Estimating the soil particle size distribution (PSD) from visible-near infrared (vis-NIR) reflectance spectra is conventionally limited to predicting discrete soil fractions (e.g., sand, silt, and clay). This approach presents significant challenges: it requires the harmonization of data from different classification systems and, by reducing the PSD to a few values, fails to reflect the entire variation in soil texture. To address these gaps, we present a novel physics-informed neural network (PINN) for the direct estimation of the continuous PSD from vis-NIR reflectance measurements. The PINN learns a continuous, differentiable, and non-parametric representation of the cumulative PSD by integrating both measurements and physical constraints imposed during training. This approach eliminates the need for harmonization and interpolation of measurements originating from different soil texture classification systems and allows the model to be trained on datasets with varying numbers of measurements per sample. Performance evaluation on 30% of the 2777 studied samples showed that the PINN achieved an RMSE of 6.77% and an R 2 of 0.97 in predicting the cumulative PSD fraction. For the texture fractions, the model achieved RMSE values of 4.72%, 3.06%, and 2.75% for sand, silt, and clay, respectively. A comparison with a physics-agnostic (i.e., physics-uninformed) version of the model revealed that both approaches performed similarly in terms of RMSE and R 2. However, the physics-agnostic model violated physical constraints even in data-rich scenarios. In contrast, the PSD obtained from the PINN maintained its physical integrity even under data sparsity conditions and consistently produced non-negative, monotonically increasing predictions that sum to 100% at the largest particle size.

Pollen is a source of protein, lipids, vitamins and minerals for bees and other flower-visiting insects. The composition of macro- and micronutrients of pollen vary among different plant species. Honey bees are long-distance foragers, collecting nectar and pollen from plants within several kilometers of their hive. Availability of pollen within the foraging range of honey bees is highly dynamic, changing seasonally, and across different landscapes. In the present study, the aim was to investigate the composition of pollen collected by honey bees in rural-urban landscape mosaics typical of Northern Europe. Samples of corbiculate pollen were collected 3-9 times during the growing season by citizen scientist bee keepers from a total of 25 observation apiaries across Denmark in 2014-2015. Palynological analysis was conducted identifying 500 pollen grains per sample to pollen type (mostly plant genus). Pollen diversity denoted the number of different pollen types in a sample, while relative abundance was calculated as the proportional representation of a pollen type, if found in >1% of the sample. The quantity of pollen types across study years and sites was measured as the occurrence of each pollen type (number of samples with the pollen type present) and abundance (total number of pollen grains). Pollen diversity was highly variable, with effects of season, year, and area of green urban spaces. In terms of quantity, a few key pollen types occurred repeatedly and abundantly in the samples. Only 17 pollen types were present in >15 samples. These pollen types were consistent across study years and different landscapes. Pollen diversity may impact colony health, and hence foraging decisions by honey bees, especially in late summer. However, the bulk of the pollen collected by colonies came from a limited number of pollen sources, regardless of year and landscape context in the rural-urban landscape mosaics of Denmark.

As the globe gets more urbanised, the question about how natural biodiversity is structured in cities becomes increasingly pertinent. To contribute to an answer, we studied species richness and spatio-temporal structure of bees in a North European metropolitan area. A gradient of 13 sites in the city of Aarhus, Denmark, was censused from April to September 2016. Forty species, i.e. 29 solitary species (40% of all individuals), ten Bombus species (28%), and Apis mellifera (32%), were sampled monthly in pan traps. (i) Information about species traits was extracted from literature, and trait values were correlated and used to characterize the fauna. Most were soil-nesters, pollen generalists, and common. (ii) Habitat diversity within five concentric circles with trap as centre and radius from 50 m to 1000 m was related to bee α site diversity. The relationship was significant only within 1,000 m for all bees and for bumblebees. Solitary bee diversity was uncorrelated with habitat diversity at all spatial levels. (iii) Spatio-temporal structure was analysed as two networks, one for bees linked to sites, and one for bees linked to months. Link patterns were analysed for levels of nestedness, modularity, and spatio-temporal β diversity. The two networks were weakly and non-significantly nested, but strongly modular, being composed of five and four modules of co-occurring bees, respectively. (iv) Finally, we studied total β diversity, βTOTAL, being the sum of species turnover, βTURN, and species loss/gain or nestedness, βNEST. For both site and season, βTURN was higher than βNEST, and site βTOTAL was higher than season βTOTAL. One reason for this metacommunity structure may be a high spatio-temporal habitat patchiness, sustaining a rich biodiversity. Thus, a few large areas may not compensate for the loss of several small patches. Consequently, establishment of many green, even small habitats is recommended.

