Sustainable agri-food processes and circular pathways in a life cycle perspectives: Where is applicative research going?

This study aims at providing a systematic and critical review on the state-of-the-art of life cycle applications from the circular economy point of view. In particular, the main objective is to understand how researchers adapt life cycle approaches for the measurement of the empirical circular pathways of agri-food systems along the overall lifespan. To perform the literature review, PRISMA protocol was considered to conduct a review by qualitative synthesis. Specifically, an evaluation matrix has been set up to gather and synthesize research evidence, by classifying papers according to several integrated criteria. The literature search was carried out employing scientific databases. Findings evidence that the most common circularity topics are about closed-loop production systems, i.e. nutrient recovery for agricultural purposes, production of renewable energy, valorization of residues and wastes as fertilizers, food waste, and agro-wastes recycling for agriculture. To evaluate the benefits/impacts of CE strategies, Life Cycle Assessment (LCA) proved to be the most common methodology applied by authors, as it can help to meet the main CE requirements slowing and closing resource loops.


Theoretical background of circular economy
Circular Economy (CE), also intended with the synonymous "circularity", is an expression that, although it is now widely used and known, remains shrouded in an aura of mystery, especially if the intent in its use is to grasp the most practical advantages in its application. For this reason too, but not only, discussion themes about CE are strongly explored at various levels and from different perspectives by researchers, academics, politicians, practitioners, and entrepreneurs. The CE concept, which can be dated from the original and renowned idea of "closing circle" [1], has been brought back to the forefront in 2010, thanks to the popular activity of Ellen MacArthur Foundation [2] which reprised, among others, the most recent cradle-to-cradle approach [3]; since then, insights on CE never stopped moving forward. However, as Borrello et al. [4] argued, the originality of new contributions is not always clear and this risks making the concept even more disorienting, although it is shareable to consider CE nowadays as a necessary concept precisely because it is still "essentially contested" [5]. One of the key issues that make the CE's discourse particularly complex is the understanding of the link between circularity and sustainability [4; 6; 7] and, although it is quite shared the vision of CE as an effective way to achieve some of the sustainability goals, often the boundaries of these two overblown terms are not so see-through and this risks to blur their meaning as in a real tangle of buzzwords [8; 9]. In concise and effective terms, the CE model, opposed to the linear economic model, would reduce and/or avoid resource depletion, wastes, and other greatest concerns is the combination of different assessment methods and merging their results in a suitable and believable way. Furthermore, evaluating CE strategies should require a systemic and synergistic approach by considering the agri-food supply chain as a whole, especially to not incur the risk of making effective only one stage nor only single portions, while neglecting the others [19; 20]. This would mean, for example, to include the analysis of pre-production and consumption stages, but also of co-products markets, and other secondary supply chain articulations. To satisfy these purposes, sustainability evaluation methods and, among them, the life cycle (LC) approaches, are particularly appreciated as a robust, science-based, and useful tool not only to measure, but also to validate CE assumptions, to help the feasibility of its implementation getting feedback for improvements and, finally to communicate innovation strategies [21]. Especially Life Cycle Assessment (LCA) is widely regarded to be the tool "par excellence" when it comes to evaluating the environmental impacts of circular based products or systems [13; 22]. The flexibility of LCA is also well appreciated, principally because allows to incorporate it in several other metrics by demonstrating feasibility and usefulness for CE purposes [23]. However, all LC methodologies, LCA (or eLCA), Life Cycle Costing (LCC), Social Life Cycle Assessment (sLCA), while are obtaining a growing consensus in the appraisal of different agricultural and food systems by measuring environmental, economic, and social impacts, separately or jointly, lastly required also to be systemic, multidisciplinary, and multicriterial. In these terms, the use of an LC framework, able to capture potentially all sustainability dimensions, can be adapted to evaluate CE strategies operationally and comprehensively, by shifting from the typically "from cradle to grave" to "from cradle to cradle" circular vision. Furthermore, to avoid partial and compartmentalized analyses in CE context, Life Cycle Sustainability Assessment (LCSA) [24; 25] is also recommended by Niero and Hauschild [19], since it could suggest elements of integration among sustainability dimensions, life cycle stages and interdependent subjects of the supply chains by preventing or avoiding burden shifting.

Goal and scope of the review
To the best of our knowledge, within the extensive scientific literature that investigated definitions of CE, discourse typologies, applications, and measurements, no recent review has explored the use of Life Cycle (LC) approaches to measure the impacts deriving from the implementation of CE strategies in the context of agricultural and food productions. Only two reviews addressed a somehow related theme) [17; 26], but the first is mainly focused on general trends in CE research related to the agri-food sector, while the second is particularly centered on bioenergy agricultural practices. Both the studies mention LC as a tool especially liked and appropriate for CE purposes without arguing about the limitations and advantages of using the tool to explain the impacts of circularity pathways.
Therefore, the originality in the scope and approach of this review is to understand how and how much the LC-based analysis is useful to evaluate if CE strategies are more sustainable than linear/traditional economic models in agri-food production systems. To address this issue, the following research questions will be answered: (1) how researchers apply LC methods to evaluate environmental, economic, and social consequences of agrifood circular processes? (2) How LC methods are combined with other approaches to CE measuring? (3) Have impact results been used to increase understanding of the sustainability implications of CE strategies? This study aims to contribute to the research on CE implementation providing an understanding of the role in life cycle approaches to measure the effectiveness of CE strategies for improving agri-food production processes sustainability. A visual diagram is provided in Figure 1 to summarize how the CE vision, i.e. the well-known butterfly diagram by [2], can be brought back to a life cycle perspective through the necessary flow of data and information to measure environmental, economic and social impacts. The paper framework is the follows: the next section describes the research methodology used in this study to conduct the systematic and critical literature review. Section 3 presents the results in terms of the main criteria used in the analysis. Section 4 and 5 argue the discussions concerning the above-mentioned research questions and by drawing the research conclusions and future research proposals.

