The first step was to calculate the relative impact weights (RIWs) by collecting the data for the country’s total GHGE, WU, EU, LU and population for a time scale of 17 years. The collected data is for calculating the relative impact weight (RIW) results for the environmental impact categories as mentioned above. In the second step, the CWIs or eco-potential points of the product were calculated by dividing the environmental impacts of the product (GHGE, WU, ALU, and EU) by the RIWs calculated in the first step. In the third and fourth step the same methodology for calculating the RIW, CWI and eco-points was applied to the environmental impacts of the protein supplies in the selected countries only. We then carry out an analysis of the eco-potential of nine conventional protein sources [i.e., peas, soybeans (tofu), groundnut, beef, cheese, pork, eggs, fish (farmed) and poultry meat] using the developed weighting methods according to geographical aspects (148 countries), resource (land, climate, water and energy uses), time scale (17 years) and population aspects. Next, we assess the applicability of the eco-potential weighting method to identify the additional potential environmental impact of alternative protein sources (i.e., crickets, mealworms, black soldier fly, housefly, cultured meat, Chlorella, and Spirulina) in relation to the aforementioned aspects and the impact of the current protein supply.

The relative environmental impacts and eco-potentials of the target products were calculated in four application steps as explained in the Table 2. In the first step (I), impact categories (country level, annual values) of arable land use, greenhouse gas emissions, water use, and energy use were selected. In the second step (II), normalized values of relative impact weights (RIW) are calculated in 4 sub-steps for the selected impact categories. In step (II), the four impact indicators (ALU, GHGE, WU, and EU) are normalized by the ALU and population in the selected countries (RIW1–4). Similarly, RIWpr1–4 (protein related) values are also calculated. For the calculation of the protein related RIW, in the first step (Ia), data on the dietary protein supply (g/capita/day) per country and the environmental impacts (GWP, WU, EU, and ALU) of the food protein sources were collected. An average of the selected protein sources was again calculated for the environmental impact indicators (GWP, WU, EU, and ALU). The average values of these impact indicators were then multiplied by the converted amount of protein supply (g/capita/year). In step (IIa), values for RIWpr1–4 are calculated by normalizing the protein-related impact indicators (ALUpr, GHGEpr, WUpr and EUpr) by ALU and population. In step (III), the eco-potential points (EPP1–4) for the target food products in the different countries are calculated by dividing the overall average environmental impacts of the product by the country’s total impact weights (RIW1–4) and similarly with the protein-related Relative impact weights (RIWpr1–4) in step (IIIa). Finally, in steps (IV and IVa) the product-specific eco-potential points (EPP1–4) were integrated by summing them to obtain the final result.

First, data were collected for total GHGE, WU, EU, ALU and population for 148 countries (Supplementary Table S1). Data for resource use and GHGE were only available for 148 of the 195 countries. The weighting method is intended to be dynamic and should therefore reflect the variations in the annual environmental impacts and indicators used as weights. To determine the sensitivity of the proposed weights to annual variations, data were collected for from 2000 to 2016 (more recent data were not available). Data on the environmental impacts of conventional and alternative protein sources were also collected (Supplementary Table S1).

Data on environmental impact indicators of global warming, water use and land use for conventional protein sources were collected from Poore and Nemecek (2018), while data on the environmental impacts of alternative protein sources were collected from different sources (Supplementary Table S1). In addition, the original environmental impact data, expressed as total average yields for each product presented in the data source, expressed in terms of its 100 g nutritional units (NU), were converted to 1,000 g or 1 kg of protein per product. The calculations behind these conversions are presented in Supplementary Table S1. The data were converted to show the environmental impact of the global average yield of 1 kg of protein per product. The environmental impact values for conventional and alternative protein sources are presented below in Tables 2, 3.

