A Location Model for the Agro-Biomethane Plants in Supporting the REPowerEU Energy Policy Program

Abstract

Biomethane represents one of the solutions towards the European Union (EU) energy transition, being capable to decarbonize the EU’s energy system and to reduce the dependence on imported natural gas, as underlined by the “REPowerEU” energy policy program. As its production is expected to expand primarily from biogenic wastes and residues, such as agricultural residues and animal effluents, it is necessary to make its deployment cost-efficient, taking into account factors such as local resources, existing infrastructure, and raw material and investment costs. From this perspective, this paper proposes a replicable predictive model for locating agro-biomethane plants according to raw material potential, relative economic factors, and territorial characteristics. To this end, an analysis was conducted in the Geographic Information System environment, based on location theory. The analysis included testing the minimum transport cost of feedstock in a case study of a rural area in Southern Italy. Three optimal locations for 2 MW size plants were selected where some key element conditions had been identified. The research findings may provide useful information for the EU policymakers in defining more specific energy planning strategies, in accordance with the REPowerEU objectives, addressing the increase in biomethane production by 2030.

1. Introduction

The European Union (EU) launched the Green Deal strategy in 2019, designed to make Europe climate-neutral by 2050 through the deep decarbonization of all economic sectors. Among the different core policies, the integration of an energy system largely based on renewable energy sources (RES) has assumed particular importance [1,2]. Along with solar and wind energy, also biomethane (BioCH₄), obtained from the upgrading of biogas derived from anaerobic digestion technology that breaks down organic matter by microbial action, can play a strategic role in this sense. Additionally, in an era of energy and geopolitical crisis, due to the Russia–Ukraine conflict, which has brought serious consequences for European economies, biomethane could be a sustainable option to replace natural gas imports from Russia. To this end, in March 2022, the European Commission (EC) issued the “REPowerEU” energy policy program that established a target of 35 Gm³ of annual biomethane production by 2030, investing 37 billion euros. By 2050, it could provide up to 167 Gm³ and cover 62% of the gas demand [3]. Currently, the most updated data indicates 15 Gm³ of biogas and 3.5 Gm³ of biomethane were produced in the EU-27, with the latter being the largest in the world [3,4]. There are 1322 biomethane facilities (April 2023), recording a significant growth in the last decade [5].

Among them, agro-biomethane plants, i.e., those fed by agro-food residual feedstock, assume particular importance, being able to foster a sustainable green transition, contribute to the energy security and independence of a territory, and promote the circular economy approach by reducing waste production, as REPowerEU strongly encourages. Indeed, it is estimated that the expansion will primarily come from biogenic wastes and residue as, member states will be required to collect their organic waste separately by 2024 [6]. There are other benefits linked to agro-biomethane, such as the use of local feedstock resources, the ease of integration into existing infrastructure, and its by-product, digestate, which can be used to produce fertiliser in substitution for synthetic fertilisers. The latter currently require great quantities of natural gas to be produced and are hence insufficient and costly due to the Russia–Ukraine conflict [7].

In this context, Italy has great potential in terms of agro-biomethane production, but it is poorly exploited compared to other RES such as wind and solar. Italy could become the first European country in the production of this green gas, above all in the rural areas. Indeed, the National Recovery and Resilience Plan—so called NEXT Generation ITA—has provided some measures that most directly affect agriculture, agro-food, and rural areas, such as Mission 2 (green revolution and ecological transition) [8]. In line with the Green Deal, these actions deepen some issues already addressed in the recent changes to the Common Agricultural Policy (CAP), focusing on ecological conversion and climate change, as well as on socio-economic development and digitalization in rural areas [9,10]. In this framework, significant investments are planned for the development of agro-biomethane (1.92 billion euros) to be implemented by maximizing the use of organic residues and the conversion of biogas plants [8]. Currently in Italy, there are over 2000 biogas plants and around 85 biomethane production plants, mainly located in the northern regions [11].

However, to produce agro-biomethane, it is important to design proper strategic planning that considers the rural areas’ territorial characters and the engagement of local communities, including economic actors, to share and approve these types of investments and/or innovative projects [12]. In this context, Cavicchi et al. [13] conducted a comparative analysis between two case studies in rural areas of a central region in Italy and Norway. This analysis was conducted for a better understanding of the impacts of different applied policy regimes and resulting policy implications on general bioenergy innovation. They revealed the need for a greater territorialisation of energy planning and policy, enabling municipalities and decentring policy instruments, to foster the emergence of bioenergy innovation systems. Consequently, to establish proper energy planning and policy of agro-biomethane development, it is very important to evaluate the plants’ location, considering the natural and anthropic characteristics of the areas where they are supposed to operate. This is also recognized by the REPowerEU energy policy program, which has stressed the importance of “close research, development and innovation gaps” in the biomethane supply chain for “addressing the main obstacles to increasing the production and use of sustainable biomethane and to facilitate its integration into the EU internal gas market” [6]. This assumes a greater value and role for rural areas, considering their economic and social challenges.

