Life Cycle Environmental Impacts and Energy Demand of Craft Mezcal in Mexico

: Agave distillates, such as tequila and mezcal, are alcoholic spirits representative of Mexican culture. In recent years, the demand for mezcal has increased, and with it the requirement for raw materials, bringing with it a series of di ﬃ culties. The objective of this study was to evaluate the potential environmental impact and energy demand of the production of young craft mezcal from an endemic agave ( Agave cupreata ) found in the central and southern Paciﬁc area of Mexico. The potential environmental impact of the mezcal studied was obtained through the life cycle analysis methodology using a midpoint approach by the ReCiPe method to calculate the potential environmental impact with SimaPro software (version 8.2.3.0., PR é Sustainability, Amersfoort, The Netherlands). The functional unit is a young craft mezcal bottle of 750 mL with 46% Vol. Alc. The stage of highest contribution to the environmental impact of mezcal was the manufacturing / processing, contributing 59.6% of them. The energy demand of the craft mezcal resulted in 163.8 MJ / bottle of 7.5 dl. The kg CO 2 eq in mezcal (1.7) is higher than beer (0.63) or white wine (1.01), but lower than whisky (2.25) or pisco (3.62). These ﬁndings could allow the search for alternatives for the development of sustainable production.


Introduction
Mezcal and tequila are distilled agave spirit drinks representative of Mexican culture. Linked to the global expansion of tequila, mezcal has increased in popularity in recent years. The production of this drink grew by 79% compared to 2017, bottling 7.14 million litres by 2019 [1].
In the specific case of the mezcal beverage, several states of Mexico, including Oaxaca, Durango, Guerrero, Michoacan, Puebla, Guanajuato, San Luis Potosi, Tamaulipas, and Zacatecas, have protected the designation of origin (PDO) "Mezcal" [1]. However, both raw materials and production practices differ noticeably between production regions, localities, and even factories, resulting in a set of highly distinctive products [2]. The mezcal production process, in general terms, begins with the cultivation of agave which can be of wild origin, semi-cultivated, or cultivated. Once the agaves are ripe, the leaves are cut, leaving what is known as the piña (succulent core). Only the piña of the agave plant is used to make mezcal. These piñas are baked, ground, and fermented. The product obtained from the ethyl fermentation is known as must, which is distilled to separate the residues and obtain the mezcal ( Figure 1). The regulation for mezcal production [3] declares certain categories according to the machinery, equipment, or infrastructure used in each production process. These range from industrialized production to craft production or even ancestral. In the same way, the classes of young or white mezcal are established: matured in glass, rested, aged, flared or distilled according to the process carried out after the distillation.
The increase of interest in mezcal has involved, as with tequila, several problems. Some authors pointed to the shortage of wild agave [4], overproduction of maguey [5], pests in plants [6], and even inappropriate waste management [7], both solid and vinasse, which are the residual liquids of distilling the fermented must [6]. The residual liquid effluents or stillage are classified as a polluting product because they are disposed of at temperatures close to 90 °C, with a pH lower than 5.0 and a high chemical oxygen demand (COD) (50-150 g O2/L) [8]. By placing the stillage on the ground, the suspended solids cause a decrease in permeability, obstructing the pores of the soil [9].
In response to the problem of scarcity of raw materials in the agave-mezcal product system, sustainable practices have been proposed, focused mainly on the production of agave. For example, agroforestry management [10,11], the selection of potential areas for agave plantations [12], and even assisted plant reforestation programs, wet nurses, and agaves [13]. To reduce the problem of waste in the production of mezcal, studies have been carried out on the production of tiles made from a bio composite material, bagasse [14], and the production of biofuels through a treatment of bagasse [15] and adobes reinforced with agave fibre in Oaxaca [16].
The efforts made to solve the problems of mezcal, to date, have focused on specific stages of the production chain of said drink. However, there are still no reports that describe the environmental impact of mezcal production with a life cycle approach or indicators on the energy performance of the process.
There are several studies on the environmental impact of alcoholic beverages approached from the methodology of life cycle assessment (LCA), including wine and beer production in the UK [17,18], Spanish wine [19,20], whisky in Sweden [21], and Peruvian pisco [22], where the greatest impact comes from the agricultural stage. On the other hand, other LCA studies analysing alcoholic beverages point out that the main impact corresponds to the glass bottle production, as in the case of red wine in Catalonia, Spain [23] or white wine in Sardinia, Italy [24]. There is, however, a lack of life cycle perspective in the production stage of drinks obtained by distilling agave plants such as tequila The regulation for mezcal production [3] declares certain categories according to the machinery, equipment, or infrastructure used in each production process. These range from industrialized production to craft production or even ancestral. In the same way, the classes of young or white mezcal are established: matured in glass, rested, aged, flared or distilled according to the process carried out after the distillation.