It’s clear from the results for the country specific RIWs, that they are dynamic over time. Many factors that change over time (e.g., GHGE, WU, ALU and EU) contribute to the temporal changes in the results. Only the factors GHGE, WU, EU, and ALU were selected because water-land-energy emissions are an important basis to reduce the GHGE from agricultural activities (Deng et al., 2021). For example, RIW2 for China reflects the dynamics associated with increasing GHG emissions and decreasing arable land (Figure 1). Similarly, the values of the other RIWs (1–4) depend on the initial impacts considered (arable land, GHGE, WU, and EU) and the values used for the normalization of the impacts (population and arable land). The dynamics of the indicated factors would also determine the dynamics of RIWs. RIWs calculated for 148 countries are included in Supplementary Table S1.

Country weighted impacts represent the impact of a product in a selected impact category, weighted by the relevant relative impact weight (RIW). Temporal dynamic changes in the country weighted impacts (country-level eco-potential) are caused by the dynamic weights as discussed above (Figure 1), but also by changes in the impacts of the product in the same category (if the changes are known). By definition “temporal dynamic changes” means the changes in the relative impact weights (country’s eco-potential) triggered by the change in the country’s total environmental impacts throughout the year. The country weighted impact (CWI) from GHGE, based on 1 kg of beef production, is shown in Figure 2. The analysis shows that the eco-potential is highest in India and lowest in the United States and the Russian Federation. The values of the eco-potential for beef production is highest in India as the country’s relative impact weight (RIW) for GHGE is lowest (Figure 1) which means there is lesser burden on the environment when the country’s total GHGE were normalized by its arable land use and population. So, producing beef in the country would relatively have a lower additional impact when compared with the other chosen countries (China, United States, and Russia). Calculated CWIs for 148 countries are provided in Supplementary Table S1.

The same normalization and weighting base defined for relative impact weights (RIW) and country weighted impacts (according to the methodological framework in Table 1) allows to obtain results in points that can be summarized to reflect the combined potential for the increased impact in the selected categories and for the country of interest. These combined eco-potential points obtained for food products or protein sources show the relative potential environmental impact of the production of the food products with respect to the country-specific situation, taking into account several impact categories (see Table 4).

The eco-potential of beef production (Figure 3) showed that, among comparable countries, India could be a representative country for beef production with the lowest incremental environmental impact. The incremental impact of beef production is much lower relative to the amount of resources normalized to population and available arable land (Figure 3). In contrast, the United States is the worst performing country, where the additional environmental impacts of one unit of beef production is associated with a higher additional environmental burden due to the country’s existing environmental impacts. Similar trends are observed for alternative protein products when the same weighting system is applied (Figure 4). Therefore, the application of higher integrated relative impact weights results in less beneficial eco-potential (potential for environmentally beneficial production of new products). The graphs for the other products can be found in Supplementary Table S1.

the same methodology was used to calculate the RIWs. The main difference here is that it focuses only on the impacts of the level of protein supplies in the different countries. The results for the country specific RIWs show that they are dynamic over time. The average environmental impacts values of the selected alternative protein sources were calculated and multiplied by the amount of protein supplies in the countries. The average environmental impacts for the four environmental impact indicators of the protein supplies (g/capita/year) in the countries were divided by the arable land use per country. The results for the RIW of global warming according to the protein supplies from alternative protein sources in Figure 5, show that China and Russia have the highest weights for total GHGE according to protein supplies from alternative protein sources. Overall, the results show that the RIW for GHGE or resource use will be higher in those countries where the relative emissions or resource use is comparatively higher than their arable land use and population and vice versa.

As calculated above, the country weighted impacts represent the impact of a product in a selected impact category, weighted by the relative impact weight (RIW). Dynamic changes over time in the country weighted impacts (country-level eco-potential) are caused by the dynamic weights as discussed above (Figure 5). In addition, changes in protein supply in countries over time also affect the eco-potential of the target product. The target product used as an example shown in Figure 6 is Tenebrio molitor (mealworm). The results show that the production of 1 kg of protein from mealworm has the highest GHG eco-potential in India compared to the other countries. A higher eco-potential indicates the highest potential for lower additional GHGE weighted by arable land per capita. Country-level weighted impacts (CWI) of a product indicates its eco-potential in a particular environmental impact category. The higher bars in Figure 6. Indicate that there is a higher probability of lower additional environmental impacts when compared with other countries.

