−1 (without use of the heat energy). The use of substrates of external provenance in DS+FER increased the energy yields dramatically, leading to a net surplus of 41.9 GJ ha−1 yr−1.
In series II, the reference system wL showed similar requirements for energy per area as the reference systems in trial series I. The digestion of residues (wL-FR) dramatically increased the total energy requirements per area, due to the additional transport of substrates and the inputs needed for the operation of the digester. The energy yield was very similar to DS+FR (21.6 GJ ha−1 yr−1). Accounting for the additional energy needs for operation of the biogas system and the amount of energy obtained by digestion indicated that a net yield of approximately 12.5 GJ ha−1 yr−1 of electrical energy was obtained. Approximately one-half of this energy was needed for production of plant products. The net energy surplus was 6.3 GJ ha−1 yr−1. The additional digestion of purchased silage maize (wL-FER) led to very similar results as described for DS+FER.
Global warming potential (GWP)
Results showed that in the systems with dairy farming, methane emissions are the most important GHG emitted by the undigested reference system US (approximately 60%), whereas nitrous oxides are the most important GHG in the FYM reference system (approximately 54%) as well as in the stockless reference system wL (71%). Furthermore, the often preferred stable housing system in organic farming systems, the FYM system, showed the highest negative impact on GHG emissions (Table 7). The main reason was the much higher nitrous oxide emissions from manure handling in FYM systems than in liquid manure systems (US). Digestion of stable wastes (DS) significantly reduced the direct GHG emissions from animal husbandry in comparison with the reference system US. The main effect of digestion of stable wastes is the reduction of emissions of methane from 1921 in US to 1159 kg CO2-equivalent emissions ha−1 in DS. Without accounting the credits for the substituted fossil fuels, the reduction was 15% for DS. However, displaced fossil fuel was the largest GHG sink. The inclusion of the credits means a reduction of 32% in the total balance of GHG emissions for DS. The additional digestion of crop residues (DS+FR) also reduced the GHG emissions from animal husbandry. However, digestion increased the GHG emissions from arable land. Two factors influenced the balance: (1) higher emissions due to additional processes for harvest and transportation of the substrates and return of the effluents as manures to the fields; and (2) higher emissions of ammonia after spreading manures. No losses of ammonia-N during digestion, and the higher ammonia concentration of effluents in comparison with undigested slurry stored in uncovered storage tanks, lead to higher emissions after spreading (DS). Furthermore, in DS+FR higher total amounts of N and higher amounts of ammonia were spread on the field as effluents, leading to higher emissions of ammonia.
Table 7. Inventory of the impact category GWP of the LCA.
1 kg CO2-equivalent emissions ha−1 yr−1.
2 Without savings/credits for the production of renewable energy.
The inclusion of credits means a total reduction of GHG emissions of approximately 50% for DS+FR in comparison with US. Additional digestion of silage maize in the treatment DS+FER increased the direct emissions from plant production (higher gaseous N emissions, higher amounts of transported effluents, higher amounts of soil-borne N2O emissions due to higher amounts of manure application) and emissions for operating the biogas digester. Simultaneously, the credits for the substitution of higher amounts of fossil energy are significantly higher. Concerning the transportation of substrates, this operation will have only a limited impact on the various emissions analyzed here, provided the mean transportation distance is 20 km.
In the trial series II, the digestion of the available biomass residues (wL-FR) was related to an increase of GHG emissions per area of farmland (+51%), mainly due to the resources necessary to operate the biogas digester (Table 7). Digestion of organic residues led to a significant decrease of direct emissions from plant production, mainly due to the lower soil-borne N2O emissions after removing the clover/grass-biomass and the cover crop biomassReference Möller and Stinner11. Accounting for the credits for the displaced fossil fuels, the total emissions per area were reduced by 59% in relation to the reference system without digestion (wL). The inputs for construction and dismantling the biogas digester slightly influenced the GHG balance, in both trial series.