To the best of the author’s knowledge, only a couple of studies have been published about the agro-biomethane plants’ location model, as discussed in detail in the Section 3, but they present some limits as they do not consider the different feedstock typologies for biomethane production, as well as its integration with other RES to design an energy diffuse system. In light of the above, this paper aims to develop a replicable location-predictive model capable of providing an initial contribution to the most sustainable choice for agro-biomethane plant locations in rural areas. This model is intended to serve as one of the tools fostering more appropriate energy planning in accordance with the most recent policy program “REPowerEU”. Specifically, starting from the most recent studies in the field of location theory, through the use of spatial analysis methods and techniques developed in the GIS environment, the main characteristics of the territorial context (specifically a case study in Southern Italy) and the various constraints (mainly remoteness and territorial planning) will be considered. Moreover, the analysis of the minimum transport cost will be conducted. The location model has been tested in a case study rural area in Southern Italy. In fact, the most recent applications, even considering complex spatial models, are scarce in the academic literature, and they often oversimplify reality, ignoring the specific territorial context factors (see Section 2.1), and do not consider an additional location-predictive model. Thus, the innovation in this research, in addition to using an interdisciplinary approach to assess the territorial complexity under the geolocalization framework, lies in filling the following gaps: (i) proposing a new and more complete location model; (ii) applying a territorialisation approach to EU energy planning and policy; (iii) focusing on agro-biomethane production for its integration into EU bioenergy policy. The findings could be used as an important tool for EU policymakers in defining specific planning strategies in the REPowerEU framework, aimed at increasing agro-biomethane production, especially in rural areas.

2. Materials and Methods

The methodology includes four major stepwise procedures, presented in Figure 1. The first three phases and sub-phases are described in the following sections while phase four is the result of the application of the previous ones.

2.1. Description of the Case Study Area

The case study area chosen for testing the location model is the province of Foggia (NUTS level 3), one of the six provinces of the Apulia region (NUTS level 2) in Southern Italy. It has a considerable territorial extension of 6965 km² (ranking as the second-largest province in Italy) and includes 61 municipalities (Figure 2). The region is predominantly rural, in particular the province of Foggia, with the exception of the major centres, and it is characterized by peripherality, ageing, and depopulation phenomena.

The province of Foggia has three sub-regions, each defined by a distinct characterization: the Gargano promontory, which rises to the east along the Adriatic Sea; the Daunian Sub-Apennines on the western side, recognized for its internal foothills and mountains; and, in the central part, the Tavoliere plain, with a uniform and flat morphology. The particularly favourable strategic position of the province of Foggia has led to its inclusion in important communication routes, both via road (such as the A14 Bologna–Taranto and A16 Napoli–Canosa highways) and via railway. Its economy, as stressed before, is essentially based on agriculture (35% of the total regional utilized agricultural area (UAA)—equal to almost 490,500 ha; the first national area of production of durum wheat accompanied by a wide range of vegetables (tomatoes, artichokes, spinach, etc.), which were 63% and 50% of the total regional UAA, respectively, in 2020) [14] and the transformation activities connected to it. As a result, this area accounts for 70% of the total regional nitrate-vulnerable zones, areas designated as being at risk for agricultural nitrate pollution. Thus, in these areas, it is necessary to reduce nitrate loss from agricultural land and to protect human health and water resources from pollution by adopting good farming practices [15,16].

Due to its position and its geomorphological characteristics and agricultural characteristics, the province of Foggia represents a site of important renewable energy power plants despite being poor in energy resources of fossil origin. On a national scale, the territory is a significant case study in terms of local potential (above all wind and photovoltaic energy, while biomass is still not totally exploited), territorial planning strategies and implemented interventions, involving several aspects: social, economic, environmental, and normative. In fact, the last few decades have seen a flourishing of incentives and regulatory concessions for renewable energies that have brought about the rapid development of a large number of wind and photovoltaic plants [12]. Figure 3 shows the electricity produced from different sources of renewable energy in 2021 (the most recent data at NUTS 2 and 3 levels): 83% of Apulian wind energy comes from the province of Foggia, which also represents more than 22% of the Italian production. Photovoltaic (PV) energy accounts for 21.4% of the regional production, with the region being the greatest producer in Italy, contributing 15.5% to the national total. Finally, Apulia’s bioenergy contribution to national electricity production is equal to 7.6%, of which 24% comes from the province of Foggia. Also noteworthy is the number of agro-biogas plants in this area, totalling nine, the highest in the region.