The same methodology (according to the methodological framework in Table 1) as discussed above is applied here to obtain the combined eco-potential of the target product shown in Figure 7. The results show that the country weighted impacts (country-level eco-potential) (GHGE, EU, ALU and WU) from the total protein production of the alternative protein sources for mealworm alone, when integrated together, give India the highest eco-potential points when compared to other countries. The eco-potential of the mealworm alternative protein source showed that India could again be a representative country for the mealworm production, which would result in the lowest additional environmental impact. The additional impact of mealworm production will be much lower relative to the amount of resources normalized to population and available arable land (Figure 7). In contrast, Russia is the worst performing country, where the additional environmental impacts of one unit of mealworm production is associated with a higher additional environmental burden due to the existing environmental impacts from protein supply in the country. Graphs for the other products are presented in Supplementary Table S2.

CWIs for GWP, WU, EU and ALU are also compared with each other in order to show the dynamics between them. For example, below in Figure 8, CWIs (1–4) for 1 kg of beef are compared with each other. Analysis of the compared results shows differences in the range of eco-potential points between the selected environmental impact categories. The CWI (GWP) for 1 kg of beef is highest in India while the CWI (WU) is the second highest. CWI (EU and LU) shows the lowest eco-potential points. This shows that CWI (GWP) has a higher contribution to the cumulated eco-potential point when aggregated together with CWI (WU, EU, and LU). This means that the GWP of 1 kg of beef, normalized for arable land use and population, has the lowest environmental impact contribution when compared with water use, energy use and land use. While the CWI (WU) for 1 kg of beef is the highest in China, the United States and the Russian Federation and the CWI (GWP) is the second highest, this shows that the CWI (WU) has the highest contribution to the cumulated eco-potential point when aggregated with the CWI (GWP, EU, and ALU). This means that WU normalized by arable land use and population from 1 kg of beef has the lowest environmental impact contribution in China, the United States and the Russian Federation while CWI (GWP) has the second lowest environmental impact contribution. CWI (EU and ALU) had the lowest points in all countries, meaning that they have the most negative environmental impacts. When all the points are added up to cumulative eco-potential point, India shows the highest due to the higher contribution of CWI (GWP and WU) compared to other countries (Figure 9).

The results of this study for the Country Weighted Impacts (CWI) of conventional protein sources for the environmental impact categories of GWP, WU and ALU were compared with unweighted environmental impact values from the scientific literature (Poore and Nemecek, 2018). Comparisons were possible for the three environmental impact categories of GWP, WU and ALU. However, comparisons were not possible for all countries and energy use impact category due to a lack of data in the literature.

CWI values are dynamic and change annually. For the analysis of comparisons between CWI derived from conventional protein sources and their country-level absolute environmental impacts, an average CWI value from 2000 to 2016 was taken for each country, and the average value for the country-level absolute environmental impacts were taken from scientific literature.

It is important to note that the CWI indicates the eco-potential for the additional production of a unit in a particular impact category. Therefore, higher CWI values represent the better option for the production with a lower relative impact than the option with a lower CWI. The analysis shows that while the absolute emissions or resource use of the product may be higher in a given country, the CWI may be lower in the same country.

The analysis of the absolute impact values for the production of different protein sources and their weighted values showed that they vary. For example, based on the LCA results for GHGE (kgCO₂eq) from soybean, India has the highest GHGE while the GHGE-related CWI is the highest in India compared to other countries. Therefore, the values are not contradictory, but rather complementary, the absolute values indicate the high absolute negative impact of production while the proposed CWI indicates a high relative eco-potential of production compared to existing resources and emissions (probability of lower relative impacts).

