**Identification of Suitable Areas for Biomass Power Plant Construction through Environmental Impact Assessment of Forest Harvesting Residues Transportation**

**Maria Pergola 1, Angelo Rita 2, Alfonso Tortora 2, Maria Castellaneta 2, Marco Borghetti 2, Antonio Sergio De Franchi 2, Antonio Lapolla 2, Nicola Moretti 2, Giovanni Pecora 2, Domenico Pierangeli 2, Luigi Todaro <sup>2</sup> and Francesco Ripullone 2,\***


Received: 3 April 2020; Accepted: 26 May 2020; Published: 28 May 2020

**Abstract:** In accordance with European objectives, the Basilicata region intends to promote the use of energy systems and heat generators powered by lignocellulosic biomass, so the present study aimed to investigate the availability of logging residues and most suitable areas for the construction of bioenergy production plants. The life cycle assessment (LCA) methodology was employed to conduct an environmental impact assessment of the biomass distribution and its transport, and spatial LCA was used to evaluate the impact of regional transport. One cubic meter kilometer (m<sup>3</sup> km<sup>−</sup>1) was used as the functional unit and a small lorry was considered for the transport. The results showed that the available harvesting residues amounted to 36,000 m<sup>3</sup> and their loading environmental impact accounted for 349 mPt m−3. The impacts of transport (4.01 mPt m−3) ranged from 3.4 to 144,400 mPt km−<sup>1</sup> forest parcel<sup>−</sup>1, mainly affecting human health (95%) and, second, the ecosystem quality (5%). Three possible sites for bioenergy plant location were identified considering the environmental impact distribution due to feedstock transport. Findings from this research show the importance of considering the LCA of biomass acquisition in site selection and can fill the knowledge gaps in the available literature about spatial LCA.

**Keywords:** bioenergy; life cycle assessment; geographic information system (GIS); harvesting residues

#### **1. Introduction**

Harvesting residues are the biomass left on fields after wood harvesting (tops, branches, and little non-marketable trunks) [1]. On average, 10% to 15% of this biomass is left on site as forest residues following harvesting operations [2] because it is expensive to harvest and transport and there are few markets for this wood material. Occasionally, some of the larger logging wood is removed as firewood for domestic consumption [3].

At the same time, the use of wood biomass is believed to be an important component of renewable energies, particularly for producing thermal energy or joint thermal and electrical energy with a view to creating smart energy cities. Bailey et al. [4], Perez-Verdin et al. [5], and Moon et al. [6] argued that the use of wood biomass residues to produce energy or fuel can encourage the rise of regional economies

and the creation of new employment opportunities. In the last decade, there has been increased awareness in using this residual biomass as a raw material for renewable energy as a response to the standards for renewable fuels and energy markets [7]. This is because the production of forest biomass energy has the potential to reduce carbon emissions when replacing fossil fuels, although several authors have reported contrasting evidence [8]; retrieve waste that would otherwise be disposed of in landfills or be incinerated; create jobs (especially in rural areas); and supply local and sustainable energy for communities, reducing their dependence on the international fuel market, as affirmed by Shabani et al. [9], Saidur et al. [10], and Ahtikoski et al. [11]. However, these residues are often not fully utilized due to the lack of demand within the immediate vicinity of the processing plant. Furthermore, transporting residues to an area with high demand is considered uneconomical [12], and significant costs are associated with the supply of forest residues from the forest. Transport also constitutes one of the major sources of air pollution, in particular, due to emissions of sulfur dioxide (SO2), nitrogen oxides (NOx), volatile non-methane organic compounds (NMVOC), primary particulate matter (PM 2.5), and carbon monoxide (CO). The latter are produced during fuel combustion, but other non-exhaust emissions of particulates, are produced during road and rail transport due to the abrasion of brakes, wheels, etc. [13]. Thus, the inclusion of some collection sites is out of the question due to long distances, harsh topographic conditions, or ecological restrictions. In any case, it is advised that 30% of the harvesting wood residues are left in place in order to restore the fertility of the forest soil [14–17].

When analyzing the whole life cycle of the product, Law and Harmon [18] and Schulze et al. [19] highlighted the fact that some aspects of production could worsen rather than contribute to mitigating climate change such as long-distance lumber transport. Farkavcova et al. [20] stated that in Europe, transport represents 22% of the total emissions, and that these emissions are constantly increasing. In addition, urban transport is responsible for around 25% of the CO2 emissions produced by all transport [21]. Referring to the forestry sector, cradle-to-grave life cycle assessment (LCA) studies have shown that transport significantly contributes to the overall results by representing 60%–70% of the overall environmental impact [20,22]. In this regard, LCA is very useful, since it is a useful tool for evaluating all environmental impacts linked with a product, process, or activity as well as the consumption and emissions of resources [23]. LCA is constantly evolving, and its application to bioenergy systems has been a key factor for the development of the process in the last few years. In the literature, bioenergy production chains have been evaluated from an environmental and energy point of view by several authors [24–30]. Particular attention has been paid to saving greenhouse gas (GHG) emissions and energy balances for the production of liquid biofuels [31]. Some reviews have considered electricity, while only one study has included heat in addition to generating electricity [24]. Referring to transportation systems, an LCA study includes the identification of direct, indirect, and supply chain emissions affecting the system. In particular, direct emissions refer to energy consumption and emissions associated with vehicle movement, namely, air emissions (CO2, CO, SOx, NOx, PM, etc.) from diesel combustion. In LCA transport, fuel use and the related produced emissions are called direct emissions because they are associated with the direct objective of the system to ease the movement of residues. Indirect processes are those that must exist for the direct process to exist, in this case, vehicles, infrastructure, and energy production services; vehicle production; infrastructure construction, management, and maintenance; and fuel and electricity production. Additionally, these indirect processes depend on a supply chain to produce materials, services, and other activities, probably far from where the vehicle acts [32]. Similarly, the direct energy input represents the energy effectively used to sustain a process (fuel and oil consumption of machineries, and energy consumption of humans during the work), while the indirect energy input stands for the energy stored in the materials used in the process (the energy value for the production of machinery and tools) [33,34].

Wood biomass residues are geographically allocated, with alterations in space–time availability. Therefore, energy and environmental evaluation requires a decision support system for efficient planning [35,36]. To plan a biomass facility, a preliminary and precise database including the distribution of residues and the seasonal variation (peak period and decreased availability period) is

essential. Logistics such as the harvesting, storage, and transport of residues are spatially interconnected and require accurate planning. A geographic information system (GIS) is an important territorial decision-making tool that allows for a precise evaluation of distributed resources for renewable energy [37–39]. The joint use of LCA and GIS, also known as spatial LCA [35], can be useful for estimating the biomass potential in a region, and enhancing the results of environmental impact assessment by counting spatial variations and considering power plant design.

