The authors have declared that no competing interests exist.
This paper presents a spatially explicit method for making regional estimates of the potential for biogas production from crop residues and manure, accounting for key technical, biochemical, environmental and economic constraints. Methods for making such estimates are important as biofuels from agricultural residues are receiving increasing policy support from the EU and major biogas producers, such as Germany and Italy, in response to concerns over unintended negative environmental and social impacts of conventional biofuels. This analysis comprises a spatially explicit estimate of crop residue and manure production for the EU at 250 m resolution, and a biogas production model accounting for local constraints such as the sustainable removal of residues, transportation of substrates, and the substrates’ biochemical suitability for anaerobic digestion. In our base scenario, the EU biogas production potential from crop residues and manure is about 0.7 EJ/year, nearly double the current EU production of biogas from agricultural substrates, most of which does not come from residues or manure. An extensive sensitivity analysis of the model shows that the potential could easily be 50% higher or lower, depending on the stringency of economic, technical and biochemical constraints. We find that the potential is particularly sensitive to constraints on the substrate mixtures’ carbon-to-nitrogen ratio and dry matter concentration. Hence, the potential to produce biogas from crop residues and manure in the EU depends to large extent on the possibility to overcome the challenges associated with these substrates, either by complementing them with suitable co-substrates (e.g. household waste and energy crops), or through further development of biogas technology (e.g. pretreatment of substrates and recirculation of effluent).
In order to reduce emissions of greenhouse gases and limit its contribution to global climate change, the Renewable Energy Directive (RED) of the European Union (EU) requires member states to source at least 10% of transport fuels from renewable sources by 2020. Achieving this target sustainably, without unintended negative consequences, is a challenge. So far, increased demand for renewable transportation fuels has primarily been met by so-called conventional biofuels, produced from food crops (e.g., wheat ethanol and rapeseed biodiesel). However, the rapid increase in conventional biofuel consumption has caused concerns over the impacts increased biofuel feedstock demand has on agricultural commodity markets, e.g., raising food prices [
The European Commission has responded to these concerns by proposing amendments to the RED [
One example of the latter is biogas produced from municipal and industrial wastes or agricultural residues through anaerobic fermentation and upgraded to vehicle fuel quality biogas. (Biogas produced through anaerobic fermentation commonly consists of 50–75% methane (CH4) and 25–50% carbon dioxide (CO2). To be used as a transportation fuel, or to be injected in the natural gas grid, the biogas needs to be dried, desulfurized and removed of CO2, raising the CH4-content above approximately 96%.)
In 2013 it is estimated that there were nearly 14,000 anaerobic digesters in the EU, producing a total of 560 PJ of biogas [
Given these policy developments, both on EU level and in individual member states, pertinent questions are: (1) How large is the potential for producing biogas from agricultural residues, such as straw and manure, in the EU, (2) how is this potential distributed across member states, and (3) what are the main limiting factors for this potential? These are the questions we set out to answer in this paper.
Several studies have already estimated the availability of agricultural waste for bioenergy in the EU (see [
A number of bottom-up studies, on the other hand, have done detailed analyses of biogas potential from agricultural residues at local to regional scale, accounting for a wide array of technical, economic and environmental constraints (see, e.g., [
Lacking in the current literature, therefore, is an EU-level analysis of the potential for biogas production from crop residues and manure, which accounts for how key technical, economic and environmental constraints affect both the absolute magnitude and spatial distribution of the potential. It is important to analyze how the stringency of these constraints affects the biogas potential since the constraints are bound to change over time, due to changes in technology (e.g., further development of processes for dry fermentation), economic factors (e.g., fuel price changes affecting the economically viable collection radius for substrates), and policies (e.g., changes in subsidies favoring small-scale biogas plants, as is currently happening in Germany and Italy).
