Environmental Impact of Agro-Food Industry and Food Consumption

By Charis M. Galanakis

Environmental Impact of Agro-Food Industry and Food Consumption covers trends associated with the impact of food production on the environment using lifecycle analysis and the standard methods used to estimate the food industry’s environmental impact. The book discusses city-scale actions to estimate the environmental impact of food systems, including the meat chain, feeding crops to farmed fish, the confectionary industry, agriculture, tea processing, cheese production, the dairy industry, cold chain, and ice cream production. Food waste and consumption in hospitality and global diets round out these interesting discussions.

Written for food scientists, technologists, engineers, chemists, governmental regulatory bodies, environmentalists, environmental technologists, environmental engineers, researchers, academics and professionals working in the food industry, this book is an essential resource on sustainability in the food industry.

• Addresses all levels of the food chain
• Provides solutions for the food industry to estimate and reduce environmental impact
• Assists members of the food industry in optimizing their current performance and reducing their environmental footprint

• Language: English
• Publisher: Academic Press
• Release date: Nov 18, 2020
• ISBN: 9780128213704

Book preview

Italy

Preface

Charis M. Galanakis¹, ², ³, ¹Research & Innovation Department, Galanakis Laboratories, Chania, Greece, ²King Saud University, College of Science, Riyadh, Kingdom of Saudi Arabia, ³Food Waste Recovery Group, ISEKI Food Association, Vienna, Austria

Sustainability and environmental impact are becoming essential items for the food industry around the world, as resources become more restricted, demand grows, and ecological problems are increased. Industrial food processing ensures that the required resources to produce food products are used most efficiently. In our times the food industry should be as efficient as possible in terms of yield and environmental impact. Besides, recent changes in the legislative frameworks, ecological concerns, and increasing attention toward sustainability have stimulated industry to reconsider their environmental management policy and face the ultimate utilization of their resources. Subsequently, there is a need for a new guide covering the latest developments concerning the ecological impact of different food industries, as well as the environmental impact of food waste.

Food Waste Recovery Group provides insights into the food and environmental science and technology sectors, and this line is publishing different books and reference documents. The books deal with innovations in traditional foods, innovation strategies in food and environmental science, sustainable food systems, nutraceuticals, nonthermal processing, food waste recovery technologies, saving food efforts, bio-based products and bio-based industries, and the valorization of food processing by-products (e.g., from coffee, cereals, meat, grapes, and olives). The group has also prepared handbooks for personalized nutrition, shelf-life, food quality, innovative food analysis, nonalcoholic drinks, and textbooks for specific food components such as dietary fiber, glucosinolates, proteins, lipids, carotenoids, and polyphenols. Following these considerations, the current book covers the environmental impact of the agro-food sector and food industries. The ultimate goal is to support professionals and enterprises that aspire to improve the efficiency of the food industry and to diminish its environmental footprint.

The book consists of 12 chapters. The assessment of the environmental performance of food and beverage production is presently carried out using numerous single- or multienvironmental issue standard methods. In Chapter 1, the human impact on the environment and the basics of the life cycle assessment (LCA) methodologies are briefly outlined. Independently of the number of mid- or end-point impact categories accounted for, all these methods give no more than a partial schematic representation of the environmental impact of a product or activity. They are useful for in-house product improvement and, except for the carbon footprint labeling, cannot be presently used for external communication to environmentally unconscious consumers.

Chapter 2 reviews the environmental impact of organic agriculture. The nutritional value of organic food, and productivity and profitability of organic farms are also addressed. Organic farming has been shown to reduce soil degradation, increase agrobiodiversity, and eliminate pesticides. Chapter 3 presents a systematic review of the studies developed on poultry and pig production, confirming the importance of feeding processes (from crop cultivation to manufacture and transport) as the largest source of environmental impact in the overall chain. Some of these studies reported that improving feeding practices may mitigate the ecological footprint of animal production. In this context, precision feeding is highlighted due to applicability in modern poultry and pig farming.

Chapter 4 focuses on the environmental impacts associated with egg production and industry to provide an update of the global situation and identify the main opportunities for reducing these impacts. The environmental impacts associated with poultry farms can be attributed to three primary sources: feed production, energy use, and emissions from housing and manure management. Also, other factors such as productivity, poultry genetics, mortality rate, and coproducts generation are essential issues to be considered.

Chapter 5 deals with the environmental impact of the pasta industry. Notably, the cradle-to-grave environmental impact of 1kg of dry pasta, produced from a medium-sized pasta factory located in the north of Italy and packed in 0.5-kg polypropylene bags, was investigated by using a well-known LCA software in compliance with the Cumulative Energy Demand, PAS 2050, IMPACT 2002+, ReCiPe 2016, and Product Environmental Footprint standard methods. All these methods allowed durum wheat cultivation and pasta cooking to be identified as the primary and secondary hotspots. Improvements in the cultivation phase were outlined by resorting to conservative farming systems with low N fertilizer inputs. The replacement of the ordinary home gas-fired hobs with a more ecosustainable pasta cooker reduced the damage to climate change and resource depletion.