According to the European commitment of the last few years, which is aimed at solving the international economic and environmental problems linked to the climate and energy supply, the Basilicata region has highlighted the importance of the agricultural and forestry sector in the development and diffusion of renewable agro-energy sources [40]. The regional commitment aims at strengthening the financial instruments to support research and experimentation as well as the involvement of interested companies in implementing pilot projects in the regional territory. In 2012, the potential supply of forest wood biomass for bioenergy production in the Basilicata region was estimated to be around 500,000 tons per year [41]. Of this quantity, the same authors identified a mean annual production of about 22,000 tons of residual biomass of forestry origin and an average annual production of about 400,000 tons of residual biomass of agricultural origin as well as an average annual unitary production of dry biomass from dedicated crops, consisting of approximately 60,000 tons on private fields and approximately 57,000 tons on public fields [41]. Thus, the regional administration has assigned a strategic role to the energy sector to relaunch the territories with the aim of creating new and qualified job opportunities and environmentally friendly development [40]. Hence, the present research is part of the "Smart Basilicata Research and Development" project. This project aims to develop innovative techniques for the management of wood biomass including their use for energy purposes. The aim of the present research was the selection of co- and trigeneration plant construction sites and the optimal residual forest woody material collection areas in the regional territory after first investigating the availability and amount of harvesting residues through an analysis of forest management plans (FMPs) still in force. GIS was used as a decision-making spatial tool for the accurate assessment of spatially logging residues for lignocellulosic bioenergy production. Additionally, the LCA methodology was employed for an environmental impact assessment of the biomass load and transport and spatial LCA was used to investigate the distribution of the impact on the regional territory.

#### **2. Materials and Methods**

#### *2.1. Case Study Description*

The study was performed in the Basilicata region, one of the most forested areas in southern Italy (356,426 hectares) in which the forest sector is governed by Regional Law No. 42 of the 30th November 1998 "Rules on Forestry" [42]. The objective of forest management planning in this region is to apply sustainable forest management guidelines, which are carried out through FMPs. In the present study, only wood biomass produced directly from forest plans was taken into consideration (i.e., the logging residues estimated as percentages of forest utilization), according to the guidelines developed by the Basilicata region for the reduction of FMPs [42].

Considering the forestry sector and despite the benefits of forest management, in the Basilicata region, as in other Mediterranean areas and South-East Europe, the seasonality in the demand for wood products such as firewood, the substantial investments required to purchase woodland lots by forest companies, and the high cost of transactions due to the slowness of the administrative and authorization procedures, has resulted in an excessive bureaucracy [43,44], which significantly reduces the gross operating margins of companies. All this does not incentivize the purchase and consequent management of woods, especially in terms of public ownership. This leads to the abandonment of forests and sometimes to the degradation phenomena that affects the capacity of forest ecosystems [45]. Therefore, in Basilicata, there are currently 83 FMPs, 35 of which are still in force. Consequently, the study was performed in these latter municipalities (Figure 1).

**Figure 1.** Study area: municipalities of the Basilicata region in which forest management plans (FMPs) are still in force.

#### *2.2. The Life Cycle Assessment (LCA) Approach*

The present study was designed to investigate the optimization of biomass supply distances in order to designate suitable sites for the implementation of cogeneration or trigeneration plants powered by biomass in the Basilicata region. An LCA analysis was performed according to the ISO 14040/44 (2006) [23,46].

The objective of the present analysis was to address the movement of the harvesting residues, by road, from the production forest parcels in the regional territory. Therefore, the system boundaries include the environmental impacts during all phases of the transport (transport operation and infrastructure), from raw material extraction to their use and, finally, disposal. Moreover, through the LCA methodology, all important emissions were quantified as well as their related environmental and health impacts and the issue of the resource consumption combined with transport.

In order to estimate the environmental impact of transport services and correlate transport datasets with other product life cycles, environmental loads are determined using the functional unit of one cubic meter kilometer (m3 km<sup>−</sup>1). A cubic meter kilometer is defined as the transport of a cubic meter of harvesting waste from a given transport service over a kilometer [47].

Data on the available harvesting residues and their distribution in the regional territory were gathered from an analysis of the 35 FMPs still in force. In particular, these residues were estimated as percentages of forest utilization, according to the guidelines developed by the Basilicata region for the redaction of FMPs. These percentages reflect the share of residues in the mean annual cut, as indicated by Cozzi et al. [48]. In order to ensure sustainable harvesting levels (max 60% of the annual increment for high forest and 90% for coppice forest), the percentage of forest utilization was between 5% and 50% of the total wood mass in the examined FMPs, and the relative percentage of residues available ranged from 9% to 20%. Data related to harvesting residues (tops, branches, and little non-marketable trunks usually left on sites) and load (forestry machines used, duration of the operation and fuel consumed) were taken from information reported in Pergola et al. [49], while data on the transportation were extrapolated from SimaPro's Life Cycle Inventory (LCI) databases and, in particular, from databases of scientific relevance and accuracy such as Ecoinvent 3 [50]. Since one of the objectives of this research was to investigate the regional distribution of the environmental impacts of the transport of woody residues, the item "Small lorry transport, Euro 0, 1, 2, 3, 4 mix, 7.5 ton total weight, 3.3 ton max payload RER S" was employed, whose emissions calculation was obtained from the literature [51] based on measurements [52].

Environmental assessment was performed using the SimaPro 8.0.4.30 (copyright PRé Consultants bv 2014) software by means of the LCA Eco-indicator 99 endpoint method, in which "environment" was defined as being affected by three types of damage [53]:


In addition, the impact assessment was performed following the endpoint approach, which expresses the total environmental impact in a single score using the point (Pt) or millipoint (mPt) as the standard unit [53].

#### *2.3. The Geographic Information System (GIS) Analysis*

The ability to analyze the environment and understand all the factors that characterize it is a prerequisite for carrying out a study on a territory's suitability, and the main tools are often represented by the GIS. The latter is itself a system of tools designed to acquire, extract, archive, manipulate, analyze, manage, visualize, and present all types of geographically referenced data, which are data from the real world [54].

In particular, geo-referenced data of the total harvesting residues load and transport impacts were imported into a project in the GIS software package, together with maps of the main road network, main electricity grid, borders of the Basilicata region and municipalities, and main protected areas, which were freely available through the RSDI Basilicata portal (http://rsdi.regione.basilicata.it/ web/guest/mappe-in-linea). The residential areas were taken from the digitalized map (1:25,000) of the Istituto Geografico Militare (IGM). To compute the spatial distribution of the transport impact, we multiplied the harvesting residues transport impacts per distance from the parcel, where buffers were generated using the buffer tool with distances of 1, 5, and 10 km. We highlighted that longer distances

from the parcels do not maximize the cost-effectiveness of the residual transport. Finally, layers were then arranged to produce the final maps using ArcMap® 10.4.1 software by Esri (Copyright © Esri). A detailed workflow of the processes is provided in the Supplementary Materials (Figure S1).