Nevertheless, the fast growth of the renewable energy sector has aroused, in several cases, the opposition of the local communities, which fear the impact on human health, the pollution of the environment, and the worsening of the quality of life. For instance, in this area, citizens raised objections against two large-sized biomass plants designed to produce heat and electric power. This has often considerably changed the previous scenarios and landscape regulatory systems, with a significant rethinking of regional strategies and environmental, landscape, and socio-economic effects [18].

2.2. Data Collection

The data collection activity was structured in three phases. The first was based on the information entered in the single applications for access to CAP contributions, provided by the Agricultural Assistance Centres of the CIA (Confederazione Italiana Agricoltura) and CONFAGRICOLTURA associations. These data concerned the farms in the province of Foggia (NUTS 3) that applied for assistance in the 2020 agricultural year. The total number of collected observations was 49,909. The database was structured from the raw data the aforementioned agricultural associations provided, including all the individual farms’ crop plans. These crop plans contained information relating to the year, the surfaces, the characteristics of the soils, and their location. The structural data of the individual farms were then indicated. Based on the raw database, some information was extracted. For each single production cadastral unit (parcel), identified within a single municipality, the cultivation practices (distinguished by type and crops), the surface area (in hectares—ha), and the characterization (organic or not organic) were indicated.

These data represented 40% of the total utilized agricultural area in the province of Foggia and were evenly distributed over the total area examined. The data was obtained in a perfectly random way because it was based on the voluntary and unsolicited behaviour of the farmers in the phase of entering the information for submitting a single CAP application. Based on the collected data set, the cadastral data were then requested, and the specific cadastral unit was identified for each single parcel. Subsequently, the individual data relating to the specific parcel were georeferenced by associating each set of data to specific geographic coordinates. For each crop, the average yields were then calculated based on the parameters the EC used for calculating the area premiums (measures 11.2 and 11.1) of the national Rural Development Programs [10].

The second phase concerned livestock farms in the province of Foggia, derived from the national data set of the zootechnical registry; these data represent the number of livestock present in one municipality in the province of Foggia. Additionally, in this case, data from the literature were used to calculate the total amount of sewage (liquid animal effluent) produced for each municipality and typology of livestock [19].

The last phase was detecting agro-food by-products. For this activity, the ORBIS data set of the Bureau Van Dick was used for identifying the processing firms: wineries, oil mills, and fruit and vegetable processing companies, for a total of 299 production units [20]. A short questionnaire was distributed among them, containing the following topics: quantity of waste and by-products produced, current use of waste and by-products, and current sale price (where present) of waste and by-products. Based on the collected data, the types and quantities of waste and by-products were then identified and divided by municipality.

The acquired data are interesting both for the way they were obtained and for their capacity to facilitate spatial and temporal analysis, providing valuable insights. In fact, compared to other research conducted so far, the data come from voluntary acquisition methods and, above all, unlike those commonly used in Italian research, they allow for updated insights and at different regional scales. In fact, in most of the Italian research studies the data used mainly come from census sources that are available on a municipal scale and updated at long intervals of time; however, in our case, the data are available on a parcel scale and can be periodically updated. This, thanks to geolocation operations, makes it possible to carry out surveys at different territorial scales. Above all, it represents an interesting method compared to often out-dated official sources and in territorial scales that are not always able to facilitate the necessary insights, especially regarding the choice of optimal localizations compatible with local characteristics.

2.3. Location Analysis Using GIS

The collected data were essential to proceed with the analysis in the context of the location theory. Location theory has been the subject of numerous studies [21,22,23,24]. The theoretical framework includes major themes such as land use, industrial production, central locations, and spatial competition [24,25]. For a long time, according to the neoclassical tradition, the study has been largely descriptive; however, the predictive and prescriptive aspect of location modelling is missing, highlighting the issue’s complexity [24], particularly in the energy sector. In this sector, complex ecological, geopolitical, and geoeconomic relationships intervene at different territorial scales [26,27], making spatial choices even more complex, especially at an operational level. In general, due to the increase of new information, technologies, growth, and new developments in methodological issues raised by classical location models [22] and various spatial models [24,28,29], recent applications are scarce and often oversimplify reality, ignoring the specific factors of the territorial context. However, as is known, multiple factors are involved in the location choice [22,23,24,29,30,31], and the various location factors lie in the way in which they intervene in the process, although they are under-investigated, especially at an operational level [22]. As discussed, and for all these reasons, the proposed methodology considers the specific geographical characteristics of the area, such as the overall potential and pre-existing constraints and limits, and is aimed at defining a predictive location model in order to minimize costs and environmental impacts in the identified area. The GIS was useful because it allowed assessments to be made based on these factors, including the anthropic infrastructures present in the area and the structures already present, which are often not considered in most research.