Acidification and eutrophication
The comparison of both reference systems FYM and US showed higher emissions with acidification potential for FYM than for US (Fig. 2). This was mainly related to higher N losses during storage of the stable wastes. Digestion of stable wastes prior to returning the effluents as manure to the fields strongly reduced the emissions with acidification potential related to animal husbandry (due to the reduction of emissions from stable wastes). However, the emissions of plant production were strongly increased, due to the fact that more N was applied, and the higher ammonia-N concentration in digested effluents. Both boosted ammoniacal-N emissions after spreading manures. As a consequence, digestion of stable wastes (DS) only slightly influenced the total balance.
Figure 2. Inventory of the impact category acidification potential of the LCA (kg SO2-equiv. emissions ha−1 yr−1).
The additional digestion of biomass residues from arable land (DS+FR) increased the emissions with acidification potential, mainly due to the fact that more manures were applied on soils as effluents of biogas digestion (higher gaseous N losses). Similar results were found for wL-FR, as well as for DS+FER and wL-FER. The influence of the credits for the replaced fossil fuels only slightly influenced the total balance in biogas systems.
Figure 3 reveals the results of the inventory of emissions with terrestrial eutrophication potential. The results showed a very similar picture compared with the acidification potential. This is due to the fact that NH3 and NOx contribute to both environmental effects.
Figure 3. Inventory of the impact category terrestrial eutrophication potential of the LCA (kg PO4-equiv. emissions ha−1 yr−1).
Groundwater pollution and nitrate leaching risk
Potential nitrate leaching assessed according to the farmgate N balance differed in both trial series, mainly depending on whether purchased substrates were introduced or not (Table 8). In trial series I, lowest values were assessed for the treatments FYM and DS+FR. The US and DS systems showed slightly higher values. A large increase was found for DS+FER. In trial series II, very similar values were found as reported for trial series I. Digestion of crop residues (wL-FR) reduced nitrate leaching potential by approximately 10%, whereas the inclusion of purchased substrates largely increased nitrate leaching potential. The assessed relation of nitrate leaching potential and measured soil mineral N contents in autumn at the beginning of the leaching periodReference Möller and Stinner11, did not fully match. For example, via the LCA methodology a higher nitrate leaching potential was found for US and DS in comparison with FYM. However, measured soil mineral N content did not differ between the FYM and the US and DS treatments. Furthermore, the slight increase of soil mineral N content in both systems with purchased substrates (DS+FER and wL-FER) did not match with the assessed large increase of the potential nitrate leaching via the LCA methodology.
Table 8. Inventory of the impact category groundwater pollution (N balance) of the LCA (kg N ha−1).
1 As published by Möller and StinnerReference Möller and Stinner11; values of soil mineral N content only serve as an additional indicator that is not part of the balance sheet.
Discussion
The results concerning the higher total GHG emissions in FYM compared to US (Table 7) matched well to other statements in the literature: Külling et al. calculated that for the daily manure amount per cow, GHG emissions from manure stored for 5–7 weeks were higher in the farmyard manure system (2.4 kg CO2 equivalents) than in the liquid manure (1.5 kg CO2 equivalents) systemReference Külling, Menzi, Sutter, Lischer and Kreuzer45. Substantial emission of nitrous oxide occurred with farmyard manure, which also had the highest methane values and, consequently, by far the highest GWPReference Külling, Menzi, Kröber, Neftel, Sutter, Lischer and Kreuzer46. Slurry stores produce only little N2O because slurry generally contains neither nitrate nor nitriteReference Hüther47. Therefore, changing from FYM to liquid slurry systems has been suggested as a GHG mitigation optionReference Sneath, Chadwick, Phillips and Pain48. However, slurry systems are often criticized in the organic farming community because of concerns about soil fertility and animal welfare. However, newer comparisons of the balances of the total soil organic dry matter (DM) inputs for manuring systems based on the use of FYM in comparison with liquid slurry indicated that slurry-based systems provided soils with higher amounts of organic DM and carbon, if in the FYM system no straw was purchased from surrounding farms to meet the needs of straw for bedding animalsReference Möller17. Large amounts of organic carbon (and N) in FYM systems get lost due to their partial decomposition during storageReference Möller17. Even when the slurry was digested prior to its application to soil, the overall soil carbon and organic DM inputs via manures, stubbles and other biomass residues were only slightly affectedReference Möller17. Concerning animal welfare of cattle, it appears that there are some alternatives to straw available, which may provide floor-comfort equal to that of strawReference Tuyttens49. Another option to improve FYM systems from an environmental point of view is to introduce the farmyard manure after removal from the stable into a biogas digester instead of the usual storage in heaps.