#### **3. Results and Discussion**

#### *3.1. Harvesting Residue Availability*

Table 1 reports the results for each municipality of the analysis of 35 FMPs, referring to the availability of harvesting residues for bioenergy production (expressed in m3) and environmental impacts (expressed in mPt) relative to harvesting residues loading and harvesting transport per kilometer.

**Table 1.** Harvesting residues available for bioenergy production and environmental impacts per studied municipality.


Referring to the availability of harvesting residues, the analysis of the 35 FMPs showed that they amounted to about 36,000 m3, and the Basilicata region could be split into three areas: the largest

supply basin was located in the central part (about 18,300 m3), followed by the south (about 7900 m3), and then the north (about 9798 m3). Harvesting residues per forestry parcel ranged from 0.85 to 668 m3.

In line with the "Smart Basilicata Research and Development" project, investigating the availability and quantity of residual forest woody materials is useful for understanding the feasibility of cogeneration or trigeneration plants powered by biomass, which could be used to produce thermal and electrical energy to create district energy systems. The latter is a growing phenomenon in many cities around the world [55] and, as stated in Perea-Moreno et al. [56], the introduction of such schemes into urban networks has important benefits such as the availability of an open energy supply grid, greater use of renewable energy sources, less reliance on imported resources and fossil fuels, greater leverage over energy supply, and the development of energy supply [57].

At the same time, harvesting residues are widespread in the regional territory, as shown in Figure 2, so there is a need to transport and concentrate them in specific areas for subsequent bioenergetic purposes.

**Figure 2.** Regional distribution of harvesting residues and relative load impacts per impact class.

#### *3.2. Loading Impacts*

Harvesting residues load impacts were, on average, equal to 349 mPt m−<sup>3</sup> and ranged from about 300 to 233,000 mPt forest parcel<sup>−</sup>1. The main impacts concern human health (284 mPt m−3), followed by resource depletion (59 mPt m<sup>−</sup>3), and ecosystem quality (6 mPt m−3). Human health damage (in total, equal to 0.173 DALY) was mainly due to the fuel consumed in the various loading operations and the impact categories most responsible for this damage were climate change (75%) and radiation (24%). Resource depletion (in total, equal to 7344 MJ surplus) was essentially affected by the materials and processes necessary for the construction of forest machines. Referring to ecosystem quality (in total, equal to 10,090 PDFm<sup>−</sup>2year−1), the most important impact categories were air acidification and water eutrophication, representing 73% of this damage (Table 2).


**Table 2.** Characterization of the total harvesting residues load impacts.

Many of the analyzed forest parcels fell in protected areas and in particular, in two national parks (Appennino Lucano-Val d'Agri-Lagonegrese and Pollino National Parks), in two regional parks (Gallipoli Cognato Piccole Dolomiti Lucane and Vulture), and in several Natura 2000 Network sites (Figure 2), so the load operations should be performed with caution to ensure their conservation for present and future generations [51]. In particular, these areas of relevant naturalistic and ecological value are subjected to a specific rule of protection and management to preserve animal and plant species; safeguard anthropological, historical values, and agro-forestry–pastoral and traditional activities; promote education, training, and scientific research activities; defend and replenish the hydraulic and hydro-geological balance; and promote the enhancement and testing of compatible production activities [58].

#### *3.3. Transport Impacts*

LCA analysis showed that the transport of 1 m<sup>3</sup> of harvesting residues caused environmental damage equal to 4.01 mPt km−1, which, in total, for the whole regional territory, corresponded to 144,421 mPt km−<sup>1</sup> (Table 1). Needless to say, the greatest impacts were recorded in the municipalities with the greatest amount of residues to transport and, therefore, in the middle area (73,587 mPt m<sup>−</sup>3) (Table 1). Similar to loading, the greatest impacts were on human health (3.81 mPt m−<sup>3</sup> km<sup>−</sup>1), but in this case, were followed by ecosystem quality (0.195 mPt m−<sup>3</sup> km<sup>−</sup>1).

Human health was mainly affected by climate change (59%) and respiratory inorganics (39%); in particular, the latter refers to "winter" smog caused by inorganic substance emissions. At the same time, eutrophication/acidification was the impact category with the greatest negative effects on the quality of the environment, representing 97% of the total impact of ecosystem quality (Table 3).


**Table 3.** Total transport damage characterization per impact categories.

Farkavcova et al. [20] stated that transport from the forest to the production site caused significant environmental impacts and, more precisely, mainly caused the consumption of fuel fossil resources, the abiotic depletion of the non-fossil resources, and the potential reduction of the ozone layer. In addition, the environmental impacts of transport are particularly due to the consumption or partial combustion of non-renewable fossil fuels as well as trace elements in fuel and tire abrasion. According to Handler et al. [59] and Sonne [60], the transport of biofuels or raw materials for bioenergy is potentially the major source of environmental impacts in the total supply chain. All of this was confirmed by Murphy et al. [22], and in accordance with these observations, several studies have reported that forest biomass transport accounts for most of the energy consumption and environmental impacts in forest biomass systems [61–63]. The results of these studies have shown that the transport of biomass significantly contributes to both the energy demand and GHG emissions, representing 70%–78% of the overall energy needs and 68%–75% of GHG emissions.

Other comparisons of this research and the results of other LCA studies were not possible since they only focused on the global warming potential (GWP) or energy demand, and therefore did not provide a complete picture; Farkavcova et al. [20] rightly advised that the whole set of indicators should be considered. Moreover, as stated by Murphy et al. [22], comparisons of results are complicated by discrepancies in system boundaries, geographic areas, and the employed characterization methods. According to Heinimann [64], LCA studies neglecting embodied burdens of road infrastructure and forest machines, called "truncated LCAs", always result in an underestimation of environmental impacts or an overestimation of environmental performance, respectively.

#### *3.4. The Territorial Distribution of Impacts*

Harvesting residues were widespread in the regional territory and, consequently, their loading and transportation for bioenergy production also had widespread impacts. Figure 3 shows the distribution of the total environmental impact for each forest parcel when transporting residues within 10 km from the source. Since there were many forest parcels (327), for a better representation of the environmental impacts, the calculation was simplified and considered the calculation of the impacts for 1, 5, and 10 km, and not for each kilometer, to best represent the environmental impacts at a territorial level. Red areas represent areas with the greatest environmental impacts, given by the sum of the various "impact rays" calculated for each forest plot. Therefore, the total impact ranged from a minimum of 3.41 mPt to a maximum of about 276,000 mPt (Figure 3).