The increase of interest in mezcal has involved, as with tequila, several problems. Some authors pointed to the shortage of wild agave [4], overproduction of maguey [5], pests in plants [6], and even inappropriate waste management [7], both solid and vinasse, which are the residual liquids of distilling the fermented must [6]. The residual liquid effluents or stillage are classified as a polluting product because they are disposed of at temperatures close to 90 • C, with a pH lower than 5.0 and a high chemical oxygen demand (COD) (50-150 g O 2 /L) [8]. By placing the stillage on the ground, the suspended solids cause a decrease in permeability, obstructing the pores of the soil [9].
In response to the problem of scarcity of raw materials in the agave-mezcal product system, sustainable practices have been proposed, focused mainly on the production of agave. For example, agroforestry management [10,11], the selection of potential areas for agave plantations [12], and even assisted plant reforestation programs, wet nurses, and agaves [13]. To reduce the problem of waste in the production of mezcal, studies have been carried out on the production of tiles made from a bio composite material, bagasse [14], and the production of biofuels through a treatment of bagasse [15] and adobes reinforced with agave fibre in Oaxaca [16].
The efforts made to solve the problems of mezcal, to date, have focused on specific stages of the production chain of said drink. However, there are still no reports that describe the environmental impact of mezcal production with a life cycle approach or indicators on the energy performance of the process.
There are several studies on the environmental impact of alcoholic beverages approached from the methodology of life cycle assessment (LCA), including wine and beer production in the UK [17,18], Spanish wine [19,20], whisky in Sweden [21], and Peruvian pisco [22], where the greatest impact comes from the agricultural stage. On the other hand, other LCA studies analysing alcoholic beverages point out that the main impact corresponds to the glass bottle production, as in the case of red wine in Catalonia, Spain [23] or white wine in Sardinia, Italy [24]. There is, however, a lack of life cycle perspective in the production stage of drinks obtained by distilling agave plants such as tequila or mezcal. To the best of the author's knowledge, no previous published studies are available that investigate impacts from a life cycle perspective.
On the other hand, Pimentel [25] points out that 25% of the world's fossil energy is used to produce food. The operation of current agro-ecosystems is based on two energy flows: the natural one corresponding to solar energy and an "auxiliary" flow, controlled by the farmer who resorts to the use of fossil fuels, either directly or indirectly, through the industrial inputs used in the production process. In addition to agro-ecosystems, studies have been carried out to find out the energy demand in other areas such as production processes. In the case of alcoholic beverages such as whisky [21], their primary energy consumption or energy obtained from either direct sources or combustible ones corresponded to 57.3 MJ/ 750 mL bottle (the energy equivalent of 1.63 L of gasoline). Olajire [26] mentions that a well-run brewery would use 1.125 MJ/ 750 mL of beer produced (the equivalent of 0.032 L of gasoline). In the case of agave distillates such as tequila or mezcal, this information was scarce within the bibliographic search carried out by the authors.
The present work evaluated the production of Agave cupreata mezcal in a vinata (the place where agave is processed to obtain a non-industrialized mezcal) of the community of Etucuaro. It is a drink of the artisan category, that is to say, it uses ancestral equipment and is of low technology in its production. It is classified as young because it is not subject to any type of post-distillation process, resulting in a colourless and translucent product [3]. The objective of the study was to identify the critical points of environmental impact from the point of view of LCA and to calculate the energy demand of the product including direct and indirect energy in order to calculate the productivity of the process. This information will serve as a basis for finding alternatives that increase the productivity of inputs and reduce the impact on the environment by improving the sustainable development of the activity.

Materials and Methods
The study evaluates two aspects of the craft mezcal production chain. First the information related to the LCA is shown and then the calculation of the energy demand. For both aspects the same stages and processes of mezcal production were taken.

