Reframing beef production in the Pampa biome: a life cycle assessment of environmental sustainability

1 Introduction

The growing demand for animal protein, coupled with population growth and changes in dietary patterns, has driven the intensification of global food systems. It is estimated that the global food system is responsible for approximately 34% of anthropogenic greenhouse gas (GHG) emissions, of which animal products contribute about 57% (Crippa et al., 2021; Xu et al., 2021; Pradhan et al., 2025). In this context, beef production takes center stage in both food security and the debate on environmental sustainability, due to its high intensity of emissions and use of natural resources.

The beef production chain is widely recognized as one of the production systems with the greatest environmental impact per unit of product. According to the global Life Cycle Assessment study conducted by Poore and Nemecek (2018), the average emission related to beef production is approximately 60 kg CO₂eq per kg of product, varying from less than 20 to more than 100 kg CO₂eq, depending on the production system adopted. Additionally, beef can have emissions that are 2–9 times higher than other animal products and more than 50 times higher compared to plant-based foods per unit of protein (Cusack et al., 2021).

These impacts stem primarily from enteric fermentation in ruminants, manure management, and feed and input use. Enteric fermentation alone contributes approximately 40%–60% of total GHG emissions in beef production systems, where methane (CH₄) is the predominant gas emitted (Gerber et al., 2013; Herrero et al., 2016; Haider et al., 2025). Furthermore, the use of fertilizers and the management of waste are responsible for nitrous oxide (N₂O) emissions, which reinforces the relevance of an integrated assessment of carbon and nitrogen flows throughout the production chain (Wang et al., 2018; Garnier et al., 2019; Baruselli et al., 2023; Singaravadivelan et al., 2023).

Land use is also one of the main impacts linked to beef production. The meat production system occupies approximately 80% of global agricultural land, despite contributing less than 20% of the calories consumed by the global population, highlighting its low land-use efficiency (Ritchie et al., 2022; Harsini and Mishra, 2025). Furthermore, beef production consumes considerable amounts of water per kg of meat produced (Ridoutt et al., 2012; Rotz et al., 2019; Johannes and Akbar, 2025).

However, scientific literature indicates wide variability among production systems. The meta-analysis conducted by Poore and Nemecek (2018), based on more than 38,700 farms in 119 countries, demonstrates that the environmental impacts of beef can vary by orders of magnitude within the same product. This difference is directly associated with production efficiency, feed management, the use of external inputs, and the duration of the production cycle, showing that improvements in system management can substantially reduce environmental impacts (Gerber et al., 2013; Tinitana-Bayas et al., 2024).

In this context, the Pampa biome stands out as a strategic area for beef production in South America, encompassing regions of Argentina, Brazil, and Uruguay. Traditionally, production in these regions is based on extensive systems sustained by native pastures, low use of inputs, and a strong dependence on local ecological and climatic conditions (Becoña et al., 2014; Montossi and Cazzuli, 2016; Dick et al., 2021). It is important to highlight that the meat produced in these systems within the Pampa biome has been associated with distinct attributes that place it among the highest quality products on the international market. Among these attributes, the most notable is the more favorable lipid profile, with higher concentrations of polyunsaturated fatty acids and a lower n-6:n-3 ratio, which is related to the nutritional quality of the meat (Vahmani et al., 2015; Klopatek et al., 2022; Kearns et al., 2023). Furthermore, production on native pastures imparts a distinct sensory identity to the meat, which is frequently associated with specific volatile compounds and the production system (environment, diet, among others) (Klopatek et al., 2022).

Different production systems around the world, such as in Japan and the United States, use different criteria to define meat quality, with marbling frequently used as an indicator in intensive systems, due to its relationship with attributes such as tenderness and juiciness (Vahmani et al., 2015). However, pasture-based systems, such as those in the Pampa biome, are distinguished by integrating nutritional quality, sensory characteristics, and aspects linked to naturalness and the mode of production—particularities that are increasingly valued by consumers and international markets. Thus, Pampa beef stands out for combining nutritional quality, sensory characteristics, and territorial identity, strengthening its importance in the global beef production landscape. However, in recent decades, there has been a growing intensification of production systems in this region, with the cultivation of sown pastures, feed supplementation, and integrated production systems to meet global food demand.

Production intensification involves important trade-offs for the environmental sustainability of the beef production chain. This is because intensive systems tend to reduce emissions per unit of product by increasing productivity and decreasing slaughter age; however, they often shift environmental impacts to other categories due to the use of fertilizers and agricultural inputs (Gerber et al., 2013; Leip et al., 2015; Poore and Nemecek, 2018; Harchaoui and Chatzimpiros, 2025). In contrast, integrated systems, such as crop-livestock and livestock-forestry integration, have gained prominence as promising strategies due to their ability to increase productivity and promote carbon sequestration (Carvalho et al., 2018; Stanley et al., 2018; Valera et al., 2024).

Life Cycle Assessment has become established as a potential tool for quantifying and comparing the environmental impacts of food production, including beef, allowing for the integrated assessment of various impact categories throughout the production chain (Schneider et al., 2024; Machado et al., 2025; Dutra et al., 2026). This methodology enables the identification of critical points and the analysis of climate change mitigation strategies and is widely applied to support decisions in complex food systems (Jacob-Lopes et al., 2021; Sala et al., 2021).

Although there have been advances in the literature on the subject, there are still bottlenecks in the integrated evaluation of beef production systems in specific biomes, such as the Pampa. Although Argentina, Brazil, and Uruguay share this biome, production systems have developed under different climatic conditions, soil characteristics, regulatory frameworks, and management practices. Importantly, the same broad production category does not necessarily correspond to an identical production model across the three countries. Systems classified as extensive, semi-intensive, intensive, or integrated may differ in forage base, supplementation strategy, stocking rate, production cycle, and input use according to national and regional production conditions (Jaurena et al., 2021). Edaphoclimatic differences across the Pampa biome can also influence pasture productivity and, consequently, forage availability and animal performance, contributing to differences in beef productivity even among systems belonging to the same broad production category (Gutiérrez et al., 2020). These within-category differences can alter productivity, resource requirements, and emission profiles and therefore warrant scenario-level comparison. The lack of comparative analyses that encompass these distinct production realities within the Pampa biome limits understanding of environmental trade-offs and hinders identification of opportunities to improve the sustainability of beef production systems.

Thus, the present study aims to quantify and compare the environmental impacts of beef production in the Pampa biome, considering different production systems from Argentina, Brazil, and Uruguay, using the Life Cycle Assessment methodology. The main contribution of this study is the integrated comparison of beef production systems across the Pampa biome using a common Life Cycle Assessment framework, enabling the identification of environmental trade-offs associated with different production and intensification strategies. By integrating multiple environmental impact categories and indicators, this study seeks to compare the environmental performance of beef production systems and elucidate the environmental burdens, trade-offs, and heterogeneities that define their sustainability. By preserving scenario-level differences within shared production categories, the analysis also seeks to identify management combinations associated with lower environmental intensity across the three national contexts. By conducting an analysis that goes beyond isolated sustainability metrics, this work aims to contribute to a robust understanding of the environmental burdens associated with beef production systems in the Pampa biome and to provide scientific evidence on the environmental trade-offs involved in different production and intensification strategies.

2 Materials and methods

This study focused on beef production in the Pampa biome, encompassing regions of Argentina, Brazil, and Uruguay. Figure 1 illustrates the extent of the biome and summarizes the environmental and productive characteristics of each country. Although all scenarios were located within the Pampa biome, differences in climate, rainfall, soil characteristics, and production systems among countries were maintained throughout the assessment, reflecting the diversity of production conditions under which beef is produced across the region.