Groundwater-dependent terrestrial ecosystems (GWDTE) have been increasingly under threat due to groundwater depletion globally. Over the past 200 years, there has been severe artificial drainage of low-lying areas in Denmark, leading to a gradual loss of GWDTE nature habitat areas. This study explores the spatial-temporal loss of Danish GWDTE using historical topographic maps. We carry out geographic information systems (GIS) overlap analysis between different historical topographic maps with signatures of GWDTE starting from the 19th century up to a present-day river valley bottom map. We then examine the changes in two protected GWDTE habitat types in different periods and hydrologic spatial locations. Results reveal a decrease in the area of GWDTE over the last 200 years. We attribute this to different human interventions, e.g., drainage, that have impacted the low-lying landscape since the early Middle Ages. We further conclude that downstream parts of the river network have been exposed to less GWDTE habitat loss than upstream ones. This indicates that upstream river valleys are more vulnerable to GWDTE decline. Therefore, as a management measure, caution should be exercised when designing these areas for agriculture activities using artificial drainage and groundwater abstraction, since this may lead to further decline. In contrast, there is a higher potential for establishing constructed wetlands or rewetting peatlands to restore balance.

Soil provides essential ecosystem services sustaining and improving human life, but mapping soil functions is an ongoing challenge. Denmark has a long history of carrying out soil assessments − originally in order to determine tax revenues for the king, and, more recently, for aiding policymakers and farmers. This knowledge has supported the development of intensive agricultural systems while maintaining the provision of ecosystem services (e.g., clean water). Getting an overview of historical soil surveys and pedological mapping approaches can generate useful information for mapping soil, identifying gaps and proposing directions for future research. In this review, we explore the evolution of soil and environmental inventories, the historical development of soil mapping methods, and how these factors contributed to a better spatial understanding of soil functions. Specifically, we discuss soil functions related to water regulation (e.g., drainage, groundwater and water surface interactions, water table), water filtering (e.g., nitrogen leaching), carbon sequestration (e.g., peatlands), agricultural production (e.g., land suitability, wheat yields), and threats related to soil degradation (e.g., soil erosion). Denmark has benefitted from a government-coordinated approach, promoting detailed and systematic national soil surveys and environmental monitoring programmes. The large databases produced in the surveys formed the basis for mapping several soil properties and functions at increasingly high resolutions over the last many years based on developments in machine learning. In contrast to methodological advances in soil mapping and relevant contributions to pedometric research, we identified a lack of spatial information on soil biodiversity. Detailed spatial information about soil functions is essential to address global issues, such as climate change, food security and water security, and the experience of mapping soil functions in Denmark can be a source of inspiration to other parts of the world.

A prerequisite for successful afforestation and reforestation is understanding the quality of a site before establishing a specific tree species. Ecograms have been widely applied to determine the suitability of different sites for different species by a simple assessment of nutrient and water availability. Their graphical representation of suitability into classes allows ecograms to be easily understood. However, ecograms have generally been mapped for small areas only and their validity has rarely been documented. The aim of this study is to map ecograms for five tree species across Denmark and validate the maps using forest stand production data. For this purpose, we classified the landscape into six nutrient classes and nine water classes based on four variables to generate the ecogram maps. Based on these classes, the generated maps depict if a tree species is unsuitable, suitable or optimal for a specific site. The absolute average misclassification for nutrient and water supply was 1 and 2 classes, respectively. Stepwise linear regression was implemented to determine if the four variables used to create the ecograms were able to predict production as observed from forest experiment and management planning data obtained from across the country. All five species used the full model to explain variation in production. However, the average production values per ecogram growing condition were not significantly different for all species. The range of $R$2 for the five species was 0.05–0.32, indicating that one ecogram template might not work for all species. The high-resolution national ecogram maps incorporate large-scale variables important for tree growth and will be beneficial when selecting new land for afforestation. The simplicity of ecograms allows for easy interpretation, meaning that foresters can quickly determine which regions of a landscape are suitable, saving time and resources.

•Denmark benefits from a strong soil survey history and the availability of soil data at a national scale.•DSM and DSA products contribute to enriching the Danish soil database and establishing a national digital soil information system.•Lack of awareness of soils in Danish society can be linked to a gap in education.•Soil sealing, soil organic matter decline, compaction and erosion are the major soil threats in Denmark.