**Figure 3.** Territorial distribution of harvesting residues transport impacts.

To our best knowledge, this paper is one of the few studies that wish to represent transport environmental impacts territorially, in order to understand how they are distributed when more displacements are involved. In conducting a complete environmental analysis, the present study considered a set of indicators, rather than a single category (e.g., GWP). Other LCA studies of the biomass supply chain [20,22,24] have referred to the movement of lumber from the forest to a bioenergy plant, while the present study, through additionally considering environmental impacts, tried to give indications for the regional administration on which areas may be the most suitable locations for a bioenergy production plant (i.e., without compromising the environment and human health).

Indeed, one of the objectives of the "Smart Basilicata Research and Development" project is the development of cogeneration or trigeneration plants powered by lignocellulosic biomass in the regional territory. As stated by Zubaryeva et al. [65], the site should be readily reachable by transport, close to service points, and achievable for the best planning of energy transport lines. In addition, the plant should be established at an acceptable distance from residential areas, natural reserves, and protected areas to diminish the potential negative impacts of plant operation and waste disposal [35].

According to Hiloidhari et al. [35], the site selection of a biomass power plant based on GIS can be carried out through two methods: (i) suitability analysis and (ii) optimization analysis. The former allows users to recognize the most appropriate site for a power plant among many candidate sites based on user-defined constraints and support criteria. On the other hand, the best analysis accounts for the relationship between biomass and power plants to determine the optimization locations of power facilities with minimal transport and distribution costs [66].

In the present study, we tried to select the most suitable sites considering the map of the impacts; the proximity to the main road and electricity networks as well as residential areas; and the presence/absence of protected areas. Therefore, as reported in Figure 4, three possible sites were identified (one for each area of the Basilicata region):


**Figure 4.** The three sites suitable for biomass power plant construction \*.

The optimization of biomass power plant location may be carried out through the modeling of the location–allocation or the modeling of the supply area including or not including the impacts, but, as stated by Cozzi et al. [48], regardless of the applied method, the selection of an appropriate biomass center should take into account several aspects (energy, environmental, and economic) to be in line with the three pillars of sustainability (economic viability, environmental protection, and social equity). From a landscape perspective, it would be effective to place the plant in urbanized areas with similar structures, in order to avoid areas typified by agricultural and forestry aspects, keeping in mind the costs for the transport of biomass, since lower costs are obtained in areas close to the biomass processing plant [48]. Furthermore, it must be considered that the studied forest particles mainly fall in protected areas, which are rural though heavily frequented areas, where the concentration of emissions during traffic congestion could enhance by 100 times [21], further damaging human health and the quality of ecosystems.

#### **4. Conclusions**

The present study aimed to identify suitable sites for locating cogeneration/trigeneration plants powered by lignocellulosic biomass in the Basilicata region based on GIS–LCA information, after investigating the quantity of harvesting residues and environmental impacts of their loading and transport. This research allows us to take further steps forward in our knowledge about spatial LCA with regard to bioenergy production. Indeed, traditional LCAs are inadequate to identify the spatial dimensions of environmental impacts, however, this becomes feasible when they are carried out applying a GIS framework. In this study, we first assessed the environmental impacts per kilometer (4.01 mPt m−3), and then built a map of cumulative impacts over a radius of 10 km for the different analyzed forest parcels, in order to identify areas with major and minor impacts. In this way, we were able to identify three areas to locate biomass plants after considering the main road network, electricity network, proximity to residential areas, and excluding protected areas. The present study represents a replicable example of how it is important to consider the environmental impact distribution of feedstock transport and not only those of bioenergy facility construction in the site selection of a biomass power plant.

Finally, we emphasize these essential aspects: only biomass residues from locally performed forest harvesting operations, or wood residues from local saw milling activities should be used for bioenergy production, and each project for bioenergy production should be preceded by a careful assessment of the potential impact of biomass removal on soil fertility and forest ecosystem biodiversity.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1996-1073/13/11/2699/s1, Figure S1: Workflow of the processes for the GIS analysis.

**Author Contributions:** Conceptualization, M.P., F.R.; Methodology, M.P., A.R.; Software, M.P., A.R., A.T.; Validation, F.R., M.C., A.R.; Formal Analysis, M.P., M.C., A.R.; Investigation, M.C., G.P.; Resources, M.C.; Data Curation, M.C., M.P.; Writing—Original Draft Preparation, M.P., A.R.; Writing—Review & Editing, A.R., F.R., M.B., D.P., L.T.; Visualization, A.S.D.F., N.M., G.P., D.P., L.T., A.L.; Supervision, F.R.; Project Administration, F.R., D.P., M.B., N.M.; Funding Acquisition, F.R., D.P., M.B., N.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the "Smart Basilicata" project (Notice MIUR n. 84/Ric 2012, PON 2007–2013 of 2 March 2012).

**Acknowledgments:** This research was carried out in the framework of the project "Smart Basilicata", which was approved by the Italian Ministry of Education, University and Research (Notice MIUR n. 84/Ric 2012, PON 2007–2013 of 2 March 2012) and was co-funded by the Cohesion Fund 2007–2013 of the Basilicata Regional authority. "Ufficio Foreste e Tutela del Territorio" of the Basilicata region is duly acknowledged for their kind permission to use data extracted from the "Forest Management Plans". The PhD program in 'Agricultural, Forest and Food Sciences' at the University of Basilicata supported O.P.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Consequential Life Cycle Assessment of Swine Manure Management within a Thermal Gasification Scenario**

#### **Mahmoud Sharara 1, Daesoo Kim 2, Sammy Sadaka 3,\* and Greg Thoma <sup>2</sup>**


Received: 20 September 2019; Accepted: 23 October 2019; Published: 25 October 2019

**Abstract:** Sustainable swine manure management is critical to reducing adverse environmental impacts on surrounding ecosystems, particularly in regions of intensive production. Conventional swine manure management practices contribute to agricultural greenhouse gas (GHG) emissions and aquatic eutrophication. There is a lack of full-scale research of the thermochemical conversion of solid-separated swine manure. This study utilizes a consequential life cycle assessment (CLCA) to investigate the environmental impacts of the thermal gasification of swine manure solids as a manure management strategy. CLCA is a modeling tool for a comprehensive estimation of the environmental impacts attributable to a production system. The present study evaluates merely the gasification scenario as it includes manure drying, syngas production, and biochar field application. The assessment revealed that liquid storage of manure had the highest contribution of 57.5% to GHG emissions for the entire proposed manure management scenario. Solid-liquid separation decreased GHG emissions from the manure liquid fraction. Swine manure solids separation, drying, and gasification resulted in a net energy expenditure of 12.3 MJ for each functional unit (treatment of 1 metric ton of manure slurry). Land application of manure slurry mixed with biochar residue could potentially be credited with 5.9 kg CO2-eq in avoided GHG emissions, and 135 MJ of avoided fossil fuel energy. Manure drying had the highest share of fossil fuel energy use. Increasing thermochemical conversion efficiency was shown to decrease overall energy use significantly. Improvements in drying technology efficiency, or the use of solar or waste-heat streams as energy sources, can significantly improve the potential environmental impacts of manure solids gasification.