Here we present a new method developed to analyze the potential for biogas production from crop residues and manure in the EU. The model provides spatially explicit biogas potentials with a resolution of 250 m, by downscaling agricultural statistics on crop production (akin to the method proposed in [
The rest of this paper is structured as follows. In the Methods section, we first describe how the regional availability of manure and crop residues are estimated and spatially downscaled. The section ends with a description of how technical, biochemical and economic constraints are represented in the biogas production model. In the Results section we present the geographical distributions of substrates and the biogas potential, and describe how the biogas potential changes depending on the stringency of the constraints. We end by discussing the usefulness of our method, the relative importance of different constraints and what conclusions can be drawn for future research on the limitations to biogas production from crop residues and manure.
Before making a detailed description of the method and the data sets used, we briefly summarize the three steps of the analysis (see also
First, available substrate amounts were estimated using subnational statistics. Second, the substrate amounts were spatially downscaled using land cover and livestock population datasets. Third, economic and technical limitations were analyzed using a model of a biogas plant. The local production potentials were upscaled to the EU level to obtain an overall potential estimate.
First, the amounts of manure and crop residues available for biogas production in the EU were estimated, using subnational agricultural statistics and several other data sources which are further discussed below. The result of this was an estimate of the annually available substrates, on the same statistical aggregation level as the agricultural statistics used (Eurostat NUTS2).
Second, the regional substrate amounts were spatially downscaled to a much finer resolution using a land cover map, Corine Land Cover [
Third and last, the biogas potential was estimated by maximizing the use of available substrates in each geographical location, taking into account technical and economic constraints such as the minimum viable plant size, reasonable collection distances for different substrates, and the DM concentration and C:N ratio of substrates. This local potential measure was then upscaled to the EU level to obtain an overall potential estimate.
The management of crop residues is highly variable depending on, e.g., competing uses, weather and climate conditions, soil properties, and crop rotations. Some residues, primarily straw, are directly used elsewhere, e.g. as animal bedding material, in mushroom production, or incinerated for energy, but a large fraction of crop residues is also left in the fields. In general, there is little data available on the produced quantities of crop residues and how they are managed, but there have been some efforts to estimate how much could be used for energy purposes using national statistics [
The production of collectable residues was estimated for wheat, barley, rye, rapeseed and turnip rape, grain maize, sugar beets, and sunflower. The production of residues was obtained by multiplying the crop harvests by residue-to-crop ratios. However, all the produced residues cannot be sustainably removed from the fields if soil organic matter is to be preserved. This constraint is further discussed in a separate section below.
Subnational agricultural production statistics are published by Eurostat according to the NUTS classification (Nomenclature of territorial units for statistics). The 28 member states of EU constitute the NUTS0 level which is further subdivided in about 130 regions on NUTS1 level and about 300 regions on NUTS2 level. For many crops in most countries, harvested amounts are published by Eurostat on NUTS2 level each year. Where such harvest data were missing, we instead estimated harvests on NUTS2 level by multiplying national-level yields by NUTS2 level data on planted areas from the detailed triannual Eurostat farm structure survey. Throughout the calculations, we used averages over the years 2009–2011 where available. See the
The crop harvests were multiplied by residue-to-crop ratios indicating how much residues are produced per unit of harvest (see
Crop | Residue-to-crop ratio | Source |
---|---|---|
Wheat | 0.9 | [ |
Rye | 1.1 | [ |
Barley | 0.7 | [ |
Oats | 0.8 | [ |
Rapeseed and turnip rape | 1.2 | [ |
Maize | 1 | [ |
Sugar beets | 0.6 | [ |
Sunflower | 2 | [ |
There are at least two reasons why only a part of the produced crop residues can be sustainably utilized. First, crop residues are an important source of soil organic matter and therefore not all residues can be removed from fields without creating unacceptable impacts on soil quality. Second, if the weather is very wet after harvest, the quality of the residues may be degraded and there may be severe soil compaction if heavy machinery is used for removal of residues. These are both serious concerns which farmers legitimately may consider before utilizing or selling their crop residues.
