Scenarios for long term conservation of Swedish semi-natural grasslands with limited climate impact

Scenarios for long term conservation of Swedish semi-natural grasslands with limited climate impact | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Scenarios for long term conservation of Swedish semi-natural grasslands with limited climate impact Johan Olof Karlsson, Karin von Greyerz, Anna Hessle, Mikaela Lindberg, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8112228/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Semi-natural grasslands are among Europe’s most species-rich habitats but are rapidly declining due to agricultural intensification, abandonment, and afforestation. Maintaining and expanding these habitats requires continued livestock grazing, raising potential conflicts with climate mitigation goals, particularly commitments to reduce methane emissions. This study explores production-side interventions that could increase grazing of semi-natural grasslands in Sweden under different caps on enteric methane emissions. Using a spatially explicit agri-food systems model (CIBUSmod), we combined regional estimates of grassland productivity with potential restoration areas to assess the impacts of the interventions on the maximum area of semi-natural grasslands that could be managed as well as effects on livestock production, cropland use, and greenhouse gas emissions. We evaluated scenarios including rearing male cattle as steers, prolonging dry periods or delaying culling of dairy cows, expanding winter lamb production, increasing horse grazing, and combinations thereof. Results show that with all interventions combined, the grazed area of semi-natural grasslands could increase by 0.5 Mha under constant methane emissions, nearly doubling current areas. Even under a 30% methane reduction, an additional 0.2 Mha (+ 38%) could be managed. Among the individual interventions, rearing steers, expanding winter lamb production, and increasing horse grazing showed the greatest potential to expand semi-natural grassland areas, although all interventions contributed when combined. Greenhouse gas emissions per hectare of grazed semi-natural grassland declined across all scenarios, but emissions per unit of edible protein generally rose, underscoring tensions between efficiency-oriented climate metrics and biodiversity goals. Achieving large-scale restoration in practice will require stronger market incentives, targeted policy support, and investments in rural infrastructure. This study demonstrates that it is technically feasible to expand grazing in semi-natural grasslands while containing climate impacts, but only if biodiversity and climate objectives are explicitly balanced when designing future food system policies. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1 Introduction Europe’s temperate semi-natural grasslands support high levels of biodiversity across several taxonomic groups but are today threatened by abandonment and traditional low-intensity agricultural practices being replaced by modern farming practices and intensification, both of which reduce grassland species diversity (Shipley et al. 2024 ). Similarly, Swedish semi-natural grasslands are characterised by high species richness, a result of centuries of traditional farming practices involving continuous nutrient removal without fertilisation or ploughing (Bengtsson et al. 2019 ; Eriksson 2022 ). Grazing livestock on these lands, serve a dual role of biodiversity conservation and meat and milk production. Since the early 20th century, there has been a dramatic decline in the extent of semi-natural grasslands in Sweden. In some landscapes, nearly all semi-natural grasslands present in the early 20th century has been lost (Cousins et al. 2015 ). This decline has largely been driven by agricultural and forestry intensification together with a decrease in the national stock of ruminants, which has resulted in a combination of abandonment, conversion to cropland, and afforestation of previous semi-natural grasslands (Eriksson & Cousins 2014 ). In turn, this has led to a loss of habitats along with landscape homogenisation (Auffret et al. 2018 ) and fragmentation of semi-natural grasslands (Ibáñez et al. 2014 ), with consequent loss of biodiversity. The area of semi-natural grasslands remaining in Sweden today (Fig. 1 ) is uncertain and depends on definitions, but 0.36 million hectares (Mha) is reported under EU Habitats Directive (Westling et al. 2020 ) and 0.42 Mha are managed with agri-environmental support (Swedish Board of Agriculture 2024 ). The area of semi-natural grasslands needed to achieve favourable conservation status in Sweden is uncertain and depends on how reference areas are defined. Under current EU Habitats Directive reporting (Westling et al. 2020 ), it is assumed that grasslands should have an areal cover of at least 20% in all landscapes with registered habitats, which would result in a seven-fold increase from today’s areas (Swedish EPA 2024a ). However, many regions have very little grassland habitat, making such restoration technically and ecologically unrealistic. Even when accounting for these constraints, at least a two-fold increase is still deemed necessary to meet EU requirements (Swedish EPA 2024a ). Achieving this would demand large-scale restoration and continuous management through grazing or mowing. Currently, grazing – primarily with cattle, sheep, or horses – is the main method of managing semi-natural grasslands. Although mowing played a historically important role in their maintenance (Lennartsson et al. 2016 ), it is now a marginal practice, covering less than 7,000 hectares in Sweden and mainly carried out for conservation purposes (Swedish Board of Agriculture 2024 ). Expanding the area of semi-natural grasslands would therefore likely depend on an increased number of livestock grazing them. Depending on the species and breeds involved, as well as the specific management and feeding strategies adopted, this could increase the total number of livestock and raise greenhouse gas emissions. However, parts of the current livestock production systems are not optimised for grazing. For example, most male cattle are currently raised as intact bulls (Gård & Djurhälsan 2023) with limited opportunities for grazing. Raising these animals as steers could allow for increased grazing (Hessle & Kumm 2011 ), but it would also reduce productivity and likely result in increased climate impact per unit meat produced. As such, there are possible trade-offs between biodiversity conservation goals, livestock production, and climate mitigation that need to be accounted for when designing strategies for biodiversity conservation in semi-natural grasslands. This has become increasingly relevant following the Global Methane Pledge (European Commission & United States of America 2021) where 160 countries, including Sweden, have agreed to collectively reduce global methane emissions by 30% from 2020 to 2030. Previous studies have explored scenarios for maintaining the current extent of semi-natural grasslands in Sweden while aiming to reduce the land use and climate impacts of diets, using food systems modelling approaches (Röös et al. 2016 ; Karlsson & Röös 2019 ). Scenarios involving a substantial expansion of semi-natural grassland area have also been developed for cattle on national level, assessing livestock numbers and associated methane emissions (Hessle & Danielsson 2023 ), as well as on county level, assessing environmental and economic outcomes (Hessle et al. 2017 ). However, no previous study has assessed scenarios for managing an expanded area of semi-natural grassland while also accounting for resource use and climate impacts within a comprehensive agri-food systems framework considering multiple species of livestock. Such an approach is essential for evaluating trade-offs and ensuring the feasibility of scenarios studied on a national level. This is particularly important in relation to the allocation of cropland for feed production versus other uses, as grazing livestock in Sweden need harvested feed during the relatively long indoor period. Moreover, existing scenarios have not considered where geographically the expansion of semi-natural grasslands could occur, nor how this spatial variation would influence potential forage production and, consequently, the number of livestock required. The aim of this study was to explore the potential for expanding grazing on Swedish semi-natural grasslands through production-side interventions, under different caps on enteric methane emissions. Regionally explicit pasture yield estimates and historic land use maps showing where the area of semi-natural grasslands could be expanded was used. The scenarios were modelled using a regionally explicit model of the Swedish agriculture and food system, to assess their impacts on the maximum area of grazed semi-natural grassland, livestock production, cropland use, and greenhouse gas emissions. 2 Materials and methods 2.1 Semi-natural grassland areas and yields In this study, semi-natural grasslands were defined in broad terms as permanent grasslands that are not practically suitable for arable crop production using standard agricultural machinery. As such, areas referred to here as semi-natural grasslands include both grasslands with minimal signs of historic productivity improvements, alongside grasslands whose land use history include varying levels of e.g. fertilisation, soil tillage and seeding (areas A and B in Fig. 1 , respectively). The reason for using this broader definition was two-fold: First, there is large variation in species richness among grasslands and many grasslands with signs of historic agricultural improvements still host a high species richness (Glimskär et al. 2023 ) and may be important complements to core habitats in maintenance of ecosystem services and biodiversity conservation (Hooftman et al. 2023 ; Aguilera Nuñez et al. 2024 ). Secondly, it aligns with the definition of ‘pasture’ on non-arable land used by the Swedish Board of Agriculture, meaning that nation-wide data is available on fine spatial resolution via the Swedish Land Parcel Identification System (LPIS), which covers all agricultural land for which farmers have applied for some kind of agricultural support. Data on biomass growth in semi-natural grasslands are very limited, with only a few studies having conducted field measurements (Spörndly & Glimskär 2018 ). In this study, we combined these field measurements with remote sensing data on topographic soil moisture index (SMI) and tree cover to estimate biomass growth across semi-natural grasslands. Both the SMI and tree cover datasets were derived from LiDAR imagery with national coverage, produced as part of Sweden’s National Land Cover Database (Swedish EPA 2020 ). These datasets were linked to semi-natural grassland parcels identified in the 2020 Swedish LPIS, by calculating the average SMI and tree cover for each parcel. Each grassland parcel was then classified into a soil moisture regime (dry, mesic, or moist) based on its SMI value (see Table S1 ). Biomass growth was estimated using the data presented in Table S1 using the lower productivity of shaded areas for the part of a parcel covered by trees. The LPIS includes crop codes indicating the land use or crop type for each parcel. Several of these codes distinguish different types of semi-natural grasslands, including pastures, meadows, and special classes such as ‘Alvar’, ‘Mosaic pastures’, and ‘Pastures with limited grass cover’. For these latter classes, the SMI was considered not to accurately reflect actual soil moisture conditions, due to characteristically thin soils and substantial areas of exposed bedrock in these landscapes. These parcels were therefore grouped into a separate category, and a lower biomass yield was assumed in accordance with Hessle and Danielsson ( 2024 ). The yields of mowed leys and pastures on arable land were estimated based on national statistics accounting for the fraction of mowed areas where regrowth is grazed following the method presented in Cederberg and Henriksson ( 2020 ). For the grazed areas, we assumed that gross biomass growth was 80% of the yield in mowed leys in accordance with Cederberg and Henriksson ( 2020 ). In the model, semi-natural grasslands were divided into four types: ‘semi-natural meadows’ (i.e. mown semi-natural grasslands), ‘semi-natural pastures’, ‘semi-natural pastures, thin soil’ (based on the LPIS classifications noted above), and ‘semi-natural pastures, wooded’ (defined as grassland parcels with over 60% tree cover). For each grassland type, estimated biomass yields were aggregated over harvest regions (i.e. the geographic unit of calculation used in the model). To account for climatic variability, the estimated biomass growth in semi-natural grasslands was adjusted based on the regional yield of mown leys derived from national statistics relative to the median ley yield across regions. This is a pragmatic approach that was also taken by Eriksson et al. ( 2009 ), but it is important to note that regional ley yields are also affected by management (e.g. fertilisation level) which confound the actual variation due to climate. To calculate the final grazed biomass per hectare (i.e. the net yield), which is decisive for the number of animals required to graze a certain area, the biomass yield estimates were adjusted for pasture utilisation rate with different factors per type of pasture (Table S1 ) derived from Ahlgren et al. ( 2022 ) and Hessle and Danielsson ( 2024 ). The estimated regional biomass productivity per pasture type was assumed to remain unchanged in the scenarios where areas expand. In other words, the restored semi-natural grasslands of a certain type in a region were assumed to be equal to existing semi-natural grasslands in terms of soil moisture, tree cover, and other factors affecting biomass productivity. Baseline areas of semi-natural grasslands were derived from LPIS data for the four grassland types described above. We did however find that estimated biomass growth in semi-natural grasslands on the islands of Öland and Gotland exceeded expected grazing demand based on animal numbers. This is likely an effect of low-yielding pastures on thin calcareous soils on the islands not being classified as ‘semi-natural grasslands, thin soils.’ We therefore assigned 80% of semi-natural grassland area in these regions to that class irrespective of LPIS classification. This is also in line with Ahlgren et al. ( 2024 ) who assumed a low yield for semi-natural grasslands on Gotland for this reason. Previous studies have also found that there are substantial areas of managed semi-natural grasslands that are not represented in LPIS data (Glimskär et al. 2023 ). Hiron et al. ( 2022 ) estimated semi-natural grassland area as well as the fraction of grasslands represented in the LPIS from a national-scale monitoring programme of grasslands. From there we derived correction factors, which were applied to the areal extent according to LPIS data to derive baseline areas. These correction factors were 1.37 for Northern Sweden (‘Norrland’), 1.23 for central Sweden (‘Svealand’), 1.18 for south Sweden (‘Götaland’), and 1.23 for the islands of Öland and Gotland. To estimate the potential for larger areas of semi-natural grasslands in different regions of Sweden we relied on a previously published dataset of potential pastures (Swedish EPA 2019 ). This dataset was created by combining information on semi-natural grasslands currently managed with land cover data and historical land use maps to distinguish areas that could potentially be cleared and/or restored to semi-natural grasslands. The preparation of this dataset also accounted for economic feasibility by only including areas that could be brought into management by creating larger enclosures together with existing semi-natural grasslands. From this dataset the total area of potential semi-natural grassland was extracted for each region used in our modelling (see section 2.3) using areal overlay. We used all enclosure sizes in the dataset (from 15 to 60 ha), which results in the largest area of potential pastures. 2.2 Scenario development Scenarios for alternative futures can support policy development and action by addressing long-term challenges characterised by high levels of uncertainty and complexity (Fauré et al. 2017 ). In this study, strategic scenarios (Börjeson et al. 2006 ) were developed by identifying a set of interventions in animal production systems with the potential to increase the area of grazed semi-natural grassland (Table 1 ). Following the finalisation of the scenarios and the production of preliminary modelling results, a workshop was held with researchers and stakeholders involved in the project. The stakeholders included representatives from a farmers’ organisation promoting pasture-based meat production from semi-natural grasslands, environmental organisations, and the retail sector. The workshop was structured in three parts: first, an overview of the scenarios and preliminary results was presented; second, participants discussed the technical design of the scenarios and the modelling approach; and finally, a broader discussion was held on the barriers and opportunities associated with each scenario, as well as the policy measures and incentives that would be required for their implementation. Participant inputs were collected by designated note-takers and via post-it notes. Based on this feedback, the scenarios and modelling assumptions were subsequently revised. The broader discussions during the workshop also informed the reflections presented in section 3.4 of this paper. Table 1 presents the final set of scenarios, and Section 2.4 elaborates on each scenario, outlining the underlying rationale and principal assumptions. Table 1 Overview of scenarios with their acronyms and a brief description Scenario Description MaxCur : Maximise use of semi-natural grasslands Currently, grazing in semi-natural grasslands is limited by access to semi-natural grasslands on a reasonable distance from the farms and lack of economic incentives (Larsson et al. 2020 ). Here we assume that such barriers are removed and that cattle, sheep and horses in current production systems graze semi-natural grasslands at their maximum potential. Steers : All male cattle raised as steers Currently most male cattle are raised as intact bulls with limited grazing (Gård & Djurhälsan 2023). Here we assume that all male cattle for slaughter are castrated and raised as low-intensive steers with a large share of their feeds coming from grazing semi-natural grasslands. CulCows : Culled cows are kept for an additional grazing season If not culled earlier because of injury or acute desease, dairy cows are culled and slaughtered after their completed final lactation period. In this scenario, dairy cows that are planned to be culled and slaughtered before the grazing season (March-May) are instead taken out of milk production and kept for another grazing season, grazing semi-natural grasslands during that time. DryCows : Prolonged dry period for dairy cows The dry period for dairy cows is usually two months, with potential to graze semi-natural pastures until one month before calving. Here, the dry period is extended to four months for all dairy cows, which allows for up to three months of potential grazing in semi-natural pastures. WinLamb : All lambs raised as winter lambs Currently around 20% of Swedish lambs are raised as winter lambs, i.e. lambs are born in the spring and graze during summer, whereafter they are kept indoors until slaughter in late winter/early spring. Winter lamb production is the production system with the highest potential for grazing semi-natural pastures (Ahlgren et al. 2022 ). In this scenario, all lambs born are assumed to be raised as winter lambs. RecHorses : Maximise semi-natural grassland grazing by horses Today, horses get a limited share of their grazing intake from semi-natural pastures (Cederberg & Henriksson 2020 ). In this scenario we assume that horses for breeding and all ponies and cold-blooded horses are able to graze semi-natural grasslands to a large extent. All : All interventions combined The combined effect of all above-described interventions. All + NatHorses : Nature conservation horses alongside the other interventions In this scenario, horses are introduced alongside the other interventions solely for the purpose of maintaining semi-natural grasslands. We assume that these horses will only be introduced in the southern parts of Sweden and are fed grazed grass for ten months of the year and fed harvested forage two month per year. 2.3 Model description The different scenarios were modelled using the CIBUSmod agri-food systems model (version 25.04; Karlsson et al. 2025 ). CIBUSmod is a biophysical mass-flow model of the Swedish agri-food system that balances demand for agricultural products with domestic production, allocating crop areas and livestock across 106 regions while minimising deviations from current land use and livestock numbers. The regions used are the smallest geographical unit for agricultural statistics in Sweden and vary in size from 100 ha to 68,000 ha of agricultural land with a median of 26,000 ha. The model calculates e.g. land use, nutrient flows and greenhouse gas emissions and includes constraints on e.g. crop rotations, regional forage supply and maximum land use to ensure agronomic feasibility. The model also includes a baseline dataset for Sweden (2016–2020), which was constructed based on national statistics on consumption, trade, land use, animal numbers and agricultural production along with industry statistics and published literature (see Karlsson et al. 2025 ). In this study, the model input was adjusted by removing the demand for milk, beef, and lamb, allowing the number of grazing animals to vary across scenarios, while demand for all other crop and livestock products was maintained at baseline levels. The total grazing livestock population was capped by a limit on methane emissions from enteric fermentation. Three different caps were implemented that reflect different priorities for biodiversity conservation in semi-natural grasslands versus climate impact mitigation. These were set at 