Science of the Total Environment 774 (2021) 145682 Contents lists available at ScienceDirect Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv Environmental impact assessment of agro-services symbiosis in semiarid urban frontier territories. Case study of Mendoza (Argentina) Natalia Bonilla-Gámez a, Susana Toboso-Chavero a,b, Felipe Parada a, Bárbara Civit c, Alejandro Pablo Arena d, Joan Rieradevall a,b, Xavier Gabarrell Durany a,b,⁎ a Sostenipra Research Group (2017SGR 1683), Institute of Environmental Sciences and Technology (ICTA-UAB), Unidad de Excelencia Maria de Maeztu (CEX2019-000940-M), Universitat Autònoma de Barcelona (UAB), 08193 Barcelona, Spain b Department of Chemical, Biological Environmental Engineering, Universitat Autònoma de Barcelona (UAB), 08193 Barcelona, Spain c Centro de Estudios para el Desarrollo Sustentable (UTN FRM), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), 5500 Mendoza, Argentina d Grupo CLIOPE, Energía, Ambiente y Desarrollo Sustentable (UTN FRM), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), 5500 Mendoza, Argentina H I G H L I G H T S G R A P H I C A L A B S T R A C T • Assessment of environmental impacts in linear and circular cities • Resource symbiosis in agro-services sys- tems in semiarid frontier territories • Symbiosis of water and energy in agro- services systems • Integration of mosaic roofs in house- holds • Mosaic roofs with photovoltaic panels, rain and graywater, and crops a r t i c l e i n f o a b s t r a c t Article history: In the next 10 years, cities will house 60% of the world's population, where in agro-urban frontier territories, Received 21 November 2020 mainly in semiarid regions, problems associated with land use and water distribution will arise. Therefore, a Received in revised form 16 January 2021 model of growth that contemplates the lowest use of resources must be proposed. The aim of this study is to de- Accepted 3 February 2021 termine the environmental impact of the use of resources in agro-services frontier territories of semiarid regions Available online 6 February 2021 in three urban growth scenarios, linear and circular systems. The study is focused on Luján de Cuyo in Mendoza, Editor: Deyi Hou Argentina. Through a life cycle assessment, environmental performance was evaluated for the current scenario and the three proposed ones. In addition, an ecoefficiency analysis was performed in relation to climate change Keywords: and water consumption and the cost of surface uses, as well as an assessment according to the multifunctionality Life cycle assessment (LCA)1 of surface use. Scenario 1 is a linear diffuse urban system, scenario 2 a circular water agro-urban system, and sce- Circular green cities2 nario 2 green plus energy a circular water and energy agro-urban system. The outcomes illustrate that scenario 2 Water footprint3 green plus energy has the least environmental impacts. Compared to the linear scenario, both circular scenarios Frontier territories4 show a substantial reduction in water consumption (38–40%) and marine eutrophication (32–47%) and curtail Agro-urban planning5 freshwater eutrophication impacts. Furthermore, household energy impacts are reduced by 39% in scenario 2 Vineyard6 green plus energy with photovoltaic panel implementation, and maximum ecoefficiency in response to climate change is reached. Additionally, the impacts of scenario 2 green plus energy are more than 42% less than those of scenario 1 in terms of the multifunctionality of surface use. This study shows that it is possible to achieve more sustainable semiarid urban frontier territories with local water and regional energy circularity. © 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). ⁎ Corresponding author at: Department of Chemical, Biological Environmental Engineering, Universitat Autònoma de Barcelona (UAB), 08193 Barcelona, Spain. E-mail address: xavier.gabarrell@uab.cat (X. Gabarrell Durany). https://doi.org/10.1016/j.scitotenv.2021.145682 0048-9697/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 1. Introduction areas (Moura De Moraes et al., 2011). These areas are more vulnerable to climate change due to fresh water resource scarcity and the con- The United Nations stated that approximately 55.3% of the world's sumption of fossil fuels for energy supply; therefore, the water footprint population was concentrated in urban settlements in 2018 and that and greenhouse gas (GHG) emissions must be reduced. The problems urban areas are projected to house 60% of the population (at least half associated with water resources are local and so are their solutions; a million inhabitants) globally by 2030 (UN, 2018). For this reason, in for energy resources, the problems associated with their use are na- addition to the expansion of urban land; the exploitation of natural re- tional but have local solutions. Therefore, in frontier territories, a vision sources and new land for agricultural use; the loss of biodiversity and of a local circular green city must be sought. Accordingly, the semiarid increased demand for food, water, and energy flows, cities have in re- zone in Argentina is representative of this type of area. cent years placed more pressure on global and local natural ecosystems Over the last four decades, the Metropolitan Area of Mendoza (Cerón-Palma et al., 2012). (AMM), similar to other Latin American cities, has experienced the ef- The most commonly used city model is that with linear input and fects of uncontrolled and unplanned urban growth (Secretariat of output flows, characterized by the import and export of resources with- Environment and Land Management, 2018). AMM is considered a fron- out closing cycles (Cerón-Palma et al., 2012). Since cities depend on tier territory, denoting the presence of urbanization over purely agricul- water, energy and material resources, they play a vital role in the sus- tural areas or green belts, and because it is located an arid zone with tainable transition by, for example, managing urban metabolism with water scarcity issues, conflicts associated with the type of use and distri- circular economy initiatives (circular cities) (Sánchez Levoso et al., bution of water resources have arisen. 2020) and facilitating the exchange of flows between cities, the natural One of the main territorial problems found in the North Oasis of environment, and urban subsystems from an industrial and urban ecol- Mendoza is related to the management of the urban-rural frontier be- ogy perspective (Cerón-Palma et al., 2012). cause it is a geographic space that is affected by rapid and continuous Factors such as population growth, urbanization, water pollution, transformation. A clear example of a frontier territory located between the use of fossil fuels and unsustainable development are increasing a diffuse city and vineyard, one of the main land uses in the area, is pressures on natural resources globally, which are further intensified shown in Fig. 1. by climate change (UN Environment, 2019). On the other hand, local Currently only independent studies of cities and agriculture exist communities depend on agriculture for employment and income gener- and not integrated studies from an environmental perspective, that con- ation, yet this cannot be developed further because of the pressures al- template a vision of nexus water-energy-food, use of unused surfaces ready placed on land and water resources, mainly