In Chapter 6, the environmental impact associated with the production of dairy products, along with details of the leading method for estimating the impact, LCA, is discussed. The extensive use of LCA to assess the environmental impact of dairy products is presented, along with an overview of critical studies in this area. The acidification, eutrophication, and global warming potential were found to be the three most assessed environmental impact categories. In addition, a case study investigating the environmental impact associated with producing fluid milk in the Republic of Ireland is presented.

Chapter 7 presents an overview of the food industry with an emphasis on raw materials of ice cream, the processes and methods followed, and their effect on the environment and human health. For example, raw material extraction of this supply chain contributes a significant impact in all the categories of LCA. In contrast the production unit and refrigeration consume high energy, which accounts for the depletion of the ozone layer.

Chapter 8 describes the cheese production system and its associated environmental aspects that can result in potential environmental effects. After that, it discusses the main methods of impact reduction in environmental management following cheese production cases. It also presents the main contributing flows, the analysis, and interpretation of the associated environmental impacts, based on the LCA method. Chapter 9 provides an overview of sugar technology and points out the essential characteristics of generated effluents. Moreover, it deals with the environmental impact of the confectionery industry, explaining the biotechnological utilization of its effluents within the biorefinery concept.

Tea is one of the most widely used drinks in the world, whereas the demand for this product is increasing day by day. However, traditional tea production does not meet this demand. Therefore the chain of tea production consumes a lot of energies and materials to achieve a higher yield of tea in cultivation and processing steps compared with traditional tea cultivation and processing. Chapter 10 provides an overview of tea’s life cycle from production into consumption/waste management before providing insights into the LCA of this supply chain.

The refrigeration of foods has significant impacts on the environment, such as direct emissions through refrigerant leakage and indirect emissions through large amounts of energy consumption. Chapter 11 investigates how cold chains of the food sector contribute to environmental issues and identifies best practices that allow decreasing the environmental impact by lowering energy consumptions and quality losses. These best practices can belong to investments in more efficient technologies or the implementation of maintenance and operational practices as well as better coordination through different actors of the chain.

Chapter 12 deals with the environmental impact of food waste. Since one-third of global produced food is lost or wasted throughout the entire supply chain, the respective environmental impacts that are attributed to these materials are also significant. For the European Union, this accounts for 88 million tonnes (Mt). Due to this high amount, 186 Mt of CO2 equivalents (CO2e) can be related to food wastage in the European Union, which accounts for 4% of the overall European Global Warming Impact. Food waste prevention, for example, by donation preventing especially the amount of meat and dairy waste, may achieve savings exceeding even 1000 kg CO2e t−1 saved food. Also, the valorization of food processing by-products could result in considerable GHG savings up to 500 kg CO2e.

Conclusively, the current book is assisting food scientists, technologists, engineers, and chemists as well as environmentalists, environmental technologists, and engineers, researchers, academics, and professionals working in the food industry. University libraries, institutes, laboratories, food companies, and regulatory bodies could use it as a textbook in undergraduates and postgraduate level multidiscipline courses dealing with food and environmental science and technology, sustainable food consumption, food processing, especially in postgraduate programs.

At this point, I would like to thank and acknowledge and contributing authors for their dedication to this book project. Their acceptance of my invitation, collaboration, and respecting deadlines is highly appreciated. I would also like to thank the acquisition editor Megan Ball, the book managers Laura Okidi and Lena Sparks, and all colleagues from Elsevier’s production team for their assistance during the preparation of this book. Finally, those collaborative editing projects of hundreds of thousands of words may contain gaps or errors. If you have any comments, revisions, or suggestions, please do not hesitate to contact me.

Chapter 1

Standard methods useable for mitigating the environmental impact of food industry

Mauro Moresi, Matteo Cibelli and Alessio Cimini,    Department for Innovation in the Biological, Agrofood and Forestry Systems, University of Tuscia, Viterbo, Italy

Abstract

The assessment of the environmental performance of food and beverage production is presently carried out using numerous single- or multi-environmental issue standard methods. All of them are compliant with the life cycle assessment (LCA) methodology but diverge along the number of impact categories and materials covered, the models used to characterize the environmental impact, and even for the free use of different normalization and weighting factors.

In this chapter the human impact on the environment and basics of the LCA methodology and main life cycle impact assessment methods are briefly outlined. Independently of the number of mid- or end-point impact categories accounted for, all these methods give no more than a partial schematic representation of the environmental impact of a product or activity. They are useful for in-house product improvement and, except for the carbon footprint (CF) labeling, cannot be presently used for external communication to environmentally -unconscious consumers.

Whether the life cycle impact of foods and beverages is mostly affected by fossil cumulative energy demand (CED) and consequently by greenhouse gas emissions, the CED indicator or CF appears to be the most straightforward and cost-effective method usable by 99% of the European food and beverage enterprises to start improving their sustainability, even with the help of secondary instead of primary data. Any additional environmental impact category or alternative method can be included, primarily if the LCA modeling is carried out using the LCA software nowadays available.