By using GIS (QGIS 3.28 software), the analysis involved the application of specific functions of query, proximity, centrality, and service zones for the various spatial information data. The main phases were the following: (a) identifying the most important factors in the specific case study, (b) elaborating vector information layers for the selected factors, (c) cartographic overlay and geo-processing operations, and (d) final elaboration of isocost areas that summarize the analysis.

2.3.1. Identifying the Most Important Factors

The model started from identifying the main factors and specific data to describe and represent them. In particular, it considered the topographical conditions of the production factors, the main feedstock (located in specific places and “weight loss” that is, when the weight of the raw material is only partially found in the finished product unit) deriving from agriculture and livestock, the accessibility of transport, the technical conditions of the potential plants, the adaptability of the network of pre-existing plants, and political conditions such as specific incentive schemes (essential for defining cost ranges) and spatial planning constraints (in particular in areas of respect, environmental and landscape protection) (see

Table 1. The main location factors considered in the case study. (Source: personal elaboration based on classification by Toschi [31]).

Main Location Factors	Description
Topographic conditions of production factors	✓ Feedstock specifically located and weight loss (detailed data on land use and livestock consistency);
	✓ Potential of bio-methane based one the specific diet for the anaerobic digester (elaboration, processing of land use and livestock consistency data);
	✓ Transport, presence of communication routes (routes data, cost distance).
Technical-organizational conditions of the farms	✓ Adaptability of existing constructions and presence of common services (infrastructure and existing methane pipeline system).
Other conditions	✓ Preferential tariffs and production subsidies (cost data and elaboration);
	✓ Territorial planning and constraints (existing constraints in areas of respect and landscape protection, areas of greater vulnerability due to the presence of zones vulnerable to nitrates);
	✓ Plant size respecting the local conditions.

).

Specifically, the pre-existing methane pipeline systems were considered, together with the environmental, landscape, and transportation costs as the main constraints. Feedstock and their potentialities of transformation in agro-biomethane played a significant role. These were distributed among all the municipalities in the province of Foggia, and through calculating the effective potential, considering a specific diet for an anaerobic digester and plant size, specific significant geographical areas were obtained through cartographic representation.

To identify a specific diet (feedstock typologies) for the agro-biomethane plant and its relative size, as stressed in Table 1, the first aspect considered was the potential availability of raw materials suitable for biomethane production and present in the case study area. Then, the best performance regarding environmental (significant reduction of GHGs’ emissions (≥80%) and mainly use of residual feedstock) and economic sustainability (potential economic return) was calculated among the small and medium plant sizes. Accordingly, the size of 2 MW, that is, 500 m³/h or 12,000 m³/day of biomethane production was considered worthwhile. Figure 4 shows the feedstock typologies and amount necessary to feed the proposed plant, while

Table 2. Estimated feedstock amount and feedstock BMP.

Feedstock	Annual Requirement	BMP (Nm3/ttw)
Triticale silage	25–35 t/year	120.0
Cattle sewage	2676 adult cows (more than 2 years old)	14.1
Poultry manure	30,000 of poultry unit	106.6
Olive biphasic pomace	6000 t/year of milled olives	88.3
Grape pomace	238 ha for vineyard cultivation	32.9

indicates the corresponding yearly amounts and their biochemical methane potentials (BMP), expressed in m³ per tonnes of total weight [32]. The specific diet was composed of 72% of agro-food residues (grape and olive biphasic pomaces) and animal effluents (cattle sewage and poultry manure). Triticale silage constituted the remaining 28%; this feedstock has the higher BMP value necessary to make the biomethane production feasible but also represents the smallest amount. This choice was made to preserve the rural areas’ richness in natural assets and to ensure the lowest environmental effects due to the realization of an industrial plant, albeit of small to medium size. This is in line with the circular economy approach, which the EU requested, aimed at reusing and recycling waste from every supply chain.