Literature search
In order to provide a comprehensive vision on how much and how well life cycle methodologies are suitable to comply with CE requirements in the agri-food sector, a systematic and critical review of the existing scientific literature was carried out. Based on the study conducted by [27], a critical review goes beyond a mere description of the literature, but extensively it should evaluate its quality seeking to identify most significant items, analyzing significant components, and synthetizing the main concepts. Embracing the same main characteristics, a systematic review differs from the previous one as seeks to systematically search for, appraise, and synthesize research evidence. Therefrom, this study, combining the strengths of these two review typologies, carry out an extensive review employing the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement [28]. PRISMA was used as a formal systematic review guideline for data collection providing a standard peer accepted methodology, to contribute to the quality assurance of the revision process and its replicability. A review protocol was developed (Figure 2), describing the search strategy, article selection criteria, data extraction, and data analysis procedure.

Figure 2.
Methodological steps of the literature search process using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram [28].
In the "identification" step of the PRISMA flow diagram (cf. Figure 2), a set of keywords was selected based on the question formulation, i.e. the research scope, which consisted of searching for all documents proposing life cycle approaches to measure the empirical circular pathways of agri-food systems. The literature search was performed in Scopus and Web of Science (WOS) databases, through the combination of main keywords using Boolean operators (AND/OR). As shown in Table 1, the following search strings were applied: ("circular economy"), ("life cycle assessment" OR "life cycle analysis" OR LCA), ("life cycle costing" OR LCC), ("social life cycle assessment" OR "S-LCA" OR SLCA OR "social-LCA), ("life cycle sustainability assessment" OR LCSA) combined with ("agr* OR food). The research has been conducted in the fields "title", "abstract", and "keywords" for the main keywords, and in "all fields" for the other terms, i.e. agr* or food. The databases were consulted in October 2020 with no time restriction. Table 1. Query used in database searching.

Database
Search strings 1

Scopus
(TITLE-ABS-KEY ("circular economy") AND TITLE-ABS-KEY ("life cycle assessment" OR "life cycle analysis" OR "life cycle costing" OR "social life cycle assessment" OR "life cycle sustainability assessment" OR lca OR lcc OR "s-lca" OR slca OR "social-lca" OR lcsa) AND ALL (agr* OR food)) Web of Science TOPIC: ("circular economy") AND TOPIC: ("life cycle assessment" OR "life cycle analysis" OR "life cycle costing" OR "social life cycle assessment" OR "life cycle sustainability assessment" OR lca OR lcc OR s-lca OR slca OR "social-lca" OR lcsa) AND ALL FIELDS: (agr* OR food) 1 Last accessed on 29 Oct 2020.
Searches on Scopus and WOS databases led to 341 and 255 articles, respectively, tfor a total of 596 papers. Duplicate papers were excluded, resulting in 464 documents, which have been subjected to a screening process. A first selection was made by using the "Refine results" tool of the databases used to exclude review and editorial material and include only English language. Then, only applicative indexed references were taken into consideration. A second screening was performed based on the content of abstracts, excluding discussion papers, or off-topic and studies that did not focus on the agri-food sector or life cycle approaches. In so doing, 122 articles were assessed for eligibility by reading the fulltext in-depth. Studies not directly focused on the issue of measuring circularity quantitatively were discarded.
Through the above-specified criteria application, the total amount of articles found was reduced to a final portfolio of 84 representative papers that were included in the qualitative synthesis. These articles were read in full and analyzed one by one for the purpose of this study.

Characterization of matrix criteria for the systematic and critical review
According to De Luca et al. [29], an evaluation matrix has been set up to synthesize research evidence, by classifying the selected papers according to several integrated criteria. As shown in Table 2, all reviewed papers have been categorized by bibliometric information (authors, year of issue, title, journal); descriptive statistics that refers to the place where the case-study is applied, field of application (i.e., the area of human activity), the main product under study, circularity topics; and relevant data on circularity assessment methods and circularity indicators. These latter criteria included the differentiation of both the methodologies into "LC tools and other life cycle approaches" and "Other methods different from LC", and indicators into "Circularity indices" and "CE assessment indicators". According to Corona et al. [30], the former indicators measure the circularity degree of a system, based on a mere material recirculation and addressed to resource efficiency. The latter assesses the effects (burden or value) of circularity, showing high potential in addressing all the CE goals at the product/service level. Here, we divided the "CE assessment indicators" into "Life cycle based-indicators" and "No life cycle based indicators" (see 3.3 Section for more details). "Circular strategy application-level" indicates the levels to which the circular strategies or interventions are applied namely: micro level (products, companies or organizations, consumers), meso level (eco-industrial parks), and macro level (regions, cities, countries, or the global economy) [31]. Finally, the last columns of the matrix are focused on the main features that qualify the life cycle approaches, e.g., functional unit, system boundary, database, LC impact assessment method, software, etc. (please see Table S1).
Once the matrix has been completed, the input data were compared and the results were qualitatively and quantitatively extracted to highlight significant information and relationships. The main highlights and conclusions of the selected studies are reported in the following section.