Study Zone
The selected zone was Etucuaro, municipality of Madero, Michoacan de Ocampo, Mexico ( Figure 2). The municipality is bordered to the north by Morelia, the state capital. According to National Institute of Statistics and Geography (NISG) [27], luvisol (38%) and regosol (35%) soils predominate in the municipality. The climates that predominate are semi-warm and temperate sub humid (33 and 27% respectively) with rainfall ranging from 800 to 1300 mm. Its vegetation is mainly composed of forest (82%) with low use of agricultural (6%) and urban land (0.17%). Most of the population (54%) is dedicated to primary activities [28]. On the other hand, Pimentel [25] points out that 25% of the world's fossil energy is used to produce food. The operation of current agro-ecosystems is based on two energy flows: the natural one corresponding to solar energy and an "auxiliary" flow, controlled by the farmer who resorts to the use of fossil fuels, either directly or indirectly, through the industrial inputs used in the production process. In addition to agro-ecosystems, studies have been carried out to find out the energy demand in other areas such as production processes. In the case of alcoholic beverages such as whisky [21], their primary energy consumption or energy obtained from either direct sources or combustible ones corresponded to 57.3 MJ/ 750 mL bottle (the energy equivalent of 1.63 L of gasoline). Olajire [26] mentions that a well-run brewery would use 1.125 MJ/ 750 mL of beer produced (the equivalent of 0.032 L of gasoline). In the case of agave distillates such as tequila or mezcal, this information was scarce within the bibliographic search carried out by the authors.
The present work evaluated the production of Agave cupreata mezcal in a vinata (the place where agave is processed to obtain a non-industrialized mezcal) of the community of Etucuaro. It is a drink of the artisan category, that is to say, it uses ancestral equipment and is of low technology in its production. It is classified as young because it is not subject to any type of post-distillation process, resulting in a colourless and translucent product [3]. The objective of the study was to identify the critical points of environmental impact from the point of view of LCA and to calculate the energy demand of the product including direct and indirect energy in order to calculate the productivity of the process. This information will serve as a basis for finding alternatives that increase the productivity of inputs and reduce the impact on the environment by improving the sustainable development of the activity.

Materials and Methods
The study evaluates two aspects of the craft mezcal production chain. First the information related to the LCA is shown and then the calculation of the energy demand. For both aspects the same stages and processes of mezcal production were taken.

Study Zone
The selected zone was Etucuaro, municipality of Madero, Michoacan de Ocampo, Mexico ( Figure 2). The municipality is bordered to the north by Morelia, the state capital. According to National Institute of Statistics and Geography (NISG) [27], luvisol (38%) and regosol (35%) soils predominate in the municipality. The climates that predominate are semi-warm and temperate sub humid (33 and 27% respectively) with rainfall ranging from 800 to 1300 mm. Its vegetation is mainly composed of forest (82%) with low use of agricultural (6%) and urban land (0.17%). Most of the population (54%) is dedicated to primary activities [28].  [29]. Figure 2. Study zone [29]. Identify the critical points of environmental impact in the craft mezcal production chain under the LCA methodology.