2.1 Life cycle assessment (LCA)

The study was conducted in accordance with the principles and requirements established by the International Organization for Standardization (ISO), adopting the Life Cycle Assessment framework consisting of four stages: (i) definition of the objective and scope; (ii) life cycle inventory (LCI); (iii) life cycle impact assessment (LCIA); and (iv) interpretation of the results (ISO, 2006a).

2.1.1 Defining the objective and scope

The main objective was to quantify and compare the environmental impacts associated with the production of chilled beef, considering representative systems of the Pampa biome, encompassing regions of Argentina, Brazil, and Uruguay. Analyses of relevant environmental impact categories were included to reflect the main impacts of beef production: global warming potential (GWP), acidification potential (AP), eutrophication potential (EP), and land use (LU), in addition to assessing water use (WU) and energy use (EU). The GWP category was used to quantify the climate change impact associated with greenhouse gas emissions within the defined LCA framework.

The scope of the study focused exclusively on production systems located in the Pampa biome to ensure consistency with the ecological, edaphoclimatic, and zootechnical conditions of the study region. Studies conducted in other biomes or geographic regions were included only for methodological comparison purposes, parameterization of specific productive components, and support for modeling the evaluated scenarios, without forming part of the primary database representative of the Pampa biome. The data collection was derived from references in classical literature (Pordomingo et al., 2012; Modernel et al., 2013; de Freitas et al., 2014; Picasso et al., 2014; Rearte and Pordomingo, 2014; Dick et al., 2015; Cardoso et al., 2016; Mogensen et al., 2016; de Assis et al., 2018; SENAR, 2018; Vale et al., 2019; Arrieta et al., 2020; Fischer and Bilenca, 2020; Mezzomo et al., 2020; Pordomingo et al., 2020; Silva et al., 2020; García-Favre et al., 2021; Preisz et al., 2021; Reis Filho et al., 2021; Tonini et al., 2021; Volpi-Lagreca et al., 2021; Maciel et al., 2025; Moreira et al., 2025; Silva Machado et al., 2026).

The functional unit (FU) adopted was 1 kg of chilled beef, chosen because it reflects the product actually made available to the market and allows comparability between the different systems evaluated. Five Argentinian scenarios were included: extensive system (scenario A); semi-intensive system based on sown pastures (scenario B); semi-intensive system based on grains (scenario C); intensive system (scenario D); and Crop-Livestock Integration (CLI) system (scenario E). Six Brazilian scenarios were also analyzed: a semi-intensive system based on sown pastures (scenario F); three CLI systems, including semi-intensive systems based on sown pastures (scenarios G and H) and grains (scenario I); an intensive system (scenario J); and a Livestock-Forest Integration (LFI) system (scenario K). Finally, two Uruguayan scenarios were included: an extensive system (scenario L) and a semi-intensive system based on sown pastures (scenario M). For the confined systems (scenarios D and J), stocking rate (AU ha^-1) and beef production (kg beef ha^-1) refer specifically to the physical area occupied by the feedlot facility during the finishing phase. Thus, these values characterize production density within the confinement area, whereas the corresponding area-based indicators in grazing systems refer to pasture area. Therefore, the area-based production-density values reported for the confined systems should not be directly compared with pasture-based stocking rates, as the hectare denominator represents the physical feedlot area in the former and the grazing area in the latter. This distinction should be considered when interpreting the area-based production indicators. Despite that, environmental impacts across all scenarios were compared using the common functional unit of 1 kg of chilled beef, with land-use calculations accounting for the land requirements associated with the different components of each production system. Table 1 presents the detailed production data for each scenario analyzed.

TABLE 1. Production data for the scenarios.

Adapted from Pordomingo et al. (2012); Modernel et al. (2013); Picasso et al. (2014); Rearte and Pordomingo (2014); SENAR (2018); Vale et al. (2019); Arrieta et al. (2020); Fischer and Bilenca (2020); Pordomingo et al. (2020); Volpi-Lagreca et al. (2021).

These scenarios represent the diversity of extensive, semi-intensive, and intensive systems used in the regions, combining variations in management, feed composition, technological level, forage seasonality, and input use. The evaluated scenarios were selected from the scientific literature to represent the diversity of beef production systems reported for the Pampa biome of Argentina, Brazil, and Uruguay. The scenarios were not intended to represent national prevalence shares or arbitrary production categories, but to encompass production configurations documented for the biome. Importantly, scenarios within the same broad production category are not equivalent. For example, the semi-intensive scenarios differ in forage base (ryegrass, forage sorghum, millet, Sudan grass, or grain supplementation), production time (24–36 months), stocking rate (1.0–5.0 AU ha^-1), and beef productivity (181.2–1,200 kg ha^-1). The intensive scenarios D and J also differ in production time, stocking rate, beef productivity, and ration composition, whereas the extensive scenarios A and L differ in productivity and resource requirements. These distinctions are further reflected in fertilizer type and application rate, crop integration, water and energy use, and direct emission flows detailed in Supplementary Tables S2–S14 of the Supplementary Material. Therefore, merging scenarios within a broad category would require averaging distinct inventories and would create a composite system that does not correspond to any of the production systems documented in the source literature. All material, energy, and emission flows were normalized by this FU.

The system’s boundaries encompassed the breeding, rearing, and fattening phases, pasture establishment and management, concentrate consumption, soil correction and fertilization, fuel use, direct water consumption for drinking and handling, enteric methane emissions, emissions from manure management, and energy and water consumption required for beef processing up to its chilled form. Therefore, the study adopted a modified cradle-to-gate system boundary, considering chilled beef at the slaughterhouse gate as the final product. This system boundary was selected to focus the assessment on the production stages directly influenced by management practices, thereby enabling a consistent comparison of representative beef production systems under a common modeling framework, in accordance with the principles of comparative life cycle assessment (ISO, 2006b; de Vries and de Boer, 2010).

The life cycle inventory included the quantities of fertilizers, lime, pesticides, seeds, feed ingredients, fuels, electricity, and other external inputs directly used within each production system. For example, the quantities of corn, soybean meal, or other feed ingredients consumed by the animals and the quantities of fertilizers applied to pastures or crops were included as inventory inputs. The exclusion applies to the processes required to produce or manufacture those external inputs before they entered the evaluated production system, such as the energy, materials, and emissions associated with fertilizer manufacture or external feed production. Consequently, differences in the quantities of these inputs among scenarios remained part of the comparative assessment, whereas their upstream production or manufacturing processes remained outside the defined system boundaries. Likewise, downstream processes occurring after the slaughterhouse gate, such as distribution, transportation to retail, storage, marketing, consumer use, and end-of-life management, were excluded from the assessment.

The adopted system boundaries were identical for all scenarios; therefore, differences among the evaluated systems arise from their foreground inventories rather than from differences in system definition. Scenario differentiation is determined by animal productivity, production cycle, feeding regime, forage base, pasture and crop integration, quantities of fertilizers and other inputs, resource consumption, and direct emissions. The excluded upstream and downstream processes were intentionally omitted from all scenarios and therefore do not alter this comparative differentiation. The boundaries of the systems analyzed are represented in Figure 2.

2.1.2 Life cycle inventory (LCI)

The inventory analysis phase consisted of quantifying the resource inputs and emission outputs associated with the life cycle of beef production. The data collection was derived from primary and secondary scientific sources on production systems typical of the Pampa biome. Supplementary Table S1 of the Supplementary Material presents a complete characterization of the main sources used to construct the scenarios, distinguishing those used to obtain productive and environmental parameters from those used for methodological support, assumption definition, and parameterization of the evaluated systems. For each scenario, data were collected on average daily gain, slaughter age, stocking rate, animal yield, dry matter intake, diet composition, pasture and forage crop productivity, fertilizer and pesticide application, diesel and electricity consumption, direct water consumption, and direct emissions of methane, nitrogen oxides, and ammonia. In the case of systems with agricultural and forestry crop production (forage sorghum, ryegrass, millet, Sudan grass, soybeans, and eucalyptus), inputs for soil preparation and cultivation, fertilization, mechanized operations, and irrigation, when applicable, were included. For Brazilian beef production scenarios, a legal reserve (LR) of 20% per hectare was considered. The inventory data for each scenario are detailed in Supplementary Table S2 through S14 of the Supplementary Material.

