Opportunities and barriers for promoting biodiversity in Danish beef production Authors: Rasmus Ejrnæs, Lars Dalby, Jesper Bladt, Skjold Alsted Søndergaard, Lasse Dümke, Camilla Fløjgaard, Lars Dyhrberg Bruun, Ditte Dalsgaard Ejrnæs, Jesper Erenskjold Moeslund, Hans Henrik Bruun Published: 2024 ABSTRACT SummaryGrazing at near-natural stocking rates is increasingly rare, whereas abandonment and overgrazing is common, despite both leading to loss of threatened species. Here, we evaluate a biodiversity-promotion strategy of a beef-producing company involving livestock grazing for conservation. Using field surveys, a national biodiversity map, and farmer interviews, we evaluate the conservation potential of farmers committing to a conservation grazing scheme. Most farmers practiced summer grazing, and—contrary to their pledge—primarily grazed areas of low conservation value, and with stocking rates far above carrying capacity. We conclude that a conservation grazing beef brand will only have real value for threatened species if committed to, and held accountable to, a scientifically informed grazing scheme. Our results point to rural traditions, legal regulations, and economic incentives as factors obstructing of a beneficial grazing practice. Hence, the results of this study call for a critical and scientifically informed approach to future land-sharing policies.Subject areas: Ecology, Agricultural science, Relation between agriculture and environment Introduction Land-use change is the most important driver of global biodiversity loss,1,2 with intensification of agricultural practices aiming at increased production of agricultural commodities per unit of land area being the most impactful.3,4 In beef production, this intensification mainly consists of using high yielding cattle breeds, which are stabled and fed high-nutrient fodder during winter, and the use of highly improved grass leys for summer pasture.5 The flip side of the intensification process has been grazing abandonment of un-improved, semi-natural grassland landscapes, which were traditionally used as pasture to hardy breeds of cattle, horse, and sheep.6,7,8,9 High conservation values associated with temperate grasslands, which historically have been maintained by low-intensity livestock grazing, have thus been put in peril.10,11 The biodiversity loss associated with grazing abandonment has led to proposals for mitigation,8,12 including the introduction of agri-environmental schemes involving subsidies for sustained grazing of high nature value non-arable farmland.13,14,15 Despite these efforts, both the extent of resumed grazing and the quality of grazing management have been assessed as inadequate for proper conservation and restoration of biodiversity associated with semi-natural habitats.13,16 In Denmark, a national mapping showed that as much as 80% of the total area of semi-natural grasslands, heathlands, and wetlands currently has no grazing management at all17 and that the actual grazing management on the remaining 20% semi-natural habitats is predominantly summer-only grazing at animal densities much higher than estimated carrying capacity.18 Agricultural intensification on a fraction of total land area has been proposed as a way to enable the setting aside of area for dedicated nature conservation or ecological restoration, while upholding a constant food production.19,20 Also, a strategy combining aspects of land sharing and land sparing into three compartments, i.e., strictly protected areas, areas under relaxed protection and extensive use, and intensive farmland, has been evaluated.21,22 The latter approach has been implemented in the recent biodiversity strategy of the European Union for 2030 in the form of targets for 10% strict and 20% relaxed protection, allowing for some exploitation.23 In regions dominated by intensive farmland, such as Denmark,24 farmers own and manage a very large proportion of land area—not only rotational fields but also semi-natural habitats. Unfortunately, production optimization leads them to either overgraze or abandon grazing in valuable habitats, emphasizing the intricate relationship between biodiversity and agricultural policy and economy.25 A large share of the North European biota is associated with ecosystems shaped by large herbivore grazing,26 and large herbivores play a key role in the natural ecosystems of the region and beyond.27,28 Through the late Quaternary, the indigenous European fauna of large herbivores was depleted by extinction29 and the ranges of most extant species contracted to small areas.30 The domesticated lineages of the aurochs and the wild horse continued for millennia to roam the landscape and exert much of the key functions of their wild counterparts.31 Domestic cattle kept for beef production may fill a missing key function in semi-natural ecosystems, supporting biodiversity conservation,32 provided their grazing and other activities are managed to levels mimicking a natural grazing function. In this respect, the obvious challenge is to uncouple the general intensification of agricultural practices from the grazing management that needs to comply with natural stocking rates and take place in natural, infertile ecosystems that are not production-optimized by drainage, fertilization, and reseeding with grass and clover