Descriptive statistics
The descriptive analysis was based on the distribution of the reviewed articles over the years and by country (based on the place of case-study application), as well as their distribution per journal, field of application (resulted from the main argument or topic of study), main reference product (which refers to the product analyzed in the case-study), and the most common topics dominating the CE literature in the agri-food sector.
The selected 84 papers were published from 2014 to 2021, as shown in Figure 3. It should be noted that the papers issued in 2021 were already available online in October 2020. The results revealed an exponential increase in the number of publications regarding the application of life cycle methodologies as circularity metrics in the agri-food sector over the last 7 years. The first publications, starting from 2014 to 2016 (one for each year), concerning energy recovery from dairy farming [32], recycling food waste for use as feed in aquaculture [33], and the waste management with energy recovery in the anchovy industry [34]. Only in the last years, more specifically in 2018-2020, strong efforts were made towards the development of studies to measure the circularity through LC approaches in agri-food systems. In particular, 16 documents were published in 2018, 23 documents in 2019, and 32 documents in 2020. Nevertheless, as mentioned by Barros et al. [35], only a few efforts can be observed towards assessing agricultural systems accounting for sustainability in a circular perspective. Thus, the peak development of this theme has yet to be reached.
According to the place of case-study application, most publications may be traced to European countries (49%), followed by China, as it can be seen in Figure 4. Indeed, the five highest-ranked origins of the reviewed articles are Spain (22.6%), Italy (14.3), United Kingdom (UK) (7.1%), China (6%), and Ireland (4.8%). These findings were consistent with results reported by Esposito et al. (2020) [26], who showed the great interest of European scholars towards the development of CE models also in the agri-food sector. Outside the European continent, China represents the major contributor in researching this topic. This is likely due to the Chinese government's request to stimulate actions in favor of the environment also via CE [36]. The interest in using LC tools to analyze CE strategy seems to be growing in Brazil and Sweden with 3 publications each (for more information, please refer to Table S1, in Supplementary Material section).  Concerning the type of contribution, 83 articles out of 84 were published in scientific peer-reviewed journals and only one in proceedings of scientific international conferences. The highest-ranked journals were Journal of Cleaner Production (21), Science of the Total Environment (10), Resources, Conservation and Recycling (8), Waste Management (5), and Sustainability (4), with 57.1% of documents considered. This is due to the scope of these journals also related to the theme of the CE. The remaining journals showcased 2 or 1 publications each. All scientific journals were about sustainability topics and environmental issues, only one specialized in agricultural systems and food production (Agricultural Systems). Figure 5 presents the main argument covered in the studies analyzed. In this review, we refer to the area of human activity or context in which the application is implemented ( Table S2). The waste and/or biomass fields of application were the most addressed by the published articles accounting for 55% of the total, of which 24 documents (29%) are strictly dedicated to wastes, 20 (24%) to biomass, and 2 (2%) to the whole of "wastes and biomass". Here, "Wastes" refers to the use and recycling of household wastes [37], wastewater [38], agricultural wastes [39; 40], food waste [41; 42; 43], and organic waste [44; 45], including recovery of nutrients [46], organic compounds, and energy [47]. This field is of great interest to European societies and academics, and constantly under a spotlight due to the recent publication of the waste management directives [48] that fall within the "Circular Economy package" adopted by the European Commission in December 2015 [49]. In the field named "Biomass", several kinds of goods (wood, garbage, crops, fruits, litter, manure, landfills gas, etc.) for energetic purposes (anaerobic digestion [e.g., 50; 51; 52; 53; 54, etc.) were included.
Around 15% of the studies were included in the "Manufacturing" field, which includes product production from raw materials (renewable or not). For instance, the manufacture of biochemicals and bio-based plastics is one of the strategies promoted by the European Union within the Europe 2020 strategy [55], as well as the production of biobased packaging is an effective and promising climate change mitigation strategy [56]. The "Agriculture" field, accounting for 11% of the total, enclosed production of fruits and vegetables [57; 58] for fresh consumption or industrial transformation [e.g., 59] (raw materials, food, and no food).
Considering the reference product analyzed in the case studies, Figure 6 shows the most common ones in the selected papers. With 11% of the total, "Food waste" is the most represented product category. In this group, the authors were considered a heterogeneous set of products. For instance, [42] attempted to identify the optimal combination of food waste prevention by analyzing the quantity of food waste generated along five different food supply chain (i.e. grain, meat, fruit and vegetables produces, milk/dairy, and seafood). [37] analyzed the avoidable food waste amounts contained in household waste (fruits and vegetables, bread and pastries, fish, meat, dairy products, eggs, meal leftovers,), for waste prevention, energy recovery, or recycling purposes. Other main reference products identified in this review included several agro-industrial products, such as tomato [ [81], coffee [82].
The need to recycle nutrients like phosphorus has been widely considered as an important issue of a CE. For instance, [83] and [84] carried out an environmental evaluation of technologies for phosphorus recovery from sewage sludge to be applied on agricultural land as fertilizers (P-based fertilizers). As argued by the authors, phosphorus in wastewater should be utilized to avoid depletion of mineral phosphorus reserves, in line with the principles of a CE. The most common "circularity topics" that emerged in this study's final portfolio were closed-loop production systems, e.g., nutrient recovery for agricultural purposes, production of renewable energy, valorization of residues and wastes, food waste, and agro-wastes recycling for agriculture, as well as reduction of input, final wastes or product losses (Figure 7, Table S3). Here, the "Waste valorization" topic, accounting for 32% of the total, is used to indicate retrieve elements from wastes or losses to be used for new purposes, such as extraction of biochemical feedstock and nutrients recovery [e.g, 85; 86; 87; 88; 89]. The second main circularity topic issued in this study refers to Energy recovery, with about 29% of the final portfolio. Incineration of material, usually biomass, with energy recovery [e.g., 90], composting for energetic purposes [e.g., 91], and anaerobic digestion [e.g., 92; 73; 79] were the recurring questions include in this topic.
To follow, the "Recycle" topic was observed in 15% of the total documents. In the present review, we refer to turning an item (products, co-products, by-products) into raw materials that can be used again, usually for a completely new product. For instance, composting and packaging recycling [e.g., 93; 94; 95]. In this topic, energetic purposes are excluded.