2.1.3 Life cycle impact assessment (LCIA)

Based on the inventory flows calculated by functional unit, a Life Cycle Impact Assessment (LCIA) was performed for each of the established scenarios, using the ReCiPe 2016 method, as described by Huijbregts et al. (2017). Four midpoint environmental impact categories were considered: global warming potential (kg CO₂eq kg^-1 of beef), acidification potential (kg SO₂eq kg^-1 of beef), eutrophication potential (kg PO₄eq kg^-1 of beef), and land use (PDF·m²eq kg^-1 of beef). In addition, water use (m³ kg^-1 of beef) and energy use (MJ kg^-1 of beef) were assessed as complementary resource-use indicators. All categories were quantified using specific characterization equations. The environmental impact characterization factors and product/substance emissions considered in this study are presented in the Supplementary Material (Supplementary Table S15).

The global warming potential (GWP) was calculated over a 100-year horizon using the warming potentials of CO₂, CH₄, and N₂O through Equation 1 (Laratte et al., 2014):where GWP is the global warming potential (kg CO₂eq), i is the time horizon (100 years), GWPi is the characterization factor of substance i (kg eq), and mi is the emission mass of substance i (kg).

G W P = ∑ i G W P i × m i (1)

The acidification potential (AP) was obtained by converting atmospheric emissions of NH₃, NO_X, and SO₂ into sulfur dioxide equivalent according to Equation 2 (Hauschild and Wenzel, 1998):where AP is the acidification potential (kg SO₂eq), APi is the equivalence factor of substance i (kg eq), and mi is the emission of substance i (kg).

A P = ∑ i A P i × m i (2)

The eutrophication potential (EP) was calculated by converting nitrogen and phosphorus loads from fertilization, volatilization, leaching, and manure management into phosphate equivalents according to Equation 3 (Jacob-Lopes et al., 2021):where EP is the eutrophication potential (kg PO₄eq), EPi is the equivalence factor of substance i (kg eq), and mi is the emission of substance i (kg).

E P = ∑ i E P i × m i (3)

Land use (LU) was estimated from the area transformed and occupied over time by pastures and forage crops, considering both permanent and rotational systems using Equations 4 and 5 (de Baan et al., 2013; Chaudhary and Brooks, 2018): where LUtrans and LUocc are the impacts of land transformation and land occupation, respectively, expressed in PDF·m²eq kg^-1; A is the transformed or occupied area (m²), FCtrans is the land transformation characterization factor (m² eq), FCocc is the land occupation characterization factor (m² eq), and t is the occupation time (year).

L U t r a n s = C F t r a n s × A (4)

L U o c c = C F o c c × A × t (5)

Water use (WU) was calculated by summing the amounts of water used in each stage of the system to produce 1 kg of beef, according to Equation 6. Only direct water use was considered for the calculation (Rotz et al., 2019).where WU is the total direct water usage (m³), Wi is the amount of direct water used in process i (m³), and n is the number of processes considered in the production system.

W U = ∑ i = 1 n W i (6)

Total energy use (EU) was defined as the sum of energy associated with diesel and electricity consumption within the production system. The flows were converted to energy units and expressed per functional unit, as per Equation 7. Only indirect energy use related to fuel and electricity consumption was considered, whereas embodied energy associated with the production of inputs, machinery, equipment, and infrastructure was not included (Pimentel, 2019; Rotz et al., 2019).where EU is the total indirect energy use (MJ kg^-1), Ej is the indirect energy from energy input j (MJ kg^-1), and m is the number of energy inputs included in the system (diesel and electricity).

E U = ∑ j = 1 m E j (7)

2.1.4 Interpretation of the results

The results interpretation phase included contribution analysis procedures, comparisons between scenarios, and consistency checks. The stages and processes that contributed most to each impact category were evaluated, identifying the main hotspots in each production system. Comparability among the Argentinian, Brazilian, and Uruguayan scenarios was facilitated by the standardization of system boundaries, functional units, impact assessment methods, characterization factors, and emission factors. While the scenarios represent distinct production systems and management strategies within the Pampa biome, the use of a harmonized Life Cycle Assessment framework enabled a consistent comparison of environmental impacts per unit of product. The adoption of common system boundaries ensures that the observed differences among scenarios reflect differences in production systems rather than inconsistencies in system definition. However, differences associated with national regulatory frameworks were maintained in the scenario modeling. In particular, Brazilian scenarios included a legal reserve corresponding to 20% of the total property area, as required by national environmental legislation (Brasil, 2012). Therefore, part of the observed differences among countries may reflect not only production efficiency and management strategies, but also institutional and regulatory conditions related to land use. The results were expressed per functional unit (1 kg of chilled beef), allowing for the identification of structural differences between extensive, semi-intensive, and intensive systems. However, environmental performance may vary according to the functional unit adopted. Alternative metrics based on land occupation, protein production, nutritional value, or economic value can provide complementary perspectives and lead to different interpretations of system performance, particularly when comparing systems with contrasting levels of productivity and management intensity. Therefore, the results of this study should be interpreted within the context of the selected functional unit.

2.2 Carbon-balance indicator

As a complementary and exploratory assessment to the Life Cycle Assessment, the carbon-balance indicator was estimated using Equation 8, following the methodology of the GHG Protocol for Agriculture (Protocol, 2014). The calculation was based on the difference between CO₂ emissions and literature-based estimates of gross vegetation CO₂ uptake associated with the vegetation components included in each production system. Gross vegetation CO₂ uptake values (kg CO₂eq ha^-1 year^-1) for ryegrass, forage sorghum, millet, Sudan grass, soybean, native pasture, Pampa forest, and eucalyptus were obtained from the literature and are presented in Supplementary Table S16, based on estimates reported by Veeck et al. (2022), Mergen et al. (2023), SEEG Brasil (2024), Sharma et al. (2025), Shirvan et al. (2025), and Xue et al. (2025). The annual values were adjusted according to the period during which each vegetation component remained in the evaluated production systems and subsequently expressed per functional unit. This indicator does not quantify permanent carbon sequestration, changes in soil organic carbon stocks, or long-term carbon storage.where CBindicator represents the carbon-balance indicator, CO _*2* emissions (kg CO₂eq) represents the carbon dioxide emissions generated within the system boundaries, and CO _*2* uptake (kg CO₂eq) represents the literature-based estimate of gross CO₂ uptake by the vegetation components included in each scenario during the production period.

C B i n d i c a t o r = C O 2 e m i s s i o n s − C O 2 u p t a k e (8)

Thus, CO₂ emissions represent the emissions generated within the system boundaries, whereas vegetation CO₂ uptake represents the estimated gross uptake associated with the vegetation components present in each scenario during the production period. The resulting carbon-balance indicator represents the arithmetic difference between these two estimates. Although the vegetation CO₂ uptake values were originally reported on an area basis, their expression according to a product-based functional unit is consistent with approaches adopted in previous LCA studies in which land-related carbon components were normalized to product output (Sabia et al., 2024; Silva Machado et al., 2026). The carbon-balance indicator was calculated independently from the Life Cycle Assessment impact categories and was not incorporated into the GWP results. Accordingly, negative values of this complementary indicator indicate only that the estimated gross vegetation CO₂ uptake exceeded the modeled system emissions under the adopted assumptions; they do not demonstrate permanent carbon sequestration, changes in soil organic carbon stocks, or a net-negative product carbon footprint.