**Keywords:** life cycle assessment; environmental impact; greenhouse gas; gasification; swine manure management

#### **1. Introduction**

Agricultural systems are significant contributors to global climate change and ecosystems degradation [1]. Local, regional, and global agreements are increasingly mandating legislative and regulatory actions to restrict emissions to mitigate the short-term and long-term environmental degradation. However, both legislative and regulatory efforts to reduce environmental impacts, especially greenhouse gas (GHG) emissions, will eventually put a more significant burden on agricultural and industrial sectors as well as increase the cost of production. Livestock production, in particular, has been recognized as a significant source of GHG emissions and a driver of both freshwater and marine water eutrophication [2,3]. Therefore, the vulnerability of livestock production and the

agriculture sector to climate change further incentivizes the search for and adoption of sustainable agricultural practices [4].

In livestock production, manure management is a significant source of direct GHG emissions, such as methane (CH4) and nitrous oxide (N2O) [5]. Land application is the most common practice to handle swine manure to use available nutrients for crop production. However, applying swine manure to crop and grass fields where nutrients are available more than agronomic crop needs or where fields have historically received large volumes of manure application increase environmental risks to surrounding ecosystems. Liquid manure management systems, relevant to swine production, are also a significant source of gaseous emissions. Liquid manure storage promotes anaerobic conditions, which transform organic matter into CH4 and ammonia (NH3). Besides, uncontrolled anaerobic and aerobic conditions initiate nitrification-denitrification processes, which convert a share of manure nitrogen to N2O, which is a potent GHG. Solid-liquid separation of swine manure has been recognized as an emission mitigation strategy. However, increased N2O and CH4 emissions have been reported during storage of manure-separated solids [6]. Transforming separated solids into a gas fuel (syngas) and a stable nutrient-rich co-product (biochar), via gasification, can potentially reduce emissions associated with manure-separated solids and generate value-added products. Furthermore, gasification-derived biochars have been shown to have adsorbing characteristics for various organic contaminants such as *p*-Cresol [7,8].

Evaluating emissions and impacts associated with this conversion strategy, i.e., gasification of swine manure solids, can facilitate adoption and expand the set of technologies available to livestock producers for manure management. Life cycle assessment (LCA) is an essential tool to assist decision-makers by evaluating the environmental performance of proposed management strategies. According to ISO standard 14040 [9], LCA considers the various input and output flow, and the corresponding environmental burdens, resulting from production, consumption, and disposal of associated product systems. Energy recovery from swine manure incineration was found to be a promising pathway to reduce GHG emissions associated with manure management [10]. Several LCA studies have reported on the performance on anaerobic digestion as manure management and energy recovery as a sole feedstock or in combination with other biomass streams [11–13]. Wu et al. [10] performed an LCA comparing GHG emissions between land application and gasification as manure management practices. The study showed that gasification has high potential to reduce GHG emissions due to the environmental benefits of syngas production and biochar application to crop field. Biochar also has been used for carbon sequestration, soil amendment and biomass waste management [14].

In a comprehensive review of swine manure conversion technologies, Sharara and Sadaka [15] highlighted the scarcity of research studies that investigated swine manure solids gasification and pyrolysis. Accordingly, it was recommended to develop an LCA of swine manure management systems. Therefore, the objective of this study is to evaluate potential environmental impacts of a manure management scenario that utilizes thermal gasification of swine manure solids as a disposal/energy retrieval strategy using consequential life cycle assessment (CLCA) methodology.

#### **2. Methods**

#### *2.1. LCA Goal and Scope*

The goal of this CLCA is to determine the impacts associated with swine manure management using gasification for manure solids. Figure 1 shows a schematic diagram of the proposed swine manure management (SMM) scenario. The scope of this CLCA covers manure management activities associated with 1000 kg of flushed swine manure at 5% dry matter content, without accounting for animal maintenance (feed, drinking water, climate control - all assumed to be unaffected by treatment), until the land application of both the liquid fraction (slurry) and the solid fraction (biochar). The functional unit (FU) is the treatment of one metric ton (1000 kg) of swine manure slurry, at 5% DM, via gasification and land application. The thermal gasification of manure solids produces three

co-products: syngas, heat, and biochar. All three co-products were modeled as displacing existing processes in the system with surpluses used to displace their alternatives beyond the system boundary. A share of the produced syngas is consumed as fuel in a boiler, while the excess syngas is considered a replacement for natural gas. The heat generated during the thermal gasification is used for drying, and biochar is land applied as a fertilizer replacement.

**Figure 1.** The scope of proposed swine manure management scenario illustrating mass balances and emissions flows (Note: black arrows represent main product flow; red arrows represent direct and indirect greenhouse gas (GHG) emissions. Blue arrow represents water evaporation. Numbers in red are direct emissions of GHG in kg CO2-eq).

The excreted manure (urine and feces) is flushed from shallow pits under the house. The flushed manure is stored in a holding pond, and stirred, before being pumped into the separation stage. Manure separation is accomplished using a screw press separator. This class of size-separators utilizes a tapered screw and a fine-mesh screen (0.75–3 mm) to fractionate the manure into solids-rich and liquids-rich fractions. The solids-rich fraction is transported to a thermochemical conversion facility that contains a dryer, a gasifier, and a gas boiler. The manure gasification is accomplished in an atmospheric, fluidized-bed gasifier to produce syngas, which is subsequently fired (burned) in the gas boiler for heat, in addition to biochar. The liquid-fraction (slurry) is stored, then transported to an agricultural field for land application. In this model, the emissions associated with land application of biochar and slurry are presented in detail separately first, then combined to estimate the total impacts of biochar-slurry land application. The total impacts represent the summation of the impacts from each substrate without including any synergistic effects. SimaPro© 8.5.2 software (PRé Consultants, The Netherlands) and ecoinvent v3.4 database [16] with IMPACT World+ midpoint method [17] were used to model the impacts. We modified the characterization factors in IMPACT World+ by adopting the most recent Intergovernmental Panel on Climate Change (IPCC, v1.02) 100-year global warming potentials (GWP 100a): CH4 biogenic: 27.8 kg CO2-eq, CH4 fossil: 30.5 kg CO2-eq and N2O: 265 kg CO2-eq [18]. The inventory of mass, energy, and emission flows, in each stage, is presented in the following sections.