30% reduction compared to baseline levels in line with the Global Methane Pledge (European Commission & United States of America 2021), no change from the baseline and a 10% increase from baseline levels. For non-grazing livestock (i.e. pigs and poultry), their populations were held constant at baseline levels across regions. To prevent drastic changes in the types of production across scenarios, the relative shares of beef and lamb production remained fixed, as did the ratio between beef and milk production. An exception to this was made for the ‘WinLamb’ scenario (see section 2.4). In all scenarios, total cropland use in each region was constrained to not exceed baseline levels to avoid potential competition with other land uses. The total area of semi-natural grassland (excl. wooded pastures) was constrained to the potential pasture area in each region (see section 2.2). However, the relative areas of the different types of semi-natural pastures were held constant, which also constrains the area of wooded pastures. The area of semi-natural meadows (i.e. mown semi-natural grasslands) was kept unchanged in the scenarios. To maximize the area of semi-natural grasslands in each scenario, the model was initially set to maximise the total biomass grazed from semi-natural grasslands. Grazed biomass rather than area was used in the optimisation goal to avoid that the optimisation algorithm prioritised grazing in the least productive regions. In a second round of optimisation, the maximized area of grazed semi-natural grassland in each region was enforced as a constraint, while minimising deviations from current crop areas and livestock numbers, to avoid unnecessary deviations from the current state. Modelling these scenarios within an agri-food systems framework ensures that they are feasible in terms of meeting feed requirements for livestock, including winter and concentrate feeds, while still maintaining food production and meeting the demand for monogastric livestock feed without increasing cropland use. 2.4 Detailed description of scenarios 2.4.1 MaxCur: Maximise use of semi-natural grasslands Currently in Sweden, grazing in semi-natural grasslands is often limited by a lack of economic incentives (Larsson et al. 2020 ). The remaining semi-natural grasslands are often small and scattered across the landscape, leading to high costs for transport, fencing, water supply, and animal monitoring (Kumm & Hessle 2020 ; Holmström et al. 2024 ), and restoration of semi-natural grasslands is expensive and associated with uncertainties regarding support payments. As these additional expenses are not fully offset by agri-environmental payments or price premiums on the market, profitability is typically low or absent for production based on semi-natural grasslands, and lower compared to other forms of livestock production (Holmström et al. 2021 ). In this scenario, we assume that these barriers have been removed and that strong economic incentives are in place, allowing cattle, sheep, and horses to maximize the use of semi-natural grasslands. This assumption also applies to all other scenarios described below, which means that this scenario forms the foundation for the other scenarios, which further increase the potential through different interventions in the production systems. In the model, the demand for grazing can be met either through grazing on arable land or in semi-natural grasslands. However, due to technical and biological limitations, such as the low energy density in the grass biomass of semi-natural grasslands or practical constraints on their use e.g. during lactation for dairy cows, CIBUSmod imposes animal category-specific limits on the maximum proportion of grazed biomass that can come from semi-natural grasslands. These limits were based on data from existing literature on typical feed rations for different animal categories (see Table S3). To account for year-to-year variability in pasture yields and avoid overestimating the potential for grazing in semi-natural grasslands, the literature values were reduced by 30%. This also applies to the maximum share of grazing from semi-natural grasslands in the different scenarios, which implies that the maximum share could not exceed 70% for any animal category, in any scenario. 2.4.2 Steers: All male cattle raised as steers Male calves can be reared either as intact bulls or castrated to become steers, which makes them calmer and more suitable for grazing. According to the Swedish Animal Welfare Ordinance (Djurskyddsförordning 2019 , 2 kap. 4 §), cattle older than six months must have access to pasture during the summer, except for bulls, which are exempt from this requirement. Currently, approximately 80% of male calves of dairy breed and 90% of calves of beef breed are raised as intact bulls (Gård & Djurhälsan 2023) kept indoors post-weaning, since raising bulls is often more profitable than keeping steers (Holmström et al. 2021 ). Castrating all male calves (from dairy systems and suckler herds) and rearing them as steers can therefore increase the number of grazing livestock available for maintaining semi-natural grasslands, without increasing the total cattle population. Steers from suckler cow systems are typically slaughtered at either 24 months in the spring or at 30 months during the autumn. The latter option allows them to graze for an additional season. In this scenario, all male calves are assumed to be castrated and raised as steers, with the slaughter age for all steers set at 30 months. 2.4.3 CulCows: Culled cows are kept for an additional grazing season The most common culling reasons in Swedish dairy cows, after udder health problems, are reduced fertility and low milk yield (Växa Sverige 2024). Cows are typically culled at the end of the lactation if the reason is not related to injury or acute disease. Currently, cows are generally slaughtered when they are culled. In this scenario, cows culled just before the grazing season (March–May) are instead taken out of milk production (“dried off”) and kept on pasture for a “retirement period” and slaughtered after the grazing season, to increase grazing in semi-natural grasslands. 2.4.4 DryCows: Prolonged dry period for dairy cows Dairy cows have high nutritional requirements during the lactation period. Grazing semi-natural pastures only during this period is therefore not a viable option for maintaining current productivity levels (Hessle & Danielsson 2024 ). However, before calving, dairy cows are usually dried off for about two months. During this period, it is possible to have them graze semi-natural pastures up until one month before calving when nutritional requirements are lower and before the cow needs to adapt to a more energy dense diet that supports lactation (EAAP 2011 ). To increase the grazing of semi-natural pastures by these animals while maintaining the cows’ efficiency during the lactation period, the dry period may be extended by lengthening the calving interval. In this scenario, the calving interval was extended from around 12 months to 14 months, resulting in a dry period of four months. This allows for up to three months of grazing in semi-natural pastures instead of one month for the cows whose dry period coincides with the grazing season. 2.4.5 WinLamb: All lambs raised as winter lambs Currently, around half of the lambs raised in Sweden are slaughtered in the autumn, while the remainder is split between slaughter in the spring and winter (Ahlgren et al. 2022 ; LRF 2025 ). When lambs are slaughtered in the autumn (called autumn lamb production), both ewes and lamb graze, but grazing only semi-natural grasslands is not possible in this production system due to their high energy requirements. When lambs are slaughtered in the spring (i.e. spring lamb production), the dry ewes graze semi-natural grasslands, but the lambs do not graze as they are born in the winter and slaughtered before time of turn-out to pasture. When spring-born lambs are slaughtered in the winter (i.e. winter lamb production), however, both ewes and lambs can graze semi-natural grasslands the entire grazing season due to the lower expected growth rate of lambs in this system, resulting in a large possible intake from semi-natural grasslands (Ahlgren et al. 2022 ). In this scenario we therefore assume that all lambs are raised as winter lambs, to maximise the amount of semi-natural grasslands maintained by these animals. In this scenario, lamb production was also allowed to increase in relation to beef production to align with the corresponding ratio in Swedish meat consumption (currently import shares are higher for lamb compared to beef). 2.4.6 RecHorses: Maximise semi-natural grassland grazing by horses In the model, the maximum share of grazing from semi-natural grasslands was set at 10% for all horses (Table S3) due to practical difficulties in keeping riding horses on semi-natural grasslands. However, most horses can meet their nutritional needs entirely from semi-natural grasslands (Jamieson & Hessle 2021 ). Moreover, keeping young horses on semi-natural grasslands with rough terrain has additional benefits, such as improving their physical strength (Sassner et al. 2022 ). This presents an opportunity to increase the extent of semi-natural grasslands managed by these animals. In this scenario we assume that horses for breeding and all ponies and cold-blooded horses can obtain up to 70% of their grazing intake from semi-natural grasslands. The total intake from grazing remains unchanged to reflect current horse-keeping practices. 2.4.7 All: All interventions combined This scenario represents the combined effect of all above-described scenarios. 2.4.8 All + NatHorses: Nature conservation horses alongside the other interventions Previous studies have suggested that year-round grazing by horses can have positive effects on biodiversity in semi-natural grasslands (Köhler et al. 2016 ; Ringmark et al. 2019 ). Moreover, horses do not produce as much methane as ruminants (Elghandour et al. 2019 ), which means that horses can contribute to biodiversity conservation through grazing at a lower climate cost. A field trial in east-central Sweden (Ringmark et al. 2019 ) found that crude protein concentrations in pastures were adequate to support year-round grazing, but that energy concentration and herbage availability were potentially limiting factors during the winter months. In this scenario additional horses kept exclusively for the purpose of managing semi-natural grasslands are introduced in southern Sweden (up to 60° north) alongside the other interventions described above. These horses are assumed to be fed only grazed grass for ten months of the year with up to 70% of the dry matter intake from semi-natural grasslands. The remaining two months, horses are assumed to be fed harvested forage. 3 Results and discussion 3.1 Semi-natural grassland areas and yields In the baseline, we estimate the area of grazed or mowed open semi-natural grassland at 0.51 Mha plus an additional 0.01 Mha grazed wooded pastures, after correcting for areas that are managed but not represented in the Swedish LPIS (Glimskär et al. 2023 ). The largest areas of semi-natural grasslands are located in the south-eastern parts of Sweden (Fig. 2 a). The total area of potential open semi-natural grasslands in the dataset based on historic land use maps (Swedish EPA 2019 ) was around 1.2 Mha (including currently managed pastures), which is an increase of 0.70 Mha from the baseline. While there is a clear geographic concentration of currently managed semi-natural grasslands, the potential additional areas are distributed across Sweden, including northern Sweden, which currently have relatively small areas of semi-natural grasslands (Fig. 2 b). The gross biomass yield in semi-natural pastures was estimated at 2.4 tonnes dry matter per hectare (t DM/ha) in the baseline as a national average across the three types of semi-natural pastures included in the model (Figure S1 ). For comparison, the gross yield of pastures on arable land were estimated at 4.3 t DM/ha. Accounting for pasture utilisation rates, net pasture yields were estimated at 1.1 t DM/ha across the three types of semi-natural pastures, as a national average. The net yields estimated here are higher than the 0.9 t DM/ha estimated by Hessle and Danielsson ( 2023 ) as a Swedish average. It should however be noted that Hessle and Danielsson ( 2023 ) only considered grasslands classified as valuable grassland habitats under the EU Habitats Directive, which is a stricter definition of semi-natural grasslands compared to the one used here. This likely explain the discrepancies in estimates. 3.2 Potential for grazing in semi-natural grasslands Results show that, under a combination of all intervention scenarios (excluding introduction of nature conservation horses; i.e. the ‘All’ scenario) and allowing a 10% increase in enteric methane emissions, it would be possible to manage 1.1 Mha of open semi-natural grassland, more than double the baseline area (Fig. 3 ). Even with strong reductions in enteric methane emissions (˗30%), which would limit livestock numbers, an area of up to 0.70 Mha could still be managed under the ‘All’ scenario. This represents an increase of 0.19 Mha compared to baseline areas. These results depend strongly on the assumed biomass productivity in semi-natural grasslands, which is uncertain due to the limited data available. Sensitivity analysis showed that assuming a 25% lower biomass productivity across all types of semi-natural grasslands resulted in an 7–33% increase in the maximum area possible to graze under the ‘All’ scenario (see S3 in Table S4). The effect was strongest under the lowest methane emissions cap (˗30%) showing that other factors constrain the area increase under the caps that allow more animals. Uncertainty in estimates of grassland area and biomass productivity is a major source of error in modelling grass-fed livestock systems (see e.g. Pfeifer et al. 2025 ). To better estimate both the livestock numbers needed to manage biodiverse semi-natural grasslands, and the food production potential of grasslands, further research and data collection are required to improve area and yield estimates. In addition to the open semi-natural grasslands, the area of wooded pastures increased from 0.01 Mha in the baseline to up to 0.04 Mha under the ‘All’ scenario with 10% increase in methane emissions. In the scenarios, the proportion of wooded pastures to open semi-natural grasslands were assumed to remain constant across regions. Sensitivity analysis showed that assuming a 10-fold increase in the proportion of wooded pastures resulted in a 16–18% increase in the maximum total area of semi-natural grasslands (incl. wooded pastures) under the ‘All’ scenario (see S4 in Table S4) due to the lower biomass productivity in wooded pastures. This would however also imply a smaller area of open semi-natural grasslands as compared to the original scenario (11–14% reduction). Introducing grazing horses for nature conservation along with the other interventions (‘All + NatHorses’), showed a large potential to increase grazed areas under the lower methane emission caps. It is however important to note that this would require many additional horses kept for nature conservation (Figure S3). With a 30% reduction in enteric methane emissions, an additional 190,000 horses would be required to reach the maximum potential of 0.84 Mha open semi-natural grasslands, which can be compared to the 360,000 horses currently kept in Sweden. None of the scenarios led to a full utilisation of all potential pasture areas (Figure S2). This was mainly explained by large areas of potential pastures in the northern and alpine regions of Sweden where the constraint on cropland expansion did not allow enough production of winterfeed for the number of animals required to graze all potential areas. Many of these areas have likely been mown for winter feed rather than grazed, a management practice that persisted longer in northern Sweden than in the south (Lennartsson & Westin 2019). Mowing as a management option was not considered here due to the economic and practical constraints on large-scale re-introduction of traditional mowing practices. On a national scale, sensitivity analysis showed that cropland area was not a constraining factor for the maximum area of semi-natural grasslands in the ‘All’ scenario (see S6 in Table S4). Likewise, the potential area of semi-natural grasslands was shown not to be a constraining factor for the maximum area possible to manage under the ‘All’ scenario on a national level (see S5 in Table S4), while for several individual regions this was a constraint (Figure S1 ). Among individual interventions, ‘Steers’ had the greatest potential to increase semi-natural grassland grazing, followed by ‘RecHorses’ and ‘WinLamb’, with some variations in the ordering of scenarios depending on the level of methane emissions allowed (Fig. 3 ). Interventions focused on dairy cows (‘DryCows’ and ‘CulCows’) had relatively smaller effects. However, all interventions contributed to the overall effect when combined. Few studies have previously evaluated the potential for expanding grazing on Swedish semi-natural grasslands. Hessle et al. ( 2021 ) estimated the area of semi-natural grasslands that could be grazed by rearing all male cattle as steers, similar to the ‘Steers’ scenario in this study. They found that an extensively raised steer could graze between 1.0 and 1.6 ha of semi-natural grassland over its lifetime. This aligns with results from our ‘Steers’ scenarios, where the national average was approximately 1.2 ha per steer, with some variations between scenarios due to differences in the spatial distribution of increased grassland areas. As a total potential, Hessle et al. ( 2021 ) estimated that all male cattle currently not grazing, or only grazing during the suckling period, could graze 0.21–0.29 Mha of semi-natural grassland if castrated and raised extensively as steers, which is similar to our ‘Steers’ scenario (± 0% CH₄ cap), where the area of semi-natural grassland increased by 0.31 Mha compared to the baseline. However, it is important to note that this figure also reflects increased use of semi-natural grasslands by other animal categories (see section 2.4.1). The potential for increased grazing in semi-natural grasslands across the scenarios was a combined effect of changes in total demand for grazed biomass and the proportion of biomass sourced from these grasslands (Fig. 4 ). The relative importance of these two factors differed across scenarios with e.g. the ‘RecHorses’ scenario having no effect on grazed biomass but increased the share from semi-natural grasslands and the ‘Steers’ scenario having a strong effect on total grazed biomass but a smaller effect on the share from semi-natural grasslands compared to the ‘MaxCur’ scenario (Fig. 4 ). Sensitivity analysis showed that the limits set on the maximum share of grazed biomass that could come from semi-natural grasslands had a decisive influence on the results. When the originally applied correction factor of -30% (see section 2.4.1) was removed the maximum area of semi-natural grasslands increased by 9–42% under the ‘All’ scenario depending on methane emissions cap (see S2 in Table S4). The effect was strongest for the lowest cap (-30% CH 4 ) where the maximum area increased from 0.73 to 1.0 Mha. The regional distribution of changes in animal numbers in the ‘All’ scenario showed a tendency of reduced livestock numbers in south-western Sweden and increased numbers in northern Sweden (Fig. 5 a), which indicates that current distribution of livestock numbers does not align with potential grassland areas. Sensitivity analysis in which the regional distribution of animal numbers was constrained to the baseline distribution however showed only a small effect (-6%) on the maximum area of grazed semi-natural grasslands in the ‘MaxCur’ scenario (see S1 in Table S4). Under the strictest methane emission cap (-30%), semi-natural grassland areas increased mainly in the northern and central parts of Sweden and on the islands of Öland and Gotland of the east coast (Fig. 5 b). When larger methane emissions were allowed, semi-natural grassland areas increased throughout most regions of Sweden. However, it is important to note here that the regional distribution of areas follows from the optimisation goal function employed (i.e. maximising the grazed biomass from semi-natural grasslands) which does not necessarily reflect biodiversity conservation priorities. Sensitivity analysis showed that changing the optimisation goal function to one that maximised the total area of semi-natural grasslands led to a 1–3% increase in the total area under the ‘All’ scenario (see S7 in Table S4). Areas increased in northern Sweden, while they decreased slightly in some regions of southern Sweden compared to the original scenario due to the lower estimated productivity in semi-natural grasslands in northern Sweden. This shows that using a different approach for prioritising which areas to graze under the set constraints results in a different regional distribution and influence the maximum area of semi-natural grassland due to regional differences in pasture productivity. However, the sensitivity analysis showed that a different prioritisation of where to expand areas had a limited impact on results in terms of the total area of semi-natural grassland possible to graze under set constraints. While a broad range of scenarios were developed in this study, they do not represent an exhaustive assessment of production-side interventions that could potentially increase the area of grazed semi-natural grasslands under maintained or decreased climate impact from the livestock sector. We chose here to focus on concrete adjustments to current production systems, rather than transformative changes. Especially for the dairy sector, which has developed fast towards increased intensity and productivity (Karlsson et al. 2023 ), more transformative changes could be envisioned to allow increased grazing for dairy cows during lactation. 