Scope
The present study included everything from obtaining raw materials (cradle) to the end of life or waste disposal (grave) as shown in Figure 3. The data for the system analysis were collected directly in the field by the authors and correspond to the production of a young mezcal batch in 2017. The analysis was performed under the ReCiPe method to calculate the potential environmental impact using Simapro software (version 8.2.3.0., PRé Sustainability, Amersfoort, The Netherlands). The impact categories evaluated for the analysis were those considered mid-point by the ReCiPe method, although the results only express those considered significant, that is, when they represent more than 75% of potential impact. The present study included everything from obtaining raw materials (cradle) to the end of life or waste disposal (grave) as shown in Figure 3. The data for the system analysis were collected directly in the field by the authors and correspond to the production of a young mezcal batch in 2017. The analysis was performed under the ReCiPe method to calculate the potential environmental impact using Simapro software (version 8.2.3.0., PRé Sustainability, Amersfoort, The Netherlands). The impact categories evaluated for the analysis were those considered mid-point by the ReCiPe method, although the results only express those considered significant, that is, when they represent more than 75% of potential impact. The functional unit (FU) is a young craft mezcal bottle of 750 mL with 46% Vol. Alc. made from a vinata in Etucuaro, Michoacan. The stages of the production process studied were: (1) obtaining raw material, (2) manufacturing/ processing, (3) blending and distribution, and (4) end of life.
In the analysis, it was considered that obtaining raw materials takes approximately seven years, from the germination of the seed until the agave reaches sexual maturity and is harvested just before flowering, when the sugar concentration is at its maximum [30]. The agave used for mezcal production was semi-cultivated, remaining three years in the nursery and four years in the wild. The processing stage together with the packaging stage lasts approximately two months. For the cooking of the agave hearts, it was identified that oak firewood obtained from the region is used because it provides organoleptic properties to the mezcal, while the double distillation uses residual wood obtained from a nearby lumber warehouse. For the milling, the inputs used by the tractor are included, but the impact of the manufacture of the equipment is not. Since data were collected in 2017, mezcal production was governed by the indications of NOM-070-SCFI-1994 [31], which allowed the production of type I mezcal with 100% agave sugars and type II which allows up to 20% use of other carbohydrates. The studied vinata produced type II mezcal, and sugarcane was added. Currently, the NOM-070-SCFI-2016 [3] is in force, which only allows the production of 100% agave mezcal. For the end of life, a landfill scenario was considered due it being the final disposal site for The functional unit (FU) is a young craft mezcal bottle of 750 mL with 46% Vol. Alc. made from a vinata in Etucuaro, Michoacan. The stages of the production process studied were: (1) obtaining raw material, (2) manufacturing/processing, (3) blending and distribution, and (4) end of life.
In the analysis, it was considered that obtaining raw materials takes approximately seven years, from the germination of the seed until the agave reaches sexual maturity and is harvested just before flowering, when the sugar concentration is at its maximum [30]. The agave used for mezcal production was semi-cultivated, remaining three years in the nursery and four years in the wild. The processing stage together with the packaging stage lasts approximately two months. For the cooking of the agave hearts, it was identified that oak firewood obtained from the region is used because it provides organoleptic properties to the mezcal, while the double distillation uses residual wood obtained from a nearby lumber warehouse. For the milling, the inputs used by the tractor are included, but the impact of the manufacture of the equipment is not. Since data were collected in 2017, mezcal production was governed by the indications of NOM-070-SCFI-1994 [31], which allowed the production of type I mezcal with 100% agave sugars and type II which allows up to 20% use of other carbohydrates. The studied vinata produced type II mezcal, and sugarcane was added. Currently, the NOM-070-SCFI-2016 [3] is in force, which only allows the production of 100% agave mezcal. For the end of life, a landfill scenario was considered due it being the final disposal site for 78.5% of the waste in Mexico [32]. It includes the transport used in production from obtaining raw materials to the distribution of the final product.
The purchase of agave in the stage of obtaining raw materials was not considered for the present analysis, neither was the recycling of waste in the end of life stage because it only represents the destiny of 9.6% of the waste in Mexico [32]. The potential impact of the production of machinery, equipment, or infrastructure in the craft mezcal category (the oven for cooking, the tractor for grinding, the fermentation vats, the type or material of the stills) was not included. Transport used after the distribution of the final product was not considered either because the location of the final consumer and the distance between the consumer and the waste disposal site is unknown.

Types of Mezcal Considering in the Study
Craft Mezcal Type II: This is the product obtained from the distillation and rectification of musts whose formulation are up to 20% of other carbohydrates allowed by the corresponding legal provisions [31].
Craft Mezcal: The formulation consists of 100% agave. The cooking of agave is in a floor oven, stone or masonry. The traditional equipment used for milling uses human strength or power tools. The distillation is with direct fire in copper stills, clay pot, or stainless steel, and the process can include the maguey fibre (bagasse) [3].
Ancestral mezcal. The production is in a rustic way with a 100% agave formulation. The cooking of agave is in a conical oven on the floor (earth). For milling, is used only the force of man or animals without electrical tools. The distillation only uses direct fire in the clay pots and must include the fibre of the maguey (bagasse) [3]. The price of this type of mezcal is higher compared to a craft mezcal. This because of the losses in the process and the added value of being made manually.

Life Cycle Inventory Analysis
The inventory considered the system inputs and outputs by production stage required per FU ( Table 1). The raw materials included the agrochemicals and water used, the inputs required to prepare the soil for transplanting, the equipment for harvesting the agave stems (also called hearts), and the transport from the field to the vinata (three trips). Inputs for cooking, milling, fermentation, and distillation were included in the processing. The blending and distribution of the final product considered a 750 mL transparent glass bottle, label and screw cap assembled by hand, for distribution a trip was considered to cover the delivery route. The end-of-life scenario was the disposal of waste in a landfill.

Energy Demand of Craft Mezcal Production
The energy demand of a production process is the energy used to create a production unit (kg, L, bottle, piece, among others). For the calculation of the total energy demand in the production of craft mezcal, the same FU was taken as for the LCA, and therefore the same stages apply. Direct and indirect energy sources were considered for each production stage. Energy equivalence data for inputs were obtained from literature reviews plus our own calculations from these or field data ( Table 2). For some data such as firewood or wood, polypropylene (PP) and cardboard, the calorific potential was used, as well as own calculations. In the processing stage, the agave input was no longer included in the calculation to avoid double counting, because it was the result of the raw material stage. Note: * The unit is the one represented for each input.
From the total energy demand used and the total product produced, the specific energy (SE) (Equation (1)) and the energy productivity (EP) (Equation (2) were calculated, which represent the relationship between a product and the energy invested in a production process, valued in megajoules (MJ) [43].