#### *2.2. Life-Cycle Inventory Assessment*

#### 2.2.1. Swine House

According to manure characteristics standard [19], the amount of total solids in as-excreted swine manure is between 5% and 10% by weight for grow-finish pigs. Manure can be collected from barns through flushing, scrapping, or using pull-plug systems that rely on gravity. Collection systems significantly impact the concentration of solids in the collected manure. In this study, the composition of swine manure solids was taken from first-hand analyses of manure solids sampled from the gravity-collected slurry in a feeder-finisher farm in Washington County, Arkansas. Table 1 shows the characteristics of swine manure solids. The accumulated manure is collected every two weeks by gravity using a pull-plug system (no energy or mechanical power is needed for drainage). During the 2-week storage, various biogenic emissions, namely NH3, N2O, CO2, and CH4, are released due to aerobic and anaerobic activities in the manure substrate. NH3 and N2O emissions as nitrogen during this period were estimated at 16% and 0.5% of total manure nitrogen [20]. Manure-related GHG emissions, i.e., CH4 and N2O were estimated using the IPCC Tier 1 approach for GHG emissions in livestock [1].

**Table 1.** Characteristics of swine manure under the current study.


#### 2.2.2. Pre-Separation Tank/Stirring and Mixing

Pre-separation storage was modeled as an opened storage tank. The projected NH3 loss is 2% of the total N in the manure [21]. IPCC guidelines for GHG emissions were used to estimate the N2O and CH4 emissions in the pre-separation tank. For the agitation/mixing step, Wesnaes et al. [21] and Nguyen et al. [22] reported that the energy requirements for pumping and stirring 1000 kg of manure slurry to be 0.5 kWh and 1.2 kWh, respectively. Thus, the total energy consumption associated with this stage (stirring and pumping) was taken as 1.7 kWh.

#### 2.2.3. Mechanical Separation and Separated Solids Transport

Moller et al. [23] estimated the power required for manure solid-liquid separation using a mechanical screen press to be 0.50 kWh per metric ton. Therefore, the energy required for separation was modeled as 0.5 kWh. The U.S. electricity mix was used to model the impacts of electric power utilization. No air emissions or water contamination are associated with manure during this stage.

The separation indices reported for screw presses [24] were used to determine the amount and composition of separation products. In that study, the solids content in the original slurry varied from 1.8% to 6.3%. As shown in Table 2, the separation index (%) is defined as the mass of a given compound in the solid fraction to the mass of that compound in the original (unseparated) slurry. The mass of the separated solids fraction (containing both solids and slurry) ranges between 5.0% and 7.3% [25,26]; the lower value (5.0%) was used in this study. The separated solid fraction (TS = 28% *weight basis*) were assumed to be transported 500 meters (0.5 km) from the separation platform to the drying and conversion facility. Emissions associated with this step were estimated per shipping unit,

i.e., ton-kilometer (tkm) using ecoinvent (v3.4) inventory (transport, tractor, and trailer, agricultural {GLO}| market for | Conseq, U) for farm operations [16].


**Table 2.** Separation indices for mechanical screw press separation [23].

<sup>1</sup> Based on data collected from ref. [23].

#### 2.2.4. Drying

Gasifying organic material requires that the moisture content should be 15% or lower [24,27]. Therefore, the separated solids fraction must be dried first. Drying can be accomplished passively by relying on solar heating and natural air circulation, or through the mechanical circulation of heated air through the wet mixture using blowers or fans. Ideally, for such a system to be efficient, the material is moved inside the dryer to ensure quick and uniform dryness. Passive drying of swine manure solids can be a source of objectionable odors and can reduce the organic carbon content of the material. Therefore, a heated air-drying technique was modeled for this study. The thermal energy required to remove 1 kg of moisture from manure was reported to be 2.3 MJ [23]. Hospido *et al.* [28] evaluated different scenarios for utilizing solid sludge from urban wastewater treatment plants (WWTP) using 1,000 kg of dried sludge as the unit basis. In their study, the electricity and heat consumption associated with sludge drying were 118 kWh and 1,638 kWh per 1 metric ton of dried sludge, respectively.

During drying, as much as 20% of the manure-N was reported to volatilize, typically as NH3 [29]. Also, C loss during drying was reported to be around 4%. In the municipal sludge drying process model, 44.3 g of volatile organic compound (VOC) emissions were reported per ton of dried sludge. The same emissions factor was used to model the manure emissions in this study.

#### 2.2.5. Gasification/Boiler

In this thermochemical conversion process, the dry manure solids are transformed at temperatures between 600 and 800 ◦C to gas (referred to as producer gas, or syngas) in addition to biochar, and a small amount of condensable material (tar) is produced. Gasification utilizes air, or another oxidizing agent, to partially oxidize the biomass C into CO and CO2. However, given the scarcity of the data on the gasification of swine manure solids, the dataset used in this study (Table 3) was compiled from available studies on swine manure solids and feedstock, such as, poultry litter, sewage sludge, cattle feedlot manure, that have similar characteristics as swine manure solids, i.e., high ash, and nitrogen content. The primary product, syngas, is combusted in a steam boiler to generate steam that is used to satisfy heating needs on the farm, e.g., the drying manure solids, and heating the farrowing crates. The syngas displaces natural gas demand and, consequently, the impacts associated with natural gas production. To account for the summer season when the heating is not necessary on the farm, we deducted 25% of the heat production to be claimed as a credit in the computational modeling.


**Table 3.** A gasification process model for 1 kg of dry (15% moisture) swine manure solids.

† HHV: higher heating value.

The cold-gas efficiency in Table 3 is the chemical energy retained in the syngas as a share of the total chemical energy in the feedstock, without considering the gas sensible heat. In this case, however, since the syngas is used to replace a heat source (natural gas), both the sensible and chemical energies in the syngas were considered. Accordingly, the conversion efficiency increases, with thermal gas efficiency (HGE) taken to be between 60% and 90%. A 70% HGE was used in this model. Accordingly, the amount of heat generated (MJ) due to gasification was calculated after subtracting the thermal energy required for the process (calculated using the pyrolysis enthalpy in Table 3 taken here to be 1.0 MJ kg<sup>−</sup>1). Gasification also yields a biochar fraction, which is utilized as a soil amendment. The biochar produced was assumed to be a nitrogen-free co-product since all nitrogen typically devolatilizer during gasification as N-species. P and K were assumed to be sequestered entirely in the biochar fraction. Table 4 presents the emission associated with the current study via the gasification facility for the swine manure solids.