in frontier/peri-urban such as roofs, ornamental to productive green spaces and naturalization areas of cities in semiarid regions where this loss of agricultural area oc- of cities. Under the premise of making AMM's economic and social curs with greater intensity and where there is water scarcity (FAO, development sustainable, given increasing populations and housing 2017). Worldwide, agriculture uses an average of 70% of all fresh demands and the loss of agricultural areas, a model of agro-urban water sources, so competition for water from cities and industry creates growth that contemplates the lowest use of resources must be a need to improve agricultural water use efficiency while producing proposed. In exploring such an approach, the following research more food and using fewer and less harmful inputs (UN Environment, questions arise. How can resources better more effectively used in the 2019). For this reason, the promotion of water-use efficiency, recycling and rainwater harvesting is becoming increasingly important to ensure water security and more equitable water allocation for different users and uses as well as the production and consumption of cleaner forms of energy (Corcelli et al., 2019; UN Environment, 2019). In addition, urban clusters (urban centers and their suburbs) have grown by a factor of approximately 2.5 since 1975, accounting for 7.6% of land worldwide in 2015 and creating urban heat islands due to effects on the hydrolog- ical cycle and soil functions (UN Environment, 2019). In this sense, sus- tainable land-use planning and management can protect high-quality, fertile agricultural soil from competing interests and maintain land- based ecosystem services (UN Environment, 2019). In this sense, the developing region of Latin America and the Carib- bean (LAC) has registered the fastest urbanization in the world, with its urban population increasing from 41% in 1950 to 80% in 2010 (IDB, 2011). This has resulted in an increase in vulnerability due to deficient urban planning processes and the abuse of environmental resources. In addition, 60 to 70% of the region's gross domestic product (GDP) is concentrated in urban centers (IDB, 2011). Cities of LAC face common challenges and problems such as i) disjointed urban growth in the periphery, which causes scarce provi- sions of essential services in the peripheries of urban areas, ii) high rates of sociospatial segregation, iii) cities with monocentric development, and iv) an absence of clear and efficient urban development policies that allow for integral development (Nadal et al., 2019). In addition, the disorderly and unbalanced occupation of the territory promotes a low-density city model, resulting in the presence of incompatible land uses and excessive urban expansion on land with agricultural capacities, resulting in the loss of productive agricultural land and thus affecting fu- ture of food security in cities and cultural identity (Secretariat of Environment and Land Management, 2018). Fig. 1. An example of an agro-urban frontier territory in Luján de Cuyo, Mendoza, Semiarid and arid regions cover approximately one-third of the Argentina. world's land area, and in South America, there are four large semiarid Source: (Google, 2020). 2 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 Metropolitan Area of Mendoza, Argentina and how can we mitigate department of Luján de Cuyo is used as a reference because it experi- competition from agro-services in terms of the use of these resources enced the greatest shift from rural or frontier land use to urban land and environmental impacts? How can agricultural production be main- use from 2001 to 2010. In other words, these areas were converted tained in ways that are compatible with urban development in frontier from rural areas with agricultural production to urban areas while territories while also addressing climate change and water eutrophica- maintaining irrigation water rights (Spotti, 2015). The Cabernet tion issues, minimizing losses in the water distribution network and Sauvignon grape variety was used as a model because previous studies taking advantage of nutrients in domestic wastewater? on water and carbon footprints were performed by the UTN FRM. To answer these questions, we aim to determine environmental im- Three possible scenarios of urban growth were examined for the pacts of the use of resources in agro-services frontier territories of semi- frontier territory of the semiarid region of Luján de Cuyo, Mendoza in arid regions with conventional linear urban growth and of systems that addition to current conditions. Fig. 2 shows the scheme of each scenario integrate agro-urban symbiosis with circular flows of water and energy. as well as the associated area (in hectares). The current scenario Our specific objectives are: (A) represents the frontier of Luján de Cuyo before urban growth, which includes conventional agriculture and a conventional urban dif- 1. To identify all inputs and outputs that participate in conventional lin- fuse system (households and green gardens). Regarding the proposed ear systems of diffuse cities, which do not integrate agriculture, at the urban growth scenarios, scenario 1 (B) contemplates conventional lin- frontier of semiarid urban regions in South America. 2. To propose future scenarios of circular green urban planning that in- ear urban planning diffuse city with gardens and green areas (phasing tegrate agriculture in frontier territories of semiarid regions. out agricultural activity). Scenario 2 (C) considers agro-urban planning, 3. To compare environmental impacts of the growth of conventional so it integrates the diffuse city with the agricultural area, and the green linear diffuse cities those that integrate a circular green city vision gardens become vineyard gardens (urban soil-based farming) with a in frontier territories, determining the actions that have had the circular flow of graywater and rainwater. For scenario 2 green plus en- greatest impacts. ergy (D), the diffuse city is integrated with the agricultural area, 4. To quantify the minimization of resource consumption through the where the green gardens become vineyard gardens (urban soil-based symbiosis of agro-urban systems in diffuse circular green cities. farming), and a roof mosaic plan is employed with the combination of different systems by integrating urban rooftop farming (vineyard) 2. Materials and methods with solar energy through the installation of photovoltaic (PV) panels on the remaining surfaces of roofs not colonized by the vineyard, with This section provides detailed information on the scenarios under a surface ratio of 50% vineyard land use and 9% PV paneling. This sce- study as well as stages of the life cycle assessment (LCA) methodology nario also considers a circular flow of graywater and rainwater. applied for the environmental impact assessment. The environmental impact assessment according to the surface multifunctionality and 2.2. Life cycle assessment ecoefficiency assessment methods is presented. The evaluation was performed by applying the LCA methodological 2.1. Case study framework based on ISO standards 14040:2006 and 14044:2006 (ISO, 2006a, 2006b). The following steps were adopted to implement the The study is focused on AMM and local vineyard activity, specifically LCA: i) goal and scope definition; ii) life-cycle inventory (LCI); iii) life- in the frontier territory of the department of Luján de Cuyo. The cycle impact assessment (LCIA) and iv) interpretation of results. Fig. 2. Scheme of the scenarios under study and areas in hectares. 