cultivars. The European Union is now demanding economic activities to be sustainable regarding their impact not only on climate, natural resources, and pollution but also on the protection and restoration of biodiversity.33 This urge has trickled down to economic activities involving agriculture and, in the context of this study, beef production employing free-ranging cattle. In the same vein, the multinational pork and beef production company Danish Crown has set up a niche production label, “Friland Naturpleje” (Free range Conservation Grazing), aiming to combine beef production from free-ranging cattle with conservation grazing. The company has ventured with independent scientists to define a set of pledges and criteria to ensure grazing management that supports the goals of biodiversity conservation. In early 2022, we published a set of conservation pledges aimed at safeguarding biodiversity on the properties and grazing lands of beef producers under the brand of “Friland Naturpleje.”34 Direct assessment of whether the pledges are upheld is outside the scope of this study. Rather, our aim was to identify opportunities and barriers for successful conservation grazing. To this end, we assessed the actual grazing practices during 2022 of 20 beef producers, sampled among those who had signed the conservation pledges for an incentive of a 15%–20% increased meat price. Based on our results, we discuss how companies such as Danish Crown can proceed in developing the brand “Friland Naturpleje” for the mutual benefit of beef production and biodiversity. Results Categories of land use The stratified random selection of producers resulted in a selection made sure to not confound geographic location and property size (Figure 1). Our field mapping of production areas of the 20 farmers to land use categories (conservation value) revealed a considerable variation among producers (Figure 2A). Map of Denmark showing farm location and size (number of cattle) for the 20 studied producers Yellow lines connect grazing ranges to the farms responsible for managing them. Percent managed land classified as each of the four conservation categories (C, = cultivated; L, = low value; M, = moderate value; and H, = high value) (A) Depicted for each of the 20 producers (a–t) separately. The dotted horizontal line marks the promised 10% for uncultivated ecosystems (see conservation pledge 1). (B) Depicted for all 20 farmers combined. In Denmark, approximately 59% of the land area was used for agricultural crops in 2021, and rotational fields take up an even higher proportion of the agricultural landscape after urban areas and state-owned property have been accounted for. For 11 out of 20 producers, cultivated fields covered more than 50% of the managed land, and for six producers it covered more than 75% of the managed land. All producers complied with the pledge of allocating at least 10% of the land to uncultivated biotopes, such as set aside, permanent grassland, semi-natural habitats, or woodland, although for half of the producers, the uncultivated areas were of low-moderate conservation value. Seven producers had more than 25% of natural areas of high conservation value (category H) and four had a share of more than 50%. Conservation value of grazing lots When pooling the total area for the whole sample of 20 producers, the share of high conservation value areas was above 70%, and the share of cultivated areas was down to 20% (Figure 2B). This reflects several very large natural areas, all of high conservation value, including Kalvebod Fælled (868 ha, state owned), Vejlerne (1,637 ha, owned by a private foundation), and North tip of the island of Fanø (437 ha, privately owned). Our summary of semi-natural habitat areas without grazing in the surrounding landscape is based on agricultural statistics and may include rare cases of un-subsidized grazing. More than half of the properties were surrounded by potential grazing land of considerably higher conservation value than the areas currently used for grazing (Figure 3). For four producers (a, b, f, h), the conservation value of currently grazed land equaled that of the surrounding landscape. Percent area in each bioscore category (from low to high conservation value: <2, 2–5, 5–8, >8) for the grazing fields of each producer (a–t) on their own property (upper graph), ranked inversely to their proportion of low bioscore area and for the ungrazed and uncultivated semi-natural areas in the corresponding surrounding landscape (lower graph). Grazing areas by length of season and grazing intensity The survey of grazing practices revealed that the majority of the grazed land was used for seasonal grazing, typically involving a grazing period of 6–8 months (Figure 4A). Short-term grazing was slightly more common for improved grass leys and long-term grazing for high conservation value areas, but otherwise the grazing periods were not clearly associated with the conservation value of the area (Figure 4B). Grazing period (A) The length of the grazing period in days per year accumulated over the total area studied. (B) The share of six grazing period length classes across the total area for each conservation class. Regarding the grazing intensity, we saw a marked difference, with more than 75% of the improved grass leys (category C) subject to a