Furthermore, the results of the present study indicated that the implementation of a biogas digestion system has the potential to be a very effective and efficient strategy in combating several of today's environmental problems in agriculture, as also reported by othersReference Amon, Kryvoruchko, Amon and Zechmeister-Boltenstern50, Reference Ghafoori, Flynn and Checkel51. The implementation of biogas systems may lead not only to reduced gaseous emissions when fossil fuels are substituted but also to indirect benefits from changed handling of animal wastes and organic crop residues. The most significant impact of the anaerobic digestion of liquid dairy cattle manure with biogas capture and utilization is the reduction of the emissions of methane from store to the atmosphere. Furthermore, biogas digestion allows the use of a lot of different substrates like clover/grass-ley, crop residues and cover crops, which represent, until now, a large unexploited energy potential. As stated by MöllerReference Möller17 and references therein, organic farming systems are very often N limited, and not carbon limited, and therefore a large potential for use of the circulating organic carbon for energy production is available in such systems. Digestion would provide additional revenue from the available organic DM in residues. Further benefits of digestion of residues are an increase of the amounts of mobile manures which can be allocated freely within the cropping system to the crops with highest N demandReference Stinner, Möller and Leithold13, Reference Möller, Stinner, Deuker and Leithold14. Regarding the effects on soil life, investigations within our own field experiments related to the soil carbon metabolism, e.g., total carbon content, water-extractable carbon, soil microbial biomass carbon and carbon substrate utilization test, did not reveal any significant differences between the performed manuring treatmentsReference Möller17, Reference Schauss, Ratering, Stinner, Deuker, Möller, Schnell, Möller, Leithold, Michel, Schnell, Stinner and Weiske52, in spite of very large differences in soil DM inputs via organic amendments (stable wastes, biogas effluents, crop residues and green manures)Reference Möller17. These results showed that the different organic manuring treatments performed during a period of 4 years did not affect measurably soil N or soil carbon metabolism. It can be concluded that digestion of residues could lead to a win–win situation by reducing GHG emissions, implementing a new product (energy), and simultaneously by increasing food yields due to a better allocation of the available N in space and time.
The net GHG emissions of 3.9 Mg CO2-equiv. ha−1 farmland in the reference system FYM and 3.2 Mg CO2-equiv. ha−1 farmland in US are lower than the 5.6–7.3 Mg CO2-equiv. ha−1 farmland (Table 7) found by othersReference Haas, Wetterich and Kopke53. Olesen et al.Reference Olesen, Schelde, Weiske, Weisbjerg, Asman and Djurhuus2 found emissions of organic dairy farms ranging between 3 and 9 Mg CO2-equiv. ha−1. The lower GHG emissions found in the present study were related to the fact that a mixed farming system with milk and plant production was modeled. Arable cropping systems emitted much less GHG per unit area than mixed farming systems with arable crops and animal husbandry, as shown, for example, by results with the reference system wL in trial series II.