3 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 2.2.1. Goal and scope definition sent to ACRE farms for irrigation, and the remaining water is sent to This research aims to quantify the environmental impacts of three the Mendoza River with collected rainwater. The processes of the different scenarios of resource supply in agro-urban frontier territories DWTP and WWTP are the same as those in the current frontier scenario. of semiarid regions under urban growth. Each frontier territory scenario The system does not use electricity to pump water into the network, as was divided into three subsystems: vineyards (V), households (HH) and water is moved by gravity. The electricity used by households comes green gardens (GG). Limits were set taking into account all input and from the Mendoza electricity network. The second scenario, including output flows of water, resources, and energy levels of the scenario use the circular green agro-urban system and scenario 2 green plus energy: phase; therefore, infrastructure and ornamental plants in the GG sub- the circular green plus PV energy agro-urban system, contemplate the system and vine plants in the V subsystem were excluded. For scenario process stages of the DWTP, diffuse city and vineyard gardens (HH 2 green plus energy, infrastructure associated with PV panel implemen- and V) and WWTP. Drinking water in the DWTP is used by households, tation in the HH subsystem is contemplated. black domestic wastewater is treated in the WWTP and sent to ACRE It should be noted that Luján de Cuyo is already applying policies and farms for irrigation, and the rest is sent to the Mendoza River. The elements of sustainability in using little or no fertilizers in vineyards and gray domestic wastewater and collected rainwater from household green gardens; using water supplies that do require prior pumping or roofs are stored in separate tanks and reused without treatment for distribution energy due to the geography of the area; and managing a the irrigation of vineyard gardens. Pumping this reused water requires wastewater treatment plant, given the low cost of soil, with seminatural electricity, unlike the rest of the processes. The processes of the DWTP systems and low resource consumption. Additionally, in Mendoza, the and WWTP are the same as those of the current frontier scenario. For recirculation of treated wastewater for agricultural irrigation is already scenario 2 green, the energy used by households comes from implemented in some areas. Argentina's electricity network. For scenario 2 green plus energy, a per- The functional unit (FU) of the study is to meet the average resource centage of the energy used by houses comes from solar energy by an needs necessary to annually supply a use phase of 1 ha (ha) of an agro- installed capacity of 3 kWp of PV panels per house rooftop, and the services frontier territory in a semiarid region. This functional unit was rest comes from the electricity network (50% vineyard and 9% PV). selected in order to be able to compare agro-services (water, food and This installed capacity is taken as a reference for this study since it rep- energy resources). The reference flows are equivalent to the average an- resents the average amount of contracted power that homes in Men- nual resource consumption per capita of a citizen in the case study area, doza use, taking into consideration what is established in the which is equivalent to 2716.7 kWh of electricity, 174 m3 of blue water, regulatory framework of Argentina. The tilt of the panels (4°) is equal and an average annual consumption level of 1 kg of vid (Vitis vinifera), to the average slope of the roofs of houses in Luján de Cuyo (7%), which translates into the need for 0.79 kWh of electricity and 1.55 m3 since through our preanalysis it was observed that if the optimal angle of blue water for their production. of the zone of 31° is used, a mounting system is required, which in- Fig. 3 shows a diagram with the system boundaries for the proposed creases environmental impacts, economic costs, is not visually attractive urban growth scenarios, divided by linear system flows (for scenario 1), and only produces 4.65% more electricity. It should be stressed that circular system flows (for scenario 2 and 2 green plus) and common since there is a regulatory framework that allows for distributed gener- system flows. The system boundaries of the current frontier scenario, ation, battery use was not considered. an agro-urban linear system before urban growth, consider services or The assumptions made in this study for Luján de Cuyo based on local process stages of the local drinking water treatment plant (DWTP), dif- data and previous research carried out by the UTN FRM are shown in fuse city (HH and GG), and wastewater treatment plant (WWTP). Supplementary Information Table A1. Drinking water from the DWTP is used by households and gardens, and wastewater from households is treated in the WWTP. After treat- 2.2.2. Life cycle inventory (LCI) ment, a large portion of the water is sent to restricted crop areas An inventory of input and output data of the studied scenarios per (ACREs, by its Spanish acronym) for irrigation, and the remaining subsystem (V, HH, and GG) was carried out, involving the collection of water is sent to the Mendoza River. Rainwater collected in the rainwater data to meet the predefined goals. Inventories for meeting the needs system is transported to the river. The DWTP employs 3 subprocesses: of the subsystems covered energy (electricity and transport), water pretreatment with chlorine, treatment with active carbon and post- (blue, green, and reuse), resources (pesticides, reagents, PV panels, in- treatment with chlorine, where electricity is required for such treat- verters, and electric installation components), and outputs (grapes, ment (Manau-Sifres, 2019). Active carbon in the DWTP is not taken wastewater, collected rainwater, etc.) and are described in detail in Sup- into account in this study. The WWTP adopts a natural biological plementary Information Table A2. In this study, local data were mainly treatment process carried out in an aerobic, anaerobic, facultative and collected from our own research done at the “Centro de Estudios para maturing lagoon. Cloacal liquid treatment takes approximately el Desarrollo Sustentable” of the UTN FRM and from other public sector 35 days, and the process does not require electricity or reagents entities in Argentina. For data from which no local information was (Departamento General de Irrigación, 2017). The electricity used by available, regional or global data were used from various studies con- households comes from the Mendoza electricity network. On the ducted at the Autonomous University of Barcelona (UAB). In the specific other side of the frontier territory is agricultural land, which includes a case of solar energy data, the SolarGis Prospect application was used. vineyard. Untreated surface water is used to irrigate the vineyard. The The source of each data point can be seen in Supplementary Information vineyard subsystem accounts for processes of cultivation and the man- Table A1 of the inventory. ual harvesting of grapes, the tillage and maintenance