Keywords

Cumulative energy demand; environmental product declaration; food and drink industry; IMPACT 2002+; life cycle assessment; life-cycle impact assessment methods; product carbon footprint; environmental product footprint; ReCiPe 2016

1.1 Introduction

Despite there is no clear consensus on the contribution of food to global anthropogenic greenhouse gas (GHG) emissions, a recent literature review estimated that global food consumption was responsible for 10.8–18.1 Tg CO2e in 2010, that is, 22%–37% of global anthropogenic emissions (Rogissart et al., 2019). The vast majority (72%–82%) of such emissions derived from the production phase (agricultural production and land-use change), including not only the emissions generated on-farm but also those linked with the manufacture of fertilizers, pesticides, equipment, and energy. The GHG emissions resulting from the post-production or post-sale steps were of minor magnitude with 2.4 or 1.0 Tg CO2e, respectively. Waste management represented the smallest emission source, while emissions embedded in food wasted at the consumption phase were no way negligible (c.1.6 Tg CO2e). About 63% of food-related GHG emissions derived from the production and consumption of animal-based products, except fish and fisheries, with 8.5±2.4 Tg CO2e in 2010 (Rogissart et al., 2019).

According to Sandström et al. (2018), food consumption of European Union (EU)-28 citizens would have generated 540 Gg CO2e in 2010, including land-use changes, of which about 160 Gg CO2e derived from land-use changes in the importing countries. The average EU production- and trade-related dietary emissions from food supply were of the order of 1070 kg CO2e cap−1 year−1 (Sandström et al., 2018). In detail, the consumption of meat and eggs covered the largest share (49%–64%) of food supply emissions, followed by that of dairy products (16%–36%); cereals, rice, and maize (2%–8%); conventional and energy drinks (<5%); and vegetable oils (<5%).

Owing to the predicted rise in global demand for food (FAO, 2009), the growth perspectives of the food and beverage industry are challenged by climate-related risks to food security, water supply, and economic growth, as well as increasing insufficiency for natural resources (NR) and costs for fuels and energy (IPCC, 2014).

Such an industry is the leading industrial sector in the EU with a critical role in terms of turnover and the total number of workers and enterprises (FoodDrinkEurope, 2019a). About 99% of them are small- and medium-sized enterprises (SMEs), the vast majority of which are not prepared to face fast-changing, complex global challenges to sustainable development.

This chapter first aims at reassessing the human impact on the environment and basics of the life cycle assessment (LCA) methodology briefly. Then, among the numerous life cycle impact assessment (LCIA) methods available, it suggests which one might be regarded as being adequate for improving the sustainability of the food and beverage SMEs initially and supporting Europe to become the first climate-neutral continent in the world by 2050 (FoodDrinkEurope, 2019b).

1.2 Human impact on the environment

The anthropogenic impact on the Earth’s environment is multiple. For instance, it includes human reproduction, overconsumption, overexploitation, pollution, burning fossil fuels, and deforestation. Some human activities may directly or indirectly cause damage to the environment, such as global warming and environmental degradation, including ocean acidification, soil erosion, biodiversity loss, air pollution, and undrinkable water. Such adverse impacts can pose an existential risk to the human race, its intensity becoming worse as the problem of human overpopulation continues.

Life on Earth is sustained by the following four-planet subsystems which are strictly interdependent: (1) the lithosphere that coincides with the external layer of the planet, (2) the hydrosphere that includes the three states of water, (3) the atmosphere that forms the gaseous layer surrounding the Earth, and (4) the biosphere that accounts for all living organisms, as well as the dead ones not decomposed yet. These subsystems interchange mass and energy through the so-called biogeochemical cycles (e.g., the carbon, nitrogen, oxygen, phosphorous, sulfur and water ones) within the general constraint that the Earth is both an open system that receives solar radiation during the day and disperses energy toward space at night, and a closed one in terms of mass (Morawicki, 2012).

The human population has increased from about 0.275 billion to 1.6 billion since the year 1000 up to the dawn of 1900, is now about 7.7 billion, and will raise to 9.4 billion in 2050 (Worldometer, 2020). Such an exponential increase in the human population has resulted in uncontrolled use of nonrenewable resources and alteration of the aforementioned cycles with loss of terrestrial and aquatic biodiversity and arable land, increase in the atmospheric concentration of GHGs, and likely effects on climate change (CC). The global volumetric concentrations of CO2, CH4, and N2O have, respectively, increased from about 370 to 410 ppm, from 1771 to 1877 ppb, and from 316 to 332 ppb over the last two decades [NOAA (National Oceanic and Atmospheric Administration), 2020].

In these conditions, permanent growth seems to be unsustainable.