2.3.2. Elaboration of Vector Information Layers

Subsequently, the processed and mapped data allowed us to identify the geographical areas of the main usable crops and, in particular, the biomass potential obtainable in detail for each municipality in the province of Foggia. The data were processed and digitalized, and a spatial vector model was obtained for each of them In this case, the data were stored in the most common and internationally recognized Environmental System Research Institute (ESRI) data format, denominated shapefile. In the vector model, the information is stored in a series of data couples representing the geographical coordinates (location) associated with non-geometric data the user chooses (defined attributes of geographic features stored in a specific table), ultimately obtaining georeferenced geometric figures [33].

2.3.3. Cartographic Overlay and Geo-Processing Operations

Subsequently, the vector information layers were subjected to cartographic overlay and geo-processing operations to identify areas of better location of the plants by considering all the main factors (as reported in Table 2) and their relationships. The main geoprocessing operations included geolocalization, overlaps and spatial relationships between different information layers, centroid calculation, merging vector layers, buffer analysis, etc. The layers of information elaborated and combined in the GIS environment contributed to the modelling and cartography representation. In the specific sector, in addition to the actual local potential, the cost distance from the places of production to the points of the pre-existing methane pipelines played an important role. The selected factors contributed to obtaining the final result, which shows the isocost areas.

2.3.4. Elaboration of Isocost Areas

The data were subjected to geo-processing and, among the various methods applied, the calculation of the isocost areas was considered optimal through a simulation that was as close as possible to the actual situation of accessibility of different raw material supply places.

The Euclidean allocation method was used to define the isocost areas. Using this function derives from the need to produce a continuous surface of travel times. It is used to extend the cost distance function within neighbouring areas [34]. Therefore, transport times are extended to various areas via Euclidean allocation. The function was used to identify whether a portion of the linear surface (s_i) from point 1 to point 2 depended on the distance (not linear but via road) and, consequently, on the time necessary to reach the plant. Accordingly, the equation for transport cost will be

$$CT=\left[\left(tp\times 2\right)+tt\right]\times \frac{C_{mt}}{P_{mt}}$$

(1)

where:

• CT: transport cost (in euros per tonnes);
• tp: the raster of travel times (obtained with the Euclidean allocation function) of the area in question (in minutes);
• tt: the time required for loading and unloading the vehicles (in minutes);
• C_mt: hourly cost of the means of transport (in euros per minute);
• P_mt: maximum load that the vehicle can transport (in tonnes).

To define isocost areas, the analysis of costs, elaborated on the basis of regulatory regimes regarding economic convenience, allowed the different geographical areas to be defined from the least to the most convenient. The isocost areas were obtained from the conjunction of points of equal cost and therefore, they represent loci of points along which the transport cost is equal. They were calculated considering the cartographic base that Open Street Map provided regarding the actual presence of transport routes in the territory and then the predetermined cost ranges. This analysis, which attempts to simulate reality, considers ordinary conditions to obtain different geographical areas with gradually increasing distances and associated monetary values. These areas are identified on the map with different colours and costs, and will be better explained in the next sections.

2.4. Business Model

An economic analysis demonstrating the identified plants’ viability needs to validate the geographical location identified above. As a preliminary remark, it must be pointed out that technologies for agro-biogas production have not yet reached grid parity. Such plants are often built in an integrated manner with farms to reduce the costs of organic waste disposal, or they are supported by public incentives. The latter can be significantly reduced if biogas plants are designed to optimise the transport costs of input materials.

The economic analysis applies a discounted cash flow (DCF)-based estimation methodology by calculating the cash flow (CF), the internal rate of return (IRR), and the net present value (NPV) as financial indicators for a medium-sized plant. This methodology has been widely used to estimate the profitability of energy plants [35,36,37,38]. The profitability of plants mainly depends on the following factors: (a) technical properties of biomass used as an input source and related energy yield; (b) revenues from the sale of the gas generated; (c) the value of any granted incentives; and (d) taxation and tax deductions. On the cost side, the initial investment costs (CAPEX), which are the most important costs, should be considered, as well as operational management costs (OPEX). The economic model used is thus described as follows:

$$NPV=-EQ+{{\displaystyle \sum }}_{t=y_s}^{y_e}\frac{EBIT{A}_{\left(t\right)}-TA{X}_{\left(t\right)}-O{F}_{\left(t\right)}}{{\left(1+w\right)}^{\left(t-ys\right)}}$$

(2)

$$y_e=y_s+n$$

(3)

$$INV=CAPE{X}_{ys}$$

(4)

$$INV=DEB+EQ$$

(5)

$$DEB=d\times INV$$

(6)

$$EQ=e\times INV$$

(7)

$$d+e=1$$

(8)

$$REV=PRO{D}_P\times {\left(1-a\right)}^{\left(t-ys\right)}\times INC+PRO{D}_P\ast PE \forall t=ys\dots ye$$