of land, irrigation, the application of agrochemicals and fruit harvesting, and electricity and transport required for such processes. In general, the system does 2.2.3. Life cycle impact assessment (LCIA) not use electricity to pump water into the network, as water is moved An LCIA was performed using SimaPro 9.1.0.8 software developed by by gravity. Pré Consultant (Pré Consultants, 2020). ReCiPe at a midpoint level For the proposed urban growth scenarios, the process stages are the (H) we used to calculate environmental impacts (RIVM, 2017). The same on both sides of the frontier territory to evaluate a transition from Ecoinvent 3.6 database (Weidema et al., 2013) was used as our main the current frontier area to other more or less sustainable concepts. The source, followed by LCA Food DK (Nielsen et al., 2003). Seven impact boundaries of the first system, the linear diffuse urban system, are the categories were considered according to previous literature (Rufí-Salís DWTP, HH and GG and WWTP. Drinking water in the DWTP is used et al., 2020) and the authors' expertise, including climate change (CC, for both households and gardens, and wastewater from households is kg CO2 eq), terrestrial acidification (TAP, kg SO2 eq), freshwater eutro- treated in the WWTP. After treatment, a large portion of the water is phication (FET, kg P eq), marine eutrophication (MEP, kg N eq), 4 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 Fig. 3. Diagram of system boundaries. freshwater ecotoxicity (FETP, kg 1.4 DCB to freshwater), land use (LOP, 2.3. Sensitivity analysis of the LCA results m2a crop eq) and water consumption (WCP, m3). Impact categories of the proposed urban growth scenarios for which The results of a LCA study can be affected by different sources of uncer- there is a difference of more than 10% are shown on a tricolored scale tainty. The sensitivity analysis is one of the procedures of analysis used to (red, yellow and green) where the color green denotes the lowest im- estimate the uncertainty of LCA results, and it assesses the influence of a pact, and the color red denotes the greatest impact. parameter (the independent variable) on the value of another (the 5 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 dependent variable) (Cellura et al., 2011). The independent variables 3.1.1. Analysis by scenario used in this study were the input data of grape productivity and green Table 1 shows the environmental impacts of meeting the average re- water (rainfall). The following sensitivity scenarios are then analyzed: source needs necessary to annually supply 1 ha of an agro-service fron- tier territory in a semiarid region for each scenario as well as the relative • A: minimum and maximum yearly data of grape productivity impact of each subsystem per scenario and impact category. The results (Cabernet Sauvignon) in the province of Mendoza. confirm that the current scenario for the frontier territory of Luján de • B: minimum and maximum annual rainfall data in 10 years (from pe- Cuyo has low environmental impacts in almost all categories, since ele- riod 2008 to 2017) in the province of Mendoza. ments of sustainability are already incorporated and there is a nonintensive use of services, as only 26% of the system is occupied by The minimum and maximum yearly data of grape productivity housing and 50% is occupied by local agriculture. The results show sig- (Cabernet Sauvignon) in the province of Mendoza was taken from the nificant differences (of more than 10%) between the linear growth sce- “Instituto Nacional de Vitivinicultura”, while the minimum and maxi- nario and the circular green scenarios, specifically in the CC, TAP, FET, mum annual rainfall data in 10 years (from period 2008 to 2017) in MEP and WCP categories when comparing the impacts of scenario 1 the province of Mendoza from the “Dirección de Agricultura y and scenario 2 green plus energy, with scenario 2 green plus energy Contingencias Climáticas”. The values used were 2.43 t/ha and 10.67 t/ having the least impact, and when comparing scenario 1 to scenario 2 ha, and 154.6 mm and 367.5 mm respectively. in the categories of FET, MEP and WCP, with scenario 2 being less im- pactful. However, in the FETP and LOP categories, the linear growth sce- 2.4. Ecoefficiency assessment method nario has the fewest impacts, which are similar (less than 10% difference) to those of scenario 2 but 26% and 29% (respectively) ISO 14045 (ISO, 2012) was applied to study the life-cycle environ- lower than those of scenario 2 green plus energy. In the FET, MEP and mental impacts of the frontier territory scenarios along with the value WCP categories, the greatest difference between the linear growth sce- of services. Considering the bidimensional nature of ecoefficiency, the nario and the circular green scenarios is observed, as scenario 1 does not environmental performance of the scenarios was represented by indica- incorporate a circular vision of graywater and rainwater reuse. Avoided tors of climate change measured in kg CO2 eq/ha and of water consump- impacts are seen in scenarios 2 (−885.84 kg P eq) and 2 green plus tion measured in m3/ha. Based on the types of system value defined in energy (−3381.87 kg P eq) in FET) due to the reuse of the gray the ISO, a monetary perspective was applied. The value assessment wastewater. was performed using the economic value of surface use (average price of grape production per year plus the average price of household rentals 3.1.2. Analysis by subsystem per year) in US$·ha−1. At the subsystem level, the household subsystem is the largest con- The relationship between both parameters was analyzed with an tributor to all impacts in the three proposed scenarios, where in the CC, ecoefficiency chart to identify the most and least ecoefficient scenarios. TAP, FETP and LOP categories, it contributes more than 93.92%. In con- It was assumed that high economic value and low environmental im- trast, the vineyard subsystem contributes no more than 6.08% in the pacts are desired trends, as carbon emissions are then minimized proposed scenarios and curtails impacts for both eutrophication catego- while providing a service with high added value. ries (marine and freshwater) due to the reuse of gray wastewater (sce- Additionally, the ReCiPe single score indicator, specifically the area nario 2: FET: −4081.23 kg P eq; MET: −0.31 kg N eq, scenario 2 green of ecosystem quality, was analyzed because it provides an overall pic- plus energy: FET: −5329.38 kg P eq; MET: −0.41 kg N eq). ture of the impacts generated to the ecosystem in a single score value. However, it is just to have another approach, since this methodology 3.1.3. Analysis by flows has a lot of uncertainty. Fig. 4 compiles the environmental impacts of the four frontier terri- tory scenarios, including all subsystems and flows. When analyzing the 2.5. Environmental impact assessment according to the multifunctionality household subsystem, it can be observed that energy is the most critical of surface use flow for the impact categories of CC, TAP, MEP, FETP and LOP due to electricity consumption. However, in scenario 2 green plus energy, To analyze environmental impacts obtained from the LCA from a there is a considerable reduction in the energy flow impacts of these multifunctionality perspective, agricultural and architectural services categories (except LOP) of more than 39% due to the