The sources reviewed in [
We used the simple approach to use a single maximum residue removal rate throughout Europe for all crop residues. Following [
The most important alternative use for crop residues is straw as bedding for animals. We estimated the amount of straw used for bedding as explained in the section on manure below and then computed the straw balance separately in each NUTS2 region by subtracting the bedding straw from the available resource, based on the assumption that straw for bedding is not moved between regions. In a few regions where the estimated demand for bedding exceeded the amount of available straw we limited the bedding use to equal the available amount.
As previously mentioned, some straw is also used for mushroom cultivation. However, we chose to omit this term since it is typically quite small (less than 1% of the total production [
Manure production was estimated for dairy cows and other cattle, breeding and fattening pigs, broilers and laying hens, divided into different manure management systems. The calculation was based on animal population statistics from Eurostat on the finest level publicly available (NUTS1 for Germany, NUTS2 for all other countries in EU28). The
In principle, it is straightforward to estimate the available manure resources by multiplying each animal population by its average excretion rate and the shares excreted in different manure management systems. However, neither of the latter two parameters are systematically and comparably measured across the EU. Therefore we must rely on incomplete survey data and expert judgement for these parameters.
We chose to use a single excretion rate for each animal class across the EU, see
Excretion values for cattle and pigs from [
Animal | Excretion |
System | Manure incl. bedding |
---|---|---|---|
Dairy cows | 5.1 | liquid | 5.1 |
solid | 10.2 | ||
Other cattle | 2.6 | liquid | 2.6 |
solid | 5.2 | ||
Breeding pigs | 0.5 | liquid | 0.5 |
solid | 1.0 | ||
Fattening pigs | 0.3 | liquid | 0.3 |
solid | 0.6 | ||
Laying hens | 8 ⋅ 10−3 | liquid | 8 ⋅ 10−3 |
solid | 8 ⋅ 10−3 | ||
Broilers | 5 ⋅ 10−3 | liquid | 5 ⋅ 10−3 |
solid | 5 ⋅ 10−3 |
One way to account for the variation between countries would be to use the excretion rates reported in the National Inventory Reports (NIRs) to the UNFCCC. In some cases these numbers may be more accurate, especially where country-specific models incorporating detailed data are used. However, it is hard to judge how the overall accuracy of the calculation would be affected given the large variations in methods used by different countries.
Another method for predicting excretion rates is to use feed digestion models for the different animals (see e.g. [
In summary, adding more detailed estimates of excretion rates is a possible extension of the method presented here. Compared to detailed model systems such as CAPRI, the approach taken here improves model transparency and ease of sensitivity analysis at the expense of potentially reduced accuracy.
Manure management systems were included in the model for two distinct reasons: (1) because some of the excretions fall on pastures and we assumed that this portion cannot be collected, and (2) because manure is often mixed with significant amounts of cereal straw or other bedding materials. This addition of dry materials into solid manure is important to consider because it changes the characteristics of the manure as a biogas substrate by increasing its DM content and C:N ratio, and affects the techniques and costs associated with transportation. Furthermore, straw bedding reduces the amount of straw available elsewhere.
Information sources on manure management practices in the EU are scarce, typically based on expert judgement rather than measurements, and either qualitative or highly uncertain. It is widely recognized that manure management practices differ significantly both between and within countries. Furthermore, comparison of different information sources is complicated since no standardized terminology is agreed upon. Some recent surveys covering all or most of the EU are found in [
In face of this data scarcity, we chose to make a simple quantitative description of manure management systems which can be expected to capture the most important variations across the EU. Three management systems were included and assumed to be identical for our purposes wherever they are used: liquid manure, solid manure, and pasture. Country-specific data on the shares of manure in these different management systems for different livestock classes (dairy cows, other cattle, breeding swine, market swine, hens, and broilers) were taken from the NIRs submitted by EU countries. We used average values over the years 2009–2011 where available. (See the
For liquid manure, we assumed that no bedding material is used. For solid manure, we assumed a single bedding material (straw for cattle and pigs, wood shavings for poultry) added in fixed proportions. In a few NUTS2/NUTS1 regions, this estimated demand for bedding straw exceeded the collectable resource, indicating that we had overestimated the straw use and/or underestimated the supply. In these cases, we assumed that no straw is transported across region boundaries and instead changed shares of straw-based solid manure into liquid manure until the straw use was sufficiently small. Manure management was adjusted only in the NUTS1/NUTS2 regions where needed.