3.3 Animal source food production and climate impacts In all scenarios except ‘MaxCur’ and ‘RecHorses’, meat and milk production decreased when methane emissions were capped at baseline levels. These reductions ranged from 1% to 18% in terms of edible protein, with the greatest decreases observed when all interventions were combined (Fig. 6 a). The reason that production decreases under constant methane emissions is that most scenarios involve reduced productivity with higher methane emissions from enteric fermentation per unit food produced. For example, raising male cattle as steers implies slower growth rates and thus more energy spent on maintenance as compared to muscle growth. Similarly for the two scenarios affecting dairy cows, both result in longer unproductive (dry) periods during a dairy cow’s lifetime, again resulting in more energy spent on maintenance as compared to milk production. Allowing a 10% increase in methane emissions enabled increased production in all scenarios except in the ‘All’ scenario. Under a 30% reduction in methane emissions, edible protein production from milk and ruminant meat declined by 32% to 44% depending on scenario. It is however also important to note that all scenarios led to a reduced demand for cropland for grazing and winter feed, especially under the strictest methane emissions cap where the number of animals was reduced (Fig. 5 c). This was a combined effect of prioritising semi-natural grasslands over pastures on arable land in grazing, increased share of grazing in feed rations for some scenarios, and, under the lower methane emissions cap, reduced animal numbers. This opens opportunities for increasing other forms of food production. For example, if using the spared cropland to cultivate cereals, accounting for the regional average cereal yields, it would be possible to increase cereal production in the ‘All’ scenario by 2.2, 0.8 and 0.4 million tonnes under the − 30%, ± 0% and + 10% methane emissions caps, respectively. In the baseline, total cereal production used for food, feed, bioenergy, and exports was around 4.9 million tonnes in the model. It should however be noted that cereal production is not feasible on all arable land currently used for leys. Enteric methane emissions per hectare of grazed semi-natural grassland ranged from 148 kg CH₄/ha (‘MaxCur’) to 107 kg CH₄/ha (‘All’). This can be compared to results from Hessle and Danielsson ( 2024 ) who estimated the number of cattle required to achieve favourable conservation status for Sweden’s semi-natural grasslands, along with the associated increase in enteric methane emissions. They found that grazing an additional 2.2 Mha of semi-natural grassland would result in emissions of 179,000 tonnes of CH₄ from enteric fermentation if the existing proportions of beef and dairy cows are maintained and all male cattle are raised as steers. This corresponds to 81 kg CH₄ per hectare of grazed semi-natural grassland, which is lower than in our scenarios. This discrepancy is mainly due to our conservative assumptions regarding the maximum proportion of semi-natural grassland allowed in the grazing regime for different animal categories (i.e. a 30% reduction of literature values; see section 2.4.1). In sensitivity analyses where this adjustment was removed (see S2 in Table S4), enteric methane emissions per hectare of grazed semi-natural grassland ranged from 79 to 101 kg CH₄/ha, depending on the methane emissions cap level, which is in line with the figure derived from Hessle and Danielsson ( 2024 ). A comparison of total climate impacts per unit of edible protein versus per unit area of grazed semi-natural grassland revealed a clear negative correlation across scenarios (Fig. 6 b). That is, scenarios with lower greenhouse gas emissions per area of grazed grassland tended to have higher emissions per unit of edible protein produced, and vice versa. This highlights a risk that strong incentives for climate impact mitigation, especially if measured by efficiency (i.e. impact per unit product), may inadvertently steer towards livestock production systems that are less conducive to biodiversity conservation at low climate cost. It is thus crucially important to account for other services than food production provided by livestock systems when moving forward with stronger force on the climate agenda. This is however often not the case today. For example, the largest dairy company in Sweden has introduced an incentive tool (FarmAhead™) with additional payments to farmers for “climate and environmental sustainability activities” (Arla 2025 ). While the scheme includes some indicators on “biodiversity,” it is skewed towards indicators for (climate) efficiency in production, which may invertedly disincentivise grazing-based production systems. Climate impacts, both per unit of protein and per area of grazed grassland, were generally higher under the 30% methane reduction cap. This is largely due to reductions in cattle and sheep production, which increased the proportion of horses. As horses were assumed to produce no edible protein while contributing relatively little to grazing semi-natural grasslands, their increased share negatively affected both climate efficiency metrics. The exception is the scenario where nature conservation horses are introduced (‘All + NatHorses’) where the 30% methene reduction cap resulted in the lowest climate impact per area of grazed semi-natural grassland (Fig. 6 b) but also the lowest production of meat and milk (Fig. 6 a). The results presented here show that it is possible to increase the area of grazed semi-natural grasslands under maintained or reduced livestock numbers, corroborating previous assessments showing that animal numbers are not the limiting factor for maintenance of semi-natural grasslands (Larsson et al. 2020 ). While the scenarios generally led to reduced animal-source food production under a given cap on methane emissions, they also opened opportunities to increase other forms of food production by reducing cropland demand for grazing and winter feed. As such these scenarios also show how a future with reduced animal-source food demand can be compatible with maintained or increased areas of semi-natural grasslands, in line with previous studies (Röös et al. 2016 ; Karlsson & Röös 2019 ). 3.4 Barriers, opportunities and policy implications The feasibility of expanding grazing in semi-natural grasslands, as assessed through the modelled scenarios, hinges not only on biophysical potential but also on economic viability, logistical practicality, market acceptance, and policy alignment. In this section we discuss the broader barriers, opportunities and policy implications for the scenarios, using insights from the workshop (see section 2.2) that was conducted within the project as a base. The stakeholder workshop underscored the complexity of these interacting factors and provided insight into the real-world challenges and opportunities associated with achieving these scenarios in practice. Economic conditions were repeatedly identified as the most important constraint across all scenarios. Stakeholders emphasised the increasing costs of inputs, veterinary services, and logistics, compounded by stagnant or insufficient agri-environmental payments. As one participant summarised, "compensation for semi-natural pastures has not developed in pace with production costs." Currently Sweden provides agri-environmental support under the CAP for the management of semi-natural pastures and meadows to promote biodiversity and preserve cultural landscapes (Swedish Board of Agriculture 2025 ) and there is also nationally funded support for restoring such lands (Swedish EPA 2024b ). These payments have been indispensable in maintaining semi-natural grasslands in Sweden during the process of intensification and structural rationalisation of agriculture during the last 50 years. However, the consensus during the workshop was that without substantially higher and more stable compensation for management of semi-natural grasslands, even favourable market prices would be insufficient to drive major changes in livestock systems. This echoes findings on the EU level where the design of CAP funding schemes, while instrumental in driving land use, has often been considered poorly adapted to the preservation of high nature value farming systems including semi-natural grasslands (Varela et al. 2025 ). One specific option mentioned during the workshop was to direct current investment support, i.e. support for new buildings and other infrastructure, which have been found disproportionally directed towards larger farms (Nilsson 2017 ), towards smaller farms and infrastructure that facilitates pasture-based systems, which could also increase the effectiveness of investment support (Nilsson 2017 ). Targeted support for keeping male cattle as steers rather than bulls was also discussed as a viable option for directing support payments towards grazing livestock. However, participants raised veterinary shortages as a barrier to steer production, citing dramatically increased costs for castration services due to low availability. One participant also highlighted risks of policies that reduce the profitability of rearing bulls, since the revenues from this activity help sustain suckler cow enterprises, which are important for grazing semi-natural grasslands today. While effective support schemes are important, most revenues on beef and lamb enterprises come from selling meat and other products (Jamieson & Hessle 2021 ). This highlights the importance of market incentives in encouraging producers to increase grazing in semi-natural grasslands, a point that was also discussed during the workshop. There is a third-party certification scheme for semi-natural pasture-based meat, which is sold in one of the major retail chains in Sweden. Participants highlighted that there is high demand from consumers and retail for this certified meat, but that information towards farmers and higher prices are needed to scale production to meet demand. Participants also highlighted the need for similar schemes in the dairy sector which is currently developing towards increased size and productivity with reduced opportunities for grazing semi-natural grasslands (Karlsson et al. 2023 ). A recurring critique across several scenarios, especially for ‘Steers’, ‘WinLamb’, and ‘CulCows’, was the lack of alignment with current market dynamics. For example, large carcasses from 30 months beef steers may not match processor or consumer demand, and that longer rearing periods increase costs and reduce capital turnover. Concerns were also raised about shifting a large share of meat supply to the autumn, limiting the availability of fresh meat during other parts of the year, which may result in a growing market share for imported meat if Swedish production fails to meet demand year-round. Workshop participants viewed ‘RecHorses’ as a scenario with limited potential in real life due to limited capacity among horse owners to supervise animals remotely, fears of injury, and practical concerns like daily riding schedules. Still, there should be some potential among breeders and professional horse operations, particularly if attractive economic incentives were in place. The ‘CulCows’ scenario was r ecognised as a practical model already emerging in some operations, with potential for broader implementation if coupled with welfare-based payments or market premiums. In this scenario “retired” dairy cows could be sold to a separate enterprise that manages the grazing period. Thereby, the operations and profitability of the dairy farm would not be affected. Although stakeholders recognised the welfare benefits of the ‘DryCows’ scenario, it was widely regarded as impractical at scale due to transition risks (e.g., from indoor feed rations to pasture) and the limited availability of semi-natural pastures on many dairy farms. While novel and promising, stakeholders warned of some risks associated with the ‘NatHorses’ scenario, e.g., bark stripping by horses damaging valuable tree species. Participants also highlighted that a viable horse meat market to economically sustain such operations would likely be a prerequisite as full reliance on support payments for biodiversity conservation was regarded unrealistic. In summary, participants saw opportunities for rural development, recreation, and landscape maintenance through increased grazing. However, these benefits are unlikely to materialise without stronger market incentives and targeted investments in social infrastructure, including veterinary services, advisory networks, and housing in rural areas. Concerns were also raised about the lack of young farmers and the challenges of maintaining existing or establishing new operations in low-service regions. 4 Conclusions This is the first study to explore scenarios of increased grazing in Swedish semi-natural grasslands using a spatially disaggregated agri-food systems modelling framework that considers regional land availability constraints, as well as other biophysical and agronomic limitations, while estimating agricultural greenhouse gas emissions. Our results demonstrate that it is technically feasible to expand the area of grazed semi-natural grasslands in Sweden without increasing methane emissions and overall climate impacts. When all interventions studied here were combined semi-natural grassland area increased by 0.5 Mha under maintained methane emissions, which represents almost a doubling of current areas. This relies on the implementation of production-side interventions supporting grazing and that semi-natural grasslands are prioritised over pastures on arable land. However, achieving this in practice will require long-term supporting policies and improved economic incentives for grassland-based production. Among individual interventions, raising male cattle as steers rather than intact bulls showed the greatest potential for increased grazing in semi-natural grasslands. The comparison between climate impact per unit of food produced and per hectare of grazed grassland highlights a fundamental trade-off. Systems optimised for low climate impact per unit food produced are often not best suited to maintain semi-natural grasslands and their associated biodiversity. Climate impact-based indicators and incentive structures tend to promote efficiency in production (van der Werf et al. 2020 ) and may unintentionally favour production systems that are less conducive to biodiversity conservation at low climate cost. To achieve biodiversity conservation in semi-natural grasslands along with other goals of the agri-food system while reducing climate and other environmental impacts it is important to work towards a shared vision for the role of livestock in future food systems (Karlsson 2022 ; Resare Sahlin & Trewern 2022 ). This can guide the process of developing policies that support livestock systems that deliver on the multitude of services expected form society. Declarations Funding This work was funded by The Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (FORMAS) [grant number 2021-01434] and the Mistra Food Futures research program [Grant ID: DIA 2018/24]. Conflicts of interest The authors declare no competing interests. Ethics approval This study did not require ethics approval under Swedish law. Consent to participate Not applicable. Consent for publication Not applicable. Code and data availability CIBUSmod along with the baseline dataset for Sweden is available as open-source at https://github.com/SLU-foodsystems/CIBUSmod and is described in detail in Karlsson et al. (2025). The specific code and input data used to run the model and produce results presented in this paper are available at https://github.com/karlssonjo/seminatgrass. Authors' contributions Conceptualization: JOK and ER, Methodology: JOK, KvG Data curation: JOK, KvG, AH, ML and AG, Formal analysis: JOK and KvG, Writing – original draft: JOK, KvG and ER, Writing – review & editing: All authors, Funding acquisition: JOK and ER References Aguilera Nuñez G, Glimskär A, Zacchello G, Francksen RM, Whittingham MJ, Hiron M (2024) Agriculturally Improved and Semi-Natural Permanent Grasslands Provide Complementary Ecosystem Services in Swedish Boreal Landscapes. Agronomy 14(3):567. https://doi.org/10.3390/agronomy14030567 Ahlgren S, Behaderovic D, Wirsenius S, Carlsson A, Hessle A, Toräng P, Seeman A, den, Braver T, Kvarnbäck O (2022) Miljöpåverkan av svensk nöt-och lammköttsproduktion . https://www.diva-portal.org/smash/record.jsf?pid=diva2%3A1718732 [2025-11-14] Ahlgren S, Wirsenius S, Toräng P, Carlsson A, Seeman A, Behaderovic D, Kvarnbäck O, Parvin N, Hessle A (2024) Climate and biodiversity impact of beef and lamb production – A case study in Sweden. Agric Syst 219:104047. https://doi.org/10.1016/j.agsy.2024.104047 Arla (2025) FarmAhead™ - How Arla farmers are rewarded for their sustainability activities . https://www.arla.com/sustainability/the-farms/arlas-sustainability-incentive-model-qa/ [2025-09-19] Auffret AG, Kimberley A, Plue J, Waldén E (2018) Super-regional land-use change and effects on the grassland specialist flora. Nat Commun 9(1):3464. https://doi.org/10.1038/s41467-018-05991-y Bengtsson J, Bullock JM, Egoh B, Everson C, Everson T, O'Connor T, O'Farrell PJ, Smith HG, Lindborg R (2019) Grasslands—more important for ecosystem services than you might think. Ecosphere 10(2):e02582. https://doi.org/10.1002/ecs2.2582 Börjeson L, Höjer M, Dreborg K-H, Ekvall T, Finnveden G (2006) Scenario types and techniques: Towards a user's guide. Futures 38(7):723–739. https://doi.org/10.1016/j.futures.2005.12.002 Cederberg C, Henriksson M (2020) Gräsmarkernas användning i Sverige. CHALMERS TEKNISKA HÖGSKOLA. https://research.chalmers.se/publication/517805 [2025-11-14] Cousins SAO, Auffret AG, Lindgren J, Tränk L (2015) Regional-scale land-cover change during the 20th century and its consequences for biodiversity. Ambio 44(S1):17–27. https://doi.org/10.1007/s13280-014-0585-9 Djurskyddsförordning (2019) :66) Svensk författningssamling. https://www.riksdagen.se/sv/dokument-och-lagar/dokument/svensk-forfattningssamling/djurskyddsforordning-201966_sfs-2019-66/#K2 EAAP (2011) NorFor - The Nordic feed evaluation system . (EAAP publication, 130). Wageningen: Wageningen Academic Publishers. https://doi.org/10.3920/978-90-8686-718-9 Elghandour MMMY, Adegbeye MJ, Barbabosa-Pilego A, Perez NR, Hernández SR, Zaragoza-Bastida A, Salem AZM (2019) Equine Contribution in Methane Emission and Its Mitigation Strategies. J Equine Veterinary Sci 72:56–63. https://doi.org/10.1016/j.jevs.2018.10.020 Eriksson C, Ekberg M, Blom S, Rehn I, Sjödahl M, Söderberg T, Dahlström A, Hallgren K, Bengtsson N (2009) Utveckling av ängs- och betesmarker – igår, idag och imorgon. (Jordbruksverket Rapport 2009:10. https://www2.jordbruksverket.se/webdav/files/SJV/trycksaker/Pdf_rapporter/ra09_10.pdf [2025-11-14] Eriksson O (2022) Coproduction of Food, Cultural Heritage and Biodiversity by Livestock Grazing in Swedish Semi-natural Grasslands. Front Sustainable Food Syst. 6 https://doi.org/10.3389/fsufs.2022.801327 Eriksson O, Cousins SAO (2014) Historical landscape perspectives on grasslands in Sweden and the Baltic region. Land 3(1):300–321. https://doi.org/10.3390/land3010300 European Commission & United States of America (2021) Global methane pledge . [2025-11-14] Fauré E, Arushanyan Y, Ekener E, Miliutenko S, Finnveden G (2017) Methods for assessing future scenarios from a sustainability perspective. Eur J Futures Res 5(1):17. 