implementation of were considered. The equation is shown below. Eq. (1) denotes the 3 kWp of solar panels per house, which cover 43.6% of the electricity total impact in relation to multifunctionality (Is), where I represent the demand of each home, reducing electricity consumption from the environmental impact per category, and SA and SU represent agricul- network and its respective impact. For the FETP and LOP categories, the tural and urban surfaces in hectares, respectively, per scenario. It should impact of resources associated with photovoltaic installation (PV panels, be noted that gardens do not provide agricultural production services or inverts and electric installation components) is very high, reducing protection or shelter from environmental conditions, so their area is not impacts due to solar electricity generation. Therefore, scenario 2 green taken into consideration. plus energy has the greatest impact in these categories. For FET, the I output flow is the most critical and is associated with domestic wastewa- IS ¼ ð1Þ ter and its nitrite, nitrate and phosphate content discharged into the ðSA þ SU Þ Mendoza River. For WCP, the water flow, associated with blue water ex- tracted from the river for drinking or irrigation, is the most critical flow, 3. Results responsible for more than 78% for all scenarios and subsystems, except for the vineyard subsystem under scenario 2 and scenario 2 green plus The environmental performance levels of the case study scenarios energy, which shows values equivalent to only 70.8% and 74.9%, respec- are shown and described in the following section. tively, since most of the vineyard is irrigated with reused water. 3.1. Life-cycle impact assessment (LCIA) 3.1.4. Water consumption Since Luján de Cuyo is a frontier territory located in a semiarid region This subsection presents the results of the LCIA for each scenario, where water is a critical resource, we emphasize the category of water subsystem and inventory flow. The most important impacts are consumption and its importance. In the graph of water consumption highlighted. shown in Fig. 4, the household subsystem is shown to consume the 6 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 Table 1 Environmental assessment of meeting the average water needs necessary to annually supply 1 ha of an agro-services frontier territory in a semiarid region. Current scenario Scenario 1 Scenario 2 Scenario 2 green plus energy V HH GG Total V HH GG Total V HH GG Total V HH GG Total CC 3.85E+03 1.21E+05 6.95E+02 1.25E+05 - 2.42E+05 1.39E+03 2.43E+05 3.76E+03 2.42E+05 - 2.46E+05 5.82E+03 1.42E+05 - 1.48E+05 3.07% 96.38% 0.55% 99.43% 0.57% 1.53% 98.47% 3.94% 96.06% TAP 1.20E+01 3.73E+02 2.29E+00 3.87E+02 - 7.46E+02 4.59E+00 7.50E+02 1.17E+01 7.46E+02 - 7.57E+02 1.82E+01 4.51E+02 - 4.69E+02 3.10% 96.31% 0.59% 99.39% 0.61% 1.55% 98.45% 3.87% 96.13% FET 4.18E-01 3.64E+03 1.27E-01 3.64E+03 - 7.28E+03 2.54E-01 7.28E+03 -4.08E+03 3.19E+03 - -8.86E+02 -5.33E+03 1.95E+03 - -3.38E+03 0.01% 99.99% 0.00% 100.00% 0.00% -56.09% 43.91% -73.24% 26.76% MEP 7.14E-02 9.00E-01 1.32E-02 9.84E-01 - 1.80E+00 2.63E-02 1.83E+00 -2.43E-01 1.49E+00 - 1.24E+00 -3.01E-01 1.27E+00 - 9.67E-01 7.26% 91.40% 1.34% 98.56% 1.44% -13.87% 86.13% -19.11% 80.89% FETP 9.26E+01 2.19E+03 2.37E+01 2.30E+03 - 4.38E+03 4.74E+01 4.42E+03 8.99E+01 4.38E+03 - 4.47E+03 1.39E+01 5.87E+03 - 6.01E+03 4.02% 95.95% 1.03% 98.93% 1.07% 2.01% 97.99% 2.32% 97.68% LOP 2.63E+01 2.23E+02 5.36E+00 2.55E+02 - 4.46E+02 1.07E+01 4.57E+02 2.52E+01 4.44E+02 - 4.69E+02 3.91E+01 6.04E+02 - 6.43E+02 10.33% 87.57% 2.10% 97.65% 2.35% 5.37% 94.63% 6.08% 93.92% WCP 1.01E+04 1.72E+04 1.07E+04 3.80E+04 - 3.43E+04 2.14E+04 5.57E+04 3.27E+02 3.43E+04 - 3.47E+04 2.02E+03 3.13E+04 - 3.33E+04 26.70% 45.17% 28.12% 61.63% 38.37% 0.94% 99.06% 6.06% 93.94% Units: kgCO2 eq (CC), kg SO2 eq (TAP), kg P eq (FET), kg N eq (MEP), kg 1.4 DCB to fresh water (FETP), m2a crop eq (LOP) and m3 (WCP). Percentages express the relative contributions in each specific impact category. The impact categories of the proposed urban growth scenarios for which there is a difference of more than 10% are shown on a tricolored scale (red, yellow and green) where green denotes the least impact and red denotes the most. greatest amount of water, mainly due to water flows (blue water) and to production shows variations for all impact categories of both scenario a lesser extent due to energy flows. Scenario 1 shows the highest 2 and scenario 2 green plus energy. The lower the grape production consumption of blue water (55,716.81 m3·year−1), while scenario 2 the lower the environmental impact, except for the FET and MEP cate- green plus energy shows the lowest water consumption of gories where the lower the grape production the higher the impact. 33,309.75 m3·year−1 followed by scenario 2 (34,665.31 m3·year−1). This is because less gray water would be reused, and therefore more This denotes savings in scenario 2 green plus energy in relation to sce- would be sent to the river. For these two impact categories, the greatest nario 1 of approximately 40.2% for water and of approximately 37.8% influence is observed due to the variation in grape productivity. for scenario 2. This reflects 14.1% less consumption than the current On the other hand, in the case of rainfall, it has a very low level of im- level in the frontier territory, where the population is 50% smaller. This pact. Only in scenario 2 green plus energy there is a variation in the WCP minimization of blue water consumption is associated with the symbiosis category, the rest of the categories for the three scenarios are not influ- of agro-services, the use of rainwater and graywater in the local reuse sys- enced. This is because, as shown in Fig. 5, only in scenario 2 green plus tem, and the generation of 43.6% of electricity for houses with PV panels. energy is rainwater reused to supply the vineyard, and in a very low Currently (as shown in Fig. 5 and based on the inventory shown in amount (5%). Table A2) in the frontier territory of Luján de Cuyo, the vineyard is supplied 96% by blue water (from the Mendoza River), which is the 3.2. Ecoefficiency assessment same source used to meet the needs of the city, and only 4% by green water (rainwater). As mentioned above, since the territory is The relationship between the environmental impacts and the eco- located in a semiarid region, it rains very little here. For this reason, nomic value of surface uses was analyzed with ecoefficiency chart scenarios 2 and 2 green plus energy propose a circular use of the (Fig. 6) to identify the best frontier territory scenario and desired trends. water flow between the city and vineyard, creating agro-services In terms of ecoefficiency in addressing climate change, scenario 2 green symbiosis. Fig. 5 also shows the different water sources used for plus energy is shown in area A, and scenarios 1 and 2 are shown in area vineyard irrigation under these circular scenarios and the associated B. On the other hand, in terms of ecoefficiency for water consumption, percentages according to their use. In scenario 2, there is no compe- all scenarios are shown in area B (scenario 2 and 2 green plus energy tition for blue water from the city to irrigate the vineyard, 96% of the at the border with A). Fig. 6 also shows the variation in the economic vineyard is irrigated with reused gray wastewater, and the remain- value of surface uses (the average price of grape production per year ing 4% is irrigated with green water. On the other hand, under plus the average price of household rentals per year) per scenario. scenario 2 green plus energy, only 10% of the vineyard's water The graph of ecoefficiency assessment using the ReCiPe endpoint consumption comes from the river (blue water), 80% is composed single score methodology follows a trend similar to that of the climate of reused gray wastewater, 5% is composed of green water and 4% change indicator. For more information see Fig. A1 in the Supplemen- is composed of collected rainwater. tary Information. 