stocking rate of 800 kg ha−1 or more and more than 85% of the area exceeding 400 kg ha−1 (Figure 5B). For the non-crop areas (category L, M, or H), 48% of the area exceeded the nature pledge of 400 kg ha−1 and 9% exceeded 800 kg ha−1 (Figure 5A). Grazing intensity generally increased with decreasing conservation value from category H to C and for the grazing lots of highest conservation value, H, 63% of the area satisfied the pledge of maximum 400 kg ha−1 (Figure 5B). Grazing intensity (A) The grazing animal density accumulated over the study areas having conservation values low, medium, and high. Dotted lines depict the percentage of the area corresponding to recommended (optimum) and maximum stocking rates of max 250 kg/ha and 400 kg/ha, respectively (see conservation pledge 4). (B) The share of five grazing animal density classes over the total area of each conservation category. The field assessment of management quality revealed that most of the area satisfied a minimum criterion corresponding to expected net biodiversity benefit (compared to abandonment), and particularly applicable to areas of conservation value, more than 60% and 75% of the area (for M and H, respectively) reached up to moderate to good management quality (Figure 6A). Less than 2% of the surveyed grazing area was assessed to represent optimal grazing management (Figure 6A). A summary of the management quality for biodiversity of grazing areas, with overgrazing, machine cutting, and abandonment being suboptimal in this regard (A) Percent area with each type of grazing management for the four conservation categories. (B) Percent nature area (“L”, “M”, “H”) with each type of grazing management. Discussion Our survey of cattle grazing management for nature conservation revealed that the combination of grazing for beef production and grazing for biodiversity conservation is challenging to integrate in contemporary agricultural practice. A high proportion of the beef producers enrolled under the conservation grazing label “Friland Naturpleje” let their cattle mainly graze improved grass leys or permanent grassland with no obvious conservation value. This is not caused by a landscape level deficit of or inaccessibility to high conservation value habitat, because we generally found ungrazed high-value areas in the vicinity of the current grazing areas, suggesting that they have been abandoned for other reasons than geographical distance. This finding is conservative, because some beef producers manage grazing areas interspersed by such distances that our mapping greatly underestimates the area of high conservation value that could potentially be grazed. To our knowledge, the most obvious reason for this pattern is that high conservation value areas are often not owned by farmers with domestic grazers. In this case, negotiations are needed regarding fencing, access to water, transportation and tending of animals, and the sharing of expenses and subsidies between landowner and manager. Also, recreational hunters renting the natural areas are generally skeptical of fences and domestic animals on their hunting grounds. It should also be mentioned that the fencing and tending of animals in high conservation value areas can be demanding due to significant topographical, hydrological, and structural variation compared to former production areas. Our results are in accordance with previous findings, where abandonment has been acknowledged as a major threat to the biodiversity of semi-natural areas in Europe.8,35,36 Despite attempts to restore the grazing process by directing subsidies to high nature value farming areas (HNV) in Europe,37 dereliction and scrub encroachment remain a threat to species and habitats of high conservation importance,38 also in lowland northern Europe.39 Grazing is a key ecological process,27,40,41 and it appears obvious that the current stock of free-ranging cattle needs to be redistributed from improved leys and set aside arable land to high conservation value nature, in order to overcome the encroachment and lingering loss of species associated with warm and open habitats.42,43 In the current situation, it seems obvious that economic incentives are too weak to stimulate private landowners to advertise their potential grazing land to cattle farmers. Abandonment of livestock grazing could be considered an opportunity for passive rewilding,44 making space for a comeback of wild large herbivores. However, absence of livestock is unlikely to lead to a natural grazing function in European ecosystems, because the herbivorous mammal fauna of Europe is in general deeply depauperated and skewed toward small-bodies species29 and because the density of extant wild herbivores is regulated (by hunting) far below natural levels.45 Despite recent re-introduction of beaver to Denmark, the native large free-ranging herbivore guild mainly comprises red deer, fallow deer, and roe deer, and the density of these herbivores at county level has been estimated to be in the range of 1–8 kg ha-1,18 which is extremely low compared to a natural baseline estimated to be in the range of 70–250 kg ha-1.18,45 Alongside the lack of wild large herbivores, additional natural processes, such as natural hydrology or fire, are absent or heavily controlled in most European landscapes. Although passive rewilding