10.1007/s40309-017-0121-9 Glimskär A, Hultgren J, Hiron M, Westin R, Bokkers EAM, Keeling LJ (2023) Sustainable Grazing by Cattle and Sheep for Semi-Natural Grasslands in Sweden. Agronomy , 13(10), 2469. https://www.mdpi.com/2073-4395/13/10/2469 Gård & Djurhälsan (2023) Genomsnittligt kvalitetsutfall för nötkreatur slaktade under 2023 . https://www.gardochdjurhalsan.se/kvalitetsutfall-slaktade-notkreatur/ [2025-11-14] Hessle A, Bertilsson J, Stenberg B, Kumm K-I, Sonesson U (2017) Combining environmentally and economically sustainable dairy and beef production in Sweden. Agric Syst 156:105–114. https://doi.org/10.1016/j.agsy.2017.06.004 Hessle A, Danielsson R (2023) Number of cattle required to obtain good ecological status in Swedish grasslands and livestock’s methane emissions . (Report 60). Swedish University of Agricultural Sciences, Skara, Sweden. https://res.slu.se/id/publ/123395 [2025-11-14] Hessle A, Danielsson R (2024) Cattle population required for favorable conservation status of management-dependent semi-natural grasslands and forests, and associated increase in enteric methane emissions. J Nat Conserv 78:126571. https://doi.org/10.1016/j.jnc.2024.126571 Hessle A, Danielsson R, Lidfors L (2021) Housed young bulls - mapping of scope and potential for nature conservation . Skara. https://res.slu.se/id/publ/112871 [2025-11-14] Hessle A, Kumm K-I (2011) Use of beef steers for profitable management of biologically valuable semi-natural pastures in Sweden. J Nat Conserv 19(3):131–136. https://doi.org/10.1016/j.jnc.2010.10.002 Hiron M, Lundquist P, Groote Ad, Glimskär A (2022) Resultat för gräsmarker i Remiil 2015–2020: markslag, landskapstyp och artrikedom inom och utanför jordbruksblock . Unpublished report Holmström K, Kumm K-I, Andersson H, Hessle A (2024) Labour in suckler cow herds – a study on enterprises in southern Sweden. Acta Agriculturae Scand Sect — Anim Sci 73(1–2):49–62. https://doi.org/10.1080/09064702.2023.2245400 Holmström K, Kumm K-I, Andersson H, Nadeau E, Segerkvist KA, Hessle A (2021) Economic incentives for preserving biodiverse semi-natural pastures with calves from dairy cows. J Nat Conserv 62:126010. https://doi.org/10.1016/j.jnc.2021.126010 Hooftman DAP, Kimberley A, Cousins SAO, Santamaría Bueno S, Honnay O, Krickl P, Plue J, Poschlod P, Traveset A, Bullock JM (2023) Could green infrastructure supplement ecosystem service provision from semi-natural grasslands? J Environ Manage 328:116952. https://doi.org/10.1016/j.jenvman.2022.116952 Ibáñez I, Katz DSW, Peltier D, Wolf SM, Connor Barrie BT (2014) Assessing the integrated effects of landscape fragmentation on plants and plant communities: the challenge of multiprocess–multiresponse dynamics. J Ecol 102(4):882–895. https://doi.org/10.1111/1365-2745.12223 Jamieson A, Hessle A (2021) Hinder och möjligheter för ökad naturbetesdrift ur ett lantbrukarperspektiv: en kunskapsöversikt . (SustAinimal Reports #1). https://res.slu.se/id/publ/115251 [2025-11-14] Karlsson JO (2022) Livestock as resource users and landscape managers-A food systems perspective. Swedish University of Agricultural Sciences, Uppsala. https://doi.org/10.13140/RG.2.2.25926.14403 Karlsson JO, Karlsson-Potter H, Lagnelöv O, Ericsson N, Einarsson R, Hansson PA (2025) CIBUSmod 25.04: A spatially disaggregated biophysical agri-food systems model for studying national-level demand- and production-side intervention scenarios. EGUsphere , 2025, 1–37. https://doi.org/10.5194/egusphere-2025-2625 Karlsson JO, Robling H, Cederberg C, Spörndly R, Lindberg M, Martiin C, Ardfors E, Tidåker P (2023) What can we learn from the past? Tracking sustainability indicators for the Swedish dairy sector over 30 years. Agric Syst 212:103779. https://doi.org/10.1016/j.agsy.2023.103779 Karlsson JO, Röös E (2019) Resource-efficient use of land and animals—Environmental impacts of food systems based on organic cropping and avoided food-feed competition. Land Use Policy 85:63–72. https://doi.org/10.1016/j.landusepol.2019.03.035 Kumm K-I, Hessle A (2020) Economic Comparison between Pasture-Based Beef Production and Afforestation of Abandoned Land in Swedish Forest Districts. Land 9(2):42. https://doi.org/10.3390/land9020042 Köhler M, Hiller G, Tischew S (2016) Year-round horse grazing supports typical vascular plant species, orchids and rare bird communities in a dry calcareous grassland. Agric Ecosyst Environ 234:48–57. https://doi.org/10.1016/j.agee.2016.03.020 Larsson C, Boke Olén N, Brady M (2020) Naturbetesmarkens framtid - en fråga om lönsamhet . (RAPPORT 2020:1). Lund: AgriFood Economics Centre. https://res.slu.se/id/publ/115477 [2025-11-14] Lennartsson T, Westin A (2019) Ängar och slåtter. Historia, ekologi, natur- och kulturmiljövård . Riksantikvarieämbetet. https://raa.diva-portal.org/smash/record.jsf?pid=diva2%3A1331194 [2025-11-14] Lennartsson T, Westin A, Iuga A, Jones E, Madry S, Murray S, Gustavsson E (2016) The meadow is the mother of the field. Comparing transformations in hay production in three European agroecosystems. Martor 21:103–126 LRF (2025) Statistikplattform kött . https://www.lrf.se/om-lrf/lrf-s-branschavdelningar/lrf-kott/statistik-och-verktyg/ [2025-10-31] Nilsson P (2017) Productivity effects of CAP investment support: Evidence from Sweden using matched panel data. Land Use Policy 66:172–182. https://doi.org/10.1016/j.landusepol.2017.04.043 Pfeifer C, Winterberg R, Leiber F (2025) Quantifying the Contributing Potential of European Grasslands to Food Protein and Organic Manure in a Circular Food System. Grass Forage Sci 80(2):e12713. https://doi.org/10.1111/gfs.12713 Resare Sahlin K, Trewern J (2022) A systematic review of the definitions and interpretations in scientific literature of ‘less but better’ meat in high-income settings. Nat Food 3(6):454–460. https://doi.org/10.1038/s43016-022-00536-5 Ringmark S, Skarin A, Jansson A (2019) Impact of Year-Round Grazing by Horses on Pasture Nutrient Dynamics and the Correlation with Pasture Nutrient Content and Fecal Nutrient Composition. Animals 9(8):500 Röös E, Patel M, Spångberg J, Carlsson G, Rydhmer L (2016) Limiting livestock production to pasture and by-products in a search for sustainable diets. Food Policy 58:1–13. https://doi.org/10.1016/j.foodpol.2015.10.008 Sassner H, Forssman K, Petersson P, Fredricson I (2022) Breeding for durability - Equine monitoring at pasture using halter attached Hoofstep sensors. 18th International Equitation Science Conference , Hartpury, UK. https://res.slu.se/id/publ/118498 [2025-11-14] Shipley JR, Frei ER, Bergamini A, Boch S, Schulz T, Ginzler C, Barandun M, Bebi P, Bolliger J, Bollmann K, Delpouve N, Gossner MM, Graham C, Krumm F, Marty M, Pichon N, Rigling A, Rixen C (2024) Agricultural practices and biodiversity: Conservation policies for semi-natural grasslands in Europe. Curr Biol 34(16):R753–R761. https://doi.org/10.1016/j.cub.2024.06.062 Spörndly E, Glimskär A (2018) Betesdjur och betestryck i naturbetesmarker . (Rapport 297). Uppsala: Sveriges lantbruksuniversitet. https://pub.epsilon.slu.se/15649/ [2025-11-14] Swedish Board of Agriculture (2024) Sveriges Miljömål - Ett rikt odlingslandskap - Betesmarker och slåtterängar . https://www.sverigesmiljomal.se/miljomalen/ett-rikt-odlingslandskap/betesmarker-och-slatterangar/ [2025-11-14] Swedish Board of Agriculture (2025) Miljöersättning för skötsel av betesmarker och slåtterängar 2025 . https://jordbruksverket.se/stod/jordbruk-tradgard-och-rennaring/jordbruksmark/betesmarker-och-slatterangar/skotsel-av-betesmarker-och-slatterangar [2025-09-19] Swedish EPA (2019) Mer naturbetesmarker och ekonomiskt bärkraftiga företag - Ett samverkansprojekt inom miljömålsrådets gemensamma åtgärdslista 2019 . https://www.naturvardsverket.se/4ac403/contentassets/260e782f99c948a0a23977ba43cd66d0/naturbetesmarker-o-barkraftiga-foretag.pdf [2025-11-14] Swedish EPA (2020) Nationella marktäckedata 2018 Teknisk rapport v1.1 . https://www.naturvardsverket.se/en/services-and-permits/maps-and-map-services/national-land-cover-database/ [2025-11-14] Swedish EPA (2024a) Översyn av referensarealer för livsmiljötyper i art- och habitatdirektivet - Redovisning av regeringsuppdrag . (NV-11038-22). https://www.naturvardsverket.se/496428/globalassets/om-oss/slutredovisade-regeringsuppdrag/redovisning-regeringsuppdrag-oversyn-av-referensarealer-for-livsmiljotyper-i-art-och-habitatdirektivet.pdf [2025-11-14] Swedish EPA (2024b) NFS 2024:3 – Föreskrifter om statligt stöd för vissa åtgärder som syftar till att bevara eller återställa biologisk mångfald . https://www.naturvardsverket.se/498b26/globalassets/nfs/2024/nfs-2024-3.pdf [2025-11-14] van der Werf HMG, Knudsen MT, Cederberg C (2020) Towards better representation of organic agriculture in life cycle assessment. Nat Sustain 3(6):419–425. https://doi.org/10.1038/s41893-020-0489-6 Varela E, Jay M, Flinzberger L, Mobarak C, Plieninger T (2025) A review of high nature value farming systems in Europe: Biodiversity, ecosystem services, drivers, innovations and future prospects. People Nat. https://doi.org/10.1002/pan3.70048 Växa S (2024) Cattle statistics 2024 . https://www.vxa.se/fakta/styrning-och-rutiner/mer-om-mjolk/statistik/ [2025-11-14] Westling A, Toräng P, Jacobson A, Haldin M, Naeslund M (2020) Sveriges arter och naturtyper i EU:s art- och habitatdirektiv: Resultat från rapportering 2019 till EU av bevarandestatus 2013–2018 . (ISBN: 978-91-620-6914-8). Bromma: Naturvårdsverket. https://www.naturvardsverket.se/publikationer/6900/sveriges-arter-och-naturtyper-i-eus-art--och-habitatdirektiv/ [2025-11-14] Supplementary Files ScenariodevelopmentmanuscriptcleanSM.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 20 Jan, 2026 Reviewers invited by journal 18 Jan, 2026 Editor invited by journal 15 Jan, 2026 Editor assigned by journal 19 Nov, 2025 First submitted to journal 17 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8112228","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":576381459,"identity":"f4a058fc-d4a5-40ae-8767-06a96dcd9fe3","order_by":0,"name":"Johan Olof Karlsson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIie2RsWrDMBRFrwhkUskqY0h+QcVgAjH9loqAspRS6JJRUJCXBK/+jECgs43WtlkNXdolc6FLoYZGdshSkNoxg84gLoiD7nsCAoEzhCh6DMzmqgsjDNS/FfRKpIhfAX4rvPpDGeTr+vMOGaL8+a3+arPbZFdrhjZzF1u9zOMSEjFdcHOh5X3aCM2Ilm6lvOExhcHYaoYoIx4bYhVlfEry3SujPWwxI7ZFX+zHp6T9KzGTqOjQiA1sMQwrzyxP6YxySaNyj24WUTbiYSr03Klc5qvklS6zMdtJ8m43JorC1M1He+VWVHfy0++cuHYKwMRzFwgEAoEjBxNdUXjJBU+fAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0003-1913-8504","institution":"Sveriges lantbruksuniversitet","correspondingAuthor":true,"prefix":"","firstName":"Johan","middleName":"Olof","lastName":"Karlsson","suffix":""},{"id":576381460,"identity":"86b6c568-2f53-4ef4-b609-1cb38c838949","order_by":1,"name":"Karin von Greyerz","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Karin","middleName":"","lastName":"von Greyerz","suffix":""},{"id":576381461,"identity":"004e548f-e0db-41c0-9ff2-e322a651de7a","order_by":2,"name":"Anna Hessle","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Hessle","suffix":""},{"id":576381462,"identity":"01aedbfb-f3bd-4335-b3ac-d7f9bd957237","order_by":3,"name":"Mikaela Lindberg","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Mikaela","middleName":"","lastName":"Lindberg","suffix":""},{"id":576381463,"identity":"7b03d74c-7419-47ed-889d-3faaf94b89fd","order_by":4,"name":"Anders Glimskär","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Anders","middleName":"","lastName":"Glimskär","suffix":""},{"id":576381464,"identity":"92facd93-957b-45b0-89d4-a93eede691b4","order_by":5,"name":"Pernilla Tidåker","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Pernilla","middleName":"","lastName":"Tidåker","suffix":""},{"id":576381465,"identity":"923eb894-796f-4df3-8118-278742c3407c","order_by":6,"name":"Jan Bengtsson","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"","lastName":"Bengtsson","suffix":""},{"id":576381466,"identity":"d473d7c3-88d3-4079-8bd7-18acb47ecc6f","order_by":7,"name":"Elin Röös","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Elin","middleName":"","lastName":"Röös","suffix":""}],"badges":[],"createdAt":"2025-11-14 08:09:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8112228/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8112228/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102389730,"identity":"4ca58e6c-6293-46c1-a865-9dd08fe0a8ec","added_by":"auto","created_at":"2026-02-11 08:34:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2384130,"visible":true,"origin":"","legend":"\u003cp\u003eAgricultural landscape mosaic in central Sweden with semi-natural grassland managed with agri-environmental support in the foreground (A), pasture on arable land that has not been ploughed for more than 30 years in the middle (B), and arable land in rotation, at present ley, in the distance (C). Photo: Johan O. Karlsson\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/aac5933d1122efec5b1bf42c.png"},{"id":102398490,"identity":"a58b38d4-f3b5-4704-a0ec-aec2f93111e2","added_by":"auto","created_at":"2026-02-11 10:23:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":154997,"visible":true,"origin":"","legend":"\u003cp\u003ea) Current areas of semi-natural grasslands across regions. b) Estimated additional areas potentially available based on historic land use maps. The total potential is the sum of currently managed grasslands and the potential additional areas. The points on the colour bars show the median area of current or potential additional semi-natural grasslands across regions and the whiskers the interquartile range.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/47fbb2dc0da95acacbfa242c.png"},{"id":102389734,"identity":"86ad572b-dec5-4fb0-b649-6db8778158ed","added_by":"auto","created_at":"2026-02-11 08:34:05","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":143653,"visible":true,"origin":"","legend":"\u003cp\u003eTotal area of managed semi-natural grasslands in the baseline (left hand bar and dotted line) and the different scenarios under the three different caps on methane emissions from enteric fermentation (indicated on top). The red line shows the total area of potential pastures (excl. wooded pastures). The scenarios are ordered from smallest to largest area of semi-natural grasslands managed for the ±0% CH\u003csub\u003e4\u003c/sub\u003e case.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/29cd98f62c36189d605b75ed.png"},{"id":102389731,"identity":"e6edebe3-4772-4d92-bab5-2bf41035cd95","added_by":"auto","created_at":"2026-02-11 08:34:05","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":137937,"visible":true,"origin":"","legend":"\u003cp\u003ea) Total demand for grazed biomass per livestock category in million tonnes dry matter. b) Share of total grazed biomass from semi-natural grasslands.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/ab700846fe12e58aa315621f.png"},{"id":102397815,"identity":"8b918363-13b9-445a-80c0-00bf961661c7","added_by":"auto","created_at":"2026-02-11 10:19:49","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":522539,"visible":true,"origin":"","legend":"\u003cp\u003eRegional changes in (a) the number of cattle, sheep and horses in terms of livestock units (LSU), (b) area of semi-natural grasslands, and (c) cropland area in the ‘All’ scenarios compared to the baseline. Each column represents a different cap on enteric methane emissions (indicated at the top). Above each map the total changes in livestock numbers or area are indicated in terms of million LSU (MLSU) or million hectares (Mha), respectively.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/a083c1a24812ee40fccdd77f.png"},{"id":102389737,"identity":"e6a5518d-63d4-43e5-8058-4a9cba991aa5","added_by":"auto","created_at":"2026-02-11 08:34:06","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":148018,"visible":true,"origin":"","legend":"\u003cp\u003ea) Relationship between meat and milk production in terms of protein and semi-natural grassland area across the different scenarios and methane emissions caps. b) Relationship between greenhouse gas emissions (GHGE) associated with cattle, sheep and horses per kg edible protein produced (x-axis) and per square meter of semi-natural grasslands grazed (y-axis) across the different scenarios (indicated by colour) and methane emissions caps (indicated by marker shape). The dashed line shows the linear least squares regression across the scenarios for the ±0% CH\u003csub\u003e4\u003c/sub\u003e cap.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/4c54930951b1cff59260c7e3.png"},{"id":102404143,"identity":"247bf610-6f90-4f36-87b3-f971de61edb5","added_by":"auto","created_at":"2026-02-11 11:01:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4434263,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/99b47d19-70ad-4f36-bf8e-fe70245fc68f.pdf"},{"id":102389736,"identity":"9680ebc8-e89d-4098-adbb-482b583b8d87","added_by":"auto","created_at":"2026-02-11 08:34:05","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":695283,"visible":true,"origin":"","legend":"","description":"","filename":"ScenariodevelopmentmanuscriptcleanSM.docx","url":"https://assets-eu.researchsquare.com/files/rs-8112228/v1/64cdcef0b9b80058020087a9.docx"}],"financialInterests":"","formattedTitle":"Scenarios for long term conservation of Swedish semi-natural grasslands with limited climate impact","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eEurope\u0026rsquo;s temperate semi-natural grasslands support high levels of biodiversity across several taxonomic groups but are today threatened by abandonment and traditional low-intensity agricultural practices being replaced by modern farming practices and intensification, both of which reduce grassland species diversity (Shipley et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Similarly, Swedish semi-natural grasslands are characterised by high species richness, a result of centuries of traditional farming practices involving continuous nutrient removal without fertilisation or ploughing (Bengtsson et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Eriksson \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Grazing livestock on these lands, serve a dual role of biodiversity conservation and meat and milk production.\u003c/p\u003e \u003cp\u003eSince the early 20th century, there has been a dramatic decline in the extent of semi-natural grasslands in Sweden. In some landscapes, nearly all semi-natural grasslands present in the early 20th century has been lost (Cousins et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). This decline has largely been driven by agricultural and forestry intensification together with a decrease in the national stock of ruminants, which has resulted in a combination of abandonment, conversion to cropland, and afforestation of previous semi-natural grasslands (Eriksson \u0026amp; Cousins \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). In turn, this has led to a loss of habitats along with landscape homogenisation (Auffret et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) and fragmentation of semi-natural grasslands (Ib\u0026aacute;\u0026ntilde;ez et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), with consequent loss of biodiversity. The area of semi-natural grasslands remaining in Sweden today (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) is uncertain and depends on definitions, but 0.36\u0026nbsp;million hectares (Mha) is reported under EU Habitats Directive (Westling et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and 0.42 Mha are managed with agri-environmental support (Swedish Board of Agriculture \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe area of semi-natural grasslands needed to achieve favourable conservation status in Sweden is uncertain and depends on how reference areas are defined. Under current EU Habitats Directive reporting (Westling et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), it is assumed that grasslands should have an areal cover of at least 20% in all landscapes with registered habitats, which would result in a seven-fold increase from today\u0026rsquo;s areas (Swedish EPA \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024a\u003c/span\u003e). However, many regions have very little grassland habitat, making such restoration technically and ecologically unrealistic. Even when accounting for these constraints, at least a two-fold increase is still deemed necessary to meet EU requirements (Swedish EPA \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024a\u003c/span\u003e). Achieving this would demand large-scale restoration and continuous management through grazing or mowing.\u003c/p\u003e \u003cp\u003eCurrently, grazing \u0026ndash; primarily with cattle, sheep, or horses \u0026ndash; is the main method of managing semi-natural grasslands. Although mowing played a historically important role in their maintenance (Lennartsson et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), it is now a marginal practice, covering less than 7,000 hectares in Sweden and mainly carried out for conservation purposes (Swedish Board of Agriculture \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eExpanding the area of semi-natural grasslands would therefore likely depend on an increased number of livestock grazing them. Depending on the species and breeds involved, as well as the specific management and feeding strategies adopted, this could increase the total number of livestock and raise greenhouse gas emissions. However, parts of the current livestock production systems are not optimised for grazing. For example, most male cattle are currently raised as intact bulls (G\u0026aring;rd \u0026amp; Djurh\u0026auml;lsan 2023) with limited opportunities for grazing. Raising these animals as steers could allow for increased grazing (Hessle \u0026amp; Kumm \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), but it would also reduce productivity and likely result in increased climate impact per unit meat produced. As such, there are possible trade-offs between biodiversity conservation goals, livestock production, and climate mitigation that need to be accounted for when designing strategies for biodiversity conservation in semi-natural grasslands. This has become increasingly relevant following the Global Methane Pledge (European Commission \u0026amp; United States of America 2021) where 160 countries, including Sweden, have agreed to collectively reduce global methane emissions by 30% from 2020 to 2030.\u003c/p\u003e \u003cp\u003ePrevious studies have explored scenarios for maintaining the current extent of semi-natural grasslands in Sweden while aiming to reduce the land use and climate impacts of diets, using food systems modelling approaches (R\u0026ouml;\u0026ouml;s et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Karlsson \u0026amp; R\u0026ouml;\u0026ouml;s \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Scenarios involving a substantial expansion of semi-natural grassland area have also been developed for cattle on national level, assessing livestock numbers and associated methane emissions (Hessle \u0026amp; Danielsson \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), as well as on county level, assessing environmental and economic outcomes (Hessle et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHowever, no previous study has assessed scenarios for managing an expanded area of semi-natural grassland while also accounting for resource use and climate impacts within a comprehensive agri-food systems framework considering multiple species of livestock. Such an approach is essential for evaluating trade-offs and ensuring the feasibility of scenarios studied on a national level. This is particularly important in relation to the allocation of cropland for feed production versus other uses, as grazing livestock in Sweden need harvested feed during the relatively long indoor period. Moreover, existing scenarios have not considered where geographically the expansion of semi-natural grasslands could occur, nor how this spatial variation would influence potential forage production and, consequently, the number of livestock required.\u003c/p\u003e \u003cp\u003eThe aim of this study was to explore the potential for expanding grazing on Swedish semi-natural grasslands through production-side interventions, under different caps on enteric methane emissions. Regionally explicit pasture yield estimates and historic land use maps showing where the area of semi-natural grasslands could be expanded was used. The scenarios were modelled using a regionally explicit model of the Swedish agriculture and food system, to assess their impacts on the maximum area of grazed semi-natural grassland, livestock production, cropland use, and greenhouse gas emissions.