2.3 Normalization and weighting

To facilitate the integrated interpretation and comparison of the environmental impact categories, normalization and weighting procedures were performed. To obtain data normalization, Equation 9 was used, in accordance with the International Life Cycle Reference Data System (ILCD) Manual (Heinrich, 2010). The impact potentials were divided by the global environmental impact normalization factor according to the European Commission (Sala et al., 2017) (Supplementary Table S17 of the Supplementary Material).where k indicates the impact category, N is the normalized indicator, S is the indicator of the characterization category, and R is the reference value.

N K = S k R k (9)

Afterwards, weighting was applied by multiplying the normalized result of each impact category by its corresponding weighting factor according to Equation 10, following the methodological approach described by Heinrich (2010) and Abbe and Hamilton (2017). Unequal weights were assigned to the six categories based on Huppes and van Oers (2011): 32% for global warming potential (GWP), 32% for land use (LU), 10% for acidification potential (AP), 10% for eutrophication potential (EP), 10% for energy use (EU), and 6% for water use (WU), totaling 100%. The weighted values of the six impact categories were subsequently aggregated to obtain the overall weighting and ranking. For the ranking within each impact category, the same weighting procedure was applied separately to each category.where Wg is the impact weighting, WgFk is the weighting factor for impact category k, Nk is the normalized indicator, and k is the impact category indicator, expressed as a percentage.

W g k = ∑ W g F k × N k (10)

The overall weighting analysis was used as a complementary tool to facilitate the integrated visualization of environmental performance across the evaluated impact categories. Therefore, the resulting ranking should not be interpreted as a definitive measure of sustainability, but rather as a synthetic representation of the relative performance patterns observed among the production systems.

2.4 Deterministic sensitivity analysis

A deterministic one-at-a-time (OAT) sensitivity analysis was performed to evaluate the robustness of the global warming potential (GWP) results. Enteric CH₄ emissions were selected because they represent the largest contribution to GWP across the evaluated beef production systems, making them the most influential inventory flow for assessing the robustness of the comparative results. Enteric CH₄ emissions were independently varied by ±20%, while all remaining inventory flows and characterization factors were kept constant. The modified inventories were subsequently characterized using the same GWP100 method adopted in the baseline assessment, and the resulting total GWP values were compared with the baseline results. The adopted approach followed the OAT sensitivity analysis, in which a single parameter is varied while all remaining model parameters are held constant to evaluate its individual influence on the results (Rivera Moncada et al., 2025).

3 Results

The results of the environmental assessment of the different beef production scenarios in the three countries that make up the Pampa biome are shown in Figures 3–7 and in Tables 1 and 2. The values showed considerable variations related to the level of production intensification, feeding strategies, use of agricultural inputs, and integration between agricultural and forestry activities in beef production.

TABLE 2. Environmental impact values for the acidification potential and eutrophication potential categories associated with the production of 1 kg of beef in different scenarios.

The global warming potential ranged from 32.59 to 52.93 kg CO₂eq kg^-1 of beef, demonstrating the strong influence of production efficiency and the production system adopted. The lowest GWP values were observed in the Argentinian semi-intensive system based on sown pastures, scenario B (32.59 kg CO₂eq kg^-1 of beef), and the Uruguayan semi-intensive system, scenario M (32.63 kg CO₂eq kg^-1 of beef). The semi-intensive based on sown pastures and intensive scenarios D, E, F, G, H, J, and K, with or without crop-livestock integration, presented intermediate GWP values, ranging from 33.09 (scenario G) to 40.25 kg CO₂eq kg^-1 of beef (scenario E). On the other hand, the highest GWP values ​​observed were in semi-intensive grain-based systems in Brazil and Argentina, scenarios I (52.93 kg CO₂eq kg^-1 of beef) and C (51.07 kg CO₂eq kg^-1 of beef), highlighting the impact of using feed inputs together with native pasture, where there is lower productivity. Furthermore, extensive systems based solely on pasture-based feeding in the Pampa biome also showed high values, which in this case were scenarios A (Argentina) and L (Uruguay), with values ​​of 48.56 and 48.60 kg CO₂eq kg^-1 of beef, respectively. These results can be observed in Figure 3.

The deterministic sensitivity analysis showed that a ±20% variation in enteric CH₄ emissions affected the GWP values of all evaluated scenarios (Figure 4; Supplementary Table S18 of the Supplementary Material). The greatest variations were observed in the extensive production systems, whereas smaller changes occurred in grain-supplemented systems. Although slight changes were observed in the relative position of some scenarios with similar GWP values, the overall comparative interpretation of climate performance among production systems remained unchanged.

Regarding the acidification potential, a wide variation was observed between the scenarios, with values ranging from 1.59E-01 to 4.63E+00 kg SO₂eq kg^-1 of beef. The extensive Uruguayan and Argentinian systems based on native pasture were those that presented the lowest AP values, scenarios L (1.59E-01 kg SO₂eq kg^-1 of beef) and A (1.59E-01 kg SO₂eq kg^-1 of beef). These scenarios did not employ external inputs and, consequently, did not generate ammonia emissions associated with the application of nitrogen fertilizers. On the other hand, the systems with the highest AP values were the semi-intensive systems based on sown pastures and the agricultural and forestry integration systems, especially scenarios E (4.63E+00 kg SO₂eq kg^-1 of beef), H (3.79E+00 kg SO₂eq kg^-1 of beef), F (3.30E+00 kg SO₂eq kg^-1 of beef) and K (3.30E+00 kg SO₂eq kg^-1 of beef), due to the use of fertilizers and pesticides. The Argentine semi-intensive grain-based system (scenario C), the Brazilian semi-intensive CLI system (scenario G), and the Brazilian and Argentine intensive systems (scenarios D and J) presented comparatively low-to-intermediate AP values, ranging between 6.83E-01 (scenario G) and 7.50E-01 kg SO₂eq kg^-1 of beef (scenario D).

The assessment of eutrophication potential showed variation between results of 8.00E-05 and 1.73E+00 kg PO₄eq kg^-1 of beef, which mainly reflects the differences in the use of nitrogen and phosphate fertilizers between the scenarios. As in AP, the lowest EP values were observed in the Uruguayan and Argentinian extensive systems, scenarios L (8.00E-05 kg PO₄eq kg^-1 of beef) and A (9.00E-05 kg PO₄eq kg^-1 of beef), due to the absence of fertilizer use. In contrast, the highest EP values were observed in systems with greater application of agricultural inputs, mainly in the Brazilian scenarios H (1.73E+00 kg PO₄eq kg^-1 of beef), K (1.70E+00 kg PO₄eq kg^-1 of beef), and F (1.66E+00 kg PO₄eq kg^-1 of beef), as well as in the Argentinian scenario E (9.56E-01 kg PO₄eq kg^-1 of beef). Intermediate values ​​were observed in the other scenarios, with a variance of 7.05E-03 (scenario C) to 7.64E-02 kg PO₄eq kg^-1 of beef (scenario B). Table 2 shows the AP and EP values for different beef production scenarios.

The impacts related to land use were relatively low when expressed per kilogram of beef, but they differed across production systems. The lowest land transformation values were observed in the Brazilian semi-intensive system based on cultivated pastures and the Argentine semi-intensive system based on grains, which presented values of 2.73E-10 (scenario G), and 4.64E-10 PDF·m²eq kg^-1 of beef (scenario C), where there was less land transformation per kilogram of product. The highest values observed were in the Brazilian semi-intensive grain-based system, scenario I (2.02E-09 PDF·m²eq kg^-1 of beef), and the Argentinian intensive system, scenario D (1.30E-09 PDF·m²eq kg^-1 of beef).