A natural terroir unit is a tract of land whose natural characteristics form a unique assemblage of factors (soil, terrain and climate) which together impart specific high-quality characteristics to an agricultural product. In order to map and describe Danish natural terroir units, we built on previous efforts to quantitatively map and describe Danish natural terroir units based on soil, terrain, climatic and historical crop yield data. Our work consists of four stages: (1) The OSACA algorithm was applied to define soil centroids and measure taxonomic distance between soil profiles and soil centroids based on a Danish soil spectral library; (2) nine Danish terron classes were established by fuzzy c-means clustering based on soil, terrain and climate information; (3) a Danish terron map was generated by Cubist regression tree models and the uncertainty of this map was assessed by a terron membership map; (4) Danish natural terroir units were described by linking historical crop yield data to the terron map. The results suggested that the OSACA algorithm and Vis-NIR spectral data could be used as an efficient tool to facilitate terron identification. The terron predictions also showed that the addition of terrain and climatic predictors improved the previously created Danish terron map. The description of Danish natural terroir units showed that seven natural terroir units could be an optimal number for Denmark for specific crop types. Further investigations are needed that link more agricultural yield data to this terron map in order to describe natural terroir units for different agricultural products. The methods developed in this study could be tested in other countries to facilitate sustainable soil management and minimize environmental risks.

Description of the datasets The file 00_MUSTB_field_data_model.docx contains the data model according to which the data collected in the context of the MUSTB field data collection were reported to EFSA. The current data model description includes some modifications with respect to the specifications published before the beginning of the project (EFSA, 2017, https://doi.org/10.2903/sp.efsa.2017.EN-1234). All the tables included in the data model are published here in csv format. The underlying schemas are also published in xsd format. Sites: General information about the sites where the data collection took place; Polygons: General information about the polygons where the botanical survey took place. Table I: Pesticide application, reporting data on experimental spraying events; Table II: Resource providing unit and landscape fitness, reporting data on abundance of flowering plants in polygons mostly within 1.5 km, but in some cases up to 3 km of the experimental colony; Table III: Master list of all hives included in the study; Table IV: Colony management, reporting the log of the beekeeper regarding input (if material was added to the hive: e.g. empty frames, chemicals for varroa treatment, sugar), output (if material was removed from the hive, e.g. honey combs, supers), queen loss, swarming, or clinical signs observed in the experimental hives; Table V: Hive inspection, reporting data on in-hive measurements in the experimental colonies. This table contained several types of data, including: Data on brood development and food provision (“cell utilization”) obtained from image analysis of combs; Data on forager activity obtained from automatic video recordings and image analysis by a bee counter; Data on hive weight obtained from automatic logging by a hive scale; Data on adult bee strength, obtained by weight assessment of combs with and without adult bees (“bees per comb data”); Table VI: SSD2, reporting data on results of laboratory analyses of pollen, pesticide residues and parasites/pathogens. These four types of laboratory analyses involved different methods, and were reported according to different standards. Therefore, a number of the fields in the technical specifications for the SSD2 table (EFSA, 2017) were not applicable for records reporting results of some analyses, in particular palynological, parasite and pathogen analyses. These fields were left empty; Table VII: Colony observation, reporting observations of honey bee waggle dances from observation hives. Orientation denotes the angle of the waggling phase relative to the vertical axis on the comb. Direction denotes the actual direction in the landscape, as calculated from the orientation of the waggle dance. In all the csv files, columns with the suffix "_desc" have been included, where relevant, to include the name corresponding to the EFSA controlled terminology used in the previous column (e.g. resUnit contains EFSA term codes while resUnit_desc contains the term names). Data storage All data collected during the project was stored in a relational database. The database was developed in .NET Entity Framework Core, ran on a PostgreSQL, and was hosted by Amazon Web Service during the whole duration of the project development. Data could be imported or entered manually in the database through a web form. Administrators could create new users and administrators, new sites, and new colonies, i.e., administrators were allowed to enter or change data of all tables. Users were allowed to enter data, and could view, retrieve, and modify their own data of all tables, except for Table III (description of experimental colonies). Administrators could view and retrieve all data. Data was retrieved in CSV and XML formats, and were structured to secure a smooth transmission of data to the Data Collection Framework of EFSA. Furthermore, data flow from the field data collection to the development of ApisRAM was secured by direct communication between the field and modelling teams. Version 2 contains the UTM coordinates in tables Sites, Polygons and Resource providing unit.