Any change in human production and consumption activities exerts quite complex either negative or positive environmental effects or pressures on the subsystems mentioned before, which can be classified into four different categories (Allinson et al., 2013):

1. materials

2. water

3. land

4. carbon and air emissions.

The pressure category of materials includes both renewable (biotic) and nonrenewable (abiotic) resources. The former takes account of the harvested products from agriculture, forestry, and fishery, while the latter fossil fuels, metallic minerals, and industrial and building minerals.

The pressure category of water accounts for the water abstracted and consumed to support human activity.

The land use category comprises agricultural, forest, and grazing areas, along with any surface used for raw material extraction, infrastructures, manufacturing, or private housing (EEA, 2010). Changes in land use owing to the urbanization of agricultural or forest land are also accounted for.

Finally, the pressure category of carbon and air emissions mainly refers to the so-called GHGs, namely, carbon dioxide, methane, nitrous oxide, hydrofluorochlorocarbons, perfluorinated chemicals, and SF6. These emissions are yearly listed in the national inventory report and published on the United Nations Climate Change website [UNCC (United Nations Climate Change), 2019]. Also, human activities produce several other contaminants (e.g., SOX, NOX, and ozone) that damage human health (HH), crops, and materials. For instance, in Italy, the GHG emissions reached about 427.7 Gg CO2e in 2017, these being chiefly made of CO2, and then of CH4 and N2O [ISPRA (Institute for Environmental Protection and Research), 2019]. The energy sector with 345.9 Gg CO2e was by far the foremost GHG source, while the industrial, agriculture, and waste sectors emitted about 32.8, 30.8, and 18.2 Gg CO2e, respectively. The category of land use, land-use change, and forestry acted as the main GHG sink (−18.4 Gg CO2e). More specifically, the GHG emissions of the agriculture sector mostly derived from CH4 (as due to enteric fermentation of cattle, animal manure management, and rice farming), and N2O emitted from fertilized agricultural soils and manure management. Those from the industrial processing were mainly due to the iron and steel industry, followed by the chemical, pulp, paper, and print ones.

The human production and consumption activities involve several interlinkages between such pressure categories, and their overall environmental consequences may be somewhere assessed by accounting for a series of different environmental impact categories. According to Finnveden et al. (2009), such impacts are generally attributed to three areas of protection (AoP), namely, ecosystem quality (EQ), HH, and NR. They are generally evaluated using the best relationships at the time available.

Fig. 1.1 shows a rough draft of the environmental impact assessment mechanism, explaining how a substance emitted undergoes several repercussions causing damages to any AoP. For instance, the emission of CO2 or CH4 into the air would first cause an increase in the absorption of infrared radiation by the atmosphere and the temperatures of the air, water, and soil compartments with a significant rise in the sea level. The fate processes include both the degradation and transport of the gas in the troposphere, stratosphere, water, and soil compartments up to the AoP (Finnveden et al., 2009). In Fig. 1.1, endpoint indicators refer to the AoP, while midpoint ones to impacts brought to bear on areas ranging from the emission to the endpoint. These types of indicators are responsible for the vast majority of difference among the currently available LCIA methodologies, as pointed out later.

Figure 1.1 Scheme of the general LCA modeling used to describe how a substance emitted undergoes a series of 1, 2, …., n impacts before damaging the environmental AoP, as well as the difference between midpoint and endpoint, as adapted from Finnveden et al. (2009). AoP, Areas of protection; LCA, life cycle assessment.

Table 1.1 shows a brief description of the leading environmental impact categories, including the areas potentially damaged (HH, natural ecosystems, NR), and scales of application.

Table 1.1

GHG, Greenhouse gas.

Source: Derived from Allinson, R., Arnold, E., Cassingena Harper, J., Doranova, A., Giljum, S., Griniece, E., et al., 2013. Assessing Environmental Impacts of Research and Innovation Policy. Study for the European Commission, Directorate-General for Research and Innovation, Brussels; EC, 2011. ILCD Handbook: Recommendations for Life Cycle Impact Assessment in the European Context, first ed. Publication Office of the European Union, Luxemburg. https://doi.org/10.278/33030 (EC, 2011); Morawicki, R.O., 2012. Handbook of Sustainability for the Food Sciences. Wiley-Blackwell, Chichester, UK.

Such impact categories are generally used to offer a panoramic view of the environmental impact of the food supply chain, as reported below.