may lead to a gradual recovery of natural vegetation and hydrology, it is expected to jeopardize the conservation of biodiversity associated with open-grazing-dependent habitats. The ownership structure of the human-dominated landscapes may represent a major obstacle to targeted distribution of livestock to areas of grazing-dependent high conservation value. In Denmark, coastal and riparian wetlands are typically divided into narrow strips of land allotted to a large number of individual landowners.46 Although intensive cultivation has ceased in many marginal wetlands and coastal areas, this has not automatically resulted in room for natural ecosystem processes. Abandoned areas are generally tied up in recreative interests—such as walking, running, swimming, biking, fishing, and hunting—and these interests are often perceived as incompatible with perimeter fences and free-roaming cattle. Forestry regulations may also hinder conservation grazing, as the Danish Forest Act largely prohibits livestock grazing in forests,47 despite the practice being recognized for its nature conservation potential.48 On top of these obstacles to natural grazing, the Danish traffic legislation forbids the passing of public roads by domestic cattle, and the watercourse legislation limits livestock access to and passage of streams and rivers. Alongside grazing abandonment and encroachment in high conservation value areas, we observed clear indications of widespread overgrazing among beef producers (Figure 5A). Cattle grazing was concentrated in the summer months, during which stocking rates often exceeded 400 kg ha−1, which is twice or more the estimated natural level.18 Overgrazing is widely acknowledged as a threat to biodiversity, mainly because large herbivores in high densities reduce the availability of plant foliage and flowers for herbivorous and flower-seeking insects.16,49,50 However, also plant species can be negatively affected by high stocking rates.51 Overgrazing is closely associated with summer grazing followed by stabling and feeding of cattle through the winter season. Under natural conditions without supplementary feeding, forage limitation during winter sets an upper limit to the carrying capacity of an area. Unfortunately, summer grazing is preferred over year-round grazing, and the producers participating in our study mentioned several reasons for this pattern. These reasons are reported below in arbitrary order. First, the producers prefer to keep their animals stabled during winter for animal welfare reasons. Second, they prefer to have cattle under close surveillance during calving in early spring. Third, several producers work with high-yielding breeds such as Limousin, Simmental, and Charolais, which all need a steady supply of protein-rich fodder to thrive and avoid weight loss. The selection of breeds might also be influenced by economic incentives due to costs for butchering per individual. This economic incentive works against the use of small-bodied robust cattle breeds, such as Galloway and Dexter. These trends of untargeted grazing, overgrazing, and abandonment are mirrored in studies of farmers enrolled in agri-environmental schemes in other European countries.52 In our understanding, the overarching reason for unfocused distribution and seasonal overgrazing is the economic incentive system of the European Union’s Common Agricultural Policy, as well as the Danish implementation thereof. Despite attempts to allocate subsidies to high nature value farmland, the highest revenue per hectare is currently achieved by farmers, when the basic agricultural subsidy can be combined with a surplus subsidy for conservation grazing.53 The basic subsidy requires the areas to be managed so intensively during summer that they can pass the field control by authorities and deemed “well-managed farmland.” A precondition for this is a near-absence of wet depressions, thorny shrubs, trees, and of tufts of heather, sedges, rushes, ferns, and other unpalatable plants. In practice, farmers achieve the required uniform condition by a combination of intense summer grazing and intermittent machine cutting of unpalatable plants, which also levels out microtopographic variation caused by ants, tussock-building sedges, dung piles, mole hills, and animal trampling. Thus, keeping areas as “well-managed farmland” is at direct odds with natural ecosystem processes that underpin a diversity of microhabitats. In contrast, conservation grazing should be considered one aspect of a holistic ecological restoration and conservation, including natural vegetation, ancient trees, natural hydrology, coastal dynamics, and carbon sequestration.54 In perspective, the agricultural subsidy schemes and the incentives they create reflect a central dilemma in European nature conservation. On one hand, traditional grazing has for millennia been pivotal to the conservation of semi-natural habitats in the absence of natural densities of large wild herbivores,55 and at the same time, agricultural optimization is in obvious conflict with conservation efforts to restore grazing as a natural ecosystem process. The current EU biodiversity strategy—aligned with the Kunming-Montreal Global Biodiversity Framework—implies a designation of 30% of terrestrial and marine areas as protected (20%) and strictly