The resulting production of manure-based biogas substrates per animal head in different manure management systems is listed in
Dry matter (DM) expressed as fraction of total weight. Volatile solids (VS) expressed as fraction of DM, and carbon (C) and nitrogen (N) expressed as fractions of VS. For conversion of methane volume to higher heating value (HHV) the factor 40 MJ m−3 was used. Compositions are based on data reported in [
DM |
VS/DM |
C/VS |
N/VS |
Methane yield |
||
---|---|---|---|---|---|---|
Cattle manure | liquid | 8 | 80 | 55 | 7 | 200 |
solid | 20 | 85 | ′′ | 3.5 | ′′ | |
Pig manure | liquid | 6 | 80 | ′′ | 10 | ′′ |
solid | 20 | 85 | ′′ | 5 | ′′ | |
Chicken manure | liquid | 30 | 70 | ′′ | 9 | 250 |
solid | 70 | 70 | ′′ | 9 | 250 | |
Crop residues | straw | 85 | 90 | ′′ | 0.5 | 200 |
maize | ′′ | ′′ | ′′ | ′′ | ′′ | |
sunflower | ′′ | ′′ | ′′ | ′′ | ′′ | |
sugar beet | 13 | ′′ | ′′ | 2.5 | 300 |
The following two sections describe our method for spatially downscaling the substrate amounts; in other words, how the subnational regional estimates were combined with geospatial datasets to produce spatially explicit density maps of all the substrates in resolution fine enough to model the production potential of biogas plants.
To estimate the spatial distribution of crop residue production we used the Corine Land Cover 2006 (CLC2006) dataset, version 17 [
The spatial distribution of crop production was estimated assuming that the above-mentioned crops are allocated to five land cover classes, weighted by the share of each land cover class likely to be cropland (see [
The spatial downscaling of manure production was done in a similar manner as for crop production, but using the FAO Gridded Livestock of the World dataset, version 2.01 (GLW2) [
The estimated manure amounts in each region were distributed proportionally to the corresponding animal populations in GLW2. Note that the GLW2 animal classes (e.g. cattle and chickens) are more aggregated than the excretion categories (e.g. dairy cows and other cattle; broilers and laying hens). In other words, the difference in spatial distribution, e.g. between dairy cows and other cattle, is only accounted for down to the resolution of Eurostat animal statistics. At finer resolution, the GLW2 dataset used does not distinguish between dairy cows and other cattle, between broilers and laying hens, or between breeding and fattening pigs.
The steps described above produced an estimate of how much manure and crop residues could theoretically be used as substrates in biogas production at a spatial scale of 250 m (by interpolating the GLW2 dataset to the same resolution as CLC2006). The next step was to estimate the potential for producing biogas from these substrates when placing a hypothetical biogas plant somewhere in the EU. In this estimate, we accounted for the following constraints:
Maximum transportation distances for substrates
Minimum viable plant size
Minimum and maximum DM content
Minimum and maximum C:N ratio
The methane yield of different substrates
The following sections provide some background on these constraints before describing how they were modeled.
Larger biogas plant sizes are typically more cost-effective given a sufficient substrate supply. When the biogas is used to produce electricity, calculations indicate considerable returns to scale, perhaps a 30% cost reduction going from 150 to 1000 kW methane production (higher heating value, HHV) [
In the base scenario, we chose a minimum plant size of 1 MW HHV, which is a typical size for power production. In alternative scenarios, we also present results for both higher and lower limits on the minimum production unit size. Requiring larger plant sizes is relevant especially if the gas is to be upgraded to vehicle fuel quality,. For biogas upgrading, the typical production scale is almost an order of magnitude larger than for power production: 1 MW HHV is a relatively small upgrading unit and many commercial units handle 5 MW HHV or more [
However, larger plant sizes require longer transportation. In areas where substrate density is low, the collection of substrates and spreading of digestate may require uneconomically long transportation distances. An absolute upper limit for transport should be when the net energy balance of the operations turns negative, i.e. when more energy is used than the produced biogas output. For manure and straw this energetic break-even transportation distance can be some 200 km [
Of course, several other conditions apart from collection distances and production scale affect the economics of biogas production. The profitability is, for example, influenced by how the gas is utilized (power or biofuel), what the selling price is, whether there is economic support (e.g. feed-in tariffs), and what the cost of capital is. However, it is a daunting task to explicitly and consistently analyze all these parameters across the whole EU since they vary both over time and among member states.