(9)

$$PE=P{E}_{2020}\times {\left(1+INF\right)}^{\left(t-2020\right)} from year 2020$$

(10)

$$COST=OPEX\times {\left(1+INF\right)}^{\left(t-ys\right)}$$

(11)

$$OPEX=\left(M+INS\right)$$

(12)

$$TAX=EBITA\times IRAP+EBIT\times IRES$$

(13)

$$EBIT=EBITA-AMM-OF$$

(14)

$$AMM=\frac{INV}{n}$$

(15)

$$w=e\times r_e+d\times r_d$$

(16)

$$r_e=r_f+2\%$$

(17)

$$IRR{ }_P\stackrel{}{\Rightarrow }0=-INV+{{\displaystyle \sum }}_{t=ys}^{ye}\frac{EBIT{A}_t-TA{X}_t-O{F}_t}{{\left(1+IR{R}_P\right)}^{\left(t-ys\right)}}$$

(18)

$$IRR{ }_E\stackrel{}{\Rightarrow }0=-EQ+{{\displaystyle \sum }}_{t=ys}^{ye}\frac{EBIT{A}_t-TA{X}_t-O{F}_t}{{\left(1+IR{R}_E\right)}^{\left(t-ys\right)}}$$

(19)

where, from a financial point of view, CAPEX and OPEX indicate, respectively, capital expenditure and operating costs, COST is the gross total cost, and AMM is the depreciation. With reference to the financial structure of the operation, EQ and DEBT express, respectively, the equity and the debt capital, while e and d express, respectively, in percentage value of the total investment, the share of equity and the share of debt; INV is the total CAPEX expenditure; EBIT (earnings before interest and taxes) and EBITA (earnings before interest, taxes, and amortisation) are the main financial ratios used for the valuation; INS is the insurance policy costs; IRES and IRAP express, respectively, the national and regional corporate income taxation; IRRe and IRRp indicate the internal rate of return on equity and the internal rate of return for the project; M is the maintenance costs; n is the period of provision of incentives (15 years); OF is the total financial charges due on the debt; PE is the electricity price; PROD_P is the electricity production of the plant; PUN is the single national price of energy; REV is the total gross revenue; w is the weighted average cost of capital (WACC); y_s is starting year, and y_e is ending year.

3. Results and Discussion

The analysis, conducted through data processing, geoprocessing and overlay operations of the produced information, made it possible to identify geographical areas for the plants’ optimal locations, considering the main specific factors of the local context. By overlapping this information, the summary maps were obtained. Specifically, Figure 5 shows the overall biomethane production potential, which is derived from the above-mentioned feedstock diet: it ranges from 0.016 Mm³ to 25 Mm³, identified by the lightest to the darkest red colour among the 61 municipalities of the province. The range was defined based on previous research [19] and the quartiles method, more suitable for describing the observed phenomenon. Consequently, the most interesting geographical areas are mainly concentrated in the northern-central area of the province due to the presence of the highest amount of feedstock and the existing methane pipeline system, which represents a technical-economic constraint Other environmental conditions included the presence of economic incentives and landscape constraints (as indicated in Table 2). Indeed, this last element is a type of infrastructural and economic constraint in the first delimitation of the geographical areas, which therefore had to fall within this boundary.

To identify the optimal location for potential production plants, it is necessary to integrate the previous analysis with an economic one. Indeed, the operation of a biomethane plant requires a high availability of raw material, which has a relatively low acquisition cost but a non-negligible transport cost. A medium-sized 2 MW plant, such as the one analysed above, needs to be constantly fed, which requires moving an average of 10 to 15 tonnes of material per day (Table 2). It follows that minimising the cost of transporting raw materials is a key factor in the plant’s profitability. Therefore, the purpose of the economic analysis presented here is to investigate the effect of raw material transport costs on profitability to provide useful input for the subsequent development of the geolocation analysis.

Table 3. Economic and technical data.

Data	Value
Capex	7,500,000 €
M	1,250,000 €/year
INS	9 €/kW
d	70%
e	30%
IRAP	5%
IRES	24%
w	5.5%
n	15 (years)
rd	4.75%
a	0.5%
rf	5% *

* The risk-free interest rate was calculated as the average return of 20-year BTP in January 2023 [39]. Technical data on biomethane plants were provided by Società Italiana Gasdotti.

shows the main economic data used for the estimation.

An Excel sheet is used to estimate cash flows of agro-biomethane plants for each year of their financial lifetime. In the economic analysis, a project IRR of 20% is used as a target, which is in line with the returns of the main renewable energy sources. The gas price at which the production of the biomethane plant is to be valued is the focus of the economic analysis as well as the most uncertain factor. As is well known, gas prices have increased dramatically in 2022 (Figure 6) due to tensions related to the supply from Russia, Italy’s main supplier of natural gas.