Circularity assessment methods based on LC tools and other life cycle approaches
According to the findings shown in Figure 8, the most common LC tool adopted by papers to assess the benefits/impacts of CE strategies is LCA. This review found 52 case studies out of 84 (62% of the total) using stand-alone LCA. LCA is considered by all authors as the most suitable methodology to assess products, services, technologies in a CE perspective, including studies on biomass for energetic purposes, food products as well as biochemical and bio-composite products, waste reduction, and waste valorization also for energy recovery, manufacturing of products from raw materials (renewable or not). Most of the papers were published in 2019 and 2020, pointing out how LCA in the agrifood sector towards CE is a quite recent topic of research.
On the contrary, only 8 studies (9.5%) deal with the LCC methodology combined with other analyses. Of these, 6 adopt LCC combined with LCA [41; 55; 44; 47; 96; 74], 1 paper combine the LCC model with LCA and Material Flow Analysis (MFA) [75], and another with externality analysis in the CE perspective [97]. Finally, no paper deals with the sLCA methodology.  Table 3, this literature review found 52 case studies applying standalone LCA to analyze the contribution of circular strategies to the principle of CE. Most of the reviewed LCA is performed following several impact evaluation methods that include multiple indicators representing up to 16 different impact categories.

As described in
In such final portfolio of papers, the most common LCA indicators were Global Warming Potential (or Climate Change or Carbon Footprint) applied in 58 papers (67% of the total), Eutrophication (for marine, freshwater, and terrestrial ecosystems) in 45 papers (55%), Human toxicity in 28 papers (35%), and Ecotoxicity in 25 papers (30%). According to Berti et al. [98], these impact categories have been documented to be the most appropriate for agricultural assessments.
The most applied method was Recipe, accounting for 38.5% of the total papers. For instance, [50], used the Recipe method in the hierarchic perspective to conducted a consequential LCA to examine several scenarios where biogas produced from poultry litter is used to generate heat and electricity or is upgraded to biomethane which can substitute natural gas. Among the impact categories (midpoint level) selected by the authors, climate change, acidification and eutrophication are recommended impact categories for LCA of bioenergy systems. Moreover, these impact categories related to carbon, nitrogen, and phosphorous flows are significant for agricultural systems. The ReCiPe midpoint method was chosen by [51], allowing for the assessment of the contribution of wood-based bioenergy plant, which utilizes local residues from wood industries and forestry operations, through several impact categories, i.e., climate change (GWP), fossil depletion (FD), ecotoxicity (FEP), human toxicity (HTP), photochemical oxidant formation (POFP), etc. [60] and [39] also used the ReCiPe method in a hierarchic perspective and at the midpoint level. In the first case study, the authors conducted a life cycle assessment of different productive uses of rooftops under Mediterranean climatic conditions, while, in the second case study, an evaluation of the environmental burdens of composting as a way to achieve a more circular valorization of wine waste.
The other two most applied methods in the literature review were CML (21.2%) and ILCD (17.3%). [80], for example, assessed four environmental impact categories using the CML method, i.e, global warming (GW), abiotic depletion (AD), acidification (AC), and eutrophication (EUT), to carry out an environmental life cycle assessment of poultry fat, poultry by-product meal and steam hydrolyzed feather meal obtained by rendering poultry byproducts. The CML-IA impact assessment method and eight of its impact categories (global warming potential (GWP100a), human toxicity (HT), fresh water aquatic ecotoxicity (FW), marine aquatic ecotoxicity (MAE), terrestrial ecotoxicity (TE), photochemical oxidation (PO), acidification (AC) and eutrophication (EU)), were applied by [106] for assessing the environmental impact of a redesigned mango food supply chain to improve environmental sustainability.
To evaluate the environmental performance of the animal feed from Camelina sativa, [99] used ILCD 2011 Midpoint + considering the following impact categories: climate change, human toxicity, terrestrial acidification, freshwater eutrophication, terrestrial eutrophication, and abiotic depletion. ILCD midpoint characterization method was also adopted by [100], to evaluate the environmental impact of the bioconversion of fruit and vegetable waste into earthworm meal to be used as new food/feed source.
Considering the importance of energy consumption in the agricultural systems, some authors also included in their analyses the cumulative energy demand (CED), an impact indicator that expresses the energy consumption throughout the life cycle of a product or a service [  Few studies used water footprint (WF) indicator [94; 106], known worldwide for the assessment of environmental performance. Among the methods involved in LCA-based water footprint, the authors adopted the Water Footprint Assessment (WFA), a method centered upon computation of the Water Stress Index (WSI) that calculates the water impact on the consumption-to-availability perspective of freshwater deprivation.
As above-mentioned, few studies adopted LCC methodology as a tool for measuring CE strategies from an economic point of view. Table 4 summarizes all the reviewed papers that implemented LCC combined with other approaches. [41] and [55] performed a conventional (C-LCC) and societal (S-LCC) life cycle costing paired with LCA. The first authors used LCC to evaluate the socio-economic impacts of producing wet animal feed, protein-concentrated animal feed, and lactic, polylactic, and succinic acid from food waste. The LCC model implemented was the unit-cost method approach, where the waste management system under study is divided into stages (e.g. collection, transport, processing) and each stage is characterized by its relevant costs, classified into budgets costs, transfers and externalities. As argued by the scholars, LCC provides critical insights into process performance, giving a platform for more targeted technology optimization. The second authors used C-LCC and S-LCC to assess the economic aspects of the use of biobased plastics in the fruit chain along the whole chain, following the methodological scheme expressed by [114] and [115]. Environmental externalities and their relative monetary value were also identified.
By combining the LCC model and externalities in the CE, [97] analyzed the benefits of using aluminum packaging in the food sector. The approach proposed by [116] was used to evaluate costs and benefit and to externalities. As discussed by the researchers, it is necessary to adopt the LCC approach as a useful economic model to guide the solutions for sustainable manufacturing and the CE vision. [75] and [44] used the economic model derived from SWOLF (solid waste optimization life cycle framework) to perform an LCC analysis to evaluate a waste management system that aims at recovering nutrients from municipal organic waste. [47] performed an LCC analysis based on the approaches described by [116] and [117] to assess the costs related to different waste management alternatives from fish canning industry. The scholars suggest that LCC can help to identify all steps that constitute an opportunity to reduce costs, helping decision-makers to choose a cost-effective project alternative. To estimate the economic implications of recovering energy and material resources from food waste, [92] applied LCC methodology following the guidelines published by [117] and [116]. Finally, [63] used LCC taking the value-added approach (VA) to evaluate the economic aspects of packaging-related food loss and waste of food-packaging systems.