1.3 Life-cycle assessment methodology

The first attempt at LCA can be traced as early as the 19th century with the coal question (Hofstetter, 1998). In the late 1960s, there was renewed interest in LCA-like tools at the US Department of Energy to deal with the packaging wastes and energy crisis. In 1969 the Coca Cola Company started to measure the consumption of resources (i.e., packaging materials and fuels) and environmental releases associated with different beverage containers (Curran, 2006). Such an inventory analysis was the basis for several other studies in the United States and Europe and led Boustead and Hancock (1979) to publish the first handbook for industrial energy analysis. As the interest in the oil crisis faded, such types of study were extended to analyze the environmental problems associated with solid waste disposal on a global scale (Menoufi, 2011). The SETAC (Society of Environmental Toxicology and Chemistry) (1991, 1993, 1998) introduced another step in the LCA methodology, namely, the impact assessment one. Moreover, in 1997 the International Standards Organization (ISO) started to regulate LCA methodology, which was lastly reviewed in 2006 (ISO, 2006a,b). Nowadays, LCA is used in several industrial, buildings, agriculture, and food sectors to support business and Research & Development strategies, product or process design, product labeling, and declarations (Cooper and Fava, 2006), as well as to compare products within the same category. LCA is nowadays quite a proper and reliable tool for assessing an environmental load of a product, process, or activity during its life cycle from the cradle to the grave (Minkov et al., 2016). Moreover, it allows the identification of any mitigation action to reduce the environmental impacts of products or processes and limit their resource use throughout all life cycle steps. Fig. 1.2 shows the main stages in a generic product life cycle, namely, the supply of raw materials, their conversion into the final product, which is distributed up to sales points to be used by the consumer with the final disposal of food scraps and packaging materials.

Figure 1.2 Schematic of the life cycle stages of a typical product, as adapted from BSI (2008b).

The working procedure of the norm ISO 14040 (ISO, 2006a) entails the following four phases (Fig. 1.3):

1. goal and scope definition

2. life cycle inventory (LCI) analysis

3. LCIA

4. life cycle interpretation.

Figure 1.3 Steps of the LCA procedure. LCA, Life cycle assessment. Adapted from ISO, 2006a. 14040—Environmental Management Life Cycle Assessment – Principles and Framework. International Organization for Standardization, Genève and Morawicki, R.O., 2012. Handbook of Sustainability for the Food Sciences. Wiley-Blackwell, Chichester, UK.

The first phase defines the following: aims of the study, system boundaries, functional or reference unit, allocation procedures, assumptions, and limitations. Generally, the most used functional units in food and agricultural products are mass (kg) or volume (L) of the final product, the content of energy (kJ), or protein content per serving (Cimini et al., 2019; Cimini and Moresi, 2016; McAuliffe et al., 2020; Roy et al., 2009; Sonesson et al., 2017). As concerning the system boundaries, LCA may deal with the cradle-to-farm gate, cradle-to-retail, cradle-to-consumer or cradle-to-grave concept in order to represent the product life cycle from the cultivation phase to the farm gate, to transformation, packaging and distribution phases, to the consumption (plate) or post-consumption phase to dispose of food scraps and packaging wastes (Fig. 1.4).

Figure 1.4 Different specifications for the system boundaries for an LCA study of a food product. LCA, Life cycle assessment. Adapted from Röös, E., Sundberg, C., Hansson, P.-A., 2014. Carbon footprint of food products. In: Muthu, S.S. (Ed.), Assessment of Carbon Footprint in Different Industrial Sectors, vol. 1. Springer, Singapore, Singapore, pp. 85–112 (Röös et al., 2014).

The second phase is the application of mass and energy balances on the system boundaries previously defined. All the input and output data are to included. In particular, one has to account for the consumption of renewable and nonrenewable energy, raw and packaging materials, water, as well as formation of the final product, coproducts, by-products, emissions to air (CO2, CH4, SO2, NOX, CO) and from managed soil (NH3, NO3−, NO2), effluents (whey, olive vegetation water) in terms of total suspended solids, biological and chemical oxygen demand, chlorinated organic compounds, and solid wastes (municipal solid wastes, food processing residues, such as citrus or tomato pomace). This phase is by far more laborious and onerous than the other phases in an LCA study, essentially because of the burden of data collection (Roy et al., 2009). Data collected or measured directly by the company are the so-called primary data, whereas those extracted from the technical literature, software, and LCA libraries are secondary data. For product-specific LCA studies, site-specific data are to be used. On the contrary, data on transport, production of packaging materials, and electricity, as well as waste material disposal can be derived from commercial LCA databases (Roy et al., 2009).

During this step, data needs to be validated and referred to the functional unit that was defined during the goal-and-scope setting. In the case of energy, all sources, such as fuels and electricity, need consideration, including efficiencies and losses. Generally, the results of the inventory analysis are to be expressed by referring to the functional unit and reported in a table accounting for all the input and output data of the different life cycle phases included within the system boundaries. As an example, see Table 5.3 regarding the inventory analysis for the Italian production of the Yellow Label pasta by Sgambaro (2014).