Højmoser repræsenterer de største tørvedækkede områder i Danmark, og en undersøgelse af disse områder kan belyse menneskelig indflydelse på de danske tørvejorder. Vi har studeret den menneskelige indflydelse ved Store Vildmose og anvendt arealanvendelsen fra topografiske kort af forskellig alder (høje målebordsblade, lave målebordsblade og 4-cm-kort), punktopmålt højdeinformation fra bymålingerne, en moderne LiDAR (Light Detection and Radar) baseret højdemodel samt boredata, som giver information om tørvens tykkelse. Formålet med undersøgelsen har været at kvantificere ændringer i Store Vildmoses overfladehøjde forårsaget af både tørvedannelse, tørvesætning (material-isering og afvanding) og tørvegravning fra 1880 til 2010.

River valley bottoms have hydrological, geomorphological, and ecological importance and are buffers for protecting the river from upland nutrient loading coming from agriculture and other sources. They are relatively flat, low-lying areas of the terrain that are adjacent to the river and bound by increasing slopes at the transition to the uplands. These areas have under natural conditions, a groundwater table close to the soil surface. The objective of this paper is to present a stepwise GIS approach for the delineation of river valley bottom within drainage basins and use it to perform a national delineation. We developed a tool that applies a concept called cost distance accumulation with spatial data inputs consisting a river network and slope derived from a digital elevation model. We then used wetlands adjacent to rivers as a guide finding the river valley bottom boundary from the cost distance accumulation. We present results from our tool for the whole country of Denmark carrying out a validation within three selected areas. The results reveal that the tool visually performs well and delineates both confined and unconfined river valleys within the same drainage basin. We use the most common forms of wetlands (meadow and marsh) in Denmark’s river valleys known as Groundwater Dependent Ecosystems (GDE) to validate our river valley bottom delineated areas. Our delineation picks about half to two-thirds of these GDE. However, we expected this since farmers have reclaimed Denmark’s low-lying areas during the last 200 years before the first map of GDE was created. Our tool can be used as a management tool, since it can delineate an area that has been the focus of management actions to protect waterways from upland nutrient pollution.

Agricultural land use and population density have been increasing around the world. Determining if physical geography is a driving factor of historical change on a larger scale has received little research interest in the past outside local-scale case studies. The aim of this study was to model historical agricultural development and population density throughout Denmark using geographically weighted regression with environmental variables and data for parishes from 1860 to 1890. We analysed rye production, sheep count, and population density on the national scale. The incorporated variables were selected to represent aspects of soil, climate, and topography. Models for rye and sheep had high explanatory power (global R-2: between 0.60 and 0.68) for both time periods whereas the model for population density had low explanatory power (global R-2: 0.09 in 1860 and 0.25 in 1890). The results indicate that historical development in agricultural geography can be explained using physical geography. However, population density is more complex due to influences of industrialization, culture and scalar structure. This questions the classical understanding that soil quality is a strong determinant of population density on its own in Denmark. We instead argue that soil quality has a dynamic multidirectional interplay with human and agricultural activity.

Predicting wheat yield is crucial due to the importance of wheat across the world. When modeling yield, the difference between potential and actual yield consistently changes because of advances in technology. Considering historical yield potential would help determine spatiotemporal trends in agricultural development. Comparing current and historical yields in Denmark is possible because yield potential has been documented throughout history. However, the current national winter wheat yield map solely uses soil properties within the model. The aim of this study was to generate a new Danish winter wheat yield map and compare the results to historical yield potential. Utilizing random forest with soil, climate, and topography variables, a winter wheat yield map was generated from 876 field trials carried out from 1992 to 2018. The random forest model performed better than the model based only on soil. The updated national yield map was then compared to yield potential maps from 1688 and 1844. While historical time periods are characterized by numerous low yield potential areas and few highly productive areas, current yield is evenly distributed between low and high yields. Advances in technology and farm practices have exceeded historical yield predictions, mainly due to the use of fertilizer, irrigation, and drainage. Thus, modeling yield projections could be unreliable in the future as technology progresses.