SE =
Energy used (MJ) Product produced (unit f or measurement) (1) Table 3 shows the percentage of environmental impact contribution by each stage of the production process of a 750 mL bottle of young craft mezcal (46% Vol. Alc.), compared to two other similar categories in the current standard (NOM-070-SCFI-2016). The category "mezcal" is usually industrialized, so comparing these results was not considered. In all three categories, the stages with the greatest environmental impact were manufacturing/processing and obtaining raw materials. The use of sugar in type II mezcal generated changes in the distribution of the impact of the process stages, unlike the Sustainability 2020, 12, 8242 8 of 17 use of mechanical mills which did not represent a substantial change in the environmental impact of the process.  Table 3 shows the percentage of environmental impact contribution by each stage of the production process of a 750 mL bottle of young craft mezcal (46% Vol. Alc.), compared to two other similar categories in the current standard (NOM-070-SCFI-2016). The category "mezcal" is usually industrialized, so comparing these results was not considered. In all three categories, the stages with the greatest environmental impact were manufacturing/processing and obtaining raw materials. The use of sugar in type II mezcal generated changes in the distribution of the impact of the process stages, unlike the use of mechanical mills which did not represent a substantial change in the environmental impact of the process.  Table 4 indicates the categories of environmental impact that were significant with a value greater than 75% in the production of mezcal for the stages of obtaining raw materials and manufacturing/processing. In type II craft mezcal, the use of sugar provides three additional impact categories regarding the category of craft and ancestral mezcal in manufacturing/processing stage. On the other hand, in the raw materials stage, it has the lowest percentage regarding the contribution of water depletion at 86.5%.  Table 3 shows the percentage of environmental impact contribution by each stage of the production process of a 750 mL bottle of young craft mezcal (46% Vol. Alc.), compared to two other similar categories in the current standard (NOM-070-SCFI-2016). The category "mezcal" is usually industrialized, so comparing these results was not considered. In all three categories, the stages with the greatest environmental impact were manufacturing/processing and obtaining raw materials. The use of sugar in type II mezcal generated changes in the distribution of the impact of the process stages, unlike the use of mechanical mills which did not represent a substantial change in the environmental impact of the process.  Table 4 indicates the categories of environmental impact that were significant with a value greater than 75% in the production of mezcal for the stages of obtaining raw materials and manufacturing/processing. In type II craft mezcal, the use of sugar provides three additional impact categories regarding the category of craft and ancestral mezcal in manufacturing/processing stage. On the other hand, in the raw materials stage, it has the lowest percentage regarding the contribution of water depletion at 86.5%.  Table 3 shows the percentage of environmental impact contribution by each stage of the production process of a 750 mL bottle of young craft mezcal (46% Vol. Alc.), compared to two other similar categories in the current standard (NOM-070-SCFI-2016). The category "mezcal" is usually industrialized, so comparing these results was not considered. In all three categories, the stages with the greatest environmental impact were manufacturing/processing and obtaining raw materials. The use of sugar in type II mezcal generated changes in the distribution of the impact of the process stages, unlike the use of mechanical mills which did not represent a substantial change in the environmental impact of the process.  Table 4 indicates the categories of environmental impact that were significant with a value greater than 75% in the production of mezcal for the stages of obtaining raw materials and manufacturing/processing. In type II craft mezcal, the use of sugar provides three additional impact categories regarding the category of craft and ancestral mezcal in manufacturing/processing stage. On the other hand, in the raw materials stage, it has the lowest percentage regarding the contribution of water depletion at 86.5%.  Table 3 shows the percentage of environmental impact contribution by each stage of the production process of a 750 mL bottle of young craft mezcal (46% Vol. Alc.), compared to two other similar categories in the current standard (NOM-070-SCFI-2016). The category "mezcal" is usually industrialized, so comparing these results was not considered. In all three categories, the stages with the greatest environmental impact were manufacturing/processing and obtaining raw materials. The use of sugar in type II mezcal generated changes in the distribution of the impact of the process stages, unlike the use of mechanical mills which did not represent a substantial change in the environmental impact of the process.  Table 4 indicates the categories of environmental impact that were significant with a value greater than 75% in the production of mezcal for the stages of obtaining raw materials and manufacturing/processing. In type II craft mezcal, the use of sugar provides three additional impact categories regarding the category of craft and ancestral mezcal in manufacturing/processing stage. On the other hand, in the raw materials stage, it has the lowest percentage regarding the contribution of water depletion at 86.5%.  Table 4 indicates the categories of environmental impact that were significant with a value greater than 75% in the production of mezcal for the stages of obtaining raw materials and manufacturing/processing. In type II craft mezcal, the use of sugar provides three additional impact categories regarding the category of craft and ancestral mezcal in manufacturing/processing stage. On the other hand, in the raw materials stage, it has the lowest percentage regarding the contribution of water depletion at 86.5%.  Table 5 indicates that, for the type II craft mezcal category, in the manufacturing/processing stage, the processes that presented a greater environmental impact were fermentation and distillation, contributing 56.3% and 40.9%, respectively. In the alternative mezcal categories, the fermentation process was less than 5% impact while distillation contributed about 90%. The second stage with the greatest impact on mezcal production was obtaining raw materials ( Table 3). The process that most contributed to the environmental impact within that stage was the production of young agave plants (55.8%). This process comprises the germination of the seed to obtain the three-year-old plant, which can be transplanted in the field.