\u003c/p\u003e"},{"header":"2 Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Semi-natural grassland areas and yields\u003c/h2\u003e \u003cp\u003eIn this study, semi-natural grasslands were defined in broad terms as permanent grasslands that are not practically suitable for arable crop production using standard agricultural machinery. As such, areas referred to here as semi-natural grasslands include both grasslands with minimal signs of historic productivity improvements, alongside grasslands whose land use history include varying levels of e.g. fertilisation, soil tillage and seeding (areas A and B in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, respectively). The reason for using this broader definition was two-fold: First, there is large variation in species richness among grasslands and many grasslands with signs of historic agricultural improvements still host a high species richness (Glimsk\u0026auml;r et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and may be important complements to core habitats in maintenance of ecosystem services and biodiversity conservation (Hooftman et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Aguilera Nu\u0026ntilde;ez et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Secondly, it aligns with the definition of \u0026lsquo;pasture\u0026rsquo; on non-arable land used by the Swedish Board of Agriculture, meaning that nation-wide data is available on fine spatial resolution via the Swedish Land Parcel Identification System (LPIS), which covers all agricultural land for which farmers have applied for some kind of agricultural support.\u003c/p\u003e \u003cp\u003eData on biomass growth in semi-natural grasslands are very limited, with only a few studies having conducted field measurements (Sp\u0026ouml;rndly \u0026amp; Glimsk\u0026auml;r \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). In this study, we combined these field measurements with remote sensing data on topographic soil moisture index (SMI) and tree cover to estimate biomass growth across semi-natural grasslands. Both the SMI and tree cover datasets were derived from LiDAR imagery with national coverage, produced as part of Sweden\u0026rsquo;s National Land Cover Database (Swedish EPA \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese datasets were linked to semi-natural grassland parcels identified in the 2020 Swedish LPIS, by calculating the average SMI and tree cover for each parcel. Each grassland parcel was then classified into a soil moisture regime (dry, mesic, or moist) based on its SMI value (see Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Biomass growth was estimated using the data presented in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e using the lower productivity of shaded areas for the part of a parcel covered by trees.\u003c/p\u003e \u003cp\u003eThe LPIS includes crop codes indicating the land use or crop type for each parcel. Several of these codes distinguish different types of semi-natural grasslands, including pastures, meadows, and special classes such as \u0026lsquo;Alvar\u0026rsquo;, \u0026lsquo;Mosaic pastures\u0026rsquo;, and \u0026lsquo;Pastures with limited grass cover\u0026rsquo;. For these latter classes, the SMI was considered not to accurately reflect actual soil moisture conditions, due to characteristically thin soils and substantial areas of exposed bedrock in these landscapes. These parcels were therefore grouped into a separate category, and a lower biomass yield was assumed in accordance with Hessle and Danielsson (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe yields of mowed leys and pastures on arable land were estimated based on national statistics accounting for the fraction of mowed areas where regrowth is grazed following the method presented in Cederberg and Henriksson (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). For the grazed areas, we assumed that gross biomass growth was 80% of the yield in mowed leys in accordance with Cederberg and Henriksson (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the model, semi-natural grasslands were divided into four types: \u0026lsquo;semi-natural meadows\u0026rsquo; (i.e. mown semi-natural grasslands), \u0026lsquo;semi-natural pastures\u0026rsquo;, \u0026lsquo;semi-natural pastures, thin soil\u0026rsquo; (based on the LPIS classifications noted above), and \u0026lsquo;semi-natural pastures, wooded\u0026rsquo; (defined as grassland parcels with over 60% tree cover). For each grassland type, estimated biomass yields were aggregated over harvest regions (i.e. the geographic unit of calculation used in the model).\u003c/p\u003e \u003cp\u003eTo account for climatic variability, the estimated biomass growth in semi-natural grasslands was adjusted based on the regional yield of mown leys derived from national statistics relative to the median ley yield across regions. This is a pragmatic approach that was also taken by Eriksson et al. (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2009\u003c/span\u003e), but it is important to note that regional ley yields are also affected by management (e.g. fertilisation level) which confound the actual variation due to climate.\u003c/p\u003e \u003cp\u003eTo calculate the final grazed biomass per hectare (i.e. the net yield), which is decisive for the number of animals required to graze a certain area, the biomass yield estimates were adjusted for pasture utilisation rate with different factors per type of pasture (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) derived from Ahlgren et al. (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and Hessle and Danielsson (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe estimated regional biomass productivity per pasture type was assumed to remain unchanged in the scenarios where areas expand. In other words, the restored semi-natural grasslands of a certain type in a region were assumed to be equal to existing semi-natural grasslands in terms of soil moisture, tree cover, and other factors affecting biomass productivity.\u003c/p\u003e \u003cp\u003eBaseline areas of semi-natural grasslands were derived from LPIS data for the four grassland types described above. We did however find that estimated biomass growth in semi-natural grasslands on the islands of \u0026Ouml;land and Gotland exceeded expected grazing demand based on animal numbers. This is likely an effect of low-yielding pastures on thin calcareous soils on the islands not being classified as \u0026lsquo;semi-natural grasslands, thin soils.\u0026rsquo; We therefore assigned 80% of semi-natural grassland area in these regions to that class irrespective of LPIS classification. This is also in line with Ahlgren et al. (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) who assumed a low yield for semi-natural grasslands on Gotland for this reason.\u003c/p\u003e \u003cp\u003ePrevious studies have also found that there are substantial areas of managed semi-natural grasslands that are not represented in LPIS data (Glimsk\u0026auml;r et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Hiron et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) estimated semi-natural grassland area as well as the fraction of grasslands represented in the LPIS from a national-scale monitoring programme of grasslands. From there we derived correction factors, which were applied to the areal extent according to LPIS data to derive baseline areas. These correction factors were 1.37 for Northern Sweden (\u0026lsquo;Norrland\u0026rsquo;), 1.23 for central Sweden (\u0026lsquo;Svealand\u0026rsquo;), 1.18 for south Sweden (\u0026lsquo;G\u0026ouml;taland\u0026rsquo;), and 1.23 for the islands of \u0026Ouml;land and Gotland.\u003c/p\u003e \u003cp\u003eTo estimate the potential for larger areas of semi-natural grasslands in different regions of Sweden we relied on a previously published dataset of potential pastures (Swedish EPA \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). This dataset was created by combining information on semi-natural grasslands currently managed with land cover data and historical land use maps to distinguish areas that could potentially be cleared and/or restored to semi-natural grasslands. The preparation of this dataset also accounted for economic feasibility by only including areas that could be brought into management by creating larger enclosures together with existing semi-natural grasslands. From this dataset the total area of potential semi-natural grassland was extracted for each region used in our modelling (see section 2.3) using areal overlay. We used all enclosure sizes in the dataset (from 15 to 60 ha), which results in the largest area of potential pastures.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Scenario development\u003c/h2\u003e \u003cp\u003eScenarios for alternative futures can support policy development and action by addressing long-term challenges characterised by high levels of uncertainty and complexity (Faur\u0026eacute; et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). In this study, strategic scenarios (B\u0026ouml;rjeson et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2006\u003c/span\u003e) were developed by identifying a set of interventions in animal production systems with the potential to increase the area of grazed semi-natural grassland (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFollowing the finalisation of the scenarios and the production of preliminary modelling results, a workshop was held with researchers and stakeholders involved in the project. The stakeholders included representatives from a farmers\u0026rsquo; organisation promoting pasture-based meat production from semi-natural grasslands, environmental organisations, and the retail sector. The workshop was structured in three parts: first, an overview of the scenarios and preliminary results was presented; second, participants discussed the technical design of the scenarios and the modelling approach; and finally, a broader discussion was held on the barriers and opportunities associated with each scenario, as well as the policy measures and incentives that would be required for their implementation.\u003c/p\u003e \u003cp\u003eParticipant inputs were collected by designated note-takers and via post-it notes. Based on this feedback, the scenarios and modelling assumptions were subsequently revised. The broader discussions during the workshop also informed the reflections presented in section 3.4 of this paper.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the final set of scenarios, and Section 2.4 elaborates on each scenario, outlining the underlying rationale and principal assumptions.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOverview of scenarios with their acronyms and a brief description\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eScenario\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDescription\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMaxCur\u003c/b\u003e: Maximise use of semi-natural grasslands\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCurrently, grazing in semi-natural grasslands is limited by access to semi-natural grasslands on a reasonable distance from the farms and lack of economic incentives (Larsson et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Here we assume that such barriers are removed and that cattle, sheep and horses in current production systems graze semi-natural grasslands at their maximum potential.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSteers\u003c/b\u003e: All male cattle raised as steers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCurrently most male cattle are raised as intact bulls with limited grazing (G\u0026aring;rd \u0026amp; Djurh\u0026auml;lsan 2023). Here we assume that all male cattle for slaughter are castrated and raised as low-intensive steers with a large share of their feeds coming from grazing semi-natural grasslands.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCulCows\u003c/b\u003e: Culled cows are kept for an additional grazing season\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIf not culled earlier because of injury or acute desease, dairy cows are culled and slaughtered after their completed final lactation period. In this scenario, dairy cows that are planned to be culled and slaughtered before the grazing season (March-May) are instead taken out of milk production and kept for another grazing season, grazing semi-natural grasslands during that time.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDryCows\u003c/b\u003e: Prolonged dry period for dairy cows\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eThe dry period for dairy cows is usually two months, with potential to graze semi-natural pastures until one month before calving. Here, the dry period is extended to four months for all dairy cows, which allows for up to three months of potential grazing in semi-natural pastures.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eWinLamb\u003c/b\u003e: All lambs raised as winter lambs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCurrently around 20% of Swedish lambs are raised as winter lambs, i.e. lambs are born in the spring and graze during summer, whereafter they are kept indoors until slaughter in late winter/early spring. Winter lamb production is the production system with the highest potential for grazing semi-natural pastures (Ahlgren et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In this scenario, all lambs born are assumed to be raised as winter lambs.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRecHorses\u003c/b\u003e: Maximise semi-natural grassland grazing by horses\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eToday, horses get a limited share of their grazing intake from semi-natural pastures (Cederberg \u0026amp; Henriksson \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In this scenario we assume that horses for breeding and all ponies and cold-blooded horses are able to graze semi-natural grasslands to a large extent.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAll\u003c/b\u003e: All interventions combined\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eThe combined effect of all above-described interventions.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAll +\u0026thinsp;NatHorses\u003c/b\u003e: Nature conservation horses alongside the other interventions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIn this scenario, horses are introduced alongside the other interventions solely for the purpose of maintaining semi-natural grasslands. We assume that these horses will only be introduced in the southern parts of Sweden and are fed grazed grass for ten months of the year and fed harvested forage two month per year.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Model description\u003c/h2\u003e \u003cp\u003eThe different scenarios were modelled using the CIBUSmod agri-food systems model (version 25.04; Karlsson et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). CIBUSmod is a biophysical mass-flow model of the Swedish agri-food system that balances demand for agricultural products with domestic production, allocating crop areas and livestock across 106 regions while minimising deviations from current land use and livestock numbers. The regions used are the smallest geographical unit for agricultural statistics in Sweden and vary in size from 100 ha to 68,000 ha of agricultural land with a median of 26,000 ha. The model calculates e.g. land use, nutrient flows and greenhouse gas emissions and includes constraints on e.g. crop rotations, regional forage supply and maximum land use to ensure agronomic feasibility. The model also includes a baseline dataset for Sweden (2016\u0026ndash;2020), which was constructed based on national statistics on consumption, trade, land use, animal numbers and agricultural production along with industry statistics and published literature (see Karlsson et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this study, the model input was adjusted by removing the demand for milk, beef, and lamb, allowing the number of grazing animals to vary across scenarios, while demand for all other crop and livestock products was maintained at baseline levels. The total grazing livestock population was capped by a limit on methane emissions from enteric fermentation. Three different caps were implemented that reflect different priorities for biodiversity conservation in semi-natural grasslands versus climate impact mitigation. These were set at 30% reduction compared to baseline levels in line with the Global Methane Pledge (European Commission \u0026amp; United States of America 2021), no change from the baseline and a 10% increase from baseline levels. For non-grazing livestock (i.e. pigs and poultry), their populations were held constant at baseline levels across regions. To prevent drastic changes in the types of production across scenarios, the relative shares of beef and lamb production remained fixed, as did the ratio between beef and milk production. An exception to this was made for the \u0026lsquo;WinLamb\u0026rsquo; scenario (see section 2.4).\u003c/p\u003e \u003cp\u003eIn all scenarios, total cropland use in each region was constrained to not exceed baseline levels to avoid potential competition with other land uses. The total area of semi-natural grassland (excl. wooded pastures) was constrained to the potential pasture area in each region (see section 2.2). However, the relative areas of the different types of semi-natural pastures were held constant, which also constrains the area of wooded pastures. The area of semi-natural meadows (i.e. mown semi-natural grasslands) was kept unchanged in the scenarios.\u003c/p\u003e \u003cp\u003eTo maximize the area of semi-natural grasslands in each scenario, the model was initially set to maximise the total biomass grazed from semi-natural grasslands. Grazed biomass rather than area was used in the optimisation goal to avoid that the optimisation algorithm prioritised grazing in the least productive regions. In a second round of optimisation, the maximized area of grazed semi-natural grassland in each region was enforced as a constraint, while minimising deviations from current crop areas and livestock numbers, to avoid unnecessary deviations from the current state.\u003c/p\u003e \u003cp\u003eModelling these scenarios within an agri-food systems framework ensures that they are feasible in terms of meeting feed requirements for livestock, including winter and concentrate feeds, while still maintaining food production and meeting the demand for monogastric livestock feed without increasing cropland use.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Detailed description of scenarios\u003c/h2\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.4.1 MaxCur: Maximise use of semi-natural grasslands\u003c/h2\u003e \u003cp\u003eCurrently in Sweden, grazing in semi-natural grasslands is often limited by a lack of economic incentives (Larsson et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The remaining semi-natural grasslands are often small and scattered across the landscape, leading to high costs for transport, fencing, water supply, and animal monitoring (Kumm \u0026amp; Hessle \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Holmstr\u0026ouml;m et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), and restoration of semi-natural grasslands is expensive and associated with uncertainties regarding support payments. As these additional expenses are not fully offset by agri-environmental payments or price premiums on the market, profitability is typically low or absent for production based on semi-natural grasslands, and lower compared to other forms of livestock production (Holmstr\u0026ouml;m et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this scenario, we assume that these barriers have been removed and that strong economic incentives are in place, allowing cattle, sheep, and horses to maximize the use of semi-natural grasslands. This assumption also applies to all other scenarios described below, which means that this scenario forms the foundation for the other scenarios, which further increase the potential through different interventions in the production systems.\u003c/p\u003e \u003cp\u003eIn the model, the demand for grazing can be met either through grazing on arable land or in semi-natural grasslands. However, due to technical and biological limitations, such as the low energy density in the grass biomass of semi-natural grasslands or practical constraints on their use e.g. during lactation for dairy cows, CIBUSmod imposes animal category-specific limits on the maximum proportion of grazed biomass that can come from semi-natural grasslands. These limits were based on data from existing literature on typical feed rations for different animal categories (see Table S3). To account for year-to-year variability in pasture yields and avoid overestimating the potential for grazing in semi-natural grasslands, the literature values were reduced by 30%. This also applies to the maximum share of grazing from semi-natural grasslands in the different scenarios, which implies that the maximum share could not exceed 70% for any animal category, in any scenario.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.4.2 Steers: All male cattle raised as steers\u003c/h2\u003e \u003cp\u003eMale calves can be reared either as intact bulls or castrated to become steers, which makes them calmer and more suitable for grazing. According to the Swedish Animal Welfare Ordinance (Djurskyddsf\u0026ouml;rordning \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, 2 kap. 4 \u0026sect;), cattle older than six months must have access to pasture during the summer, except for bulls, which are exempt from this requirement. Currently, approximately 80% of male calves of dairy breed and 90% of calves of beef breed are raised as intact bulls (G\u0026aring;rd \u0026amp; Djurh\u0026auml;lsan 2023) kept indoors post-weaning, since raising bulls is often more profitable than keeping steers (Holmstr\u0026ouml;m et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Castrating all male calves (from dairy systems and suckler herds) and rearing them as steers can therefore increase the number of grazing livestock available for maintaining semi-natural grasslands, without increasing the total cattle population. Steers from suckler cow systems are typically slaughtered at either 24 months in the spring or at 30 months during the autumn. The latter option allows them to graze for an additional season. In this scenario, all male calves are assumed to be castrated and raised as steers, with the slaughter age for all steers set at 30 months.