Regarding land occupation, the lowest values were found in the semi-intensive systems based on sown pastures—the Brazilian and Argentine systems, scenarios G and B—which had values of 3.86E-12 and 9.38E-12 PDF·m²eq kg^-1 of beef, respectively. In contrast, the scenarios with the highest occupancy per kilogram of beef were the Uruguayan extensive system, the Brazilian semi-intensive grain-based system, and the Brazilian intensive system, with values of 3.57E-11 (scenario L), 3.18E-11 (scenario I), and 3.07E-11 PDF·m²eq kg^-1 of beef (scenario J), respectively. The remaining scenarios had intermediate land occupation, with values ranging between 1.26E-11 and 2.97E-11 PDF·m²eq kg^-1 of beef. The results regarding land use for both transformation and occupation can be seen in Table 3.

TABLE 3. Environmental impact values for land use in transformation (LUtrans) and for land use in occupation (LUocc) (PDF·m²eq kg^-1 of beef) in different beef production scenarios.

About water use, the results showed that the systems that used the most direct water for beef production were the Brazilian systems with pasture and soybean cultivation, scenarios I (4.55E-01 m³ kg^-1 of beef), H (4.02E-01 m³ kg^-1 of beef), and G (3.89E-01 m³ kg^-1 of beef). The lowest direct water use values were observed in the Brazilian scenarios K (1.53E-01 m³ kg^-1 of beef), with ryegrass and eucalyptus cultivation, and F (1.55E-01 m³ kg^-1 of beef), with ryegrass cultivation, followed by the Uruguayan scenario M (1.63E-01 m³ kg^-1 of beef), with ryegrass cultivation, and the Argentinian scenario E (1.67E-01 m³ kg^-1 of beef), with ryegrass and soybean cultivation. The other systems presented intermediate values ​​of direct water use, varying between 1.78E-01 (scenario J) and 2.91E-01 m³ kg^-1 of beef (scenario A).

Regarding energy use assessment, the system with the highest indirect energy use was the Argentinian CLI, scenario E, which had an expenditure equal to 1.18E+00 MJ kg^-1 of beef, followed by the Brazilian scenarios H (1.14E+00 MJ kg^-1 of beef), I (1.13E+00 MJ kg^-1 of beef), and G (1.09E+00 MJ kg^-1 of beef). The systems that used the least amount of indirect energy were the Uruguayan and Argentinian extensive systems, scenarios L (5.15E-01 MJ kg^-1 of beef) and A (5.62E-01 MJ kg^-1 of beef), as well as the Argentinian semi-intensive grain-based system, scenario C (5.85E-01 MJ kg^-1 of beef). In the remaining systems, intermediate indirect energy use values ​​were observed, ranging from 5.91E-01 (scenario K) to 6.94E-01 MJ kg^-1 of beef (scenario B). Figure 5 shows the water and energy use values ​​for each production scenario.

For the complementary carbon-balance indicator, considerable variation was observed among scenarios, with values ranging from −230.71 to 26.19 kg CO₂eq kg^-1 of beef. The lowest value was observed in scenario E (−230.71 kg CO₂eq kg^-1 of beef), corresponding to the crop-livestock integration system in Argentina, mainly due to the high literature-based estimate of gross vegetation CO₂ uptake attributed to soybean during the production period. Other scenarios with comparatively high estimated vegetation CO₂ uptake were M (−105.46 kg CO₂eq kg^-1 of beef), I (−69.29 kg CO₂eq kg^-1 of beef), K (−55.01 kg CO₂eq kg^-1 of beef), and H (−49.51 kg CO₂eq kg^-1 of beef). In contrast, the semi-intensive systems—based on grain in Argentina and sown pastures in Brazil—presented positive indicator values, indicating that modeled emissions exceeded estimated gross vegetation CO₂ uptake, as observed in scenarios C (26.19 kg CO₂eq kg^-1 of beef) and G (13.04 kg CO₂eq kg^-1 of beef). Negative indicator values indicate that estimated gross vegetation CO₂ uptake exceeded modeled emissions within the calculation framework; they do not represent permanent carbon sequestration, changes in soil organic carbon stocks, or a net-negative product carbon footprint. Figure 6 illustrates these results. In addition, Supplementary Table S19 of the Supplementary Material presents the modeled emission values, estimated gross vegetation CO₂ uptake, and resulting carbon-balance indicator for each scenario.

In summary, the overall weighting of the evaluated scenarios demonstrates that the environmental sustainability of beef production in the Pampa biome is strongly influenced by the interaction between animal productivity and the intensity of input use. The overall weighting reflects a better score for extensive systems in Argentina and Uruguay and a worse score for integrated Brazilian systems based on ryegrass pasture and Sudan grass. However, if we analyze each category individually, we will see changes in this ranking. Figure 7 illustrates the weighting assigned to each category and indicator analyzed in this study, as well as the overall weighting of the thirteen systems evaluated.

The results for the evaluated categories indicate that extensive systems based exclusively on native pasture showed lower dependence on external inputs, although some also exhibited relatively high land use and GWP per unit of product. Systems with greater production intensity achieved higher productivity, but their GWP values and land-use performance varied according to the specific production configuration. Integrated and semi-intensive systems based on Sudan grass and sown pastures presented higher AP and EP values, but also showed lower carbon-balance indicator values due to the estimated gross vegetation CO₂ uptake associated with the vegetation components included in these systems. These results illustrate the different environmental profiles of the evaluated production strategies and help explain the variation observed in the overall weighting analysis, which should be interpreted as a synthetic representation of the overall patterns observed among impact categories rather than as a definitive assessment of environmental sustainability.

4 Discussion

Among the different animal protein production systems, beef production has one of the highest emission intensities per unit of product, due to the low feed conversion of ruminants and the high contribution of enteric methane to total GHG emissions (Mottet et al., 2017). Additionally, beef production also has a substantial influence on land use and on processes that generate environmental degradation linked to agricultural expansion, loss of biodiversity, and loss of nutrients to the aquatic and atmospheric environment (Paz et al., 2020; Fernández et al., 2024). In this sense, Life Cycle Assessment has proven to be a decisive tool for quantifying the environmental impacts of different production systems, including food systems, allowing for the analysis of different production strategies and the identification of mitigation opportunities throughout the production system (Tinitana-Bayas et al., 2024; Cadéro et al., 2025).

In this context, the results of our study demonstrate that the environmental burdens of beef production in the Pampa biome are heavily influenced by the structure of production systems, such as the level of intensification, feed management, and production efficiency. The global warming potential ranged from 32.59 to 52.93 kg CO₂eq kg^-1 of beef, resulting in an approximately 62% difference between the lowest and highest impact scenarios. Geographic factors alone are not responsible for determining variations in climate performance, as these factors are influenced by animal management, pasture management, input management, and the production period. In practical terms, our results demonstrate that small changes in aspects of the production system structure can generate substantial differences in environmental degradation related to beef production.

The lowest GWP values were observed in the semi-intensive systems based on sown pastures (scenarios B and M), while the intensive confinement systems (scenarios J and D) also presented comparatively low values, corroborating the idea that productive efficiency and shorter production cycles can contribute to lower emissions per unit of product (Mottet et al., 2017; Willers et al., 2017; Dick et al., 2015). In intensive systems, greater productive efficiency allows the emissions generated during animal maintenance and growth, particularly enteric CH₄ emissions, to be distributed over a larger amount of beef produced, while shorter production cycles reduce the period over which these emissions accumulate. Conversely, the highest values were assigned to scenarios I and C, characterized by the use of grains along with pasture, but without efficient productivity gains to compensate for the additional emissions from the implementation of grains in the diet and the longer period of time the animals remain in the production cycle. Although grain and concentrate supplementation is intended to improve production efficiency, when feed conversion, growth rate, and slaughter age do not improve, additional input use may not offset the accumulation of enteric emissions over the production cycle (Poore and Nemecek, 2018; Ruviaro, 2012). Similarly, extensive systems based exclusively on native pasture, represented by scenarios A and L, also showed high climate impact values, reinforcing that even low-input systems can present higher emissions when animal performance is limited. This pattern corroborates international studies showing that emissions from beef production are strongly linked to the efficiency and duration of the production cycle, since longer cycles concentrate more enteric emissions over time (Mottet et al., 2017; Willers et al., 2017; Dick et al., 2021). Evidence from Southern Brazil demonstrated that beef production in extensive systems with native pastures had higher carbon emissions (22.52 kg CO₂eq kg^-1 of live weight) when compared to improved pastures (9.16 kg CO₂eq kg^-1 of live weight) (Dick et al., 2015).