Interpreting the Potential Impact of Craft Mezcal Production
The stage with the greatest impact on mezcal production in the categories mentioned in Table 3 was manufacturing/processing. In the production of type II craft mezcal, the potential impact of the manufacturing/processing stage was related to the use of sugarcane. Table 6 reveals that within the fermentation process, the sugarcane input contributed 96% of the environmental impact and had a representation of more than 95% in 13 of 18 impact categories. For the distillation process, the wood used contributed 94% of the environmental impact and represented over 95% in 10 of 18 impact categories. For the young agave plant, the use of the herbicide provided 82% of the environmental impact and contributed over 95% in 10 of 18 impact categories.
Unlike some LCA studies applied to alcoholic beverages, where their agricultural stage provided the greatest environmental impact, in the present study it was found that for mezcal the production the manufacturing/processing stage provides it. For example, Gazulla et al. [20] reported that the impact of their agricultural stage on wine production is due to greenhouse gas (GHG) emissions produced by the use of fertilisers. They found a use of 0.7 kg per year of fertiliser per 750 mL bottle. For mezcal production, they use 0.003 kg of fertiliser per bottle of the same size, but unlike grapes, agave uses fertiliser only in the first three years of the seven-year production cycle.  Total significant categories  13  10  10 Note: * Percentage contribution in processing for each input.
In mezcal production, the environmental impact resulted from the addition of sugarcane used in the processing stage. Figure 4 shows the origin of the impact of sugarcane obtained from the Simapro software database (version 8.2.3.0., PRé Sustainability, Amersfoort, The Netherlands). It is observed that irrigation and the use of nitrogenous fertilisers are the inputs with the greatest impact in the agricultural phase, contributing approximately 41.9%. Within its industrial part, transport stands out. In total, 0.344 kg of CO 2 eq was generated per kg of sugarcane. Unlike some LCA studies applied to alcoholic beverages, where their agricultural stage provided the greatest environmental impact, in the present study it was found that for mezcal the production the manufacturing/processing stage provides it. For example, Gazulla et al. [20] reported that the impact of their agricultural stage on wine production is due to greenhouse gas (GHG) emissions produced by the use of fertilisers. They found a use of 0.7 kg per year of fertiliser per 750 mL bottle. For mezcal production, they use 0.003 kg of fertiliser per bottle of the same size, but unlike grapes, agave uses fertiliser only in the first three years of the seven-year production cycle.
In mezcal production, the environmental impact resulted from the addition of sugarcane used in the processing stage. Figure 4 shows the origin of the impact of sugarcane obtained from the Simapro software database (version 8.2.3.0., PRé Sustainability, Amersfoort, The Netherlands). It is observed that irrigation and the use of nitrogenous fertilisers are the inputs with the greatest impact in the agricultural phase, contributing approximately 41.9%. Within its industrial part, transport stands out. In total, 0.344 kg of CO2eq was generated per kg of sugarcane. In the category of craft and ancestral mezcal subject to NOM-070-SCFI-2016 that do not use sugar in their process, they present a different distribution of environmental impact. The input that contributes most is the wood used during distillation within the processing stage. Obtaining this input involves a forestry process and an industrial process. Figure 5 shows the nitrogen products, electricity and resins as the inputs with the greatest impact on the production of waste wood, generating a total of 52 kg of CO2eq per m 3 of waste wood. In the category of craft and ancestral mezcal subject to NOM-070-SCFI-2016 that do not use sugar in their process, they present a different distribution of environmental impact. The input that contributes most is the wood used during distillation within the processing stage. Obtaining this input involves a forestry process and an industrial process. Figure 5 shows the nitrogen products, electricity and resins as the inputs with the greatest impact on the production of waste wood, generating a total of 52 kg of CO 2 eq per m 3 of waste wood. In a comparative study of LCAs in various Peruvian pisco producers, Vázquez-Rowe et al. [22] show that the impact on the distillation stage varies significantly depending on the fuel. When using wood, about 15 g CO2eq are generated per 500 mL bottle, while fossil fuels generate about 350 g. In the case of the craft mezcal studied, a comparison was made between two fuels that cover the energy demand in the FU distillation (72.43 MJ). Table 7 shows the significant categories of GHGs generated if 5 kg of wood or 1.6 kg of propane is used. Values of less than 1 kg of contribution in the impact category were not considered significant. Wood was found to have a lower impact on the environment than propane. It is important to mention that the difference between the amount of fuel used is not negligible and its choice depends on other factors, such as costs, storage, transport, and other considerations of each producer. Eriksson et al. [21] reported that 24.9% of the energy in their process is invested in distillation, which comes mostly from renewable sources, such as biomass combustion, thus reducing the environmental impact of this stage. In the stage of obtaining raw materials, the use of the herbicide has great representation during the obtaining of the young agave plant, a process that lasts three years and uses 0.07 kg of herbicide per 750 mL bottle of mezcal. In other studies, the use of fertilisers or pesticides is more represented than herbicides, such as in the production of pisco [22] or wine [18]. Figure 6 shows the most representative inputs in the impact of the production of 1 kg of herbicide, whose process generates In a comparative study of LCAs in various Peruvian pisco producers, Vázquez-Rowe et al. [22] show that the impact on the distillation stage varies significantly depending on the fuel. When using wood, about 15 g CO 2 eq are generated per 500 mL bottle, while fossil fuels generate about 350 g. In the case of the craft mezcal studied, a comparison was made between two fuels that cover the energy demand in the FU distillation (72.43 MJ). Table 7 shows the significant categories of GHGs generated if 5 kg of wood or 1.6 kg of propane is used. Values of less than 1 kg of contribution in the impact category were not considered significant. Wood was found to have a lower impact on the environment than propane. It is important to mention that the difference between the amount of fuel used is not negligible and its choice depends on other factors, such as costs, storage, transport, and other considerations of each producer. Eriksson et al. [21] reported that 24.9% of the energy in their process is invested in distillation, which comes mostly from renewable sources, such as biomass combustion, thus reducing the environmental impact of this stage. In the stage of obtaining raw materials, the use of the herbicide has great representation during the obtaining of the young agave plant, a process that lasts three years and uses 0.07 kg of herbicide per 750 mL bottle of mezcal. In other studies, the use of fertilisers or pesticides is more represented than herbicides, such as in the production of pisco [22] or wine [18]. Figure 6 shows the most representative inputs in the impact of the production of 1 kg of herbicide, whose process generates 10.2 kg of CO 2 eq. Sustainability 2020, 12, x FOR PEER REVIEW 12 of 17 Figure 6. Inputs with the largest impact on the herbicide production. Table 8 shows a comparison of CO2 emissions for various alcoholic beverages. The studies corresponding to distilled beverages show values greater than 1.5 kg of CO2 except as reported by Leivas et al. [44] for gin. The difference in beverage emissions is most likely due to variations in the product life cycle or system boundaries of each study. For example, gin has the lowest emissions value (0.58 kg CO2) because its agricultural stage does not generate impacts since it obtains its botanical resources from the wild [41]. On the other hand, Vázquez-Rowe et al. [22] show for Peruvian pisco the average emissions of several wineries, with 3.37 kg of CO2 being the highest value.