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.4.3 CulCows: Culled cows are kept for an additional grazing season\u003c/h2\u003e \u003cp\u003eThe most common culling reasons in Swedish dairy cows, after udder health problems, are reduced fertility and low milk yield (V\u0026auml;xa Sverige 2024). Cows are typically culled at the end of the lactation if the reason is not related to injury or acute disease. Currently, cows are generally slaughtered when they are culled. In this scenario, cows culled just before the grazing season (March\u0026ndash;May) are instead taken out of milk production (\u0026ldquo;dried off\u0026rdquo;) and kept on pasture for a \u0026ldquo;retirement period\u0026rdquo; and slaughtered after the grazing season, to increase grazing in semi-natural grasslands.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.4.4 DryCows: Prolonged dry period for dairy cows\u003c/h2\u003e \u003cp\u003eDairy cows have high nutritional requirements during the lactation period. Grazing semi-natural pastures only during this period is therefore not a viable option for maintaining current productivity levels (Hessle \u0026amp; Danielsson \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). However, before calving, dairy cows are usually dried off for about two months. During this period, it is possible to have them graze semi-natural pastures up until one month before calving when nutritional requirements are lower and before the cow needs to adapt to a more energy dense diet that supports lactation (EAAP \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). To increase the grazing of semi-natural pastures by these animals while maintaining the cows\u0026rsquo; efficiency during the lactation period, the dry period may be extended by lengthening the calving interval. In this scenario, the calving interval was extended from around 12 months to 14 months, resulting in a dry period of four months. This allows for up to three months of grazing in semi-natural pastures instead of one month for the cows whose dry period coincides with the grazing season.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e2.4.5 WinLamb: All lambs raised as winter lambs\u003c/h2\u003e \u003cp\u003eCurrently, around half of the lambs raised in Sweden are slaughtered in the autumn, while the remainder is split between slaughter in the spring and winter (Ahlgren et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; LRF \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). When lambs are slaughtered in the autumn (called autumn lamb production), both ewes and lamb graze, but grazing only semi-natural grasslands is not possible in this production system due to their high energy requirements. When lambs are slaughtered in the spring (i.e. spring lamb production), the dry ewes graze semi-natural grasslands, but the lambs do not graze as they are born in the winter and slaughtered before time of turn-out to pasture. When spring-born lambs are slaughtered in the winter (i.e. winter lamb production), however, both ewes and lambs can graze semi-natural grasslands the entire grazing season due to the lower expected growth rate of lambs in this system, resulting in a large possible intake from semi-natural grasslands (Ahlgren et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In this scenario we therefore assume that all lambs are raised as winter lambs, to maximise the amount of semi-natural grasslands maintained by these animals. In this scenario, lamb production was also allowed to increase in relation to beef production to align with the corresponding ratio in Swedish meat consumption (currently import shares are higher for lamb compared to beef).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e2.4.6 RecHorses: Maximise semi-natural grassland grazing by horses\u003c/h2\u003e \u003cp\u003eIn the model, the maximum share of grazing from semi-natural grasslands was set at 10% for all horses (Table S3) due to practical difficulties in keeping riding horses on semi-natural grasslands. However, most horses can meet their nutritional needs entirely from semi-natural grasslands (Jamieson \u0026amp; Hessle \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Moreover, keeping young horses on semi-natural grasslands with rough terrain has additional benefits, such as improving their physical strength (Sassner et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This presents an opportunity to increase the extent of semi-natural grasslands managed by these animals. In this scenario we assume that horses for breeding and all ponies and cold-blooded horses can obtain up to 70% of their grazing intake from semi-natural grasslands. The total intake from grazing remains unchanged to reflect current horse-keeping practices.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003e2.4.7 All: All interventions combined\u003c/h2\u003e \u003cp\u003eThis scenario represents the combined effect of all above-described scenarios.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e2.4.8 All +\u0026thinsp;NatHorses: Nature conservation horses alongside the other interventions\u003c/h2\u003e \u003cp\u003ePrevious studies have suggested that year-round grazing by horses can have positive effects on biodiversity in semi-natural grasslands (K\u0026ouml;hler et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Ringmark et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Moreover, horses do not produce as much methane as ruminants (Elghandour et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), which means that horses can contribute to biodiversity conservation through grazing at a lower climate cost. A field trial in east-central Sweden (Ringmark et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) found that crude protein concentrations in pastures were adequate to support year-round grazing, but that energy concentration and herbage availability were potentially limiting factors during the winter months. In this scenario additional horses kept exclusively for the purpose of managing semi-natural grasslands are introduced in southern Sweden (up to 60\u0026deg; north) alongside the other interventions described above. These horses are assumed to be fed only grazed grass for ten months of the year with up to 70% of the dry matter intake from semi-natural grasslands. The remaining two months, horses are assumed to be fed harvested forage.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"3 Results and discussion","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Semi-natural grassland areas and yields\u003c/h2\u003e \u003cp\u003eIn the baseline, we estimate the area of grazed or mowed open semi-natural grassland at 0.51 Mha plus an additional 0.01 Mha grazed wooded pastures, after correcting for areas that are managed but not represented in the Swedish LPIS (Glimsk\u0026auml;r et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The largest areas of semi-natural grasslands are located in the south-eastern parts of Sweden (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea).\u003c/p\u003e \u003cp\u003eThe total area of potential open semi-natural grasslands in the dataset based on historic land use maps (Swedish EPA \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) was around 1.2 Mha (including currently managed pastures), which is an increase of 0.70 Mha from the baseline. While there is a clear geographic concentration of currently managed semi-natural grasslands, the potential additional areas are distributed across Sweden, including northern Sweden, which currently have relatively small areas of semi-natural grasslands (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe gross biomass yield in semi-natural pastures was estimated at 2.4 tonnes dry matter per hectare (t DM/ha) in the baseline as a national average across the three types of semi-natural pastures included in the model (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). For comparison, the gross yield of pastures on arable land were estimated at 4.3 t DM/ha.\u003c/p\u003e \u003cp\u003eAccounting for pasture utilisation rates, net pasture yields were estimated at 1.1 t DM/ha across the three types of semi-natural pastures, as a national average. The net yields estimated here are higher than the 0.9 t DM/ha estimated by Hessle and Danielsson (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) as a Swedish average. It should however be noted that Hessle and Danielsson (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) only considered grasslands classified as valuable grassland habitats under the EU Habitats Directive, which is a stricter definition of semi-natural grasslands compared to the one used here. This likely explain the discrepancies in estimates.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Potential for grazing in semi-natural grasslands\u003c/h2\u003e \u003cp\u003eResults show that, under a combination of all intervention scenarios (excluding introduction of nature conservation horses; i.e. the \u0026lsquo;All\u0026rsquo; scenario) and allowing a 10% increase in enteric methane emissions, it would be possible to manage 1.1 Mha of open semi-natural grassland, more than double the baseline area (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Even with strong reductions in enteric methane emissions (˗30%), which would limit livestock numbers, an area of up to 0.70 Mha could still be managed under the \u0026lsquo;All\u0026rsquo; scenario. This represents an increase of 0.19 Mha compared to baseline areas.\u003c/p\u003e \u003cp\u003eThese results depend strongly on the assumed biomass productivity in semi-natural grasslands, which is uncertain due to the limited data available. Sensitivity analysis showed that assuming a 25% lower biomass productivity across all types of semi-natural grasslands resulted in an 7\u0026ndash;33% increase in the maximum area possible to graze under the \u0026lsquo;All\u0026rsquo; scenario (see S3 in Table S4). The effect was strongest under the lowest methane emissions cap (˗30%) showing that other factors constrain the area increase under the caps that allow more animals. Uncertainty in estimates of grassland area and biomass productivity is a major source of error in modelling grass-fed livestock systems (see e.g. Pfeifer et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). To better estimate both the livestock numbers needed to manage biodiverse semi-natural grasslands, and the food production potential of grasslands, further research and data collection are required to improve area and yield estimates.\u003c/p\u003e \u003cp\u003eIn addition to the open semi-natural grasslands, the area of wooded pastures increased from 0.01 Mha in the baseline to up to 0.04 Mha under the \u0026lsquo;All\u0026rsquo; scenario with 10% increase in methane emissions. In the scenarios, the proportion of wooded pastures to open semi-natural grasslands were assumed to remain constant across regions. Sensitivity analysis showed that assuming a 10-fold increase in the proportion of wooded pastures resulted in a 16\u0026ndash;18% increase in the maximum total area of semi-natural grasslands (incl. wooded pastures) under the \u0026lsquo;All\u0026rsquo; scenario (see S4 in Table S4) due to the lower biomass productivity in wooded pastures. This would however also imply a smaller area of open semi-natural grasslands as compared to the original scenario (11\u0026ndash;14% reduction).\u003c/p\u003e \u003cp\u003eIntroducing grazing horses for nature conservation along with the other interventions (\u0026lsquo;All +\u0026thinsp;NatHorses\u0026rsquo;), showed a large potential to increase grazed areas under the lower methane emission caps. It is however important to note that this would require many additional horses kept for nature conservation (Figure S3). With a 30% reduction in enteric methane emissions, an additional 190,000 horses would be required to reach the maximum potential of 0.84 Mha open semi-natural grasslands, which can be compared to the 360,000 horses currently kept in Sweden.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eNone of the scenarios led to a full utilisation of all potential pasture areas (Figure S2). This was mainly explained by large areas of potential pastures in the northern and alpine regions of Sweden where the constraint on cropland expansion did not allow enough production of winterfeed for the number of animals required to graze all potential areas. Many of these areas have likely been mown for winter feed rather than grazed, a management practice that persisted longer in northern Sweden than in the south (Lennartsson \u0026amp; Westin 2019). Mowing as a management option was not considered here due to the economic and practical constraints on large-scale re-introduction of traditional mowing practices. On a national scale, sensitivity analysis showed that cropland area was not a constraining factor for the maximum area of semi-natural grasslands in the \u0026lsquo;All\u0026rsquo; scenario (see S6 in Table S4). Likewise, the potential area of semi-natural grasslands was shown not to be a constraining factor for the maximum area possible to manage under the \u0026lsquo;All\u0026rsquo; scenario on a national level (see S5 in Table S4), while for several individual regions this was a constraint (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAmong individual interventions, \u0026lsquo;Steers\u0026rsquo; had the greatest potential to increase semi-natural grassland grazing, followed by \u0026lsquo;RecHorses\u0026rsquo; and \u0026lsquo;WinLamb\u0026rsquo;, with some variations in the ordering of scenarios depending on the level of methane emissions allowed (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Interventions focused on dairy cows (\u0026lsquo;DryCows\u0026rsquo; and \u0026lsquo;CulCows\u0026rsquo;) had relatively smaller effects. However, all interventions contributed to the overall effect when combined.\u003c/p\u003e \u003cp\u003eFew studies have previously evaluated the potential for expanding grazing on Swedish semi-natural grasslands. Hessle et al. (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) estimated the area of semi-natural grasslands that could be grazed by rearing all male cattle as steers, similar to the \u0026lsquo;Steers\u0026rsquo; scenario in this study. They found that an extensively raised steer could graze between 1.0 and 1.6 ha of semi-natural grassland over its lifetime. This aligns with results from our \u0026lsquo;Steers\u0026rsquo; scenarios, where the national average was approximately 1.2 ha per steer, with some variations between scenarios due to differences in the spatial distribution of increased grassland areas. As a total potential, Hessle et al. (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) estimated that all male cattle currently not grazing, or only grazing during the suckling period, could graze 0.21\u0026ndash;0.29 Mha of semi-natural grassland if castrated and raised extensively as steers, which is similar to our \u0026lsquo;Steers\u0026rsquo; scenario (\u0026plusmn;\u0026thinsp;0% CH₄ cap), where the area of semi-natural grassland increased by 0.31 Mha compared to the baseline. However, it is important to note that this figure also reflects increased use of semi-natural grasslands by other animal categories (see section 2.4.1).\u003c/p\u003e \u003cp\u003eThe potential for increased grazing in semi-natural grasslands across the scenarios was a combined effect of changes in total demand for grazed biomass and the proportion of biomass sourced from these grasslands (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The relative importance of these two factors differed across scenarios with e.g. the \u0026lsquo;RecHorses\u0026rsquo; scenario having no effect on grazed biomass but increased the share from semi-natural grasslands and the \u0026lsquo;Steers\u0026rsquo; scenario having a strong effect on total grazed biomass but a smaller effect on the share from semi-natural grasslands compared to the \u0026lsquo;MaxCur\u0026rsquo; scenario (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSensitivity analysis showed that the limits set on the maximum share of grazed biomass that could come from semi-natural grasslands had a decisive influence on the results. When the originally applied correction factor of -30% (see section 2.4.1) was removed the maximum area of semi-natural grasslands increased by 9\u0026ndash;42% under the \u0026lsquo;All\u0026rsquo; scenario depending on methane emissions cap (see S2 in Table S4). The effect was strongest for the lowest cap (-30% CH\u003csub\u003e4\u003c/sub\u003e) where the maximum area increased from 0.73 to 1.0 Mha.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe regional distribution of changes in animal numbers in the \u0026lsquo;All\u0026rsquo; scenario showed a tendency of reduced livestock numbers in south-western Sweden and increased numbers in northern Sweden (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea), which indicates that current distribution of livestock numbers does not align with potential grassland areas. Sensitivity analysis in which the regional distribution of animal numbers was constrained to the baseline distribution however showed only a small effect (-6%) on the maximum area of grazed semi-natural grasslands in the \u0026lsquo;MaxCur\u0026rsquo; scenario (see S1 in Table S4).\u003c/p\u003e \u003cp\u003eUnder the strictest methane emission cap (-30%), semi-natural grassland areas increased mainly in the northern and central parts of Sweden and on the islands of \u0026Ouml;land and Gotland of the east coast (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). When larger methane emissions were allowed, semi-natural grassland areas increased throughout most regions of Sweden. However, it is important to note here that the regional distribution of areas follows from the optimisation goal function employed (i.e. maximising the grazed biomass from semi-natural grasslands) which does not necessarily reflect biodiversity conservation priorities. Sensitivity analysis showed that changing the optimisation goal function to one that maximised the total area of semi-natural grasslands led to a 1\u0026ndash;3% increase in the total area under the \u0026lsquo;All\u0026rsquo; scenario (see S7 in Table S4). Areas increased in northern Sweden, while they decreased slightly in some regions of southern Sweden compared to the original scenario due to the lower estimated productivity in semi-natural grasslands in northern Sweden. This shows that using a different approach for prioritising which areas to graze under the set constraints results in a different regional distribution and influence the maximum area of semi-natural grassland due to regional differences in pasture productivity. However, the sensitivity analysis showed that a different prioritisation of where to expand areas had a limited impact on results in terms of the total area of semi-natural grassland possible to graze under set constraints.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhile a broad range of scenarios were developed in this study, they do not represent an exhaustive assessment of production-side interventions that could potentially increase the area of grazed semi-natural grasslands under maintained or decreased climate impact from the livestock sector. We chose here to focus on concrete adjustments to current production systems, rather than transformative changes. Especially for the dairy sector, which has developed fast towards increased intensity and productivity (Karlsson et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), more transformative changes could be envisioned to allow increased grazing for dairy cows during lactation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Animal source food production and climate impacts\u003c/h2\u003e \u003cp\u003eIn all scenarios except \u0026lsquo;MaxCur\u0026rsquo; and \u0026lsquo;RecHorses\u0026rsquo;, meat and milk production decreased when methane emissions were capped at baseline levels. These reductions ranged from 1% to 18% in terms of edible protein, with the greatest decreases observed when all interventions were combined (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). The reason that production decreases under constant methane emissions is that most scenarios involve reduced productivity with higher methane emissions from enteric fermentation per unit food produced. For example, raising male cattle as steers implies slower growth rates and thus more energy spent on maintenance as compared to muscle growth. Similarly for the two scenarios affecting dairy cows, both result in longer unproductive (dry) periods during a dairy cow\u0026rsquo;s lifetime, again resulting in more energy spent on maintenance as compared to milk production. Allowing a 10% increase in methane emissions enabled increased production in all scenarios except in the \u0026lsquo;All\u0026rsquo; scenario. Under a 30% reduction in methane emissions, edible protein production from milk and ruminant meat declined by 32% to 44% depending on scenario.\u003c/p\u003e \u003cp\u003eIt is however also important to note that all scenarios led to a reduced demand for cropland for grazing and winter feed, especially under the strictest methane emissions cap where the number of animals was reduced (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec). This was a combined effect of prioritising semi-natural grasslands over pastures on arable land in grazing, increased share of grazing in feed rations for some scenarios, and, under the lower methane emissions cap, reduced animal numbers. This opens opportunities for increasing other forms of food production. For example, if using the spared cropland to cultivate cereals, accounting for the regional average cereal yields, it would be possible to increase cereal production in the \u0026lsquo;All\u0026rsquo; scenario by 2.2, 0.8 and 0.4\u0026nbsp;million tonnes under the \u0026minus;\u0026thinsp;30%, \u0026plusmn;\u0026thinsp;0% and +\u0026thinsp;10% methane emissions caps, respectively. In the baseline, total cereal production used for food, feed, bioenergy, and exports was around 4.9\u0026nbsp;million tonnes in the model. It should however be noted that cereal production is not feasible on all arable land currently used for leys.