Although the functional unit adopted by Dick et al. (2015) differs from that used in the present study (1 kg of chilled beef), both studies indicate that improvements in productive efficiency are associated with lower greenhouse gas emissions per unit of product. However, this result reflects GHG emission intensity per unit of product and should not be interpreted as evidence that intensive systems necessarily generate lower GHG emissions per unit of land. Different reference metrics can provide contrasting perspectives on the environmental performance of beef production systems. For example, Molossi et al. (2020) reported that, in the Brazilian Cerrado, extensive grazing presented a carbon footprint of 19.26 kg CO₂eq kg^-1 BW, compared with 16.25 kg CO₂eq kg^-1 BW for a grain-supplemented system. Conversely, when total GHG emissions were expressed per unit of land, the extensive system presented 2,376 kg CO₂eq ha^-1, whereas the grain-supplemented system reached 4,876 kg CO₂eq ha^-1. Similar contrasting patterns were observed for pasture re-seeding and pasture fertilization, with emissions exceeding 5,100 kg CO₂eq ha^-1 despite lower product-based carbon footprints than the extensive system. The authors attributed the higher emissions per hectare of the intensified strategies to their doubled stocking density (Molossi et al., 2020). These findings illustrate that product- and land-based metrics provide complementary but distinct perspectives: the former reflects GHG emission intensity associated with producing an equivalent quantity of beef, whereas the latter reflects the concentration of emissions within a given land area. Therefore, the higher GWP observed for extensive systems in the present study should be interpreted specifically as higher GHG emissions per kilogram of beef under the adopted functional unit, rather than as evidence of a greater climate burden per hectare.

The values ​​found in this study are consistent with life cycle assessment estimates for beef production on a global scale. Several studies report that beef production is one of the food products with the most heterogeneous greenhouse gas emission values, a finding that is evident across different producers and production systems. According to Poore and Nemecek (2018), beef production generally emits 20–60 kg CO₂eq kg^-1 of beef, depending on the production system matrix. Furthermore, comparative analysis among countries belonging to the Pampa biome revealed important structural patterns.

At the regional level, in the Argentinian scenarios, it is observed that the GWP presented the greatest difference (32.59–51.07 kg CO₂eq kg^-1 of beef), which highlights the coexistence of traditional extensive systems and more intensified systems. A study by Arrieta et al. (2020), modeling 75 livestock production systems distributed across eight regions of Argentina, presented greenhouse gas emission values ​​around 17.4 kg CO₂eq kg^-1 of live weight. Despite methodological differences between studies, the results similarly highlight the influence of animal performance, feeding strategies, and production management on the environmental performance of beef systems.

In the Brazilian scenarios, our values were more homogeneous for semi-intensive scenarios based on sown pasture and eucalyptus (33.94–36.48 kg CO₂eq kg^-1 of beef). In southern Brazil, there is a greater predominance of semi-intensive systems based on sown pastures and crop-livestock integration in production areas (Dick et al., 2015; D’aurea et al., 2021; EMBRAPA, 2022; Barreto et al., 2025). Although our results showed homogeneity, this does not characterize an environmental convergence of Brazilian production systems. Furthermore, an environmental assessment of beef production in southern Brazil, considering seven production scenarios, reported a carbon footprint ranging from 18.47 to 37.18 kg CO₂eq kg^-1 of live weight (Ruviaro, 2012). Although the absolute values differ from those observed in the present study, both assessments reveal substantial variability among production systems. Furthermore, another study evaluating 17 farms representative of Brazilian beef production demonstrated a carbon footprint of 8.63–50.88 kg CO₂eq kg^-1 of beef (D’aurea et al., 2021). Therefore, the uniformity found for the Brazilian scenarios in this study requires caution in interpretation, as it may not reflect the true absence of diversity among the production systems of the Brazilian Pampa.

Furthermore, it is important to emphasize that, in the Brazilian scenarios evaluated in this study, 20% of the total area was designated as a legal reserve, as established by Brazilian environmental legislation. The presence of this unproductive area reduces the area available for food production, which can influence the values expressed per unit area and thus affect the environmental intensity of production systems when compared to scenarios in other countries within the Pampa biome, where this requirement is not considered. In this way, a portion of the observed differences between the Brazilian scenarios and the Argentinian and Uruguayan systems may be attributable to regulatory differences in land use, rather than solely to the productive strategies adopted. Therefore, comparisons among countries should be interpreted considering both productive efficiency and institutional factors associated with environmental regulations, particularly when evaluating indicators related to land occupation and land-use intensity.

Regarding the Uruguayan scenarios, there was a considerable reduction in GWP of approximately 33% between extensive and semi-intensive systems, revealing that the introduction of sown pastures can decrease the environmental impact and increase production efficiency (Modernel et al., 2013; Picasso et al., 2014). According to Becoña et al. (2014), emissions from beef production in 20 grazing systems in Uruguay ranged from 11.4 to 32.2 kg CO₂eq kg^-1 of live weight. Furthermore, the authors argue that by implementing strategies that increase grazing efficiency, meat production would increase by 22%, and consequently, emissions would be reduced by 28% per kg of meat. These findings reinforce the importance of grazing management and productive efficiency as key determinants of environmental performance in pasture-based systems. This performance is consistent with our results, reinforcing the idea that proper management and sown pastures are effective and well-established strategies when seeking shorter production cycles combined with increased production per unit of land. However, even if the carbon footprint results are favorable, this does not, in itself, represent more sustainable beef production. Production systems with reduced GWP values ​​can allocate environmental burdens to other categories, such as energy use, land use, and nutrient use.

Finally, the variations presented in the scenarios of this study largely highlight the different development trajectories and technological maturity of beef production in the three countries analyzed. Although Argentina, Brazil, and Uruguay are part of the Pampa biome, they present differences in climatic conditions, regulatory frameworks, and production strategies that have shaped distinct trajectories of beef production development. In Brazil, the adoption of public policies and research programs focused on the sustainable intensification of food production has promoted the growth of integrated production systems, such as crop-livestock integration (CLI) and livestock-forestry integration (LFI), which aimed to maximize productivity, recover degraded pastures, and increase carbon sequestration in the soil (Carvalho et al., 2018; Valera et al., 2024). On the other hand, Argentine meat production has historically maintained a strong base in extensive systems, although in recent decades there has been an increase in intensification through the use of feed supplements and confinement systems, mainly to meet global demand for animal protein (Arrieta et al., 2020; Fischer and Bilenca, 2020). In Uruguay, animal protein production continues to be characterized by traditional pasture systems, with little adoption of intensification technologies compared to Brazil and Argentina. This characteristic expresses the importance of these systems for the conservation of traditional production and ecosystems in this region of the country (Becoña et al., 2014; Montossi and Cazzuli, 2016).