3.1.5. Sensitivity analysis 3.3. Multifunctional analysis The results obtained from the sensitivity analysis of the LCA by vary- ing the input data of rainfall and grape production in the province of In analyzing the environmental impacts of each scenario in relation Mendoza are shown in Supplementary Information Table A3. Grape to the multifunctionality (agricultural and architectural services) of 7 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 Fig. 4. Environmental impacts of the five scenarios for frontier territories according to each subsystem (V: vineyard, HH: households and GG: green gardens) and flow. surfaces, the term multifunctionality is used to indicate that agricul- impacts of scenario 2 green plus energy are more than 42% lower ture can produce various commodity and noncommodity outputs in than those of scenario 1, and those of scenario 2 are more than 47% addition to food. Table A1 shows the amount of local endogenous re- lower. As there are more forms of multifunctionality and services sources obtained per hectare (grape production, PV energy produc- in the frontier scenario, the environmental impacts are reduced. tion and rainwater harvesting) for the three urban growth For example, under scenario 2 green plus energy, for which there is scenarios. The results of the proposed scenarios (see Table 2) show more surface area for vineyard services, there is less environmental relevant differences of more than 10% for all impact categories. The impact in all seven categories evaluated. 8 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 Fig. 5. Vineyard irrigation water sources for the current and proposed circular scenarios. 4. Discussion (CC, TAP, FETP and LOP) are almost uniform (less than a 10% difference), but the functions of the frontier territory are very different. Under sce- 4.1. Future circular frontier territories in semiarid regions nario 1, the frontier territory only provides one function, services, while under scenario 2, it is multifunctional, providing agro-services The frontier territory scenarios proposed in this study for a semiarid without this causing a significant increase in impacts. The household region simulate urban growth of approximately 50% of the population subsystem is the largest contributor in all impact categories. In the im- relative to the current situation scenario. When comparing these sce- pact categories of CC, TAP, FETP and LOP, it contributes more than narios, materials and construction stages for the buildings and water in- 93.92% and is associated with the critical flow of energy. This is the frastructure are similar and therefore not quantified in this LCA study, case because more than 60% of Argentina's electricity mix comes from which instead focuses on the use phase. fossil sources (mainly natural gas with 57%), unlike other countries in As shown in Table 1, scenario 2 green plus energy has the lowest en- the region, such as Brazil, which draws more than 60% of its matrix vironmental impacts globally for the seven categories analyzed from renewable sources (mainly hydro) (IEA, 2019a, 2019b). This sig- followed by scenario 2. This highlights the importance of a circular nificantly reduces carbon dioxide emissions and thus the impacts asso- and symbiosis vision that integrates graywater and collected rainwater ciated with climate change. For this reason, Fig. 4 shows a decrease of from the city into agriculture and clean energy through the use of pho- more than 39% in the impacts of energy flows (except for the LOP cate- tovoltaic panels. Under scenarios 1 and 2, impacts of the four categories gory) for scenario 2 green plus energy by generating 43.6% of the Fig. 6. Ecoefficiency of frontier territories for climate change (CC) and water consumption (WCP) against the economic value. Black arrows indicate desired trends. 9 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 Table 2 Environmental impact per scenario in relation to multifunctionality (Is). Impact/scenario Current scenario Scenario 1 Scenario 2 Scenario 2 green plus energy CC (kg CO2 eq) 1.65E+05 4.68E+05 2.46E+05 1.17E+05 TAP (kg SO2 eq) 5.09E+02 1.44E+03 7.57E+02 3.72E+02 FET (kg P eq) 4.79E+03 1.40E+04 −8.86E+02 −2.68E+03 MEP (kg N eq) 1.30E+00 3.51E+00 1.24E+00 7.68E-01 FETP (kg 1,4-DCB) 3.03E+03 8.51E+03 4.47E+03 4.77E+03 LOP (m2a crop eq) 3.35E+02 8.79E+02 4.69E+02 5.10E+02 3 WCP (m ) 5.00E+04 1.07E+05 3.47E+04 2.64E+04 Agricultural and urban surface values: current scenario: SA = 0.50, SU = 0.26; scenario 1: SA = 0, SU = 0.50; sce- nario 2: SA = 0.48, SU = 0.50, scenario 2 green plus energy: SA = 0.74, SU = 0.52. electricity demand of households with PV panels. It should be noted that In the case of 50% urban growth in the frontier territory, a notable this scenario is based on Argentine legislation on distributed generation, improvement in climate change impacts is seen, achieving ecoefficiency which establishes that the installed capacity cannot be greater than the in scenario 2 green plus energy and providing a clear vision of the circu- power already contracted by a home. Under this premise, an installed lar economy. Carbon dioxide equivalent emissions are mainly associ- capacity of 3 kWp only covers 10% of the roof of a house, leaving the re- ated with energy, once again highlighting the critical point of maining 40% in disuse. If this legislation were to be modified, up to 40% electricity consumption in homes and the importance of generating more PV panels could be installed, and more electricity could be gener- clean energy. On the other hand, in terms of water consumption, no pro- ated than the demand of a house, achieving energy self-sufficiency and posed scenario, either linear or circular, achieve maximum ecoefficiency exporting the remaining energy. in water consumption; however, scenarios 2 and 2 green plus energy In Luján de Cuyo, the energy impact is low because water is distrib- are on the cusp of achieving ecoefficiency and make economic improve- uted by gravity and does not need to be pumped, and the local wastewa- ments in relation to scenario 1. ter treatment system is atypical, does not require energy or reagents and The above results of scenario multifunctionality in terms of surface already adopts a circular mode of treated wastewater use for the irrigation use show that the scenarios with the largest agricultural and architec- of ACREs, unlike those used in other areas of Latin America where there is tural surfaces, scenario 2 green plus energy followed by scenario 2, are no water treatment or wastewater reuse and where major differences are those with the lowest environmental impacts across all categories. observed in some categories between scenarios 1 and 2. In other coun- tries, the impact of rainwater and graywater recirculation