Valuing the production of the considered plant at the average price of 2022 (1.40 €/m³), the plant would not need any incentives because the IRR would be 80%, with an NPV of 73.5 million euros. In contrast, assuming a more stable gas market, less subject to speculation, the average price of 2021 (0.415 €/m³) and 2020 (0.2 €/m³) could be used as references. In this case, with only the sale of the gas produced, the plant would not be able to generate resources to fully cover the OPEX, and therefore, there would not even be sufficient resources to cover the CAPEX. If the energy were only sold at the average prices of 2020 or 2021, the plant would not make a profit. As is known, these plants are subsidised throughout Europe, and the average subsidy is estimated at 60 €/MWh [41]. It is possible to calculate the incentive needed to reach the profitability target (IRR 20%) as a function of the transport costs of the raw material, that is, the distance between the availability of the raw material and the plant’s location. The authors assumed a minimum and maximum distance, ranging from 0–15 to 31–45 km, respectively. At a gas price equal to the average price in 2021 (0.415 €/m³), the incentive needed to achieve an IRR of 20% would be at least 26 €/MWh for a distance of 15 km or 41 €/MWh for a distance of 45 km. These values are lower than the aforementioned average incentive currently granted in the EU. However, assuming a return of the gas market to 2020 prices (0.2 €/m³), the 60 €/MWh incentive would be sufficient to achieve an IRR of 20% if the feedstock was sourced at an average distance of no more than 30 km, reaching 70 €/MWh for the longest distance.

In light of this economic analysis, it is then possible to elaborate the final geoprocessing map, shown in Figure 7. This is considered simultaneously with the geographical area of the greatest potential, which is also the closest to the existing pipeline system (Figure 5), and the overall cost distance within an optimal distance range. Specifically, this latter constraint was calculated from the geographical area of greatest potential and delivery times using the existing road network, provided by the Open Street Map service for calculating the standard transport. Following the result of the economic analysis, the final isocost geographical areas were determined for the municipalities with maximum potential. These were calculated through simulations that yielded areas of similar distances, ranging from the minimum distance (0–15 km) to a maximum distance (31–45 km). These areas involve points with similar delivery times and costs within the area of the same colour-coded region. As the colour becomes darker, all points within that specific area have increasing times and costs, considered acceptable based on the economic evaluations elaborated in the described business model. Figure 7 shows specific points by calculating centroids that identify the most suitable plant locations, represented by seven triangles. In particular, three plant locations have been identified in correspondence with geographical areas with lighter green colours (within a maximum of 15 km), characterized by optimal conditions simultaneously. The latter encompass: (a) easier access to the raw material, (b) minimised distance from the pre-existing plant system and transport costs, and (c) more efficient, sustainable transport and environmentally friendly location decisions (such as minimising the emissions from transportation and production, respecting local landscape constraints etc.).

A further element of competitive advantage is represented by the use of by-products of the agro-biomethane plants, represented by the digestate, which can be used as a fertiliser. Generally, a typical farm incurs a total cost of 1250 €/ha for an ordinary fertilization management, where the higher cost is due to the use of urea and the fertilizer 6-12-16 (constituting 66% of the total). On the contrary, this cost will decrease to 805 €/ha if the digestate is used instead of synthetic fertilizers, along with the advantage of stabilising the business plan by avoiding the fluctuation of fertilizer prices. A further benefit is that of decreasing tillage to a minimum, eliminating ploughing and grubbing [42].