If we consider soybean production in Canada (which has the lowest GHGE per FU among compared countries), then the absolute impact of tofu production is low (could be related to the geographical suitability and advances in production technologies). At the same time, Canada’s CWI is low, indicating a low GHG related eco-potential. This implies that an additional unit of production would result in a relatively higher additional environmental burden in Canada. However, not all the impacts and their CWI are related in the same way. This shows that normalizing the impact values to the available resources and using the normalized values as weights it is possible to define the country-level eco-potential of a product that is not related to the absolute impact value of production.

The proposed method for estimating the eco-potential of new products could be related to the eco-efficiency methods. While the concept of eco-efficiency reflects the current value of a product in relation to its function, it does not have a long-term strategy that can deal with rebound effects. Therefore, an increase in eco-efficiency may lead to an increase of consumption (rebound effect) and therefore a decrease in overall sustainability (Bjørn and Hauschild, 2012). Eco-efficiency focuses only on the resource consumption and waste emissions related to the current state of production, and rarely on the state required for long-term sustainability (Bjørn and Hauschild, 2012). Production may be considered eco-efficient in terms of the functional unit, but its relative impact in terms of the overall impact on the environment and available resources may be quite “costly.”

The proposed eco-potential weighting system offers the possibility of an additional indicator that defines the relationship between environmental impacts and the available capacity of the environment at the country level in a dynamic approach. It is not intended to replace the environmental impacts of midpoint impact categories, nor is it intended to act as a substitute for eco-efficiency indicators.

In our methodology, we are not mentioning that there is an additional burden due to the country’s existing environmental impacts. We normalize our results available arable land use per capita. So, the impact weights for country’s depend on various factors, not just the total emissions. When a country’s environmental impacts are higher than its current arable land use per capita, the burden shifts towards the impacts from food production/consumption. Even if a new product has a small impact, it adds to the burden per person’s available land. This means more land will be needed with additional production. This burden shift can happen within a country, not to other countries.

Our approach aims to provide complementary measurements that include the relationship with the country’s environment. For example, the production of a protein source is not beneficial to the ecosystem components, as all human activities are inherently harmful to the environment (considering only absolute impact values). On the other hand, country weighted eco-potential indicators can help to reduce environmental burden by shifting the perspective from product-based sustainability to ecological sustainability, thus contributing to a reduction in impacts relative to population, area and country-level impacts in a relevant category. In addition, countries with higher emissions and resource use defined per capita and arable land use can significantly reduce their environmental footprint by adopting alternative production systems (e.g., alternative proteins).

This study proposes a new methodology for calculating the eco-potential of alternative protein sources taking into account the environmental pillar of sustainability. Eco-potential is defined in the study as an additional environmental impact expressed relative to the existing environmental impact and the country’s environmental capacity. This methodology helps to identify the relative environmental impacts of alternative protein sources in different countries, according to the country’s total emissions and resource use (GHGE, WU, EU and ALU) per capita and on an annual basis.

The eco-potential scores for alternative protein sources varied over time and space. As the values for the relative impact weight (RIWs) based on GHGE normalized by population and arable land increased over time (2000–2016), so did the additional environmental impact burden of the alternative protein sources food products (expressed as CWI for 1 kg of beef) in Figure 2. This indicates the potential for the application of an eco-potential assessment methodology based on regional normalization and weighting factors. Comparison of the results with the scientific literature revealed inconsistencies in the representation of the results, which can be explained by the different assessment approaches.

Future studies should consider implementing the proposed weighting system for policy and decision makers, building on the concept of relative environmental sustainability (Bjørn and Hauschild, 2012). The eco-potential points obtained for the chosen protein sources could also further be used by the policy and decision makers when considering its production in a specific country based on its relative environmental impacts.