**Table 4.** Emissions to the air resulting from the gasification process (per 1 kg of dry matter of swine manure solids).

#### 2.2.6. Biochar Transport and Land Application

The environmental emissions associated with biochar transport to the field were considered with a transportation distance assumed to be 10 km (6.2 miles), which is slightly more than the upper bound on average manure hauling distances, i.e., between 1.6 and 6.4 km (1 and 4 miles). Biochar land application is beneficial both as a fertilizer/soil conditioner and as a carbon sequestration option [36]. In this study, the benefits of biochar application to the soil were determined as the avoided synthetic fertilizers due to the presence of P and K in the biochar. Nutrient credits were assigned for P and K in biochar as 80% equivalency of commercial fertilizer, according to the Wisconsin study [37]. Additional benefits of biochar application include improved water holding capacity and reduced N2O emissions. However, due to the scarcity of quantifiable data on these benefits and the strong dependence on the crop, soil and climate conditions, these additional benefits are not considered in this study, i.e., no GHG emissions from the biochar field application were considered. The amount of avoided P2O5 and K2O fertilizers, and sequestered CO2 were determined, to be 0.71 kg P2O5, 0.66 kg K2O, and 3.43 kg CO2 per ton of functional unit.

#### 2.2.7. Post-Separation Tank (Liquid-Fraction)

The separated slurry is stored in an exposed tank until it is transported to a field for application. During storage, the organic fraction of this slurry transforms, resulting in GHG emissions. The following section describes the computations for the various emissions. IPCC guidelines [38] were used to estimate CH4 emissions in the swine house, and the estimating method used for the pre-separation tank was also used here to estimate emissions during post-separation storage of the liquid slurry. It should be noted that the volatile solids loading (VS) in this storage step is much lower than in the pre-separation tank, i.e., 21.5 kg per functional unit. Similarly, NH3 and N2O emissions were estimated using emission factors outlined for pre-separation tank emissions.

#### 2.2.8. Liquid Fraction Transport, Mixing with Biochar, and Land Application

The liquid fraction transportation distance to the application field was assumed to be equal to that for biochar, 10 km. This distance has been used before in a similar study to model the impacts of dairy cow slurry digestion and land application [39]. The energy requirement for slurry and biochar mixing is 1.2 kWh ton−1, and the land application energy requirements and emissions were modeled using the vacuum spreader model available in the ecoinvent v3.4 database (Ecoinvent Centre, 2019). The impacts of slurry application (without the biochar) are presented in the following section. Table 5 below presents the summary of inputs and emissions for the functional unit as well as nutrient credits associated with land application of liquid slurry and biochar. The avoided N fertilizer value was calculated from N availability in a liquid slurry, using Delin et al. [39], as 52% equivalency of commercial nitrogen fertilizer.


**Table 5.** Summary of emissions, energy, and transportation requirements as well as an avoided burden for the functional unit.


**Table 5.** *Cont*.

NH3 devolatilization resulting from manure land application is among the primary sources of N emissions in the agricultural sector. Rates of NH3 emissions vary significantly with variability in manure slurry characteristics, soil type, and weather conditions. Misselbrook et al. [40] studied the influence of manure type (cattle, and pig manure), and land type (arable and grassland) on NH3 emissions. They reported NH3 emissions between 6.0 and 21.5% of the total ammoniacal nitrogen (TAN) in the pig manure. Sommer and Hutchings [41] reported NH3 emissions to be 5% of total NH4 in an applied slurry with trail hose application, and 8-10% of total NH4 under broad spreading. According to literature, an estimated 39% of TAN in swine slurry devolatilizes as NH3 during spring season land application [42]. In this study, NH3 devolatilization was modeled as 20% of TAN in the slurry. The TAN was taken from Buckley *et al.* [43] to be 75% of the total N in the swine manure slurry (S.D. = 17%).

According to the Intergovernmental Panel on Climate Change (IPCC) guidelines [18], the emission factor for N2O resulting from organic amendments application (EFN2O) is 0.01 kg N2O-N kg N−1. Rochette et al. [44] estimated the cumulative C loss (as CO2) due to swine slurry application to spring maize plots to be 63% of the original slurry C. In this study, the N and P leaching through the soil profile was assumed as 35% and 10% of manure N and P, respectively [22].

#### **3. Results and Discussions**

#### *3.1. Impact Assessment*

Table 6 presents a summary of the cumulative potential environmental impacts of the swine manure management scenario according to selected categories. Positive impact characterization values indicate an added environmental burden, while negative values represent avoided burden. Detailed descriptions are addressed in the following sections.



\* eq Represents equivalent.

#### *3.2. Global Warming Potential (GWP)*

The proposed manure management scenario has net emissions of 166 kg CO2-eq emitted per ton of swine manure slurry treatment. A detailed representation of the contribution of each stage to the cumulative GHG emissions is shown in Figure 2. Emissions during manure storage under slatted floors in the house and during external storage represented the majority of the GHG emissions, with the two stages contributing 42.1% and 35.1% respectively of the total emissions. This significant contribution is attributed to the high levels of N2O and CH4 emissions during these two steps, with both gases having a significantly higher impact on global warming potential. Similarly, the third-largest contributing stage to GHG emissions is post-separation slurry storage, i.e., 22.4% of scenarios of GWP.

Manure solids gasification and syngas combustion (in a boiler), represented as one coupled process (Figure 2), contribute - 3.54% of the total GWP. The net negative contribution here indicates that the avoided GHG emissions by syngas combustion, instead of natural gas, completely offset the combined emissions from syngas combustion and those associated with gasifier electricity consumption. Even though the low hot gas efficiency, 70%, and the low boiler efficiency, 78%, was used in this model, the overall ratio of avoided natural gas use resulted in a net negative GWP.

GWP for drying manure solids, 6.68 kg CO2-eq, represented 4.02% of overall GWP emissions. Despite being an energy-intensive process, the low GWP contribution here for drying is attributed to the fact that the process heat is recycled from the gasification-boiler output heat, which reduces the overall energy requirement for drying and thus the impact. The following stages: pumping, stirring, separation, and transportation cumulatively contributed 3.49% of the total GHG emissions. Land application of liquid slurry and biochar, which is a co-product of thermal gasification, contributed net negative GHG emissions (- 5.92 kg CO2-eq) due to the credit of displacing synthetic fertilizer. One thing to note is that CO2 emission during land application is accounted for as biogenic CO2 emission. The land application represents a 3.57% reduction of total GHG emissions.