The life cycle of the great majority of food products may involve multifunctional processes with multiple outputs. Exempli gratia, dairy cows yield not only milk but also meat, wheat cultivation gives rise to grains and straw, and dairy processing may produce milk powder, cream, cheese, and whey. In such cases, ISO 14044 (ISO, 2006b) suggested expanding the system boundaries to include the coproducts or to increase the level of detail in the life cycle. When both approaches cannot be applied, allocation problems arise. Thus, the environmental burden caused by multiple processes or products may be partitioned by referring to physical relationships between the inflows and outflows of the system or economic allocation in the proportion of the relative market prices of products and by-products. In the case of food products, weight, volume, and economic values are the allocation criteria most commonly used. In other cases, economic allocation appears to be more suitable as in the concurrent recovery of platinum, palladium, and gold from cyanide leach solutions or gold recovery from copper-zinc ores. The worst case would be the attribution of the overall environmental load to gold recovery only (Mogensen et al., 2009) or wheat grains when the straw is not harvested but left in the field. It is also possible to account for multiple allocation criteria, even if the ISO norm 14044 (ISO, 2006b) suggested to carry out a sensitivity analysis aimed at comparing the pros and cons of the different allocation coefficients generated using physical- or economic-based allocation criteria.

The third phase in LCA relates the outcomes of the LCI to their potential impact on HH, environmental degradation, and resource exhaustion. Following ISO norm 14040 (ISO, 2006a), this phase includes the following steps:

1. Selection of the most appropriate environmental impact categories, category indicators, and their characterization models for the goal and scope of the LCA. Table 1.1 lists the most common midpoint impact categories involved in the environmental impact assessment of food-production systems.

2. Classification of the environmental impact by assigning the LCI results in the impact categories of choice. For instance, in this step, the amount of GHGs emitted to the air is allocated to the impact category climate change.

3. Characterization of the environmental impact by calculating each jth category indicator (ICj) using the following aggregation equation:

(1.1)

where Ψi is the generic ith emission to air, water, and soil, and Fi,j the relative characterization factor, according to the characterization model at the moment retained as the most proper and reliable one for the impact category of concern. In the case of CC the amount of each GHG emitted is multiplied by its most updated global warming potential to express its contribution in kg of CO2e to the impact category climate change. Impact categories may have global, regional, or local scales (Table 1.1). The characterization step allows the LCI results of products or processes to be directly compared by assessing the relative impacts for each impact category.

4. Presentation of characterized data employing an LCIA profile using the category indicators selected, or an inventory of elementary flows not assigned to any impact category.

5. Optional features of an LCIA, such as normalization, grouping, weighting, and data quality (ISO, 2006a). The characterized impact score of the jth impact category (ICj) may be normalized concerning a standard reference (ICRj), such as the impacts caused by one person living in Europe or the world during 1 year, via the following equation:

(1.2)

where ICRj is also defined as the jth normalization factor.

In this way, it is possible to compare the normalized scores of all the impact categories accounted for.

Grouping consists of selecting the indicators according to some criterion, such as the local, regional, or global impact.

Weighting represents a way of calibrating the relative importance to the impacts considered in the LCA employing different specific weights. It is generally of the linear type according to the following equation:

(1.3)

where EI is the overall environmental impact indicator, βj and Nj are the weighting factor and normalized indicator for the jth impact category, respectively.

As concerning data quality, site-specific data are required for raw materials, water and energy inputs, air emissions, effluents, and solid wastes generated. Moreover, at least 95% of the input materials and energy sources are to be taken into account in the LCI.

The fourth phase aims at discussing the results of the LCI and LCIA, and drawing conclusions, and suggestions. According to ISO (2006a), it involves the following:

1. Identification of the most significant issues resulting from LCI and LCIA steps. For instance, the inventory data may point out the importance of the energy source used, emissions, wastes formed; while the impact assessment the most relevant impact categories involved, as well as the main hotspots in the product life cycle (namely, the cultivation phase and processing).

2. Evaluation checks for the trustworthiness of the LCA results by assessing the completeness of data taken into account, result sensitivity to specific variation in the input-output data, and consistency check concerning the goal and scope.

3. Conclusions, limitations, and recommendations for the audience of the LCA.

Finally, the guidelines provided by ISO (2006a) ask for an impartial, transparent, and detailed reporting of the LCA results, the interpretation of which is to be correspondent with the goals of the study. Any restatement of the goal and scope, system boundaries, functional unit, criteria for allocations, data, choice of impact categories, and category indicators must be included in the report. A graphical representation of the LCI and LCIA results is recommended, but not mandatory.

1.4 State-on-the-art of life cycle impact assessment methodologies

The standardization of the LCA methodology (ISO, 2006a,b) has not given rise to a worldwide accepted LCIA method yet. All methods currently available are ISO compatible, but they provide different results in a consequence of various factors, namely,

• the number of impact categories accounted for;

• the number of compounds included;

• the criteria and models required for the characterization, normalization, and weighting steps;

• the regional, continental, or global validity; and

• the temporal validity of the data used in the modeling.

Any attempt to identify a recommended best practice has been so far unsuccessful.

A few LCA-compliant standard methods refer to a single environmental area of protection only, while other ones account for more than one impact category. Generally, each single methodology refers to a database including the majority of essential compounds and materials usable in the LCA studies. Specific software tools, such as Gabi (www.gabi-software.com) and SimaPro (https://simapro.com/), are available to perform the LCIA phase in the LCIA method of choice.