Energy Demand for Craft Mezcal Production
Agro-ecosystems are demanding energy from various sources, from solar energy for biomass development, to energy from fossil sources when they require heating such as in greenhouses or agricultural machinery. In the case of mezcal production, the various stages demand various sources of energy, either for primary use or for the processing of its inputs. Table 9 shows the energy demand of type II craft mezcal production obtained from Table 2, by life cycle stage and by its relationship with the various inputs, as well as the percentage contribution of the latter. A total of 163.8 MJ/bottle (750 mL) was identified as required. The stage with the highest energy requirement was manufacturing/processing. This stage is a relatively short process that lasts from one to two months, which, nevertheless, demands 73.4% of the whole production process to obtain the mezcal product, due to the use of firewood and wood. The stage of obtaining raw  Table 8 shows a comparison of CO 2 emissions for various alcoholic beverages. The studies corresponding to distilled beverages show values greater than 1.5 kg of CO 2 except as reported by Leivas et al. [44] for gin. The difference in beverage emissions is most likely due to variations in the product life cycle or system boundaries of each study. For example, gin has the lowest emissions value (0.58 kg CO 2 ) because its agricultural stage does not generate impacts since it obtains its botanical resources from the wild [41]. On the other hand, Vázquez-Rowe et al. [22] show for Peruvian pisco the average emissions of several wineries, with 3.37 kg of CO 2 being the highest value.