protected (10%) land.23 Although rewilding with introduction of wild and feral large herbivores is an option for the ecological restoration of strictly protected areas, conservation grazing with domestic animals remains an important near-natural scenario for protected areas with limited human exploitation.56 This relies on the assumption that grazing practices in these areas follow ecological guidelines suited to mimic a natural grazing process, as opposed to current practices guided by agricultural tradition and counter-productive incentives. It should be acknowledged that livestock-grazed areas not necessarily offer habitat for the entire biota, e.g., not carrion-associated fauna and funga, and therefore should be considered a supplement to, not replacement for, strictly protected areas with rewilding of near-intact ecosystems.56 We have evaluated the current management practices applied by producers of beef for a company with the ambition to make a significant contribution to halting the loss of biodiversity. We found a considerable variation in the spatial focus among producers, with most beef-cattle grazing improved grass leys with little or no conservation value, despite appreciable areas of semi-natural grazing-dependent habitat on the property or within its close vicinity. Summer grazing was found to be standard practice, with only a few cases of year-round grazing. Consequently, stocking rates frequently exceeded natural carrying capacity by far, with overgrazing as a consequence. In response to our evaluation, Danish Crown has postponed their commercial market release of conservation-friendly beef, until solutions to the main issues have been found. The present study has uncovered a basic conflict between beef production optimization and biodiversity-promoting grazing management, which begs the formulation of a set of minimum criteria for conservation grazing and the incorporation of these criteria in the rule sets for agricultural subsidies to conservation grazing. This change to the common agricultural policy is necessary to ensure a positive net effect on biodiversity of agri-environmental efforts. We suggest that these minimum criteria cover the following aspects: (1)Protection of existing habitat and structures on the property of producers. Protection of existing habitat and structures on the property of producers. (2)Grazing focused spatially on areas of significant conservation value. Grazing focused spatially on areas of significant conservation value. (3)Restoration of natural grazing process—i.e., whole-year grazing without supplementary feeding and stocking rate within natural bounds. Restoration of natural grazing process—i.e., whole-year grazing without supplementary feeding and stocking rate within natural bounds. (4)Restoration of natural processes associated with hydrology, coastal dynamics, wildfire, and plant growth. Restoration of natural processes associated with hydrology, coastal dynamics, wildfire, and plant growth. Limitations of the study Although we sampled the cattle farmers using a stratified random approach in order to represent the variation of beef producers adopting the conservation beef program of Danish Crown, the low sample size (20 farmers) does come with a level of uncertainty. Also, the expert assessment of conservation value of areas and conservation value of area management is limited by the expertise of the involved experts and the data available for assessment. To increase the accuracy and reduce the uncertainty of these steps, we have based the assessments on very coarse ranking scales. It should also be noticed that although our study evaluates the degree of conservation focus of a beef-producing commercial company with a declared intent to mitigate the biodiversity crisis, we have not studied the response of species and habitats to area management. Future studies will be needed to evaluate whether better targeted conservation grazing can contribute to the restoration of favorable conservation status of species and habitats. Resource availability Lead contact Further information and requests for resources should be directed to the lead contact, Hans Henrik Bruun (hhbruun@bio.ku.dk). Materials availability This study did not generate new unique reagents. Data and code availability •Data used in this study have been deposited at Zenodo.org as tab separated files and are publicly available as of the date of publication. Data used in this study have been deposited at Zenodo.org as tab separated files and are publicly available as of the date of publication. •R code used in this study has been deposited at Zenodo.org and is publicly available as of the date of publication. R code used in this study has been deposited at Zenodo.org and is publicly available as of the date of publication. •All analyses are described in detail in STAR Methods; any additional information required to reanalyze the data are available from the lead contact upon reasonable request All analyses are described in detail in STAR Methods; any additional information required to reanalyze the data are available from the lead contact upon reasonable request Acknowledgments The research was partly funded and carried out as an assessment of biodiversity compliance by meat producers commissioned by Danish Crown to Aarhus University. We thank the cattle