The differences in climate performance observed among the evaluated production systems were further supported by the deterministic sensitivity analysis. Although all scenarios responded to the imposed ±20% perturbation in enteric CH₄ emissions, the magnitude of the response differed according to the emission profile of each production strategy. Extensive systems exhibited the greatest sensitivity because enteric CH₄ represented the predominant contribution to total GWP, whereas grain-supplemented systems showed a lower response due to the greater contribution of other inventory flows associated with input use and management practices. Likewise, the intensive systems maintained comparatively lower GWP values despite the imposed perturbation, indicating that their environmental performance was not determined solely by variation in enteric CH₄ emissions. Overall, these findings indicate that the comparative interpretation of GWP among the evaluated production systems remained robust despite variations in a key inventory parameter.

Although the most productive systems showed a reduction in climate intensity, the results also reveal a substantial change in environmental impacts for other categories. The acidification potential showed an extremely high oscillation between the scenarios, with values between 1.59E-01 and 4.63E+00 kg SO₂eq kg^-1 of beef produced, representing a difference of almost 30 times between the extreme systems. The lowest values were found in extensive systems based exclusively on native pastures, where fertilizers and pesticides were not applied. Conversely, the highest values were observed in scenarios E, H, F, and K, all characterized by greater use of nitrogen fertilizers and more intensive agricultural management. These results demonstrate the strong influence of the nitrogen cycle on acidification-related emissions, since fertilizers contribute substantially to NH₃ and NO_X emissions associated with food production (Gerber et al., 2013; Leip et al., 2015; Poore and Nemecek, 2018). Similar patterns have been reported in previous studies. Dick et al. (2015) observed acidification potentials of 2.80E-03 and 3.80E-03 kg SO₂eq kg^-1 of live weight for native and improved pastures, respectively, while Morais et al. (2023) reported values ranging from 2.47E-02 to 4.73E-02 kg SO₂eq kg^-1 of live weight gain in confined systems. Despite differences in methodological approaches, the studies consistently indicate that greater dependence on external inputs and agricultural management tends to increase acidification potential. Therefore, the higher values observed in scenarios E, H, F, and K reflect the environmental trade-offs associated with nitrogen-dependent intensification strategies.

This pattern was also observed in the assessment of eutrophication potential. While extensive systems showed extremely low values, below 1.00E-04 kg PO₄eq kg^-1 of beef, more intensive systems showed values above 1.60E+00 kg PO₄eq kg^-1 of beef, especially in scenarios H, K, and F, where fertilizer application and crop cultivation were more intensive. The magnitude of this difference highlights the decisive role of nutrient losses associated with pasture, crop, and forestry management in this impact category. Previous studies have reported similar results, indicating that fertilizer use and nutrient losses during soil management account for a substantial share of eutrophication impacts in beef production systems (Dick et al., 2015; Poore and Nemecek, 2018). Studies have observed eutrophication potentials of 1.17E-02 kg PO₄eq kg^-1 of live weight for extensive systems and 6.72E-03 kg PO₄eq kg^-1 of live weight for improved systems (Dick et al., 2015), while Morais et al. (2023) reported values ranging from 1.25E-03 to 2.30E-03 kg PO₄eq kg^-1 of live weight gain in confined systems. In Thailand, Ogino et al. (2016) observed lower eutrophication potential in extensive systems than in intensive systems. Although these studies adopted different functional units from the one used in the present study, they consistently demonstrate that greater dependence on fertilizers and external inputs tends to increase eutrophication potential, corroborating the patterns observed in the scenarios evaluated in this study.

Concerning land use, the results of this study indicate that productivity is an important factor influencing the impact per unit of product, although land-use performance also varied according to the specific characteristics of each production system. The intensive systems did not consistently present the lowest land-use values, as land transformation and occupation also reflected the contribution of pasture, feed, and forage production throughout the production cycle. On the other hand, some extensive systems based solely on native pasture showed relatively higher land-use values, mainly due to lower stocking rates and beef productivity. This result illustrates an important trade-off in extensive beef production because, although these systems use fewer external inputs, they require more area and time to produce the same amount of beef under a product-based functional unit. However, these results should be interpreted according to the metric adopted. Because land use was expressed per kilogram of beef produced, systems with higher productivity tend to present lower land occupation per unit of product, whereas extensive systems with lower stocking rates tend to present higher values. This result, therefore, reflects land-use efficiency relative to beef output and should not be interpreted as a direct measure of environmental pressure or ecological value per hectare. Therefore, alternative land-based metrics could provide a different comparative perspective between extensive and intensive systems and should be considered when interpreting this impact category. Studies report that beef production has one of the highest land-use intensities among all food sources, reflecting the low energy conversion efficiency of animal protein production compared with other protein sources (Flachowsky et al., 2017; Chaudhary and Brooks, 2019; Järviö et al., 2021; Fernandes et al., 2022).

The results also indicate that improvements in productive efficiency can substantially reduce land occupation per unit of product across different production contexts within the Pampa biome. In this sense, productive efficiency can contribute to lower land-use impacts per unit of product, although the results indicate that this relationship also depends on the land requirements associated with pasture, feed, and crop components within each production system. However, productive efficiency should not be interpreted independently of the structural conditions under which production systems operate. In the Brazilian scenarios, for example, 20% of the total farm area was allocated to legal reserve in accordance with national environmental legislation, reducing the area available for productive activities. Therefore, part of the differences observed among countries may reflect not only production strategies and productive efficiency, but also institutional and regulatory conditions that shape the productive capacity of the systems being evaluated. Consequently, comparisons among countries in this impact category should consider both productive and regulatory factors that influence land occupation and land availability for beef production.

The water and energy use indicators evaluated provided information on the intensity of natural resource use in different production systems. Direct water use showed considerable fluctuations between the evaluated scenarios, with values ranging from 1.53E-01 to 4.55E-01 m³ kg^-1 of beef. The highest values were observed in scenarios I, H, and G, which showed the greatest intensity of direct water use. Conversely, the lowest values occurred in scenarios F, K, and M, suggesting greater efficiency in direct water use per unit of product. These results highlight the differences in system productivity, stocking rate, and pasture management. The study conducted by Drastig et al. (2025) reports that the largest share of water use is related to beef production and animal feed production, making pasture management a decisive factor for the efficient use of water resources. A study by Klopatek et al. (2022) obtained higher water use values ​​in beef production systems where the diet was based on pasture and grains (6.78E-01 m³ kg^-1 of beef) compared to beef production based solely on pasture (4.68E-01 m³ kg^-1 of beef), corroborating our findings, where scenario H had a lower water use than scenario I.

The same pattern was observed in the evaluation of indirect energy use, with variance between values ​​of 5.15E-01 and 1.18E+00 MJ kg^-1 of beef produced. The highest energy use occurred in scenarios E, H, and I, while the lowest values ​​were found in scenarios L, A, and C. This performance is directly related to the degree of mechanization and the use of energy inputs in production systems. Systems characterized by agricultural management, soil preparation, fertilization, and mechanized operations showed higher use of diesel and electricity, which resulted in higher indirect energy use per unit of product. On the other hand, extensive systems based on native pasture showed lower dependence on energy inputs and consequently lower energy use values. According to Dick et al. (2021) and Morais et al. (2023), one of the main sources of energy use in beef production is the use of fossil fuels in agricultural operations and pasture management, especially in more intensive systems. In this context, the partial replacement of fossil fuels with renewable energy sources, such as the electrification of equipment, the use of biofuels, and solar energy, on rural properties can reduce energy use and emissions related to production systems, becoming an important opportunity to mitigate the environmental impacts of beef production (IEA, 2023; Singh et al., 2025).