in scenarios 2 4.2. Avenues for improvement and 2 green plus would be more relevant. Water consumption savings observed for scenarios 2 and 2 green By making use of the roofs of houses for urban rooftop farming, more plus energy due to closing cycles using graywater and rainwater to irri- efficient land use can be achieved by taking advantage of existing space, gate vineyards and minimized losses in the distribution system in com- as other studies have proven in the case of Europe (Toboso-Chavero parison to the current scenario and scenario 1 (linear systems) et al., 2018). Under scenario 2 green plus energy, 35% more vines can represent not only 37.8% or more less water extracted from the river be planted (0.26 ha more) than under scenario 2 by applying a roof mo- but also a substantial improvement in the quality of water in the dis- saic cultivating in 50% of roof area and installing PV panels on 9% of the charge river, since the system avoids sending contaminants in remaining roofs (equivalent to an installed capacity of 3 kWp). Almost graywater generated by households, specifically nitrate, nitrite and half of household electricity can be generated with PV panels. As a com- phosphate, which contribute to eutrophication. In addition, blue water plement to this study, architectural and engineering studies should be saved in the Mendoza River serves as an ecological flow or provides eco- carried out based on crop weights to see if houses require any additional logical services to the environment. Additionally, by eliminating garden roof reinforcements when using soil substrate rather than hydroponics. irrigation, the consumption of blue water in a semiarid region and the Additionally, the positive impacts of urban rooftops, for example, on impacts of making this water potable can be minimized. On the other building insulation and energy efficiency through their contributions hand, if infrastructure were taken into account in the LCA, by sending to building cooling and heating systems, should be considered in the fu- less wastewater to the WWTP in scenarios 2 and 2 green plus energy, ture. Studies indicate that insulation effects can generate savings of 3 to the lifetime of the plant and of pipes and tanks of the water distribution 5% (Palma, 2012). network would be extended. Since Mendoza is located in a semiarid region, we initially antici- In scenario 2, gray wastewater generated by houses exceeds the de- pated water to be a major obstacle, and this was thus the focus of the mand required by the vineyard for irrigation, so the system does not re- study. However, we found energy to be one of the flows with the quire an additional use of collected rainwater to achieve a self-sufficient greatest impact; therefore, it is necessary from a nexus perspective to system in the use of water. However, in the case of scenario 2 green plus improve efficiency levels. For this reason, modifying current legislation energy, domestic graywater used together with collected rainwater so that more photovoltaic panels can be installed per household is nec- cannot supply vineyards, so a small amount of blue (surface) water is essary. In addition, the minimization of energy impacts from the reduc- required. In semiarid regions where water scarcity poses problems for tions that green roofs generate in terms of climate control should be urban growth and where food security cannot be put at risk, both sce- quantified to reach full water, energy and food circularity, as well as narios decrease competition for water resources between the city and quantifying the environmental impact of different local energy saving agriculture by 90% (or more); in this case, the vineyard reaches agro- measures. Other means of minimizing the use of water in households services symbiosis. and the use of pesticides in vineyards should also be analyzed. 10 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 In addition, other ways to integrate new landscape, ecosystem ser- CRediT authorship contribution statement vice, biodiversity, identity and ethnological indicators must be explored. Natalia Bonilla Gámez: Writing – original draft, Investigation, Data 5. Conclusion curation. Susana Toboso-Chavero: Investigation, Methodology. Felipe Parada: Investigation, Resources. Bárbara Civit: Conceptualization, This investigation demonstrates the environmental impacts of the Project administration. Alejandro Arena: Resources, Validation. Joan use of resources under different frontier territory scenarios in Luján de Rieradevall: Conceptualization, Project administration. Xavier Cuyo, Argentina. Scenarios 2 and 2 green plus energy are designed to Gabarrell Durany: Supervision, Project administration. overcome the current frontier of agro-services by adopting a more circular vision to naturalize this barrier and achieve less aggressive Declaration of competing interest urban growth. In addition, under scenarios 2 and 2 green plus energy, there are annual yields of local endogenous resources per hectare of The authors declare that they have no known competing financial 4275 kg and 6637 kg, respectively, for grape production, of 788 m3 for interests or personal relationships that could have appeared to influ- rainwater harvesting and of 119,041 kWh for PV energy production ence the work reported in this paper. for scenario 2 green plus energy. When comparing the environmental impact of scenario 1 with that of conventional urban growth, under Acknowledgment scenarios 2 and 2 green plus energy, which integrate a green circular vi- sion by maintaining agriculture and integrating it with gardens (and The authors acknowledge the assistance of Lorena Corica, INAHE roofs for scenario 2 green plus energy), a substantial reduction in CONICET, for their collaboration in providing information and data of water consumption and marine eutrophication is observed and impacts Luján de Cuyo, Claudio Rigoni, EPAS, for providing information on on freshwater eutrophication are curtailed. Scenario 2 green plus en- Mendoza's wastewater and Joan Muñoz, ICTA- UAB, for sharing access ergy has the least environmental impacts for the seven categories ana- to Simapro for data analysis. lyzed followed by scenario 2, showing that symbiosis in the use of The authors are grateful to the National Commission for Scientific rainwater and graywater from the local system in the integrated vine- and Technological Research (Chile) for the grant awarded to F. Parada yard is effective as well as the implementation of PV panels to generate [grant number PFCHA-CONICYT 2017 – Folio 72180248]; and the solar energy. Spanish Ministry of Education, Culture and Sports for the grant to By integrating green plus energy systems between the vineyard and Susana Toboso [FPU16/03238]. This research did not receive any specific city in scenario 2 and replacing the gardens with vineyard gardens and grant from funding agencies in the public, commercial, or not-for-profit incorporating a roof mosaic of 50% vineyard and 9% PV panels, there is a sectors. decrease in the impact of energy flows (except for the LOP category) and a minimization of blue water consumption relative to scenario 1. Appendix A. Supplementary data This minimization of water use occurs due to the reuse of rainwater and graywater, water savings in the irrigation of green gardens, reduced Supplementary data to this article can be found online at https://doi. losses due to the distribution of irrigation water in a closed cycle, and re- org/10.1016/j.scitotenv.2021.145682. duced electricity use from the network. An ecoefficiency analysis proved useful in demonstrating that in References terms of the water consumption indicator, the circular green frontier Cellura, M., Longo, S., Mistretta, M., 2011. Sensitivity analysis to quantify uncertainty in territory scenarios (scenario 2 green plus energy followed by scenario life cycle assessment: the case study of an Italian tile. Renew. Sust. Energ. Rev. 15, 2) are almost ecoefficient, contrary to the linear urban growth scenario. 