Looking ahead, however, there are further benefits. The use of digestate increases the organic matter and the availability of phosphorus and potassium. This means a better yield of crops, even from the second harvest, and a reduction in future needs of fertilizer because the soil will be more fertile. In conclusion, this exploratory study has aimed to present a framework and a model for identifying the best potential agro-biomethane sites, selecting the optimal ones for a given number of plants (seven) in a rural area. This approach utilises QGIS 3.28 software and an economic evaluation, based on a minimum raw material transport cost and on the use of digestate as fertiliser. This methodological proposal fills the gap in the academic literature on location theory because it goes beyond the traditional descriptive approach [24,28,29], focusing on aspects that are scarcely explored, such as the specific local context factors, the different existing relationships, and the predictive view of location modelling. Having available data and sources, as expressly indicated and elaborated, this method lends itself to useful applications, validations, and integrations in other rural regional contexts. In addition, it allows in-depth analyses to be carried out on a regional scale because the data acquired, processed, and geoprocessed satisfy this condition. Considering that these data are detailed below the municipal scale, and with the aid of geovisualization operations, it is possible to acquire all the information requested simultaneously. Furthermore, this research represents an alternative and integrative approach to the research studies of Thompson et al. [43] and Ferrari et al. [42]. In the first case, the authors evaluated some technical and economic challenges, which are not addressed in Thompson et al. [43], by integrating their analysis with a business sustainable model and focusing on a diversified feedstock diet, and not only on animal manure, for the agro-biomethane. In the second case, although the authors selected criteria and constraints based on a critical analysis conducted by Ferrari et al. [42], this study overcomes one of limits revealed by the same authors, as it has examined the potential of agro-biomethane production in rural areas according to the different financial opportunities established by the current EU policy program. Finally, it is noteworthy that rural areas are characterized by a rapid growth in renewable resources, leveraging specific geographical features such as the abundance of natural resources, constituting a suitable context for such activities. The choice of the province of Foggia as the case study is based on the typology of available biomass resources but also several factors, explored in this study, which is oriented towards promoting more sustainable future choices.

Policy Implications

Considering the REPowerEU objectives for the biomethane supply chain, such as “closing research, development and innovation gaps” for “addressing the main obstacles to increasing the production and use of sustainable biomethane and to facilitate its integration into the EU internal gas market”, these research findings have several policy implications. First, to support the REPowerEU public funding competition tenders, setting necessary sustainable criteria may be helpful in driving projects towards a more sustainable and locally based direction. This is above all beneficial for rural areas’ policymakers, aiding them in their territorial energy planning. Second, in a context of excessive exploitation of energy sources that are not always compatible with the local environmental and landscape values, and in the absence of coordination and planning in terms of renewable energies, these findings can promote greater awareness and local acceptance of energy alternatives that are actually compatible with local characteristics. In fact, dialogue with local communities and all the stakeholders necessarily involved in the agro-biomethane location model application is critical for policy-makers in establishing targets and designing the energy system to reach them. Third, the use of this location model may also offer useful information to community planners, government agencies, utility companies, farms, extension specialists, and others who are interested in developing an agro-biomethane supply chain in an energy system based on RES. Fourth, within an international context where energy supply chains are increasingly subject to changes in geopolitical relationships, the agro-biomethane supply chain could certainly make an important contribution to the energy independence of many countries and support the economic development of many inland and rural areas. Within this process, an important role can be played by the simplification and improvement of integrated and sustainable connections to the transport and distribution network to facilitate the introduction of agro-biomethane for consumption, along with the use of its by-product, digestate, as a substitute of synthetic fertilisers.

4. Conclusions

Based on the proposal model developed by the authors (integrating location theory with the business model), it has been possible to obtain strategic spatial information and identify some geographical rural areas suitable for the best sustainable location of an agro-biomethane plant with a power of 2 MW. Specifically, three optimal plant locations have been selected in the case study in the province of Foggia (Southern Italy), where some conditions have been identified as key elements. These encompass: (a) the high amount of the feedstock available for the envisaged agro-biomethane plant, (b) the proximity of the plant to the existing pipeline system, and (c) no more than 30 km from feedstock sources, assuming gas prices return to pre-pandemic levels. Specifically, the economic analysis has shown that an efficient location is essential to increase the plant’s profitability and to reduce the amount of incentives needed to ensure these projects’ economic sustainability.

The model is unique in that as it combines methodologies from different scientific disciplines, each of which is not particularly complex to develop, which is the real added value of this analysis. Given the relative simplicity of the basic methodologies used, there are no particular obstacles to replicating this model in different territorial contexts. Nevertheless, as a great deal of data specific to the area under analysis are required, the limits to its application may stem from the difficulty of finding precise data on electricity production at provincial level, on the exact location of the natural gas network, and on the production of agro-food waste or by-products. Often, as in this case, different sources have to be combined to obtain these data, with all the methodological limitations that this entails. Future developments of the model could consider extending the analysis to the logistical management of the raw material, both from an economic and geographical point of view, as well as the possibility of examining alternative uses of the biomethane produced with respect to feeding it into the grid, and whether these alternatives have an impact on the choice of location.

In conclusion, these results could be a useful policy tool for defining integrated planning strategies in the field of energy and ecological transition. Such strategies are aimed at the decarbonisation of territories and the mobilisation of territorial stakeholders’ engagement towards renewable energies as an alternative to fossil fuels. Above all, this latter factor is particularly important for the rural areas that, as in this case study, have experienced widespread implementation of large-scale renewable energy plants but without effective socio-economic and environmental benefits for their communities.