(LCA=Life Cycle Assessment, LCC= Life Cycle Costing, MFA= Material Flow Analysis, C-LCC= Conventional Life Cycle Costing, S-LCC= Societal -Life Cycle Costing, SWOLF= Solid Waste Optimization Lifecycle Framework, VA= value-added).
In many case studies reviewed, CE strategies were assessed through LCA combined with other "life cycle-type" approaches, i.e. methods not directly ascribed to typical LC framework (i.e, LCA, LCC, and sLCA), but that approached the evaluation process in a life cycle perspective. Among the other methodological approaches most applied, there were Material Flow Analysis (MFA), Input-output (IO) analysis, and Carbon Footprint, implemented coherently with principles and methodological steps of an LC-based approach.
Five case studies adopted MFA accounting combined with LCA to evaluate the circularity of systems. Following the MFA modeling principles of Brunner and Rechberger [118], material flows of a system are measured in terms of their mass. [37] compared the environmental benefits of household food waste prevention to the benefits from various waste management strategies concerning recycling rates, energy efficiency, and emission efficiency, by using MFA model combined with published LCA results. The authors suggest that the most effective food waste management strategy seems to be a combination of prevention and recycling strategies. However, for mitigating climate change, the prevention of food waste clearly stood out as the most effective strategy. [119] studied the optimal configuration of a waste management system that valorizes the municipal organic waste (OW) in Cantabria, performing an MFA of the system and an LCA of the unit processes concerning the treatment of solid OW and the land application subsystem. The closed-loop perspective of the system analyzed by the authors is given by the application of products generated from the OW (compost, digestate, etc.) to land, which results in a reduction in the consumption of the industrial fertilizers. Their results indicated that an enhanced circularity of resources does not necessarily entail the decrease of both the overall consumption of natural resources and the emission of environmental burdens of the system. [46] also considered the circularity of nutrients within a system that handles the organic waste generated in Cantabria. They concluded that improving nutrient circularity paradoxically leads to eutrophication impacts, and increasing the source separation rates of OW has a positive effect on the carbon footprint of the system. [91] developed an LCA approach based on MFA to calculate the potential environmental impact of combined energy recovery and nutrient recycling from horse manure through anaerobic digestion in a centralized plant, replacing unmanaged composting. The authors indicated that anaerobic digestion is suitable to reduce potential environmental impact in comparison to unmanaged composting, mainly due to biogas substituting the use of fossil fuels. Finally, [73] propose an approach based on MFA and LCA for measuring the environmental performance of the anaerobic treatment of dairy processing effluents based on the CE principles. Their results showed that the recovered energy from AD provides 20% of the energy requirements of the factory reducing the total carbon footprint emissions by 13% compared to the baseline scenario.
This analysis found three studies applying Input-output (IO) technique combined with LCA. According to Miller and Blair [120], IO analysis uses a top-down, economic method to capture product and service flows from one industrial sector to all other sectors within one country, region, or multi-regions. As argued by Corona et al. [30], such a topdown approach has been applied by the LCA community to compensate for the shortcomings of process-based LCA (e.g., expanding the scope from the product level to national/global level). [77; 78], developed a hybrid life cycle assessment model integrating process-based LCA with IO analysis. In the first study, the authors implemented such an approach to evaluate the feasibility and potential benefits of a novel bio-fertilizer technology that utilizing paddy rice residues through composting. The bio-fertilizer can recycle the nutrients in residues to replace synthetic fertilizer within a circular rice production system. In the second study, the authors used a hybrid IO-LCA model to assess the environmental, social, and economic impacts of modifying conventional bioplastics production with a potato pulp residue left over from starch production to produce biocomposite. [121] also applied an IO-based hybrid LCA model to estimate total waste generation throughout the supply chain in Spain.
Only two studies applied the Carbon footprint of a product by using the LCA approach. [57] used the LCA concept for greenhouse gas emissions (LCA-GHG) to evaluate and compare GHG emissions of large-scale and individual farming in rice production, while [69] used LCA for analyzing carbon footprints in traditional and biogas-based circular economic models of pig farming.