\u003c/p\u003e \u003cp\u003eEnteric methane emissions per hectare of grazed semi-natural grassland ranged from 148 kg CH₄/ha (\u0026lsquo;MaxCur\u0026rsquo;) to 107 kg CH₄/ha (\u0026lsquo;All\u0026rsquo;). This can be compared to results from Hessle and Danielsson (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) who estimated the number of cattle required to achieve favourable conservation status for Sweden\u0026rsquo;s semi-natural grasslands, along with the associated increase in enteric methane emissions. They found that grazing an additional 2.2 Mha of semi-natural grassland would result in emissions of 179,000 tonnes of CH₄ from enteric fermentation if the existing proportions of beef and dairy cows are maintained and all male cattle are raised as steers. This corresponds to 81 kg CH₄ per hectare of grazed semi-natural grassland, which is lower than in our scenarios. This discrepancy is mainly due to our conservative assumptions regarding the maximum proportion of semi-natural grassland allowed in the grazing regime for different animal categories (i.e. a 30% reduction of literature values; see section 2.4.1). In sensitivity analyses where this adjustment was removed (see S2 in Table S4), enteric methane emissions per hectare of grazed semi-natural grassland ranged from 79 to 101 kg CH₄/ha, depending on the methane emissions cap level, which is in line with the figure derived from Hessle and Danielsson (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA comparison of total climate impacts per unit of edible protein versus per unit area of grazed semi-natural grassland revealed a clear negative correlation across scenarios (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb). That is, scenarios with lower greenhouse gas emissions per area of grazed grassland tended to have higher emissions per unit of edible protein produced, and vice versa. This highlights a risk that strong incentives for climate impact mitigation, especially if measured by efficiency (i.e. impact per unit product), may inadvertently steer towards livestock production systems that are less conducive to biodiversity conservation at low climate cost. It is thus crucially important to account for other services than food production provided by livestock systems when moving forward with stronger force on the climate agenda. This is however often not the case today. For example, the largest dairy company in Sweden has introduced an incentive tool (FarmAhead\u0026trade;) with additional payments to farmers for \u0026ldquo;climate and environmental sustainability activities\u0026rdquo; (Arla \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). While the scheme includes some indicators on \u0026ldquo;biodiversity,\u0026rdquo; it is skewed towards indicators for (climate) efficiency in production, which may invertedly disincentivise grazing-based production systems.\u003c/p\u003e \u003cp\u003eClimate impacts, both per unit of protein and per area of grazed grassland, were generally higher under the 30% methane reduction cap. This is largely due to reductions in cattle and sheep production, which increased the proportion of horses. As horses were assumed to produce no edible protein while contributing relatively little to grazing semi-natural grasslands, their increased share negatively affected both climate efficiency metrics. The exception is the scenario where nature conservation horses are introduced (\u0026lsquo;All +\u0026thinsp;NatHorses\u0026rsquo;) where the 30% methene reduction cap resulted in the lowest climate impact per area of grazed semi-natural grassland (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb) but also the lowest production of meat and milk (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe results presented here show that it is possible to increase the area of grazed semi-natural grasslands under maintained or reduced livestock numbers, corroborating previous assessments showing that animal numbers are not the limiting factor for maintenance of semi-natural grasslands (Larsson et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). While the scenarios generally led to reduced animal-source food production under a given cap on methane emissions, they also opened opportunities to increase other forms of food production by reducing cropland demand for grazing and winter feed. As such these scenarios also show how a future with reduced animal-source food demand can be compatible with maintained or increased areas of semi-natural grasslands, in line with previous studies (R\u0026ouml;\u0026ouml;s et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Karlsson \u0026amp; R\u0026ouml;\u0026ouml;s \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Barriers, opportunities and policy implications\u003c/h2\u003e \u003cp\u003eThe feasibility of expanding grazing in semi-natural grasslands, as assessed through the modelled scenarios, hinges not only on biophysical potential but also on economic viability, logistical practicality, market acceptance, and policy alignment. In this section we discuss the broader barriers, opportunities and policy implications for the scenarios, using insights from the workshop (see section 2.2) that was conducted within the project as a base.\u003c/p\u003e \u003cp\u003eThe stakeholder workshop underscored the complexity of these interacting factors and provided insight into the real-world challenges and opportunities associated with achieving these scenarios in practice.\u003c/p\u003e \u003cp\u003eEconomic conditions were repeatedly identified as the most important constraint across all scenarios. Stakeholders emphasised the increasing costs of inputs, veterinary services, and logistics, compounded by stagnant or insufficient agri-environmental payments. As one participant summarised, \u003cem\u003e\"compensation for semi-natural pastures has not developed in pace with production costs.\"\u003c/em\u003e Currently Sweden provides agri-environmental support under the CAP for the management of semi-natural pastures and meadows to promote biodiversity and preserve cultural landscapes (Swedish Board of Agriculture \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) and there is also nationally funded support for restoring such lands (Swedish EPA \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e). These payments have been indispensable in maintaining semi-natural grasslands in Sweden during the process of intensification and structural rationalisation of agriculture during the last 50 years. However, the consensus during the workshop was that without substantially higher and more stable compensation for management of semi-natural grasslands, even favourable market prices would be insufficient to drive major changes in livestock systems. This echoes findings on the EU level where the design of CAP funding schemes, while instrumental in driving land use, has often been considered poorly adapted to the preservation of high nature value farming systems including semi-natural grasslands (Varela et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). One specific option mentioned during the workshop was to direct current investment support, i.e. support for new buildings and other infrastructure, which have been found disproportionally directed towards larger farms (Nilsson \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), towards smaller farms and infrastructure that facilitates pasture-based systems, which could also increase the effectiveness of investment support (Nilsson \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Targeted support for keeping male cattle as steers rather than bulls was also discussed as a viable option for directing support payments towards grazing livestock. However, participants raised veterinary shortages as a barrier to steer production, citing dramatically increased costs for castration services due to low availability. One participant also highlighted risks of policies that reduce the profitability of rearing bulls, since the revenues from this activity help sustain suckler cow enterprises, which are important for grazing semi-natural grasslands today.\u003c/p\u003e \u003cp\u003eWhile effective support schemes are important, most revenues on beef and lamb enterprises come from selling meat and other products (Jamieson \u0026amp; Hessle \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This highlights the importance of market incentives in encouraging producers to increase grazing in semi-natural grasslands, a point that was also discussed during the workshop. There is a third-party certification scheme for semi-natural pasture-based meat, which is sold in one of the major retail chains in Sweden. Participants highlighted that there is high demand from consumers and retail for this certified meat, but that information towards farmers and higher prices are needed to scale production to meet demand. Participants also highlighted the need for similar schemes in the dairy sector which is currently developing towards increased size and productivity with reduced opportunities for grazing semi-natural grasslands (Karlsson et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA recurring critique across several scenarios, especially for \u0026lsquo;Steers\u0026rsquo;, \u0026lsquo;WinLamb\u0026rsquo;, and \u0026lsquo;CulCows\u0026rsquo;, was the lack of alignment with current market dynamics. For example, large carcasses from 30 months beef steers may not match processor or consumer demand, and that longer rearing periods increase costs and reduce capital turnover. Concerns were also raised about shifting a large share of meat supply to the autumn, limiting the availability of fresh meat during other parts of the year, which may result in a growing market share for imported meat if Swedish production fails to meet demand year-round.\u003c/p\u003e \u003cp\u003eWorkshop participants viewed \u0026lsquo;RecHorses\u0026rsquo; as a scenario with limited potential in real life due to limited capacity among horse owners to supervise animals remotely, fears of injury, and practical concerns like daily riding schedules. Still, there should be some potential among breeders and professional horse operations, particularly if attractive economic incentives were in place.\u003c/p\u003e \u003cp\u003eThe \u0026lsquo;CulCows\u0026rsquo; scenario was \u003cb\u003er\u003c/b\u003eecognised as a practical model already emerging in some operations, with potential for broader implementation if coupled with welfare-based payments or market premiums. In this scenario \u0026ldquo;retired\u0026rdquo; dairy cows could be sold to a separate enterprise that manages the grazing period. Thereby, the operations and profitability of the dairy farm would not be affected. Although stakeholders recognised the welfare benefits of the \u0026lsquo;DryCows\u0026rsquo; scenario, it was widely regarded as impractical at scale due to transition risks (e.g., from indoor feed rations to pasture) and the limited availability of semi-natural pastures on many dairy farms.\u003c/p\u003e \u003cp\u003eWhile novel and promising, stakeholders warned of some risks associated with the \u0026lsquo;NatHorses\u0026rsquo; scenario, e.g., bark stripping by horses damaging valuable tree species. Participants also highlighted that a viable horse meat market to economically sustain such operations would likely be a prerequisite as full reliance on support payments for biodiversity conservation was regarded unrealistic.\u003c/p\u003e \u003cp\u003eIn summary, participants saw opportunities for rural development, recreation, and landscape maintenance through increased grazing. However, these benefits are unlikely to materialise without stronger market incentives and targeted investments in social infrastructure, including veterinary services, advisory networks, and housing in rural areas. Concerns were also raised about the lack of young farmers and the challenges of maintaining existing or establishing new operations in low-service regions.\u003c/p\u003e \u003c/div\u003e"},{"header":"4 Conclusions","content":"\u003cp\u003eThis is the first study to explore scenarios of increased grazing in Swedish semi-natural grasslands using a spatially disaggregated agri-food systems modelling framework that considers regional land availability constraints, as well as other biophysical and agronomic limitations, while estimating agricultural greenhouse gas emissions.\u003c/p\u003e \u003cp\u003eOur results demonstrate that it is technically feasible to expand the area of grazed semi-natural grasslands in Sweden without increasing methane emissions and overall climate impacts. When all interventions studied here were combined semi-natural grassland area increased by 0.5 Mha under maintained methane emissions, which represents almost a doubling of current areas. This relies on the implementation of production-side interventions supporting grazing and that semi-natural grasslands are prioritised over pastures on arable land. However, achieving this in practice will require long-term supporting policies and improved economic incentives for grassland-based production. Among individual interventions, raising male cattle as steers rather than intact bulls showed the greatest potential for increased grazing in semi-natural grasslands.\u003c/p\u003e \u003cp\u003eThe comparison between climate impact per unit of food produced and per hectare of grazed grassland highlights a fundamental trade-off. Systems optimised for low climate impact per unit food produced are often not best suited to maintain semi-natural grasslands and their associated biodiversity. Climate impact-based indicators and incentive structures tend to promote efficiency in production (van der Werf et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and may unintentionally favour production systems that are less conducive to biodiversity conservation at low climate cost.\u003c/p\u003e \u003cp\u003eTo achieve biodiversity conservation in semi-natural grasslands along with other goals of the agri-food system while reducing climate and other environmental impacts it is important to work towards a shared vision for the role of livestock in future food systems (Karlsson \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Resare Sahlin \u0026amp; Trewern \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This can guide the process of developing policies that support livestock systems that deliver on the multitude of services expected form society.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis work was funded by The Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (FORMAS) [grant number 2021-01434] and the Mistra Food Futures research program [Grant ID: DIA 2018/24].\u003c/p\u003e\n\u003ch2\u003eConflicts of interest\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eEthics approval\u003c/h2\u003e\n\u003cp\u003eThis study did not require ethics approval under Swedish law.\u003c/p\u003e\n\u003ch2\u003eConsent to participate\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eConsent for publication\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eCode and data availability\u003c/h2\u003e\n\u003cp\u003eCIBUSmod along with the baseline dataset for Sweden is available as open-source at https://github.com/SLU-foodsystems/CIBUSmod and is described in detail in Karlsson\u003cem\u003e\u0026nbsp;et al.\u003c/em\u003e (2025). The specific code and input data used to run the model and produce results presented in this paper are available at https://github.com/karlssonjo/seminatgrass.\u003c/p\u003e\n\u003ch2\u003eAuthors' contributions\u003c/h2\u003e\n\u003cp\u003eConceptualization: JOK and ER,\u003cbr\u003e\u0026nbsp;Methodology: JOK, KvG\u003cbr\u003e\u0026nbsp;Data curation: JOK, KvG, AH, ML and AG,\u003cbr\u003e\u0026nbsp;Formal analysis: JOK and KvG,\u003cbr\u003e\u0026nbsp;Writing – original draft: JOK, KvG and ER,\u003cbr\u003e\u0026nbsp;Writing – review \u0026amp; editing: All authors,\u003cbr\u003e\u0026nbsp;Funding acquisition: JOK and ER\u003cbr\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAguilera Nu\u0026ntilde;ez G, Glimsk\u0026auml;r A, Zacchello G, Francksen RM, Whittingham MJ, Hiron M (2024) Agriculturally Improved and Semi-Natural Permanent Grasslands Provide Complementary Ecosystem Services in Swedish Boreal Landscapes. Agronomy 14(3):567. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agronomy14030567\u003c/span\u003e\u003cspan address=\"10.3390/agronomy14030567\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhlgren S, Behaderovic D, Wirsenius S, Carlsson A, Hessle A, Tor\u0026auml;ng P, Seeman A, den, Braver T, Kvarnb\u0026auml;ck O (2022) \u003cem\u003eMilj\u0026ouml;p\u0026aring;verkan av svensk n\u0026ouml;t-och lammk\u0026ouml;ttsproduktion\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.diva-portal.org/smash/record.jsf?pid=diva2%3A1718732\u003c/span\u003e\u003cspan address=\"https://www.diva-portal.org/smash/record.jsf?pid=diva2%3A1718732\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhlgren S, Wirsenius S, Tor\u0026auml;ng P, Carlsson A, Seeman A, Behaderovic D, Kvarnb\u0026auml;ck O, Parvin N, Hessle A (2024) Climate and biodiversity impact of beef and lamb production \u0026ndash; A case study in Sweden. Agric Syst 219:104047. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.agsy.2024.104047\u003c/span\u003e\u003cspan address=\"10.1016/j.agsy.2024.104047\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArla (2025) \u003cem\u003eFarmAhead\u0026trade; - How Arla farmers are rewarded for their sustainability activities\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.arla.com/sustainability/the-farms/arlas-sustainability-incentive-model-qa/\u003c/span\u003e\u003cspan address=\"https://www.arla.com/sustainability/the-farms/arlas-sustainability-incentive-model-qa/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-09-19]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuffret AG, Kimberley A, Plue J, Wald\u0026eacute;n E (2018) Super-regional land-use change and effects on the grassland specialist flora. Nat Commun 9(1):3464. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-018-05991-y\u003c/span\u003e\u003cspan address=\"10.1038/s41467-018-05991-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBengtsson J, Bullock JM, Egoh B, Everson C, Everson T, O'Connor T, O'Farrell PJ, Smith HG, Lindborg R (2019) Grasslands\u0026mdash;more important for ecosystem services than you might think. Ecosphere 10(2):e02582. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ecs2.2582\u003c/span\u003e\u003cspan address=\"10.1002/ecs2.2582\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026ouml;rjeson L, H\u0026ouml;jer M, Dreborg K-H, Ekvall T, Finnveden G (2006) Scenario types and techniques: Towards a user's guide. Futures 38(7):723\u0026ndash;739. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.futures.2005.12.002\u003c/span\u003e\u003cspan address=\"10.1016/j.futures.2005.12.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCederberg C, Henriksson M (2020) Gr\u0026auml;smarkernas anv\u0026auml;ndning i Sverige. CHALMERS TEKNISKA H\u0026Ouml;GSKOLA. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://research.chalmers.se/publication/517805\u003c/span\u003e\u003cspan address=\"https://research.chalmers.se/publication/517805\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCousins SAO, Auffret AG, Lindgren J, Tr\u0026auml;nk L (2015) Regional-scale land-cover change during the 20th century and its consequences for biodiversity. Ambio 44(S1):17\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13280-014-0585-9\u003c/span\u003e\u003cspan address=\"10.1007/s13280-014-0585-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDjurskyddsf\u0026ouml;rordning (2019) :66) Svensk f\u0026ouml;rfattningssamling. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.riksdagen.se/sv/dokument-och-lagar/dokument/svensk-forfattningssamling/djurskyddsforordning-201966_sfs-2019-66/#K2\u003c/span\u003e\u003cspan address=\"https://www.riksdagen.se/sv/dokument-och-lagar/dokument/svensk-forfattningssamling/djurskyddsforordning-201966_sfs-2019-66/#K2\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEAAP (2011) \u003cem\u003eNorFor - The Nordic feed evaluation system\u003c/em\u003e. (EAAP publication, 130). Wageningen: Wageningen Academic Publishers. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3920/978-90-8686-718-9\u003c/span\u003e\u003cspan address=\"10.3920/978-90-8686-718-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElghandour MMMY, Adegbeye MJ, Barbabosa-Pilego A, Perez NR, Hern\u0026aacute;ndez SR, Zaragoza-Bastida A, Salem AZM (2019) Equine Contribution in Methane Emission and Its Mitigation Strategies. J Equine Veterinary Sci 72:56\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jevs.2018.10.020\u003c/span\u003e\u003cspan address=\"10.1016/j.jevs.2018.10.020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEriksson C, Ekberg M, Blom S, Rehn I, Sj\u0026ouml;dahl M, S\u0026ouml;derberg T, Dahlstr\u0026ouml;m A, Hallgren K, Bengtsson N (2009) Utveckling av \u0026auml;ngs- och betesmarker \u0026ndash; ig\u0026aring;r, idag och imorgon. (Jordbruksverket Rapport 2009:10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www2.jordbruksverket.se/webdav/files/SJV/trycksaker/Pdf_rapporter/ra09_10.pdf\u003c/span\u003e\u003cspan address=\"https://www2.jordbruksverket.se/webdav/files/SJV/trycksaker/Pdf_rapporter/ra09_10.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEriksson O (2022) Coproduction of Food, Cultural Heritage and Biodiversity by Livestock Grazing in Swedish Semi-natural Grasslands. Front Sustainable Food Syst. 6\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fsufs.2022.801327\u003c/span\u003e\u003cspan address=\"10.3389/fsufs.2022.801327\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEriksson O, Cousins SAO (2014) Historical landscape perspectives on grasslands in Sweden and the Baltic region. Land 3(1):300\u0026ndash;321. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land3010300\u003c/span\u003e\u003cspan address=\"10.3390/land3010300\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEuropean Commission \u0026amp; United States of America (2021) \u003cem\u003eGlobal methane pledge\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003c/span\u003e\u003cspan address=\"http://www.ccacoalition.org/en/resources/global-methane-pledge\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaur\u0026eacute; E, Arushanyan Y, Ekener E, Miliutenko S, Finnveden G (2017) Methods for assessing future scenarios from a sustainability perspective. Eur J Futures Res 5(1):17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s40309-017-0121-9\u003c/span\u003e\u003cspan address=\"10.1007/s40309-017-0121-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlimsk\u0026auml;r A, Hultgren J, Hiron M, Westin R, Bokkers EAM, Keeling LJ (2023) Sustainable Grazing by Cattle and Sheep for Semi-Natural Grasslands in Sweden. \u003cem\u003eAgronomy\u003c/em\u003e, 13(10), 2469. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.mdpi.com/2073-4395/13/10/2469\u003c/span\u003e\u003cspan address=\"https://www.mdpi.com/2073-4395/13/10/2469\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eG\u0026aring;rd \u0026amp; Djurh\u0026auml;lsan (2023) \u003cem\u003eGenomsnittligt kvalitetsutfall f\u0026ouml;r n\u0026ouml;tkreatur slaktade under 2023\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gardochdjurhalsan.se/kvalitetsutfall-slaktade-notkreatur/\u003c/span\u003e\u003cspan address=\"https://www.gardochdjurhalsan.se/kvalitetsutfall-slaktade-notkreatur/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHessle A, Bertilsson J, Stenberg B, Kumm K-I, Sonesson U (2017) Combining environmentally and economically sustainable dairy and beef production in Sweden. Agric Syst 156:105\u0026ndash;114. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.agsy.2017.06.004\u003c/span\u003e\u003cspan address=\"10.1016/j.agsy.2017.06.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHessle A, Danielsson R (2023) \u003cem\u003eNumber of cattle required to obtain good ecological status in Swedish grasslands and livestock\u0026rsquo;s methane emissions\u003c/em\u003e. (Report 60). Swedish University of Agricultural Sciences, Skara, Sweden. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://res.slu.se/id/publ/123395\u003c/span\u003e\u003cspan address=\"https://res.slu.se/id/publ/123395\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHessle A, Danielsson R (2024) Cattle population required for favorable conservation status of management-dependent semi-natural grasslands and forests, and associated increase in enteric methane emissions. J Nat Conserv 78:126571. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jnc.2024.126571\u003c/span\u003e\u003cspan address=\"10.1016/j.jnc.2024.126571\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHessle A, Danielsson R, Lidfors L (2021) \u003cem\u003eHoused young bulls - mapping of scope and potential for nature conservation\u003c/em\u003e. Skara. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://res.slu.se/id/publ/112871\u003c/span\u003e\u003cspan address=\"https://res.slu.se/id/publ/112871\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHessle A, Kumm K-I (2011) Use of beef steers for profitable management of biologically valuable semi-natural pastures in Sweden. J Nat Conserv 19(3):131\u0026ndash;136. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jnc.2010.10.002\u003c/span\u003e\u003cspan address=\"10.1016/j.jnc.2010.10.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHiron M, Lundquist P, Groote Ad, Glimsk\u0026auml;r A (2022) \u003cem\u003eResultat f\u0026ouml;r gr\u0026auml;smarker i Remiil 2015\u0026ndash;2020: markslag, landskapstyp och artrikedom inom och utanf\u0026ouml;r jordbruksblock\u003c/em\u003e. Unpublished report\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolmstr\u0026ouml;m K, Kumm K-I, Andersson H, Hessle A (2024) Labour in suckler cow herds \u0026ndash; a study on enterprises in southern Sweden. Acta Agriculturae Scand Sect \u0026mdash; Anim Sci 73(1\u0026ndash;2):49\u0026ndash;62. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/09064702.2023.2245400\u003c/span\u003e\u003cspan address=\"10.1080/09064702.2023.2245400\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolmstr\u0026ouml;m K, Kumm K-I, Andersson H, Nadeau E, Segerkvist KA, Hessle A (2021) Economic incentives for preserving biodiverse semi-natural pastures with calves from dairy cows. J Nat Conserv 62:126010. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jnc.2021.126010\u003c/span\u003e\u003cspan address=\"10.1016/j.jnc.2021.126010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHooftman DAP, Kimberley A, Cousins SAO, Santamar\u0026iacute;a Bueno S, Honnay O, Krickl P, Plue J, Poschlod P, Traveset A, Bullock JM (2023) Could green infrastructure supplement ecosystem service provision from semi-natural grasslands? J Environ Manage 328:116952. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvman.2022.116952\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvman.2022.116952\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIb\u0026aacute;\u0026ntilde;ez I, Katz DSW, Peltier D, Wolf SM, Connor Barrie BT (2014) Assessing the integrated effects of landscape fragmentation on plants and plant communities: the challenge of multiprocess\u0026ndash;multiresponse dynamics. J Ecol 102(4):882\u0026ndash;895. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1365-2745.12223\u003c/span\u003e\u003cspan address=\"10.1111/1365-2745.12223\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJamieson A, Hessle A (2021) \u003cem\u003eHinder och m\u0026ouml;jligheter f\u0026ouml;r \u0026ouml;kad naturbetesdrift ur ett lantbrukarperspektiv: en kunskaps\u0026ouml;versikt\u003c/em\u003e. (SustAinimal Reports #1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://res.slu.se/id/publ/115251\u003c/span\u003e\u003cspan address=\"https://res.slu.se/id/publ/115251\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarlsson JO (2022) Livestock as resource users and landscape managers-A food systems perspective. Swedish University of Agricultural Sciences, Uppsala. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.13140/RG.2.2.25926.14403\u003c/span\u003e\u003cspan address=\"10.13140/RG.2.2.25926.14403\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarlsson JO, Karlsson-Potter H, Lagnel\u0026ouml;v O, Ericsson N, Einarsson R, Hansson PA (2025) CIBUSmod 25.04: A spatially disaggregated biophysical agri-food systems model for studying national-level demand- and production-side intervention scenarios. \u003cem\u003eEGUsphere\u003c/em\u003e, 2025, 1\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/egusphere-2025-2625\u003c/span\u003e\u003cspan address=\"10.5194/egusphere-2025-2625\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarlsson JO, Robling H, Cederberg C, Sp\u0026ouml;rndly R, Lindberg M, Martiin C, Ardfors E, Tid\u0026aring;ker P (2023) What can we learn from the past? Tracking sustainability indicators for the Swedish dairy sector over 30 years. Agric Syst 212:103779. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.agsy.2023.103779\u003c/span\u003e\u003cspan address=\"10.1016/j.agsy.2023.103779\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarlsson JO, R\u0026ouml;\u0026ouml;s E (2019) Resource-efficient use of land and animals\u0026mdash;Environmental impacts of food systems based on organic cropping and avoided food-feed competition. Land Use Policy 85:63\u0026ndash;72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.landusepol.2019.03.035\u003c/span\u003e\u003cspan address=\"10.1016/j.landusepol.2019.03.035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumm K-I, Hessle A (2020) Economic Comparison between Pasture-Based Beef Production and Afforestation of Abandoned Land in Swedish Forest Districts. Land 9(2):42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land9020042\u003c/span\u003e\u003cspan address=\"10.3390/land9020042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eK\u0026ouml;hler M, Hiller G, Tischew S (2016) Year-round horse grazing supports typical vascular plant species, orchids and rare bird communities in a dry calcareous grassland. Agric Ecosyst Environ 234:48\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.agee.2016.03.020\u003c/span\u003e\u003cspan address=\"10.1016/j.agee.2016.03.020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLarsson C, Boke Ol\u0026eacute;n N, Brady M (2020) \u003cem\u003eNaturbetesmarkens framtid - en fr\u0026aring;ga om l\u0026ouml;nsamhet\u003c/em\u003e. (RAPPORT 2020:1). Lund: AgriFood Economics Centre. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://res.slu.se/id/publ/115477\u003c/span\u003e\u003cspan address=\"https://res.slu.se/id/publ/115477\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLennartsson T, Westin A (2019) \u003cem\u003e\u0026Auml;ngar och sl\u0026aring;tter. Historia, ekologi, natur- och kulturmilj\u0026ouml;v\u0026aring;rd\u003c/em\u003e. Riksantikvarie\u0026auml;mbetet. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://raa.diva-portal.org/smash/record.jsf?pid=diva2%3A1331194\u003c/span\u003e\u003cspan address=\"https://raa.diva-portal.org/smash/record.jsf?pid=diva2%3A1331194\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLennartsson T, Westin A, Iuga A, Jones E, Madry S, Murray S, Gustavsson E (2016) The meadow is the mother of the field. Comparing transformations in hay production in three European agroecosystems. Martor 21:103\u0026ndash;126\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLRF (2025) \u003cem\u003eStatistikplattform k\u0026ouml;tt\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.lrf.se/om-lrf/lrf-s-branschavdelningar/lrf-kott/statistik-och-verktyg/\u003c/span\u003e\u003cspan address=\"https://www.lrf.se/om-lrf/lrf-s-branschavdelningar/lrf-kott/statistik-och-verktyg/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-10-31]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNilsson P (2017) Productivity effects of CAP investment support: Evidence from Sweden using matched panel data. Land Use Policy 66:172\u0026ndash;182. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.landusepol.2017.04.043\u003c/span\u003e\u003cspan address=\"10.1016/j.landusepol.2017.04.043\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePfeifer C, Winterberg R, Leiber F (2025) Quantifying the Contributing Potential of European Grasslands to Food Protein and Organic Manure in a Circular Food System. Grass Forage Sci 80(2):e12713. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gfs.12713\u003c/span\u003e\u003cspan address=\"10.1111/gfs.12713\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eResare Sahlin K, Trewern J (2022) A systematic review of the definitions and interpretations in scientific literature of \u0026lsquo;less but better\u0026rsquo; meat in high-income settings. Nat Food 3(6):454\u0026ndash;460. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s43016-022-00536-5\u003c/span\u003e\u003cspan address=\"10.1038/s43016-022-00536-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRingmark S, Skarin A, Jansson A (2019) Impact of Year-Round Grazing by Horses on Pasture Nutrient Dynamics and the Correlation with Pasture Nutrient Content and Fecal Nutrient Composition. Animals 9(8):500\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR\u0026ouml;\u0026ouml;s E, Patel M, Sp\u0026aring;ngberg J, Carlsson G, Rydhmer L (2016) Limiting livestock production to pasture and by-products in a search for sustainable diets. Food Policy 58:1\u0026ndash;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.foodpol.2015.10.008\u003c/span\u003e\u003cspan address=\"10.1016/j.foodpol.2015.10.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSassner H, Forssman K, Petersson P, Fredricson I (2022) Breeding for durability - Equine monitoring at pasture using halter attached Hoofstep sensors. \u003cem\u003e18th International Equitation Science Conference\u003c/em\u003e, Hartpury, UK. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://res.slu.se/id/publ/118498\u003c/span\u003e\u003cspan address=\"https://res.slu.se/id/publ/118498\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShipley JR, Frei ER, Bergamini A, Boch S, Schulz T, Ginzler C, Barandun M, Bebi P, Bolliger J, Bollmann K, Delpouve N, Gossner MM, Graham C, Krumm F, Marty M, Pichon N, Rigling A, Rixen C (2024) Agricultural practices and biodiversity: Conservation policies for semi-natural grasslands in Europe. Curr Biol 34(16):R753\u0026ndash;R761. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cub.2024.06.062\u003c/span\u003e\u003cspan address=\"10.1016/j.cub.2024.06.062\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSp\u0026ouml;rndly E, Glimsk\u0026auml;r A (2018) \u003cem\u003eBetesdjur och betestryck i naturbetesmarker\u003c/em\u003e. (Rapport 297). Uppsala: Sveriges lantbruksuniversitet. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pub.epsilon.slu.se/15649/\u003c/span\u003e\u003cspan address=\"https://pub.epsilon.slu.se/15649/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwedish Board of Agriculture (2024) \u003cem\u003eSveriges Milj\u0026ouml;m\u0026aring;l - Ett rikt odlingslandskap - Betesmarker och sl\u0026aring;tter\u0026auml;ngar\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.sverigesmiljomal.se/miljomalen/ett-rikt-odlingslandskap/betesmarker-och-slatterangar/\u003c/span\u003e\u003cspan address=\"https://www.sverigesmiljomal.se/miljomalen/ett-rikt-odlingslandskap/betesmarker-och-slatterangar/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwedish Board of Agriculture (2025) \u003cem\u003eMilj\u0026ouml;ers\u0026auml;ttning f\u0026ouml;r sk\u0026ouml;tsel av betesmarker och sl\u0026aring;tter\u0026auml;ngar 2025\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://jordbruksverket.se/stod/jordbruk-tradgard-och-rennaring/jordbruksmark/betesmarker-och-slatterangar/skotsel-av-betesmarker-och-slatterangar\u003c/span\u003e\u003cspan address=\"https://jordbruksverket.se/stod/jordbruk-tradgard-och-rennaring/jordbruksmark/betesmarker-och-slatterangar/skotsel-av-betesmarker-och-slatterangar\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-09-19]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwedish EPA (2019) \u003cem\u003eMer naturbetesmarker och ekonomiskt b\u0026auml;rkraftiga f\u0026ouml;retag - Ett samverkansprojekt inom milj\u0026ouml;m\u0026aring;lsr\u0026aring;dets gemensamma \u0026aring;tg\u0026auml;rdslista 2019\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.naturvardsverket.se/4ac403/contentassets/260e782f99c948a0a23977ba43cd66d0/naturbetesmarker-o-barkraftiga-foretag.pdf\u003c/span\u003e\u003cspan address=\"https://www.naturvardsverket.se/4ac403/contentassets/260e782f99c948a0a23977ba43cd66d0/naturbetesmarker-o-barkraftiga-foretag.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwedish EPA (2020) \u003cem\u003eNationella markt\u0026auml;ckedata 2018 Teknisk rapport v1.1\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.naturvardsverket.se/en/services-and-permits/maps-and-map-services/national-land-cover-database/\u003c/span\u003e\u003cspan address=\"https://www.naturvardsverket.se/en/services-and-permits/maps-and-map-services/national-land-cover-database/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwedish EPA (2024a) \u003cem\u003e\u0026Ouml;versyn av referensarealer f\u0026ouml;r livsmilj\u0026ouml;typer i art- och habitatdirektivet - Redovisning av regeringsuppdrag\u003c/em\u003e. (NV-11038-22). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.naturvardsverket.se/496428/globalassets/om-oss/slutredovisade-regeringsuppdrag/redovisning-regeringsuppdrag-oversyn-av-referensarealer-for-livsmiljotyper-i-art-och-habitatdirektivet.pdf\u003c/span\u003e\u003cspan address=\"https://www.naturvardsverket.se/496428/globalassets/om-oss/slutredovisade-regeringsuppdrag/redovisning-regeringsuppdrag-oversyn-av-referensarealer-for-livsmiljotyper-i-art-och-habitatdirektivet.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwedish EPA (2024b) \u003cem\u003eNFS 2024:3 \u0026ndash; F\u0026ouml;reskrifter om statligt st\u0026ouml;d f\u0026ouml;r vissa \u0026aring;tg\u0026auml;rder som syftar till att bevara eller \u0026aring;terst\u0026auml;lla biologisk m\u0026aring;ngfald\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.naturvardsverket.se/498b26/globalassets/nfs/2024/nfs-2024-3.pdf\u003c/span\u003e\u003cspan address=\"https://www.naturvardsverket.se/498b26/globalassets/nfs/2024/nfs-2024-3.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Werf HMG, Knudsen MT, Cederberg C (2020) Towards better representation of organic agriculture in life cycle assessment. Nat Sustain 3(6):419\u0026ndash;425. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41893-020-0489-6\u003c/span\u003e\u003cspan address=\"10.1038/s41893-020-0489-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVarela E, Jay M, Flinzberger L, Mobarak C, Plieninger T (2025) A review of high nature value farming systems in Europe: Biodiversity, ecosystem services, drivers, innovations and future prospects. People Nat. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/pan3.70048\u003c/span\u003e\u003cspan address=\"10.1002/pan3.70048\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eV\u0026auml;xa S (2024) \u003cem\u003eCattle statistics 2024\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.vxa.se/fakta/styrning-och-rutiner/mer-om-mjolk/statistik/\u003c/span\u003e\u003cspan address=\"https://www.vxa.se/fakta/styrning-och-rutiner/mer-om-mjolk/statistik/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWestling A, Tor\u0026auml;ng P, Jacobson A, Haldin M, Naeslund M (2020) \u003cem\u003eSveriges arter och naturtyper i EU:s art- och habitatdirektiv: Resultat fr\u0026aring;n rapportering 2019 till EU av bevarandestatus 2013\u0026ndash;2018\u003c/em\u003e. (ISBN: 978-91-620-6914-8). Bromma: Naturv\u0026aring;rdsverket. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.naturvardsverket.se/publikationer/6900/sveriges-arter-och-naturtyper-i-eus-art--och-habitatdirektiv/\u003c/span\u003e\u003cspan address=\"https://www.naturvardsverket.se/publikationer/6900/sveriges-arter-och-naturtyper-i-eus-art--och-habitatdirektiv/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [2025-11-14]\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":true,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"agronomy-for-sustainable-development","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ASDE","sideBox":"Learn more about [Agronomy for Sustainable Development](https://www.springer.com/journal/13593)","snPcode":"13593","submissionUrl":"https://www2.cloud.editorialmanager.com/asde/default2.aspx","title":"Agronomy for Sustainable Development","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8112228/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8112228/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSemi-natural grasslands are among Europe\u0026rsquo;s most species-rich habitats but are rapidly declining due to agricultural intensification, abandonment, and afforestation. Maintaining and expanding these habitats requires continued livestock grazing, raising potential conflicts with climate mitigation goals, particularly commitments to reduce methane emissions. This study explores production-side interventions that could increase grazing of semi-natural grasslands in Sweden under different caps on enteric methane emissions. Using a spatially explicit agri-food systems model (CIBUSmod), we combined regional estimates of grassland productivity with potential restoration areas to assess the impacts of the interventions on the maximum area of semi-natural grasslands that could be managed as well as effects on livestock production, cropland use, and greenhouse gas emissions. We evaluated scenarios including rearing male cattle as steers, prolonging dry periods or delaying culling of dairy cows, expanding winter lamb production, increasing horse grazing, and combinations thereof. Results show that with all interventions combined, the grazed area of semi-natural grasslands could increase by 0.5 Mha under constant methane emissions, nearly doubling current areas. Even under a 30% methane reduction, an additional 0.2 Mha (+\u0026thinsp;38%) could be managed. Among the individual interventions, rearing steers, expanding winter lamb production, and increasing horse grazing showed the greatest potential to expand semi-natural grassland areas, although all interventions contributed when combined. Greenhouse gas emissions per hectare of grazed semi-natural grassland declined across all scenarios, but emissions per unit of edible protein generally rose, underscoring tensions between efficiency-oriented climate metrics and biodiversity goals. Achieving large-scale restoration in practice will require stronger market incentives, targeted policy support, and investments in rural infrastructure. This study demonstrates that it is technically feasible to expand grazing in semi-natural grasslands while containing climate impacts, but only if biodiversity and climate objectives are explicitly balanced when designing future food system policies.\u003c/p\u003e","manuscriptTitle":"Scenarios for long term conservation of Swedish semi-natural grasslands with limited climate impact","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-11 08:34:00","doi":"10.21203/rs.3.rs-8112228/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-01-20T15:16:10+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-18T19:43:59+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Agronomy for Sustainable Development","date":"2026-01-15T12:13:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-19T06:28:07+00:00","index":"","fulltext":""},{"type":"submitted","content":"Agronomy for Sustainable Development","date":"2025-11-18T02:11:44+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"agronomy-for-sustainable-development","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ASDE","sideBox":"Learn more about [Agronomy for Sustainable Development](https://www.springer.com/journal/13593)","snPcode":"13593","submissionUrl":"https://www2.cloud.editorialmanager.com/asde/default2.aspx","title":"Agronomy for Sustainable Development","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"561a053a-f324-4f2d-893d-94810d21f12e","owner":[],"postedDate":"February 11th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-05T16:02:09+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-11 08:34:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8112228","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8112228","identity":"rs-8112228","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}