Within the first group of LCIA methods the following ones can be mentioned:

• The cumulative energy demand (CED) method was developed in the early 1970s after the first oil price crisis (Boustead and Hancock 1979; Pimentel 1973) to measure nonrenewable and renewable energy sources directly and indirectly consumed along the product life cycle (Frischknecht et al., 2007). The only category indicator is the CED one that is expressed in MJ equivalents (MJe) and can be split into the nonrenewable (e.g., fossil and nuclear) and renewable (e.g., hydro, solar, wind, geothermal, and biomass) CED indicators.

• The publicly available specification (PAS) 2050 method was developed by the British Standard Institute under the sponsorship of the Carbon Trust and UK Department for Environment, Food and Rural Affairs to assess the life cycle GHG emissions of goods and services BSI (2008a,b). This method makes use of the single impact category climate change, and disregards other potential social, economic, and environmental impacts that may be associated with the life cycle of products. This method allows the GHG emitted by different products to be compared, and such information to be communicated. It was revised in 2011 (BSI, 2011) to reflect advances in theoretical knowledge and practical experience of PAS 2050 users since the method had been published. The PAS 2050 methodology was aligned as much as possible with other international footprint methods. Besides, the following significant changes were introduced: (1) provision for the development and application of additional requirements to enable a more specific assessment of GHG emissions within sectors or product groups, (2) the inclusion of emissions from biogenic sources (e.g., biomass), and (3) greater clarity on the treatment of recyclable materials.

• The Bilan Carbone is an LCA methodology developed in 2004 by ADEME (i.e., the French Agency for the Environment and Energy Management) for quantifying the GHGs directly and indirectly emitted along the life cycle of goods, services, or districts. This method is today coordinated and disseminated by the Association Bilan Carbone (https://www.associationbilancarbone.fr/). By the end of 2013, more than 8000 Bilan Carbone inventories were carried. A ready-to-use Excel spreadsheet allows the GHG emissions deriving from the process under study (e.g., building heating, manufacturing energy consumption, freight shipments, passenger travel, production of raw materials, and waste treatments) to be easily calculated, ranked by source, and used for adopting appropriate mitigation actions. The methodology guide (ADEME, 2007a) and emission factor manual (ADEME, 2007b) can be freely downloaded from the ADEME website (https://www.bilans-ges.ademe.fr/en/accueil/contenu/index/page/calculation_methods/siGras/0).

• To estimate the GHG emissions, three distinct areas are considered. The Internal area regards the GHGs emitted in consequence of the use of energy sources (fuel, electricity, and gas) and exogenous CO2, as well as leakages of refrigeration fluids. The Intermediate area includes the GHGs emitted along with the manufacture of all raw, ancillary, and packaging materials acquired. Finally, the Global area concerns the transportation of all materials acquired, as well as the disposal of solid wastes and wastewaters. The 100-year time-horizon global warming potentials coincided with those released by IPCC (2001). As compared to PAS 2050, the Bilan Carbone results are expressed in kg of carbon equivalents that can be straightforwardly converted in kg CO2e, the transport of consumers and employees from home to sale points and from home to a place of work is included, and allocation should be based on physical rules in the case of multiple outputs.

• The GHG Protocol is an LCA method established in 1998 by the World Business Council for Sustainable Development and World Resources Institute in partnership with companies, nongovernmental organizations (NGOs), and state representatives (Bhatia et al., 2011). The standards included detailed guidelines, which can be freely downloaded from the GHG Protocol website (www.ghgprotocol.org). Emission sources are classified by scope. In detail, Scope 1 comprises all direct GHG emissions from internal sources (e.g., mobile machinery, stationary equipment, enteric fermentation, manure management, crop residue combustion, and soil carbon). Scope 2 accounts for all the emissions deriving from the consumption of purchased electricity and steam, generation of electricity, heating, and cooling. Scope 3 refers to all indirect emissions excluded from Scope 2 (e.g., product transport, production of chemicals, and packaging materials). The impact score is finally estimated by accounting for the 100-year global warming potentials extracted from Forster et al. (2007). The GHG Protocol allows public tracking of the GHG inventory for a given product, as well as its emission reductions.

• The Australian Wine Carbon Calculator is a tool derived from the GHG Protocol and using the Australian Government-endorsed emission factors (AWCC, 2012). It was sponsored by the Winemakers’ Federation of Australia (now Australian Grape & Wine), South Australian Wine Industry Association, and Winegrape Council of South Australia. It was established in 2009 to measure the Australian winery carbon footprint (CF) at the site or facility level. The methodology guide, as well as the ready-to-use Excel spreadsheet, can be freely downloaded from the Australian Wine Research Institute website (//www.awri.com.au/industry_support/sustainable-winegrowing-australia/carbon-calculator/). The revised Excel spreadsheet refers to the latest versions of Australian National Greenhouse Gas Accounts (Australian Government, 2015) and 100-year global warming potentials (Myhre et al., 2013). Moreover, such a spreadsheet does not attempt to estimate offsite Scope 3 emissions (packaging, distribution, and oak), since they can be calculated through other accepted LCIA methodologies.