When considering the complementary carbon-balance indicator, the results reveal interesting differences between the production systems evaluated. On the one hand, integrated systems presented some of the lowest indicator values, especially the Argentinian scenario E with crop-livestock integration, which showed a value of −230.71 kg CO₂eq kg^-1 of beef, substantially lower than the other scenarios analyzed. Other integrated systems, such as scenarios I, K, and H, also demonstrated negative indicator values, ranging from −49.5 to −69.3 kg CO₂eq kg^-1 of beef. Other systems also presented negative indicator values, including the intensive scenarios D and J, in which estimated gross vegetation CO₂ uptake during the pasture phase exceeded modeled emissions when the complete production cycle was considered. On the other hand, positive indicator values were observed in scenarios C and G, indicating that modeled emissions exceeded the estimated gross vegetation CO₂ uptake.

These results were associated with the combined contribution of vegetation components with high literature-based estimates of gross CO₂ uptake, particularly soybean and ryegrass, together with the period during which these components remained in the production cycle. In this context, negative indicator values indicate that the estimated gross vegetation CO₂ uptake exceeded the modeled emissions within the adopted calculation framework. However, these values represent gross vegetation CO₂ uptake during the production period and do not correspond to direct measurements of permanent carbon sequestration or changes in soil organic carbon stocks. Moreover, the subsequent fate and permanence of the carbon taken up by vegetation in biomass, soil, harvested products, or the atmosphere were not assessed. Therefore, negative values of this indicator should not be interpreted as evidence of a net-negative product carbon footprint or long-term climate mitigation.

Alves The results highlight the contribution of agricultural, pasture, and forestry components to estimated gross vegetation CO₂ uptake within the evaluated production systems. Studies conducted by Dick et al. (2015) and Mergen et al. (2023) reported a high capacity for carbon accumulation and stabilization under well-managed grazing systems, supporting the relevance of vegetation components to carbon dynamics in these systems. However, unlike the exploratory indicator applied in the present study, evidence of carbon accumulation or stabilization requires consideration of carbon stocks, permanence, and temporal dynamics. Therefore, the carbon-balance indicator should be interpreted as a complementary and exploratory comparison between estimated gross vegetation CO₂ uptake and system emissions during the production period, rather than as a measure of permanent carbon sequestration.

Finally, considering the results obtained in this study, it is evident that the pursuit of a balance in sustainability is not a trivial matter, as it demands analytical caution, critical analysis, and consideration of the multiple trade-offs involved. Even within the Pampa biome, a region internationally recognized for its traditional pasture-based production systems and the distinctive quality of its beef, substantial heterogeneity exists among production systems, highlighting the complexity of defining sustainable pathways for beef production. It is also important to recognize that the environmental performance patterns observed in this study are conditioned by the functional unit adopted, namely, 1 kg of chilled beef. Because this metric expresses impacts per unit of product, systems with higher productivity tend to present lower impacts per kilogram of beef, whereas less productive systems tend to present higher impacts. Alternative functional units based on land occupation, protein production, nutritional value, or economic value may lead to different interpretations of environmental performance, particularly when comparing systems with contrasting levels of productivity and management intensity (Wang et al., 2024). Therefore, the results presented here should be interpreted as comparisons of environmental intensity per kilogram of chilled beef rather than as definitive assessments of the overall sustainability of the evaluated production systems.

5 Limitations, uncertainties, and future recommendations

Although this study provides a comprehensive assessment of the environmental impacts associated with beef production in the Pampa biome, some methodological limitations should be considered when interpreting the results. Firstly, limitations regarding data collection for the construction of life cycle inventories. The modeling of production systems does not include some input flows, such as emissions related to the transportation and manufacturing of inputs and equipment, the use of vaccines and medicines, and procedures and activities after the departure of refrigerated meat from the slaughterhouse, such as transportation to retail, marketing, and final consumption. Although these flows can contribute to the overall environmental impact of production systems, studies indicate that, in beef production systems, the main sources of emissions are linked to enteric fermentation, manure management, and animal feed production (Gerber et al., 2013; Herrero et al., 2016). Therefore, although the common system boundaries adopted allow a consistent comparison among the evaluated scenarios, the inclusion of these omitted processes could modify the magnitude of the estimated impacts and potentially influence differences among systems with contrasting dependence on external inputs. Furthermore, Life Cycle Assessment studies applied to beef production indicate that modeling these systems is subject to uncertainties associated with variability in management practices, availability of primary data, and regional production differences, which can influence the resulting estimates of environmental impact (Poore and Nemecek, 2018; Rotz et al., 2019).

With reference to the impact categories considered, this work focused on specific environmental indicators and complementary indicators related to resource use, as in Rotz et al. (2019), not involving a complete water footprint assessment like water footprint methodologies, nor a detailed energy balance analysis of production systems. Although water use was considered in the systems, and energy use was estimated based on diesel and electricity usage, neither indirect water flows nor embodied energy throughout the production chain were taken into account. These flows could contribute to more integrated and comprehensive sustainability analyses and reveal important differences in these indicators across different production systems (Mekonnen and Hoekstra, 2012; Ridoutt et al., 2012).

Another limitation of this study is related to the exploratory carbon-balance indicator, which was based on literature-derived estimates of gross vegetation CO₂ uptake rather than direct measurements of carbon stock changes. These annual estimates were adjusted according to the period during which each vegetation component remained in the production systems. Therefore, the results represent estimated gross vegetation CO₂ uptake during the production period and do not correspond to direct measurements of long-term carbon storage in soil or biomass. Furthermore, carbon permanence and additionality were not directly evaluated. Although the 100-year horizon was adopted for greenhouse gas characterization through the GWP100 approach, this horizon was applied exclusively to the climate change impact category and does not imply permanent carbon storage over the same period. Future studies incorporating direct measurements of carbon dynamics may improve the understanding of the long-term climate implications associated with different beef production systems in the Pampa biome.

Finally, another limitation of the study is related to the absence of an integrated assessment of the three dimensions of sustainability: environmental, economic, and social. The assessment carried out focused only on the environmental dimension of the production chain, which is common in Life Cycle Assessment studies applied to food production (Poore and Nemecek, 2018; Stanley et al., 2018; Arrieta et al., 2020). However, the application of economic and social analyses, such as Social Life Cycle Assessment, can contribute to verifying aspects such as economic viability, income generation, working conditions, and social impacts on food production (UNEP, 2020; Schneider et al., 2024; Machado et al., 2025). Therefore, future research that integrates these dimensions could promote a more refined assessment of the sustainability of beef production, especially in the Pampa biome. Furthermore, the interpretation of the environmental performance of the evaluated systems is conditioned by the functional unit adopted, and alternative approaches based on land occupation, protein production, nutritional value, or economic value could provide complementary perspectives.

6 Conclusion

The results found in this study confirm that the assessment of the environmental sustainability of beef production in the Pampa biome depends on an integrated analysis of the different impact categories, since overall weighting can mask important trade-offs between production systems. No production system presented superior performance across all environmental indicators. While some semi-intensive and intensive systems showed comparatively lower climate change impacts, land-use performance varied among production strategies, and trade-offs remained across other environmental categories. Conversely, extensive systems showed lower impacts in some categories, but greater impacts in others. In this context, integrated and semi-intensive systems based on sown pastures emerged as intermediate alternatives, combining productive gains with favorable performance in multiple environmental indicators, although they are also associated with trade-offs related to the use of inputs and resources.

Thus, the results obtained provide scientific evidence on the environmental performance of contrasting beef production systems in the Pampa biome and improve the understanding of the trade-offs associated with different production and intensification strategies. Furthermore, these findings may inform future discussions on sustainability assessment frameworks and environmental performance criteria for beef production systems. Therefore, this study highlights the environmental implications of different beef production strategies and reinforces the importance of integrated assessments capable of capturing the complexity and heterogeneity of production systems within the Pampa biome.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frfst.2026.1840100/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Michele Michelin, University of Minho, Portugal

Reviewed by: Ernesto Viglizzo, Independent Researcher, Santa Rosa, Argentina

Régio Marcio Toesca Gimenes, Federal University of Grande Dourados, Brazil