farmers for access to their properties and for information of the timing and position of their herds on different pens. Author contributions Conceptualization: R.E. and H.H.B.; formal analysis: L.D., J.B., and R.E.; funding acquisition: R.E.; investigation: R.E., S.A.S., L.D., L.D.B., and D.E.; methodology: R.E., J.B., C.F., and J.E.M.; project administration: R.E.; resources: J.B. and L.D.; supervision: R.E.; writing—original draft: R.E. and H.H.B.; writing—review and editing: L.D., J.B., S.A.S., L.D., C.F., L.D.B., D.E., and J.E.M. Declaration of interests The authors declare to have been free to select methods, collect data, and draw conclusions based entirely on scientific evidence and methods, without any interference from the funder. They declare to have no conflict of interests with this work. STAR★Methods Key resources table REAGENT or RESOURCESOURCEIDENTIFIERDeposited dataData and codeThis paperhttps://doi.org/10.5281/zenodo.8147250Software and algorithmsRR Core Team (2023)57N/AQGISQGIS.org (2021)58N/A Method details The purpose of the conservation pledges, commissioned by Danish Crown,34 was to work as a commitment and brand for producers of beef to follow a best practice conservation grazing regime and enable a science-based and trustworthy branding of their meat products. We were further commissioned by Danish Crown to validate the conservation grazing practice of farmers committed to the scheme and this validation was carried out during August-September 2022. We selected a sample of 20 out of the 160 involved producers for a field visit, including recording of grazing practices and biodiversity values. We selected the 20 producers randomly but stratified to represent the variation observed in size and geographic location of their properties (Figure 1). Some of the selected producers had all grazing areas close to their farm, while others had their grazing areas far from their farm, in nature areas of other landowners (Figure 1). Conservation pledges The conservation pledges included four different aspects of biodiversity commitment. Some of these aspects refer to areas of conservation value, which we define according to a recent national mapping of conservation value.17 The biodiversity map is based on a development of the High Nature Value farming map for Denmark,37 extended to cover all terrestrial ecosystems. This map builds on a combination of verified observations and expert-drawn habitat areas of red-listed species combined with a number of indicative landscape features (proxies) with significant positive correlation to the observed occurrence of red-listed species in Denmark.17 The mapped conservation value is represented by a so-called bioscore ranging from 0 to 20, being the sum of the species score (0–9) and a landscape proxy score (0–11). A pragmatic interpretation of the bioscore scale is: <2 = no conservation value, 2–5 = low conservation value, 5–8 = moderate conservation value and >8 = high conservation value. The conservation pledges were negotiated with Danish Crown to reflect their ambitions, while still delivering guarantee for biodiversity improvement compared to status quo. The pledges may be summarized as. (1)Area reservation Area reservation All producers promise to reserve a minimum of 10% of their property and other managed land for uncultivated ecosystems dominated by native plant species, such as abandoned fields, permanent grassland or other semi-natural habitats, such as meadow, mire or woodland. (2)Protection of existing habitats Protection of existing habitats All producers promise to protect existing habitat of conservation value on their property against destruction. This pledge includes the protection of hollow trees and trees above 200 years of age or exceeding 150 cm DBH (trunk diameter at breast height), the protection of areas of moderate-high conservation value (bioscore >5) against machine cutting or clearing of self-grown woody plants. Furthermore, this pledge includes protection against planting, sowing, drainage, fertilization and application of pesticides in areas reserved under heading 1) above. (3)Spatially focused grazing Spatially focused grazing All producers promise to deploy their livestock in areas of highest possible conservation value, implying that areas on the property of moderate-high conservation value (bioscore >5) should be prioritized for grazing before areas of lower conservation value. (4)Best practice grazing for conservation Best practice grazing for conservation All producers promise to approach near-natural grazing by applying precautionary principles in the grazing management. This is to avoid damage caused by overexploitation. Grazing should take place within the range of 70–250 kg ha−1 of grazing animals over the year. Densities should not exceed 400 kg ha−1 at any given time during the year. This preferred range of 70–250 kg ha−1 is relatively high compared to figures from rewilding projects in Europe, which suggests 40–180 kg ha-1,45 but remains in accordance with Danish recommendations based on experienced carrying capacity under whole year grazing without supplementary feeding.17,18 The grazing period should last a minimum of 180 days and the herb layer should be low and open (<10 cm) in April, before spring regrowth starts. Openness implies a minimum of litter and clear