#### *3.3. Fossil Fuel Use*

Cumulative fossil fuel energy use in this scenario was - 58.0 MJ per functional unit. Figure 3 details the individual contribution of manure management stages to overall fuel consumption. Manure storage steps, from an energy perspective, were all-passive and therefore had no fossil fuel expenditure or saving. The maximum energy burden was associated with the drying stage, which represented 62.0% of total fossil fuel energy input, followed by the slurry transportation stage, which represented 24.6% of total fossil fuel energy input. The gasification-boiler stage was attributed with the net negative energy use of - 95.7 MJ, by offsetting natural gas firing to produce the credited amount of thermal energy. The energy demand for the drying process, 107 MJ, represents the electricity demand in the dryer, which cannot be met through the gasification-boiler stage supply. The primary energy saving in this scenario, - 135 MJ, was attributed to the consequences of slurry-biochar land application. This savings is from the avoided synthetic fertilizers and the fossil fuel energy used in their production. For illustration, production of 1 kg N fertilizer requires 88.0 MJ of energy using global unit process of ecoinvent v3.4 database [16]. Similarly, production of 1 kg of P2O5 and K2O require 20.1 and 18.4 MJ of fossil fuel energy in their production.

**Figure 3.** The net contribution of each stage to the cumulative fossil fuel energy use (MJ).

#### *3.4. Water Depletion*

This category indicates the total water use from different water sources: lakes, rivers, and wells. In this study, total water depletion was a process credit, i.e., avoided water depletion of 0.015 m3 per functional unit. This credit is an indirect water-saving resulting from displacing synthetic fertilizers with the slurry-biochar mixture. The difference between land application impacts on water depletion, - 0.112 m3, and total impact, 0.111 m3, is attributed to all the energy-positive stages in the scenario. However, the savings accrued by displaced fertilizers outweighed the combined water depletion potential for these stages.

#### *3.5. Marine Eutrophication*

This mid-point impact category expresses the amounts of nutrients emitted, expressed in units of kg N equivalent, which potentially reach marine water causing eutrophic conditions. The studied scenario had a net positive (a burden) of marine eutrophication, 88.5% of which is attributed to the slurry-biochar land application. This results from nitrate (NO3) leaching, and NH3 emissions following land application. Considering the full lifecycle, 80% of marine eutrophication potential is attributed to NO3 leaching, while the remainder is due to NH3 emissions. The eutrophying effect of NH3 occurs through the formation of acid rains that deposit back in water bodies causing N enrichment. Swine houses and pre-separation storage are together responsible for 5.7% of total marine eutrophication potential due to their NH3 emissions.

#### *3.6. Freshwater Eutrophication*

Given that P is the limiting nutrient for most freshwater bodies, introducing P to rivers and lakes results in eutrophying conditions. In this study, 98.5% of total freshwater eutrophication potential is attributed to the impacts of slurry-biochar application. The leaching of 10% of P from the slurry is responsible for this impact.

#### *3.7. Model Sensitivity to Thermochemical Conversion Parameters*

To improve understanding of the implications of the proposed thermochemical conversion system (drying-gasification-boiler) on swine manure treatment, the conversion parameters, i.e., hot-gas efficiency (HGE) and boiler efficiency was varied to represent two additional alternatives. The first set represents low-efficiency conditions: HGE and boiler efficiencies at 60% and 68%, respectively. The second, a high-efficiency scenario, shows HGE and boiler efficiency at 80% and 88%, respectively. Figure 4 shows the impacts of the performance levels on the gasification-boiler stage. A 10% increase in the performance of both the gasifier and the boiler yielded a decrease in the GWP for this stage by 2.6 kg CO2-eq (from 2.4 kg CO2-eq to -0.2 kg CO2-eq), and a corresponding increase in fossil fuel energy saving by 40.5 MJ (from -11.9 MJ to - 52.4 MJ). A 10% drop in the efficiencies increased GWP, from 2.4 to 4.6 kg CO2-eq, and a change from a fossil fuel energy use of -11.9 MJ to an energy expenditure of 23.5 MJ. The non-linear response in the efficiency scenarios is because the overall efficiency for the gasification-boiler is the product of the conversion and the boiler efficiencies. No noticeable changes were observed in the other impact categories with changes in the efficiencies.

For the full treatment system, increasing the thermochemical conversion efficiency by 10% relative to the baseline led to a 1.5% decrease in GWP and an increase of the fossil fuel savings of 52.4 MJ. These findings suggest that the range of sensitivity for the thermal conversion system has a marginal impact on the GWP for the entire management scenario. It is worth noting, however, that the combined GWP for the separation, drying, and gasification-boiler stages, 0.89 kg CO2-eq, is lower than the difference in GWP between pre-separation storage, 58.2 kg CO2-eq, and post-separation storage, 37.3 kg CO2-eq. The separation-drying-gasification-boiler combination can be considered an emission reduction measure for manure storage. From an energy perspective, the gasification system has a beneficial impact on the total energy use in manure management, notwithstanding high energy

requirements for drying. Improvements to thermal conversion efficiency (gasification and syngas firing) combined with improvements to the drying technology can significantly improve the overall environmental performance for swine manure management via thermochemical conversion.

**Figure 4.** Impacts of gasification-boiler performance on (**A**) global warming potential (GWP 100a), and (**B**) fossil fuel energy use (MJ).

#### **4. Implications of the Study**

The findings in this investigation contribute to the ongoing discussion on manure management best practices. Given the swine manure composition and the management techniques practiced on the farm, the thermochemical conversion is a challenging technique to dispose of wet swine manure. Improvements to the solid-liquid separation system that reduces moisture content in solid fraction can potentially improve the environmental process of the proposed system.

From GHG emissions and energy use perspectives, the land application step of swine manure management appears beneficial due to the credits from replacing synthetic fertilizer consumption. However, in regions of intensive swine production where manure land application regulations are strict, thermochemical conversion can be an alternative approach to utilize manure. Adopting innovative sludge drying technologies, i.e., biodrying technology [16], can significantly reduce the drying energy demand, and consequently, the GHG emissions. Also, more studies towards a better understanding of biochar agronomic value could potentially help in incentivizing the thermochemical conversion of swine manure solids.

#### **5. Conclusions**


**Author Contributions:** All authors contributed equally and significantly to the data collection, manuscript writing, and editing.

**Funding:** USDA National Institute of Food and Agriculture. Agriculture and Food Research Initiative Competitive Grant No. 2011-68002-30208.

**Acknowledgments:** Acknowledgment is due to the Agriculture and Food Research Initiative for their support of the Competitive Grant No. 2011-68002-30208 from the USDA National Institute of Food and Agriculture. Their financial support is much appreciated.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