Energy Demand for Craft Mezcal Production
Agro-ecosystems are demanding energy from various sources, from solar energy for biomass development, to energy from fossil sources when they require heating such as in greenhouses or agricultural machinery. In the case of mezcal production, the various stages demand various sources of energy, either for primary use or for the processing of its inputs. Table 9 shows the energy demand of type II craft mezcal production obtained from Table 2, by life cycle stage and by its relationship with the various inputs, as well as the percentage contribution of the latter. A total of 163.8 MJ/bottle (750 mL) was identified as required. The stage with the highest energy requirement was manufacturing/processing. This stage is a relatively short process that lasts from one to two months, which, nevertheless, demands 73.4% of the whole production process to obtain the mezcal product, due to the use of firewood and wood. The stage of obtaining raw materials, despite a duration of about seven years, requires 21.4% of the total energy for the production of agave hearts. This is due to the impact generated by the use of herbicides. The agave used in this study is semi-cultivated, which could explain why its greatest impact stage is not obtaining raw materials, as in other studies where a greater use of agrochemicals is required. In craft mezcal production, the main source of energy came from burning biomass. Other sources such as electricity or petroleum derivatives, e.g., agrochemicals and hydrocarbons for transportation, had less participation. Notes: * Total value of the previous stage so not counted in order to avoid double counting; ** Σ Others, included the following categories: use of chainsaw, fertiliser, pesticide, water, seeds, tractor, diesel transport, electric power, firewood and wood for agave heart cooking and distillation, PP screw caps, sticky paper labels, cardboard boxes.
The energy demand for mezcal production differed from other beverages such as whisky reported by Eriksson et al. [21] with 57.3 MJ/ 750 mL bottle. However, whisky production only considered energy obtained from direct sources (electricity and fossil fuels), while in this study, mezcal included the energy required for input production. Table 10 shows the comparison of SE and EP in alcoholic drinks and spirits around the world, including water as a reference drink. The mezcal production process had an SE value of 21.8 MJ/dl and an EP of 0.05 dl of mezcal for each MJ. Of the references cited, the beer production analysed by Olajire [26] had the lowest value for SE (0.19) and the highest for EP (5.18), which means that more product is obtained with less energy. It should be noted that differences in the values of SE and EP may be due to differing processes, inputs or equipment used, their origin, and the source and type of energy (direct or indirect).

Conclusions
This study identified the critical points of environmental impact and energy demand for craft mezcal production, resulting in the manufacturing/processing stage for both cases, unlike other studies whose main impact comes from obtaining raw materials. During the comparison of mezcal categories, it was found that changing a key input, such as sugarcane, wood, or herbicide, redistributes the percentage contribution to environmental impact. The energy demand of craft mezcal production has a value of 163.8 MJ/bottle (750 mL) equivalent to 4.62 L of gasoline. The CO 2 emissions of spirits whose process is industrialised and whose life cycle includes everything from obtaining raw materials to the end of life show higher values than spirits whose process is carried out by hand or whose raw materials are obtained from the wild. In stricto sensu, the non-technical and ancestral production of mezcal is difficult to compare because of the variants in machinery, equipment, and infrastructure. In mezcal's life cycle, for example, some agaves are only obtained in the wild, while others, because they are endemic species, only develop in the space that meets the required characteristics, and there are also species that only reproduce through seed, like the agave mentioned in this study. However, in general terms, mezcal production has homogeneous stages. For this reason, we consider that this study could be valid for others vinatas or palenques with same technification level.

Recommendations
Based on the results, we believe that it is possible to reduce the environmental impact and energy consumption for mezcal production, so the following recommendations are made for future research.

•
Carry out a study in other vinatas or palenques of the equivalent category of mezcal to compare in other regions the critical points of environmental impact.

•
Compare production processes with alternatives that may have less impact, for example, the use of renewable energies, organic agriculture, recycling, or waste management.

•
Establish a guide of good environmental practices for craft mezcal production.