Reviewed circularity assessment indicators
As shown in Table 5, only 8 articles deal with the CE assessment through specific indicators. As previously stated, in this review we refer to the classification of CE assessment indicators proposed by [30], who identify indicators that measure the circularity degree of a system, based on a mere material recirculation and addressed to resource efficiency, and indicators that assess the effects (burden or value) of circularity. Here, by LCbased indicators, we refer to the life cycle impact categories indicators retrieved from LCA, the LCC indicators when utilized for evaluating CE strategies, and stand-alone indicators based on life cycle approaches.
Purposes and advantages derived by the use of CE indicators have been widely argued in the literature [31; 122; 123; 124]. Due to the complexity inborn in the circularity economic paradigm, CE indicators combining different metrics can deliver simplified results.
Some CE indicators examined in this review were developed to assess the circular degree of a system. Within this topic, [46] developed indicators to study the circularity of nutrients within a system that handles organic waste (OW) generated in Spanish. More in detail, the circularity indicators of carbon (CIC), nitrogen (CIN), and phosphorus (CIP) have been applied to a circular integrated waste management system, which encompasses not only waste management, but also the processing and consumption of the components recovered from waste and the external raw materials. As argued by the authors, enhancing the circularity of these nutrients seems to be a suitable strategy for closing their natural biogeochemical cycles by avoiding the accumulation of nutrients in one of the Earth's subsystems at a rate faster than the ecosystems can sustain. The authors jointly evaluated the main environmental impacts associated with the emissions of these elements by using a set of indicators based on LCA, i.e., global warming, marine eutrophication, and freshwater eutrophication.
[125] used the energy return on investment (EROI) ratio, and a CE perspective, to develop an energy return on investment -circular economy index (EROIce) to quantify the amount of nutritional energy recovered from the food loss in the Spanish food supply chain. The EROIce index, based on a food waste-to-energy-to-food approach (the energy recovered from food loss is reintroduced into the food supply chain in form of food), was developed starting from the calculation of primary energy demand (PED) of each stage of the food supply chain under study. Here, we consider this index as a "life-cycle based indicator" since it is based on the evaluating of energy flows along the entire food supply chain (agricultural production, processing, and packaging, distribution, and consumption).
Advancements in the assessment of CE strategies at the product level have been suggested by [81], who coupled two sets of indicators via multi-criteria decision analysis, i.e., material circularity based-indicators -namely, material re-utilization score (MRS) and material circularity indicator (MCI) -and a selection of life cycle based-indicators (climate change, abiotic resource depletion, acidification, particulate matter, and water consumption). The MRS is the metric used to quantify material re-utilization developed by Cradle to Cradle Products Innovation Institute (C2C), while the MCI is the main index developed by the Ellen MacArthur Foundation (EMF) and Granta to measure how well a product performs in the CE context. The authors suggest exploring the application of the multicriterial analyses of LC-based indicators (including the socio-economic dimension) to address CE trade-offs and rebound effects. In a complementary manner, [82] proposed a set of indicators integrating techno-environmental and CE criteria to guide the design and development of new food packaging solutions within the new plastics economy. In detail, the authors considered nine indicators based on the CE guidelines developed by EMF, which focus on the materials and energy used in manufacturing as well as on end-of-life waste management, and four LCA based-indicators, i.e., climate change, depletion of fossil fuels, and metals, and primary energy demand to assess the environmental impacts of packaging from cradle to grave. [73] also developed an approach for measuring the material and environmental circularity performance of the anaerobic treatment of dairy processing effluents. Material CE performance was assessed by the "Material circularity performance" indicator (MCPI), suggested by [126], which enables to evaluate to what extent the demand of resource or energy flows reduced when the circularity loops are closed. On the other hand, the environmental performance was estimated by the "Environmental Circularity Performance Indicator" (ECPI) based on the ratio of the total environmental benefits and costs.
Combining LCA (global warming potential, acidification potential, eutrophication potential, and ReCIPE single score) and LCC (value-added) indicators, [47] suggested a method to assess the eco-efficiency of canned anchovy products with the eco-efficiency index (EEI), which enables the translation into economic terms of the environmental damage caused by the manufacture of a specific product.
[108] proposed a new set of hybridized sustainability indicators, drawn from the principles of green chemistry and resource (material and energy) circularity, to evaluate the environmental performance of bio-based products, bio-based packaging films, and mulch films in comparison with their commercial counterparts. These metrics are demonstrated via the application of a comparative LCA, incorporating the hybridized indicators including hazardous chemical use, circular-process feedstock intensity (CPFI), circularprocess waste factor (CPWF), process material circularity (PMC), product renewability (PR), and circular-process energy intensity (CPEI). Through a set of LCA indicators, the authors also highlighted the resource and energy hotspots and toxicity to the environment and human health, and the quantification of impacts from the minimization of resources. Last but not least, [127] used emergy-based circular economy indicators (no life cyclebased) to assess the sustainability of the urban eco-system. These indicators were developed by using Emergy accounting (EMA), which accounts for different categories of supporting contribution to the systems, including renewable and non-renewable energy and material resources, information and knowhow, and finally labor and services. The authors consider EMA indicators as valuable tools to evaluate the implementation rate of CE patterns.