As a first approximation, we therefore chose to only require a maximum collection radius and minimum plant size as described above. Since there are already thousands of commercial production plants operating under constraints similar to those assumed here, we presumed that more plants could be established in other locations, given costs, support systems, and energy prices not too unlike the present.
Biogas processes can be divided into two main categories: wet and dry fermentation. The DM concentrations in reactors are typically around 10% for wet fermentation and in the range 15–35% for dry fermentation [
We chose to model wet digestion in the base scenario since this is the most commonly used technology and considerably more mature than dry fermentation. Still, it is worth to mention two technical options for lowering the DM concentration: (1) to dilute the substrates with fresh water and (2) to recirculate liquids from the reactor effluents. Neither of these methods are necessarily simple fixes. Dilution of very dry substrates may require very substantial amounts of fresh water. Recirculation can help lower the DM concentration and increase the methane yields, but it also causes accumulation of both organic and inorganic substances in the reactor which may inhibit the digestion process [
There is also an economic aspect to the DM concentration, or rather the C concentration. In biogas plants using very dilute substrates such as pig slurry, co-substrates increasing the average C content can improve economic viability by increasing production per unit reactor volume [
In the base scenario, maximum and minimum DM concentrations of 12% and 0% were assumed. For the reasons outlined above, both parameters were tested with a wide range of higher values in alternative scenarios.
Anaerobic digestion requires a balance between C and N content. The C:N ratio is an often-mentioned parameter because ammonia inhibition [
Recent experiments on mixtures of dairy manure, chicken manure and rice straw [
In the base scenario, we chose a relatively wide range for possible C:N ratios for substrates, requiring it to be in the range 10–35.
The methane yield of substrates depends on their physical and biochemical properties and on digestion process parameters such as temperature and retention time. The physical and biochemical properties of substrates can also be changed through pretreatment, for example steam explosion, grinding or extrusion, or biological pretreatment. Pretreatment is particularly useful for some lignocellulosic substrates such as straw, which may otherwise be hard to digest [
After estimating the regional substrate amounts and spatially downscaling them, two more steps were taken to assess the limitations to biogas production from manure and crop residues. First, a stylized model of a biogas plant was introduced to derive a location-specific measure of the biogas potential, given the constraints discussed in the previous sections. Second, a weighted average of this measure over the whole EU was used as an estimate of the overall potential and its limitations.
Consider a biogas plant placed in a point
As explained above, some of the available substrate quantities
For simplicity, we assumed that each substrate has a methane yield which is not affected by co-substrates, or in other words that there are no synergy effects of substrate mixtures. With this assumption, the biogas production is a linear function of the used substrate quantities, allowing us to formulate the production maximization as a linear optimization problem as follows:
Here, the available substrate quantities at the point
There are two possible outcomes of this formulation: Either (1) the problem can be solved and the substrate flows
As a benchmark for the production potential, we used the theoretical biogas production obtained if all the available substrates could be digested, i.e.
Assumed values for the constraint parameters are listed in
Alternative scenarios were defined by varying two of the constraint parameters at a time, while keeping the others equal to the base scenario values.