The following LCIA methods point out specific environmental aspects of human production and consumption activities, namely,

• The ecological footprint (EF) is a measure of human demand on the Earth’s ecosystems, that is, the amount of land and sea areas necessary to supply the resources a human population consumes and assimilate resulting wastes. It was conceived by Rees (1992) but later detailed in a book by Wackernagel and Rees (1996). The method was developed through a consensus, committee-based process by the Global Footprint Network. The EF Standards are designed to ensure that all EF studies, including subnational populations, products, and organization, are accurate and transparent in terms of data sources, conversion factors, study boundaries, and communication of findings (Lin et al., 2016, 2019). The EF calculation accounts for the following components: (1) energy land, namely, the forest land needed to sequester anthropogenic CO2 emissions generated by the production of goods or services; (2) cropland, that is the area required to cultivate all crops consumed by humans and livestock; (3) grazing land, that is the area needed to feed livestock; (4) forest land, that is the area required to supply wood for fuel, construction, and paper; (5) built-up land, that is the area occupied by human activities; and (6) fishing ground, that is the water area supporting sustainable fish catching for food. Similarly to LCA, this method converts the demand for energy, biomass (food and fiber), building materials, and water involved in resource production and waste assimilation into a normalized measure of land area (i.e., global hectares) using appropriate global average yields and equivalence factors (Borucke et al., 2013; Kitzes et al., 2007; Lin et al., 2016, 2019).

• The water footprint (WF) is a global calculation standard developed in 2009 by the Water Footprint Network. It was conceived in 2002 by Hoekstra as an indicator alternative to water use and later refined by Chapagain and Hoekstra (2004) and Hoekstra and Chapagain (2008). The WF represents the total volume of freshwater used to produce the goods and services consumed by an individual or a community or produced by the business. It includes three items. The volume of freshwater sourced from surface or groundwater resources denotes the so-called blue water footprint. It may be evaporated and incorporated into a product or taken from one body of water and returned sooner or later to another one. The green water footprint coincides with the volume of water from precipitation that is stored in the root zone of the soil and evaporated, transpired, or incorporated by plants. Finally, the grey water footprint is the volume of freshwater needed to dilute pollutants to meet specific water quality standards. Grey water refers to virtual water consumption. The WF is a geographically explicit indicator since it measures the volumes of water used and polluted locally. WF assessment is a four-phase process (i.e., goals and scope, accounting, sustainability assessment, and response formulation) that estimates the green, blue and grey WFs (Hoekstra et al., 2011), assesses the sustainability of water use by identifying, and prioritize the most effective mitigating actions for reducing the WF.

As concerning the multienvironmental issue LCIA methods, the environmental load of products or systems may be assessed by selecting different impact categories by two different schools of thought (Jolliet et al., 2003), such as:

1. The classical impact assessment methods, also known as problem-oriented methods or midpoint methods. Such methods try to limit uncertainties by estimating the category indicators at the first stages in the cause–effect chain, which are clustered in the so-called midpoint categories according to selected themes (e.g., CC and eco-toxicity).

2. The damage-oriented methods or endpoint methods, where the cause–effect chain is modeled up to the endpoint, concerning the high uncertainty levels.

By referring to the International Reference Life Cycle Data (ILCD) Handbook (EC, 2010), the following LCIA methodologies fall within the first group:

• The Centre of Environmental Science (CML) 2002 is a midpoint-oriented method developed in 1992 by the CML at the University of Leiden (the Netherlands). It included both obligatory and optional impact categories with global validity, except for acidification and photo-oxidant formation that referred to Europe (Guinée et al., 2001). Such a method relied on the characterization factors of about 800 substances, often for more than one impact category.

• The Environmental Design of Industrial Products (EDIP) was developed in 1996 by the Institute for Product Development at the Danish Technical University (Wenzel et al., 1997) and improved in 2003 (Hauschild and Potting, 2005). The EDIP97 method normalized and weighted environmental impacts based on political, environmental targets. EDIP2003 accounted for regional information on nonglobal emission-related impact categories at the midpoint (photochemical ozone formation, acidification, eutrophication, ecotoxicity, human toxicity, and noise). This method included ~500 substances, their characterization factors regarding often more than one impact category.

• The Tool for the Reduction and Assessment of Chemical and other Environmental Impacts (TRACI) was developed by the US Environmental Protection Agency (EPA) to characterize the environmental conditions in the United States as a whole or per state (Bare et al., 2003). Such a method mostly referred to the impact categories used in the Building for Environmental and Economic Sustainability method, developed by the National Institute of Standards and Technology of the US Department of Commerce and widely applied in the US building sector (Kneifel, 2018). The normalization factors were based on US emissions and resources in the year 1999. Although the original TRACI method included over 900 substances, the updated version 2.1 covered up to 3000 substances [EPA (Environmental Protection Agency),