4697–4705. https://doi.org/10.1016/j.rser.2011.07.082. We also found energy for houses to be one of the most critical aspects Cerón-Palma, I., Sanyé-Mengual, E., Oliver-Solà, J., Montero, J.I., Rieradevall, J., 2012. Bar- riers and opportunities regarding the implementation of rooftop eco.Greenhouses due to Argentina's electricity mix, with mainly fossil fuel sources. By (RTEG) in Mediterranean cities of Europe. J. Urban Technol. 19, 87–103. https://doi. installing PV panels on the rooftops of houses under scenario 2 green org/10.1080/10630732.2012.717685. plus energy, ecoefficiency in response to climate change can be reached. Corcelli, F., Fiorentino, G., Petit-Boix, A., Rieradevall, J., Gabarrell, X., 2019. Transforming rooftops into productive urban spaces in the Mediterranean. An LCA comparison of The findings of this study in relation to the impact of circular green agri-urban production and photovoltaic energy generation. Resour. Conserv. Recycl. cities to sustainable development could contribute to the international 144, 321–336. https://doi.org/10.1016/j.resconrec.2019.01.040. discussions and to local and national policies, and therefore be applied Departamento General de Irrigación, 2017. ACRE y la Planta Depuradora Paramillos to other areas in Latin America, like the Guajira region in Venezuela (Argentina). FAO, 2017. El futuro de la agricultura y la alimentación, El futuro de la agricultura y la and Colombia, the dry regions of Paraguay, Argentina, Peru, Chile and alimentación. doi:https://doi.org/10.1515/nleng-2015-0013. Bolivia and the semi-arid region of northeastern Brazil. Google, 2020. Google Maps [WWW Document]. URL https://www.google.com/maps/@- Our LCA has allowed us to highlight certain aspects; however, to vi- 32.9830712,-68.8680215,752m/data=!3m1!1e3 (accessed 5.10.20). IDB, 2011. Sostenibilidad Urbana en América Latine y el Caribe (Washington D.C). sualize other positive impacts, such as ecosystem services (ES) and the IEA, 2019a. Electricity Generation by Source, Argentina 1990–2017 [WWW Document]. multifunctionality of vineyard systems in scenarios 2 and 2 green plus URL https://www.iea.org/countries/argentina (accessed 6.21.20). energy, this study should be extended to topics such as the positive im- IEA, 2019b. Electricity generation by source, Brazil 1990–2017 [WWW Document]. URL https://www.iea.org/data-and-statistics?country=BRAZIL&fuel=Electricity and pacts of the landscape, ES, biodiversity, and ethnological and popular heat&indicator=Electricity generation by source (accessed 6.25.20). culture features. Vineyards shape important economic, cultural, and ISO, 2006a. ISO 14044:2006- Environmental Management- Life Cycle Assessment- Re- ecological systems. Wine production (growing, making, and selling) quirements and Guidelines. leads to the formation of vineyard landscapes, which are physical and ISO, 2006b. ISO 14040, Environmental Management- Life Cycle Assesment- Principles and Frameword. cultural landscapes. Similar to other agricultural systems, they can ISO, 2012. ISO 14045: Environmental management- Eco-efficiency assessment of product serve as multifunctional landscapes that not only produce grapes but systems- Principle, requirements and guidelines Management. also serve as wildlife habitats, sequester carbon and support the rich tra- Manau-Sifres, R., 2019. Eco-efficiency of the Water Production. Universitat de Barcelona. Moura De Moraes, A., Morante Trigoso, F., Fedrizzi, M.C., 2011. LA PROBLEMÁTICA DE ditions of those living among or visiting them (Winkler et al., 2017; OBTENCIÓN DE AGUA POTABLE EN LA REGIÓN SEMIÁRIDA BRASILEÑA Winkler and Nicholas, 2016). UTILIZANDO SISTEMAS FOTOVOLTAICOS PARA BOMBEO DE AGUA. Av. en Energías With this study, we demonstrate that a more sustainable frontier Renov. y Medio Ambient. 15. can be promoted in Mendoza, Argentina in terms of water, food and en- Nadal, A., Rodríguez-Cadena, D., Pons, O., Cuerva, E., Josa, A., Rieradevall, J., 2019. Feasibil- ity assessment of rooftop greenhouses in Latin America. The case study of a social ergy use through a vision of circularity and symbiosis with the mainte- neighborhood in Quito, Ecuador. Urban For. Urban Green 44, 126389. https://doi. nance and integration of agriculture alongside urban growth. org/10.1016/j.ufug.2019.126389. 11 N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada et al. Science of the Total Environment 774 (2021) 145682 Nielsen, P., Nielsen, A., Weidema, B., Dalgaard, R., Halberg, N., 2003. LCA Food Data Base. food-energy-water nexus of the urban roof mosaic. J. Ind. Ecol. https://doi.org/ Palma, I.C., 2012. Strategies for Sustainable Urban Systems: Introducing Eco-Innovation in 10.1111/jiec.12829. Buildings in Mexico and Spain. UN, 2018. The World’s Cities in 2018- Data Booklet. New York. Pré Consultants, 2020. Simapro 9.1.0.8. UN Environment, 2019. Global environment outlook – GEO-6: summary for RIVM, 2017. ReCiPe 2016 v1.1 A Harmonized Life Cycle Impact Assessment Method at policymakers, global Environment outlook – GEO-6: summary for policymakers. Nai- Midpoint and Endpoint Level. Report I: Characterization. Bilthoven. robi. https://doi.org/10.1017/9781108639217. Rufí-Salís, M., Petit-Boix, A., Villalba, G., Sanjuan-Delmás, D., Parada, F., Ercilla-Montserrat, Weidema, B., Bauer, C., Hischier, R., Mutel, CNemecek, T., Reinhard, J., Vadenbo, C.O., M., Arcas-Pilz, V., Muñoz-Liesa, J., Rieradevall, J., Gabarrell, X., 2020. Recirculating Wernet, G., 2013. Overview and Methodology. Data Quality Guideline for the water and nutrients in urban agriculture: an opportunity towards environmental sus- Ecoinvent Database Version 3. Ecoinvent Report 1 (v3). St. Gallen. tainability and water use efficiency? J. Clean. Prod. 261. https://doi.org/10.1016/j. Winkler, K.J., Nicholas, K.A., 2016. More than wine: cultural ecosystem services in vine- jclepro.2020.121213. yard landscapes in England and California. Ecol. Econ. 124, 86–98. https://doi.org/ Sánchez Levoso, A., Gasol, C.M., Martínez-Blanco, J., Durany, X.G., Lehmann, M., Gaya, R.F., 10.1016/j.ecolecon.2016.01.013. 2020. Methodological framework for the implementation of circular economy in Winkler, K.J., Viers, J.H., Nicholas, K.A., 2017. Assessing ecosystem services and urban systems. J. Clean. Prod. 248. https://doi.org/10.1016/j.jclepro.2019.119227. multifunctionality for vineyard systems. Front. Environ. Sci. 5. https://doi.org/ Secretariat of Environment and Land Management, 2018. Area Metropolitana de Men- 10.3389/fenvs.2017.00015. doza Sostenible. Hacia un desarrollo sostenible del territorio, Mendoza. Spotti, S., 2015. Lo rural pierde con lo urbano. Toboso-Chavero, S., Nadal, A., Petit-Boix, A., Pons, O., Villalba, G., Gabarrell, X., Josa, A., Rieradevall, J., 2018. Towards productive cities: environmental assessment of the 12