Parameter | Symbol | Min | Max |
---|---|---|---|
DM concentration | 0 | 12% | |
C:N ratio | CN | 10 | 35 |
Plant size | 1 MW (HHV) | – | |
Collection radius | – | 15 km | |
Residue removal rate | – | 40% |
Finally, the overall potential
This quantity was approximated by sampling each point in a 20 km grid over the whole analysis area (yielding
The overall relative potential
More importantly, the error in the measure
The estimated total available crop residues and manure in the base scenario, after accounting for maximal removal rate and other uses of crop residues, amounted to almost 200 million metric dry tonnes (Tg DM) per year, as shown in
Black bars: Estimated available substrate quantities in EU28, excluding Greece, after accounting for the maximum removal rate from fields (40%, see
However, the substrate composition varies widely across the EU, as seen in
The panels show estimated available amounts of (a) crop residues, (b) cattle manure, (c) pig manure, (d) chicken manure, (e) sum of liquid manure, and (f) sum of solid manure. The estimates account for the maximum sustainable removal of crop residues, and exclude manure that falls on pastures. Straw used for bedding is excluded from the crop residues density and included in the solid manure densities. Striped areas are not analyzed.
(a): DM concentration in the available substrates. (b): C:N ratio of the available substrates. Both maps concern the base scenario (see parameters in
The technical and economic constraints previously described lead to very substantial limitations on the biogas production potential from the substrate resources identified above. In the base scenario, we found that about three quarters of the manure substrates and one fifth of the crop residues could be used. The utilized share of each substrate in the base scenario is shown in
The fraction of substrates utilized in the simulations varied considerably between countries since they have different composition and distribution of substrates. For example, regions with large amounts of crop residues tend to have very dry substrate mixtures and can therefore not utilize much of them if constrained to using wet digestion technology and assuming there are no additional, wetter co-substrates such as household waste or energy crops. In some regions (e.g. Denmark, the Netherlands, Belgium, and parts of Germany, France and Italy) it seems possible to utilize almost all substrates, while in others (e.g. Poland, Hungary, Romania, Finland, parts of UK and France) only a small fraction could be used. The maps showing substrate densities (
To explore how the biogas potential depends on the economic and technical constraints discussed above, we formulated alternative scenarios by varying different combinations of two parameters at a time, in rather wide ranges of values. We express the results of these scenarios as the total biogas potential relative to the biogas potential in the base scenario, thereby focusing attention on the relative response to changes in parameters rather than on absolute numbers. Here, we present results from a selection of the alternative scenarios which exhibited large differences compared to the base scenario. Interpretations of these results are further discussed in the Discussion section below. Figures illustrating the full range of analyzed alternative scenarios are included in the
We first turn to the spatial density of substrates.
The potential is expressed as a fraction of the base scenario potential. Each cell in each grid represents a different scenario, with all parameters as in the base scenario except for the two named parameters.
We noted above that only around one quarter of the crop residues and three quarters of the manure are utilized in the base scenario. The single most important constraint explaining this is the maximum DM content of substrate mixtures. As seen in
Relaxing the maximum DM constraint also affects the sensitivity of the estimated potential to other model constraints. For example,
Another notable limitation is the minimum C:N ratio of the substrate mixture.
Panel (a) shows the base case with production 0.7 EJ/year (see parameters in
Given these results it is not surprising, as shown in
In this paper, we present a new method for assessing the potential for biogas production from manure and crop residues. To our knowledge, it is the first spatially explicit biogas potential model applied to the whole EU (except Greece, where land cover data is missing). The detail level of this model is somewhere in the middle between the least detailed studies, reporting total substrate amounts in a country or region, and the most detailed studies which, in addition to many different substrate sources, explicitly account for road networks, gas grids, investment costs, etc.
The model can be used to study the effects on the biogas potential of various technical and economic constraints. For example, we have shown that the biogas potential can be very different depending on the stringency of the constraints on minimum C:N ratio and maximum DM concentration of the substrate mixtures. Furthermore, we have shown that these two constraints are interrelated: Allowing a higher DM concentration always increases the biogas potential, but the relative effect is particularly large under stricter constraints on the minimum C:N ratio. This proves the general point that the biogas potential depends nonlinearly on the constraint parameters, or in other words, that the effect of changing two parameters simultaneously can be very different from the sum of effects from changing the parameters one at a time.