As shown by results presented in Figure 1, digestion of stable wastes and crop residues allowed a net self-sufficiency concerning energy inputs, or even a net energy surplus from using such kind of wastes. Using biomass waste from food production for energy production in a biogas digester eliminates the negative side-effects of open stable waste storage (Table 7) or soil incorporation of large amounts of biomass in autumn prior to the main leaching and denitrification periodReference Möller and Stinner11 and generates useful heat and electrical energy as well. Also other studies highlight the value of bio-energy from waste products, because they are not related to any substantial land use change and its emissionsReference Moriarty and Honnery54, Reference Searchinger, Heimlich, Houghton, Dong, Elobeid, Fabiosa, Tokgoz, Hayes and Yu55. For example, Thyø and WenzelReference Thyø and Wenzel56 concluded that biogas from manure implies by far the highest reduction of GHG emissions per unit of services provided to society. The comparison of the net energy yields from organic residues obtained within the present cropping system to dedicated energy maize silage for biogas digestion showed that the obtained net energy yields from digested wastes from food and animal production correspond to approximately 20–25% of the yields on a per hectare basis obtained with dedicated maize silageReference Koch, Foth, Faulstich, von Haaren, Jädicke, Michaelis and Ott57, Reference Gerin, Vliegen and Jossart58, or even approximately 50% in comparison with agro-fuels from rape seedReference Koch, Foth, Faulstich, von Haaren, Jädicke, Michaelis and Ott57, however, without displacement of production of food crops. Owing to the implementation of crop rotations with a high percentage of fertility-building crops and large amounts of organic residues, organic farming systems mostly have a large surplus of organic carbon for soil humus production, which could be partially used for production of energyReference Möller17. Also biogas digestion mainly metabolized the easily degradable organic carbon compounds of biomass, while the lignocellulosic residues, which are needed for enhancing and sustaining soil qualityReference Reijnders59, Reference Lal60 were not degraded or degraded only to a low extent in a biogas digesterReference Asmus, Linke and Dunkel61.
When biogas systems are designed to use dedicated energy crops (DS-FER and wL-FER), a large increase in net energy yield (i.e., gross energy output minus energy inputs) could be achieved. However, the production of energy from dedicated main crops means competition between food production and the production of energy, which has led to the discussion of potential problems regarding land use competitionReference Moriarty and Honnery54, Reference Searchinger, Heimlich, Houghton, Dong, Elobeid, Fabiosa, Tokgoz, Hayes and Yu55, Reference Azar, Simpson, Toman and Ayres62. A further point of criticism addresses the emissions and energy balance of energy obtained from dedicated energy crops. Some authors reported a low net energy yield and an unsatisfactory long-term sustainability performance of energy produced from dedicated plantationsReference Searchinger, Heimlich, Houghton, Dong, Elobeid, Fabiosa, Tokgoz, Hayes and Yu55, Reference Bastianoni and Marchettini63, Reference Pimentel64. Double crop sequences of annual crops with an intercalated cover crop could also be used to improve feedstock production, while overcoming many of the liabilities associated with the use of dedicated energy crops. By pairing a cool-season cover crop with warm-season crops, dedicated biomass double crop sequences could provide soil cover for most of the year, offering the potential for increased biomass yields and reduced soil and nutrient losses compared with annual crop monoculturesReference Grass and Scheffer65, Reference Anex, Lynd, Laser, Heggenstaller and Liebman66.
The acidification potential of a system, expressed in the present system as kg SO2-equivalents, represents its contribution to the acidification of natural ecosystems like forests or lakes. The effects of biogas digestion of stable wastes on acidification and eutrophication was much less than found for the impact categories GHG and fossil fuel consumption. Furthermore, calculated acidification in SO2-equivalents (Fig. 2) was much lower than the acidification potential of 107 kg SO2 ha−1 found in the literatureReference Haas, Wetterich and Kopke53. Similarly, the computed eutrophication potential (Fig. 3) was much lower than the values calculated for organic farms of 13.5 kg PO4 ha−1 by Haas et al.Reference Haas, Wetterich and Kopke53. The obtained results indicated that acidification was reduced when stable wastes were digested. This reduction was mainly induced by reduced N emissions during storage of stable wastes. Digestion of crop residues increased acidification and eutrophication in both trial series, due to the partial ammonification of organically bound N during digestion, increasing the ammonia-N emissions after field spreading of biogas effluents.
Farmgate balances for N as an indicator for the potential groundwater pollution with nitrate indicated a significant impact of the treatment (Table 8). However, the comparison of data obtained from the farmgate balance approach and the experimentally measured data of soil mineral N content in autumn at the beginning of the leaching period (Table 8) showed that the N surplus approach, as commonly used for LCA studies, is only an indicator for the potential nutrient loss, but not for the actual nutrient loss, as also stated by othersReference Oenema, van Liere and Schoumans67. For example, cover cropping is a singly very effective measure to reduce the risk of groundwater pollutionReference Möller and Stinner11, which is not considered as a factor in LCA studies. The farmgate N balance was only slightly influenced by cover cropping; however, soil mineral N content was strongly influenced. Therefore, farmgate N balances for assessing potential groundwater pollution were only a weak indicator for groundwater pollution.