Application of mineral nitrogen (N) fertilizer is not uniform at the field scale when applied with disc spreaders. Some of the distribution variation is due to potentially predictable mechanisms, but the variation is also influenced by stochastic events and multiple interactions, making each combination of field, spreader and application event a unique case. This may be one of the reasons why the agronomic, economic and environmental consequences of uneven fertilizer distribution at the field scale remains poorly understood and insufficiently quantified. Previous studies have addressed the problem statistically or empirically in regularly shaped experimental plots or a subsection of a field. While these efforts provide valuable contributions to our understanding, they neglect critical areas of the field with regards to fertilizer application: Wedges, interfaces between the in-field and the headland and in-field obstacles. In this modelling study, we present an attempt to describe the variation in fertilizer application at the field scale and assess the consequences on winter wheat grain yields, grain N yields and N losses to the environment. By combining GIS and agroecosystem modelling, we assess yield and environmental effects of two spreaders with different working widths (24 and 48 m) in four field polygons selected to represent a relevant span with regard to size and geometry for Danish conditions. The effects of the combinations were assessed for two soils, a coarse sandy soil and a sandy loam. Both accuracy (average N input rate relative to target) and precision (evenness of distribution) was found to decrease in small (4 - 6 ha) and geometrically irregular fields compared to large (40 - 50 ha) and regular fields. Increasing the working width from 24 to 48 m increased the variation in small fields, but not in large. Grain yields were negatively affected by distribution variation, while there was a poor correlation with average applied N rate. In contrast, grain N yields were insensitive to distribution variation, but showed strong correlation with average N input rate. N leaching was affected by both the amount and distribution of applied N. Across all field x spreader x soil scenarios, field scale grain yields decreased between 0-877 kg DM/ha and annual N leaching increased by up to 9 kg N/ha when fertilizer application was simulated with a spreader rather than even application. In general, the agronomic results from the large and regular field resembled the results from previous studies much better than results from the smaller and/or irregular fields. If not accounting for the impact of field size and shape distribution in the landscape, the effects of uneven N application, and thereby the potential gains by improving spreader performance, may therefore be underestimated.

Defining homogeneous zones by soil, climate, and landscape is a first step in creating terroir units. Applications in delineating homogeneous zones are commonly interested in all land use types (urban, forest, agriculture, etc.) or vegetation specific. Terron mapping is one method for creating non-vegetation homogeneous zones specifically for agricultural management and environmental assessment. Previously, terrons were defined as areas similar in soil and landscape. In this study, we implement climate as a key component in terron delineation and a workflow is developed that automates the necessary steps in generating and mapping terron classes. The workflow is flexible by allowing the user to define different input variables, method of modeling, output resolution, and hierarchical terrons. To assess the workflow, terron classes are modeled and mapped in Denmark: 1) using fuzzy c-means with various soil and climate gridded covariates to identify national regions, and 2) using k-means to develop regional terrons within the national regions based on soil and landscape gridded covariates. The elbow method was applied to determine the optimal number of national terron classes, dividing Denmark into three national regions. Each national region was further divided into nine regional terrons. The resulting regional terron map is comprised of 27 terrons. National and regional terrons are defined at 304 m and 30.4 m resolutions, respectively. A dendrogram produced from regional terron centroids was used to generalize terrons into groups of three to five terrons. From the generalization, seven groups are used to determine terrons that are most promising for crop production. The regional terron map constitutes a useful tool enabling future land-use management decisions and the development of terroirs for Danish crops.

Spatial assessment of terroir is creating a new possibility for enhancement of high quality agro-food product and to minimize negative environmental effects such as soil degradation and associated risks. The classification and mapping of particular terroir units could be a competitive marketing tool with a major impact on farmers' incomes. For this purpose, Cane and McBratney (2005) proposed the terron concept to establish combined soil and landscape entities as the first investigative step to identify terroirs. The main objective of the present work was to assemble various environmental factors (i.e. soil, terrain and climate), to identify and then to map terrons in Denmark. First, for representing soil factors, a national soil spectral library was utilized to measure taxonomic distances between 34 Danish reference soil profiles and the Danish national soil profile database (586 soil profiles). Second, the terrain and climate factors for each soil profile location were then compiled as represented by relative slope position, valley depth, valley bottom flatness, vertical distance to the channel network, number of frost days, annual number of growing days, global solar radiation, and precipitation. Third, nine Danish terron classes were established by fuzzy c-means clustering based on an integrated matrix including all soil, terrain and climate factors whereby each terron class is characterized by soil, terrain and climate as a whole entity. Finally, the spatial distribution of Danish terrons was mapped using Cubist regression rules. The results were compared with a soil map derived from the same profile database. We concluded that the map of terrons described natural environment quantitatively and formally in terms of soil, landscape and climatic information better than just a soil class or soil attribute map. Further investigations are needed to discover whether the tenon classes give better predictions of landscape-dynamic processes and allow better management options than soil alone. This study also demonstrated several advantages of using soil spectral data and ancillary data to identify and map terrons. The next step will be to validate the tenon map by incorporating crop yield data and social factors to delineate natural Danish terroir units.