There is a considerable gap between the current biogas production from crop residues and manure and the potential estimated in this paper. A detailed analysis of the causes for this gap would be valuable if the goal is to understand how more of these substrates could be utilized. This question can partly be resolved both by complementing and comparing our work with other methods and by extending the model presented in this paper.
There are many potential limitations to the biogas production from crop residues and manure which we have not been able to cover in this paper, some of which may be hard to incorporate into our model. The profitability of biogas production depends, for example, on the cost of capital, energy prices, available subsidies and other support, the opportunity cost of the investment, and on the availability of equipment and expertise for digesting substrate mixtures rich in crop residues and manure. Furthermore, the inherent uncertainty in these factors is an economic risk that should be considered by potential investors. In principle, it is possible to include these factors in our model by making a more explicit economic model, but doing so would require significant effort since all the above-mentioned factors vary widely among EU member states.
Other issues, however, do seem more straightforward to address by refining our method. We believe that it would be useful to improve data quality and the model’s detail level where there is large uncertainty in data sources, where our results show a strong sensitivity to parameter changes, and where we know that important omissions have been made. Specifically, we identify four areas where better data or model extensions would be valuable.
First, it would make sense to add other substrates to the model, for example dedicated energy crops and suitable industrial or municipal waste. Doing so would have multiple effects on the model results since additional substrate streams would reduce overall transportation needs, increase the feasible production scale, and not least, change the chemical composition of available substrates (e.g., the overall DM content or C:N ratio). In this way, adding more substrates to the model may lead to increased utilization of the substrates already included, most notably the dry crop residues.
Second, since the biogas potential is rather sensitive to the maximum DM constraint, it could be useful to refine the model representation of DM constraints, perhaps by explicitly modeling both wet and dry digestion technology. Such a process model could include options for dilution or recirculation in wet digestion, pretreatment of lignocellulosic substrates, and development of better digester microbes.
Third, since results are also sensitive to the minimum C:N ratio constraint, we believe that a more detailed representation of the complex biochemical and technical issues related to chemical composition can lead to more accurate potential estimates.
Fourth, data on manure excretion and management systems in EU are scarce, incomplete and inconsistent. Since the utilization of crop residues is also limited by access to wet manure substrates, estimates could likely be changed substantially with improved manure data.
We have presented a new approach to estimation of the EU-wide potential for sustainable biogas production from crop residues, manure, and, given relevant data, also from other substrates. Our results indicate that the main limitation to biogas production from crop residues is neither the actual availability of residues, nor their maximum sustainable removal from cropland or their transportation to biogas plants. Hence, refinements to these parts of our model would likely be of limited value. Rather, to make more accurate estimates of the potential with this model, there is a need for more data and analysis on manure excretion and management, the spatial distribution and costs for potential co-substrates, and the technical and biochemical constraints to digestion of substrate mixtures rich in crop residues.
This PDF file contains some details on data sources, e.g. the exact Eurostat tables and National Inventory Reports to the UNFCCC that were used, and how we mapped different statistical nomenclatures to each other. It also contains results from 21 two-parameter sweeps, most of which are not included in the main text. The file also contains a link to a web site (
(PDF)
We are grateful for valuable input to the development of the model presented in this paper from Christel Cederberg and Göran Berndes, Chalmers, Mikael Lantz, Lund University, and Johan Torén, SP Technical Research Institute of Sweden, as well as from the reference group to the f3 project “Biogas from agricultural wastes and residues”. Financial support from the Swedish Knowledge Centre for Renewable Transportation Fuels (f3), the Swedish Energy Agency, the Norden Top-level Research Initiative subprogramme “Effect Studies and Adaptation to Climate Change” through the Nordic Centre of Excellence for Strategic Adaptation Research (NORD-STAR), and Göteborgs Handelskompanis deposition is gratefully acknowledged. We thank two anonymous reviewers for constructive comments.