visibility of bare ground or moss cover when the sward is inspected visually in vertical projection. Treatment with anthelmintics is prohibited when the animals are in conservation areas and two weeks before they are released onto conservation areas. Field surveys and area categorization ‘Danish Crown’ provided maps of the fields managed by the producers. From these maps, our field inventory was planned with the aim of assessing the value of all fields of potential conservation value – both grazed and ungrazed – and to investigate the vegetation of fields with uncertain status, such as improved grass ley or spontaneous vegetation without data from previous monitoring og mapping. Based on visual inspection of aerial photos and bioscores of the national biodiversity map, we selected 53% of area of the uncultivated fields for field investigation, while the remaining 43% was classified using the national biodiversity map. All field inventories were carried out during July and August 2022. We classified all visited areas into four possible categories according mainly to the observed vegetation (Figure S1). (1)Cultivated field dominated by sown crops, including grass and clover cultivars (C = “cultivated”). Cultivated field dominated by sown crops, including grass and clover cultivars (C = “cultivated”). (2)Spontaneous vegetation of low conservation value, defined as an “unlikely habitat of rare and red-listed species”. Although dominated by spontaneous plant cover, this category often also contains grass and clover cultivars. Such areas are typically drained and fertilized and heavily affected by intensive land use in the recent past (L = “low value”). Spontaneous vegetation of low conservation value, defined as an “unlikely habitat of rare and red-listed species”. Although dominated by spontaneous plant cover, this category often also contains grass and clover cultivars. Such areas are typically drained and fertilized and heavily affected by intensive land use in the recent past (L = “low value”). (3)Spontaneous vegetation of moderate conservation value – defined as a “possible habitat of rare and red-listed species”. Such areas typically host a wide range of wild forbs, grasses, sedges and rushes, but usually common species with good capacity for dispersal and for competition under fertile conditions (M = “moderate value”). Spontaneous vegetation of moderate conservation value – defined as a “possible habitat of rare and red-listed species”. Such areas typically host a wide range of wild forbs, grasses, sedges and rushes, but usually common species with good capacity for dispersal and for competition under fertile conditions (M = “moderate value”). (4)Spontaneous vegetation of high conservation value – defined as a “likely or confirmed habitat of rare and red-listed species”. Such areas typically include natural and semi-natural grassland, heathland, wetland or woodland with a cover of plants, mosses and lichens indicative of infertile habitats with long temporal continuity in time (H = “high value”). Spontaneous vegetation of high conservation value – defined as a “likely or confirmed habitat of rare and red-listed species”. Such areas typically include natural and semi-natural grassland, heathland, wetland or woodland with a cover of plants, mosses and lichens indicative of infertile habitats with long temporal continuity in time (H = “high value”). All areas were post-hoc revised according to their reported land use, aerial photos and their score in the national biodiversity map, using the following rule set. (1)Areas not covered by field surveys remained “C” if they were intensively managed (e.g., ploughed regularly, data from the national map of agricultural fields37) or the median bioscore was below 5, changed to “L” if they were extensively managed (i.e., not in rotation and with low or no fertilization37) and the median bioscore was above 5, changed to “M” if they were extensively managed and the median bioscore was between 5 and 8, and raised to “H” if they were extensively managed and the median bioscore was >8. Areas not covered by field surveys remained “C” if they were intensively managed (e.g., ploughed regularly, data from the national map of agricultural fields37) or the median bioscore was below 5, changed to “L” if they were extensively managed (i.e., not in rotation and with low or no fertilization37) and the median bioscore was above 5, changed to “M” if they were extensively managed and the median bioscore was between 5 and 8, and raised to “H” if they were extensively managed and the median bioscore was >8. (2)Areas classified in the field as spontaneous vegetation (not C), but not further investigated, were “L” if the median bioscore was <5, “M” if the median bioscore was between 5 and 8, and “H” if the median bioscore was >8. Areas initially classified in the field as L or M were re-classified to H, in case the median bioscore of the area was above 9. We interpreted such cases as strong indication that the area is habitat for red-listed species, despite the fact that we observed none during our visit in August. Of the areas visited in the field, 3.2% were reclassified to “H” based on the national bioscore. Overall, we found a strong positive association between the field-classified grades (before re-classification) and the median