Discussion
Although it is not the only existing circularity indicator, it can undoubtedly be said that the Material Circularity Indicator is one of the most robust tools for assessing the CE, and since the development of the original methodology, published in 2015 by the EMF, the similarities between Life Cycle Assessment and Material Circularity Indicator became evident. Both methodologies use system boundaries that encompass all phases of a product's life cycle, from creation to end of life. What differentiates the two approaches, however, is that the assessment of circularity cannot be limited to one life cycle, as the circular pattern of one will inevitably influence the next. Slavishly quoting the methodology for measuring circularity: "the economic and environmental benefits from using such materials do not commonly rest with the initial product but instead accrue through the successive use of the product or material over the multiple life cycles that they enable" [128:13]. So first step to marry these two methodologies should be to extend the boundaries of the system by integrating into the horizon of the analysis product losses, recycling and reuse in the next cycle, transport, and all processes that allow closing the loop of the LCA methodology according to a circular approach.
The literature review found that only 20% of studies use a cradle-to-grave [e.g., 42; 55] system boundary, only three of these explicitly refer to a cradle-to-cradle or system expansion approach [46; 68; 56] and only one of these integrate a circularity indicator [46]. Most studies focus on a partial system boundary and, therefore, life cycle analysis is more limited to assessing the environmental profile of the process or co-product [e.g., 51].
The studies do not, therefore, aim at a true "circular strategy" since circularity is not really measured in most of them. Most articles use indicators relating to the use of material and energy resources but this is not enough to define the degree of circularity of a process or product, just as circularity alone cannot define sustainability.
LCA can assess the environmental impacts of a process and, through an eco-design approach, allows for the implementation of strategies to reduce these impacts, including a reduction in the use of resources and by considering the burden-shifting phenomenon whereby a change in one stage of the life cycle influences another one [129]. What attributional LCA cannot assess are rebound effects, a key element in sustainability assessments because it takes into account changes in production and consumption when the availability of a resource change (positively or negatively) [130].
For these reasons, the assessment of circularity must pass through a multi-component approach that takes into account not only circularity itself but also other characteristic elements. The Ellen MacArthur Foundation identifies as complementary analyses, to the evaluation of circularity, the evaluation of risk factors such as Material Price Variation, Material Supply Chain, Material Scarcity and toxicity, and the evaluation of impact factors such as Energy Usage and CO2 Emissions, Water and Toxicity [128].
The analysis of the papers shows the opposite situation, where the assessment of impact factors becomes the main driver for measuring sustainability [e.g., 57], while circularity rather remains a goal to be achieved but hardly ever explicitly measured.
If we acknowledge the need to expand the boundaries of the system from a cradleto-cradle perspective, at the same time, we should be aware that stating that a process serves to make the usual business model more circular does not automatically prove that this is the case: it has to be proven.
While the use of classic impact assessment methods, as already mentioned, can give us an assessment of resource and energy use, it does not allow a complete evaluation of circular strategies, which are often based on other fundamental factors such as product lifespan or functional unit, understood as the unit of measurement of product use. We can make the same product, using the same amount of resources but, by increasing the efficiency of those resources by extending the life of the product, we can contribute positively to increasing the circularity of the process. This will probably not be detected by an LCA, especially if it does not take into account the use phase of the product.
This, moreover, may not apply to materials of biological origin, whose shelf life depends on factors not directly under anthropogenic control. However, in the case of food and agricultural products, extending the shelf life of products intended for consumption can make a significant contribution to the circularity of the process, reducing waste and thus increasing production efficiency. Moreover, as the results of the review show, lost food or products of biological origin that are no longer usable can easily be valorized in different ways, returning to the production cycle as fertilizers or generating bioenergy or bio-components with high added value.
The element that, however, appears clear and seems to be unavoidable, is that an LCA complementary to a circularity assessment framework should always assess the whole life cycle of a product and should consider its possible extensions, expanding also the time boundaries of the study by considering at least more than one life cycle.
A "Circular LCA" [128] also take into account a detailed Life Cycle Inventory analysis and consider "resource use" as one of mandatory impact indicator, or use specific supporting methodologies such as Material flow analysis which takes into account the flows and stocks of materials and substances entering and leaving a defined system. Some works explore the use of methodologies in conjunction with LCA or standalone methodologies that take a life-cycle perspective. Above all, material flow analysis is a methodology that lends itself well as a supporting methodology for the assessment of circularity [37; 119]. Other methodologies such as IO Analysis applied to the life cycle of a product allow the assessment of environmental impacts to be integrated with the impacts of positive or negative economic shocks and possible concatenated reactions on the economy [78].
As proposed by Santagata et al., [127] Emergy Accounting can also be a valuable support for measuring circularity. The authors of the paper also point out that conventional analysis methodologies (Life Cycle Assessment, Material Flow Accounting, Cost-Benefit Analysis, among others) do not allow to capture all aspects of CE, proposing emergy accounting as a possible solution, however, this methodology is also at the center of many scientific controversies and does not enjoy a consensus shared by the scientific community.
It might seem that the LCA methodology can make little contribution to the development of circularity metrics, but this is not the case because LCA is a well established and standardized methodology while circularity assessment indicators are in development and their development is not in opposition to LCA but an ideal complement to it.
It should not be forgotten that talking about LCA is probably reductive since it should be part of a multi-objective framework (i.e. LCSA) aimed at analyzing the integrated sustainability of a process, product, system, or organization [29].
In this direction, as already discussed some studies have already explored the possibility of also using the Life Cycle Costing methodology in the assessment of circularity [41]. While the main circularity indicators are essentially based on the increase in the utility of resources within an economic model, an approach that assesses the life cycle value flows of a product, process, system or organization is a fundamental complement to both circularity and sustainability assessment. What has already been discussed for the LCA also applies to the LCC, so the community of experts in life cycle methodologies and the CE must accompany the approach of these methodologies, resolving the technical and scientific issues [21] that have only briefly been discussed so far.
While the methodologies discussed so far mainly refer to environmental and/or economic metrics, the role that the large-scale economic model change will bring to the social level cannot be neglected. Therefore, the efforts of the scientific community mustn't neglect the social sustainability linked to the adoption of a new circular model. And while the LCA and LCC methodologies are now scientifically established, the social counterpart of life cycle analysis, sLCA is still in an embryonic state, so it is essential to manage the growth of this methodology so that it is consistent with new business models.
sLCA is the latest tool developed within the family of Life Cycle methodologies. Since the nineties, life cycle scholars felt the importance and urgency of taking into account social impacts [131], with an implied sustainability approach borrowed from the three pillars model, meaning that sustainability is composed by three dimensions (environment, economy, and society).
However, sLCA is still not definitively codified in an agreed and consensus-based protocol. From its beginnings, a plethora of social impact assessment methods has been proposed for sLCA, paying attention to the most diverse aspects, such as the social performances, the presence of hot spots, the accounting of risks, the consequences of a scenario, the externalities of a system, and the participation of [132; 133]. Moreover, while LCA and LCC are focused on the impacts caused by the functioning of a system (whatever it is a product or a service), very often, in sLCA studies, the focus is on companies' behaviors. This entails, therefore, evaluating a wide range of impacts, also not directly attributable to a life cycle (such as corruption, child labor, collective bargaining, fair wages, among others). This is due to the epistemological eclecticism of social sciences, inevitably reflected in sLCA studies [132]. Recently, the second edition of the sLCA guidelines have been published [134], however, life cycle methodologies are still striving to reach an epistemological alignment, with LCA and LCC approaching impacts assessment in a postpositivist way, quantifying cause-effects relationships, and sLCA mainly devoted to the interpretation of social performances according to stakeholders' perceptions and behaviors.
In general, many academics and scholars describe social sustainability as the most conceptually elusive pillar in sustainable development discourse [135].
Among the papers selected for this review, no one applies the sLCA or another specific methodology for social impacts assessment. Rather, some kind of social impacts are explicitly associated, in some few cases, with economic performances, as it is the case of [41; 55; 77]. [41] and [55] who implemented the societal LCC to evaluate high-value products from food wastes, the former, and bio-based plastics the latter. Societal LCC is a "welfare-economic", meaning that takes into account marketed goods and the effects on the society's welfare caused by exchanges that would otherwise not be accounted for, i.e. by identifying environmental externalities and measuring their relative monetary value [55]. [77], applied a hybrid life cycle assessment model to estimate social-economic impact through a multi-regional input-output database (Exiobase), with engineering process data of conventional and circular rice production systems. The indicators used in this case were "gross value added" and "employment" in each system, therefore, taking into account "the social significance" of economic performance. [45], in their study, defined social issues as "product or process-related aspects of operations that affect human safety and community welfare". The authors illustrated the possible social criteria suitable in the study of chemical/ lipid processing, i.e. the satisfaction of social needs (SN), Work satisfaction (WS), Healthcare security coverage (HcS), Employee turnover (EmpT), Working hours (WH), Employee complaints (EmpC), and Risk Assessment. However, none of these mentioned criteria is calculated in their study.
All the other papers reviewed, just mention the importance of considering the social aspects of the life cycle, mentioning sometimes the social acceptance or desirability [51], the social perception [68].
The lack of a specific, stand-alone, social impact assessment method in the papers here reviewed, but very often in sustainability assessments in general, let us make a reflection about the taxonomy and interconnections between "sustainability dimensions". Are they perfectly separable, or are they -at least -partly stackable? Do they have the same importance, or is there a hierarchy among them? These questions still remain open in the academic debate among life cycle scholars and practitioners.