The construction and demolition of installations in a biogas plant produce hardly any damage to the environment, as also found by other studiesReference Hartmann68.
Calculations performed in the present study used a lot of data available in the literature. In a few cases, the database was too weak to consider it in the calculations. For example, in the literature, there are some indications that ammonia was lost from biomass left on the field surfaceReference Andren69–Reference Whitehead, Lockyer and Raistrick74. Published data ranged between 2.7 and 45.0% of biomass-N. Due to a lack of reliable data, those ammonia-N losses from plant biomass left on the field were not included in present LCA calculations. An inclusion would boost values of the inventory categories GHG, acidification and eutrophication of the reference systems, mainly for the reference system wL in trial series II (due to the high emissions reported for mulched clover/grass leys), but also for the reference systems FYM and US of trial series I, as well as for DS, where cover crops were left on some of the fields on the soil surface until mid–end winter. Another topic concerns the methane emissions from biogas plants. GHG balances were calculated for the biogas systems assuming that the final stores of the digested effluents are hermetically sealed and only 1.8% of the methane produced in the digester permeated the covers and were lost. However, most of the biogas plants in Germany did not have a covered effluents store. Calculations indicated for trial series I that the GHG-reducing effect of stable wastes digestion (DS) would turn negative, if approximately 28% of the produced methane was lost to the atmosphere. In the DS+FR system, emissions of c. 19% of the methane would mean that the GHG balance turned negative; in DS+FER, approximately 11%. In trial series II, with 10% emissions of methane in wL-FR and 9% in wL-FER, the GHG balance of biogas digestion would turn negative. A further uncertainty concerns the emission factor for N2O emissions of soil-applied liquid and solid manures. Present calculations used the IPCC default value of 1.00% of the soil-applied N. Most of the manures were spread to soil in late winter and spring, when soils were moist. More recent results suggest that the N2O emission factor for the organic manures with very moist soils was 2.5-fold higher than the default IPCC emission factorReference Senbayram, Chen, Mühling and Dittert75. A higher emission factor would affect mainly the treatments with high levels of soil applied N, e.g., DS-FER and wL-FER. A further factor of uncertainty is the emission factor for N2O emissions from solid manure heaps. The IPCC default value of 2% of the N amounts introduced into the store32 was based on one study by Külling et al.Reference Külling, Menzi, Sutter, Lischer and Kreuzer45, the uncertainty range was assessed with the factor 2.32 Adoption of another default value would affect mainly the results for the FYM treatment, as the total GHG balance of this system was significantly affected by the total N2O emissions during the storage phase.
Conclusions
The results of the present study indicated that the implementation of biogas systems may lead not only to reduced gaseous emissions when fossil fuels are replaced, but also to indirect benefits from changed handling of animal wastes and organic crop residues. Digestion of stable wastes has the potential to reduce mainly the net emissions of GHG, whereas the other effects per unit area on the other impact categories, like eutrophication and acidification potential, are rather small. Digestion of crop residues, cover crops and dedicated energy crops also lead to a reduction of net GHG emissions; however, it has the potential to increase the risk of emissions with acidification and eutrophication potential due to higher gaseous N emissions after manure spreading. Digestion of residues could lead to a win–win situation by producing a new product (energy), by reducing GHG emissions, and simultaneously by increasing food yields per unit area due to a better allocation of the available N in space and time. By using the residues available in an organic cropping system, approximately 25% of the net energy yields achievable with conventionally cropped maize dedicated for biogas digestion, and approximately 50% of the net energy yields achievable with conventional agro-fuel production by oil crops, could be obtained without replacement of food production.
Acknowledgement
This research was supported by the ‘Deutsche Bundesstiftung Umwelt’.