bioscore of the areas (Figure S2). Areas classified in the field as spontaneous vegetation (not C), but not further investigated, were “L” if the median bioscore was <5, “M” if the median bioscore was between 5 and 8, and “H” if the median bioscore was >8. Areas initially classified in the field as L or M were re-classified to H, in case the median bioscore of the area was above 9. We interpreted such cases as strong indication that the area is habitat for red-listed species, despite the fact that we observed none during our visit in August. Of the areas visited in the field, 3.2% were reclassified to “H” based on the national bioscore. Overall, we found a strong positive association between the field-classified grades (before re-classification) and the median bioscore of the areas (Figure S2). Grazing management assessment We also assessed the quality of grazing management in the field. For this aspect, we relied on indicators of heterogeneity and variation associated with natural grazing, such as a variety of flowering herbs, micro-topographic variation, variation in vegetation height, grazing shelters caused by thorny shrubs or unpalatable herbs. Grazing management was scored from 1 to 6, with 1 = optimum grazing function, 2–3 = suboptimal grazing function, 4 = minimum grazing function, 5–6 = moderate to severe overgrazing, 7 = machine cut and 8 = abandoned/ungrazed. Optimum grazing implies a variation in vegetation height, an open sward with abundant flowering, scattered woody plants and natural micro-topographic variation, including ant hills, tufts of sedges and rushes, wet depressions and large boulders. Overgrazing implies that foliage and flowers are absent from most of the area and that the area appears highly homogeneous. Machine cutting management further implies a leveled surface and absence of woody plants and absence of variation in vegetation height. Areas were visited twice: Once by a zoologist with expertise in arthropods and once by a botanist with expertise in vascular plants and macrofungi. Both experts surveyed the areas for species of conservation value. The botanist took lead on the categorization of areas in conservation classes (C, L, M and H) and the entomologist led on the assessment of grazing management. Spatial analyses of surroundings To evaluate the degree of spatial focus in the selection of areas for grazing management, we delineated a 1 km buffer around the grazing areas of each producer and summarized the median bioscore of all unmanaged semi-natural habitats within this area. These vicinity bioscore values were eventually – for each producer – compared to the median bioscores of the managed grazing areas. Mapping of grazing density We aimed at estimating the competition exerted by cattle upon herbivorous insects by calculating density of herbivores (kg ha−1) in months with high insect activity. For this purpose, we first contacted all producers to collect information regarding the breed of cattle, the number of adult and juvenile individuals, the length of grazing period and the precise physical boundaries for all grazed areas to enable calculation of grazing densities. Secondly, we looked up all reported observations in the country of species in four large insect groups (Lepidoptera, Diptera, Hymenoptera and Coleoptera) in the Danish species portal (https://arter.dk). The frequency of insect observations over time indicated high insect activity from mid-April (Julian Day 117) to mid-September (Julian Day 278) (Figure S3). We therefore decided to use this time frame as the period where high density cattle grazing could jeopardize the herbivorous insect fauna. We multiplied the number of individuals with tabulated weights for breeds (Table S1) and calculated the stocking rate as the sum of kg ha−1 for all grazing days within the time frame divided by number of days (161 days). Quantification and statistical analysis Data handling and analysis was done using R ver. 4.3.157 and QGIS.58 Code and data used are available from Zenodo.org and is also listed in the key resources table. Test for significance was done using ANOVA and Tukey HSD from the R package “stats” (see Figure S2). Published: November 19, 2024 Footnotes Supplemental information can be found online at https://doi.org/10.1016/j.isci.2024.111422. Supplemental information Document S1. Figures S1–S3 and Table S1 References Associated Data Supplementary Materials Document S1. Figures S1–S3 and Table S1 Data Availability Statement •Data used in this study have been deposited at Zenodo.org as tab separated files and are publicly available as of the date of publication. Data used in this study have been deposited at Zenodo.org as tab separated files and are publicly available as of the date of publication. •R code used in this study has been deposited at Zenodo.org and is publicly available as of the date of publication. R code used in this study has been deposited at Zenodo.org and is publicly available as of the date of publication. •All analyses are described in detail in STAR Methods; any additional information required to reanalyze the data are available from the lead contact upon reasonable request All analyses are described in detail in STAR Methods; any additional information required to reanalyze the data are available from the lead contact upon reasonable request