Conclusions
This critical and systematic literature review provided a picture of the state-of-theart of applications of the life cycle approach in the assessment of circularity of processes and products. Lights and shadows emerged: the relationship between circularity measurement and life cycle analyses still seem to be primordial, even though these two aspects are absolutely complimentary. CE measurement and life cycle methodologies should be like Manzoni's "The Betrothed" whereas as things stand, they seem almost like the protagonists of an arranged marriage, who are getting to know each other in order to decide whether to continue their lives together or separate.
The methodological development in this field is constantly evolving and new tools are increasingly being tested by the scientific community to identify the most effective ones, as well as to provide a measurement of circularity that takes into account sustainability issues. However, the scaffolding on which this methodological development must be based has already been built, and scholars cannot ignore it because there is a risk of making an irreparable mistake. In particular, experts in life cycle methodologies must strive to adopt some key elements to ensure that the results obtained fit perfectly with the measurements of circularity and that these can even be largely based on a common basis. The effort must also go in the direction of operability of the framework for measuring circularity and sustainability, so that it does not have the opposite effect of an assessment structure that is so complex that it is hardly usable, thus thwarting efforts to create new models of sustainable agri-food production and consumption.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1, Tables S1, S2, S3.  This research is also co-funded by the European Commission, European Social Fund and by Calabria Region, through the post-doctoral research fellowship awarded to Dr. Iofrida. The authors are the only responsible for this scientific article; the European Commission and the Calabria Region decline any responsibility for the use that may be made of the information contained therein.