Terrain attributes are commonly used as predictors of soil organic carbon (SOC) in digital soil mapping. However, there are no fixed rules in the selection of suitable grid size and models with different attribute combinations. Past studies have used a few empirical, as well as pedological guidelines to determine scale dependency of terrain attributes on SOC prediction. The aim of this paper was to evaluate the scale dependency of terrain attributes using varying grid sizes and select the most important attributes and optimum grid size for SOC prediction. A 7500 km(2) area located in Denmark was selected; a total of 2,514,820 prediction models were generated in Cubist data-mining tool in which 8570 SOC observations and 22 terrain attributes at 71 different grid sizes ranging from 12.8 m to 2304 m were used as inputs. Terrain attributes were derived from the Light Detection And Ranging (LiDAR) based digital elevation model (DEM) (1.6 m x 1.6 m grid size) and was subsequently resampled to different resolutions by simple mean aggregation. Relative importance and usage of each terrain attribute in each prediction model were computed and only the top 5 attributes were reported for different attribute combinations. The results showed that the relative contribution of terrain attributes to predict SOC distribution varied by grid sizes, and by grid size and attribute combinations. Overall, Relative Slope Position (rsp), Channel Altitude (chnl_alti), and Standard Height (standh) were the three most important terrain attributes in the five-attribute-model at all grid resolutions and the remaining two attributes being Normalized Height (normalh) and Valley Depth (vall_depth) at resolutions finer than 30 m, and elevation and Channel Base (chnl_base) at resolutions coarser than 30 m. The models at 88 m and 92.8 m grid size (nearest to the 90 m SRTM data resolutions) and 30.4 m (nearest to the 30 m TM satellite image resolution) were validated. We observed that the model performance was dependent on grid size, and by attribute combinations. For example, for a 4-attribute model that used rsp, chnl_alti, elevation and vall_depth, the best performance was for 30.4 m compared to 88 m and 92.8 m grid sizes. We found that for modeling SOC distribution, the three terrain attributes rsp, chnl_alti, and standh were found to be the most important at all resolutions and should be considered as important variables in future SOC modeling studies in young moraine landscapes.

Water erosion on agricultural land and sediment delivery to streams are a major threat to soil productivity and surface water quality. Climate change and different national and international societal drivers now require Denmark to take action to protect soil and water resources. In this study, we adapted the spatially distributed sediment transport model WaTEM using the best data available at national scale. To calibrate and validate the model, sediment yield data from 31 catchments and 189 slope units in Denmark were compared with the model output, which was produced at a fine spatial resolution of 10 x 10 m. Residual analysis and cross-validation were used to identify potential catchment outliers and assess model robustness. We obtained a median Nash & Sutcliffe model efficiency range of 0.06-0.6 for the Danish environment Based on the equifinality concept, an ensemble of 100, NSE-weighted model realisations for acceptable transport capacity coefficients was used to assess model uncertainty. The comparison between rill survey data and predicted values indicates that although the model captures well the spatial variability in erosion, it may underestimate the long-term average of soil erosion. Based on the modelling, 71% of the agricultural land in Denmark is mapped as stable in terms of the amount of erosion and deposited material. Overall, 6.1% of the farmland is estimated to have unsustainable erosion and 0.9% of the farmland exceeds erosion rates of 7.5 t ha(-1) a(-1). Sediment export into surface water in Denmark equals 92,000 t a(-1), corresponding to an average sediment yield of 2.7 t km(-2). The performance of WaTEM is considered satisfactory in this study. Importantly, modelled water erosion exceeds the perceived erosion risk in Denmark. Strengthened by distributed uncertainty assessment at national scale, our study provides an important national knowledge base for engaging land users and regulators in the process of targeted erosion mitigation planning that is required to comply with national and EU regulation. Future investigations concerning the deviation between predicted and observed data and specific catchment parameters for the non-behavioural catchments are required as well as studies that include the establishment of catchment sediment budgets. (C) 2019 Elsevier B.V. All rights reserved.