An Integrated Assessment of European Soil Health and Restoration potential

An Integrated Assessment of European Soil Health and Restoration potential | 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 Article An Integrated Assessment of European Soil Health and Restoration potential Irene Calderón-Sanou, Cristiano Ballabio, Claudia Breitkreuz, and 10 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8660880/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Soils host a significant proportion of biodiversity on Earth providing ecosystem functions vital to human well-being, making it imperative to include them and their ecological features when addressing sustainability goals. We performed a comprehensive assessment of soil health across Europe by explicitly integrating biotic and abiotic indicators alongside soil degradation processes. We further identified areas with high restoration potential, quantifying the potential positive changes achievable when mitigating degradation processes. Our results show that 93% of soils in Europe are either degraded (62%) or in a moderate state (31%), with only 7% having a good (6.5%) or high (0.5%) health status. We found Southeast Europe to have the highest restoration potential, particularly for forests and annual crops. By providing spatially explicit indicators of soil health and restoration potential, our approach offers valuable guidance to support sustainable soil management and inform policies aimed at enhancing soil health across Europe. Biological sciences/Ecology/Ecological modelling Biological sciences/Ecology/Ecosystem ecology Earth and environmental sciences/Ecology/Macroecology soil biodiversity soil functioning biotic indicators soil degradation machine learning Figures Figure 1 Figure 2 Figure 3 Introduction Soils are essential components of terrestrial ecosystems, hosting 59% of Earth's species 1 and providing critical functions such as nutrient cycling, carbon sequestration, water purification and food production 2 . However, soil biodiversity and its ecosystem functions face ongoing threats from a diverse range of degradation processes including land-use and climate change, pollution, overexploitation, and urbanisation 3–6 . Despite the significant consequences of inaction and a growing recognition of its importance, soils continue to be underestimated and frequently ignored in current nature conservation and environmental policies, failing to achieve policy relevance comparable to air or water 7,8 . This is particularly true for the European Union (EU), where the proposal for a framework directive to protect soils in 2007 was blocked by a minority of member states 9 , and new soil protection legislation is now under debate as part of the EU Green Deal. A key barrier to soil legislation is the perceived lack of actionable data to inform policymaking 8 . Although national-scale data remain scarce, ongoing efforts at the European level to standardize data collection have created the opportunity to establish comprehensive soil health assessments that integrate soil biodiversity and functioning 10,11 . These assessments can be directly translated into decision support tools, guiding policymakers in the development of effective strategies for the protection and sustainable use of soils. An effective assessment of soils requires an integrative approach that considers multiple ecological dimensions 12 and provides insights into the consequences of environmental impacts 13 . The concept of soil health refers to the physical, chemical, and biological condition of the soil determining its capacity to function as a vital living system and to provide ecosystem services 14 . It aims at evaluating and enhancing the sustainable use of soils 15,16 and supports the design of legislation that fosters healthy soils 7,8,10 . A healthy soil thus becomes a sociopolitical priority, standing in contrast to soil degradation, which is defined as the diminishing capacity of the soil to provide ecosystem goods and services 17 . Soil health can be assessed using a classification system along a gradient from ‘degraded’ to ‘healthy’, referred to here as soil health status. Such classification offers an operational system for integrating soil health into complementary policies (Fig. 1). For instance, identifying degraded soils can be directly used to guide restoration priorities by offering differentiated paths for improvement. Furthermore, distinguishing soils with a moderate health status from those with good status allows for prioritizing mitigation efforts or establishing specific conservation measures, as soils with moderate status may be more vulnerable to degradation. By identifying soils in good condition, nature conservation strategies may be prioritised, not only through protected areas but also through legislation that promotes sustainable land use, such as in agriculture or managed forests. While similar operational frameworks have been successfully implemented for other compartments (e.g., water ecosystems through the Water Directive framework 18 ), challenges arise for soils due to the absence of standardised measured ecological criteria and difficulty establishing reference conditions accounting for soil variation across habitats and biogeographical regions. Soil health indicators must encompass physical, chemical, and biological components. Notably, the European Commission and international organizations (e.g., FAO, UNESCO) increasingly recognize soil biodiversity as essential for maintaining ecosystem functioning and soil health 6,11,19 . However, biological aspects are frequently missing or indirectly addressed in assessments 15 , limiting a comprehensive understanding of soil functionality. Current EU soil assessments predominantly classify soils based on soil degradation processes such as erosion, nutrient depletion, carbon losses, and pollution with heavy metal loads 20,21 . These are measured through physical and chemical measurements, such as soil pH, changes in carbon content, nutrient surpluses, elemental analyses, and erosion rates 22 , along with critical thresholds that serve as baselines above or below which each degradation process indicator reduces the soil’s capacity to provide goods and services 20 . While these indicators effectively identify degraded soils, they fail to capture the status of the rest of soils considered ‘healthy’, that is, soils that have not yet crossed degradation thresholds, nor they provide a path for ecological restoration strategies. Biotic indicators, in contrast, can be directly linked to soil biodiversity and their functional roles to support ecosystem functioning. These include diversity metrics, which reflect both the intrinsic value of biodiversity and the distinct contributions of coexisting species to ecosystem functioning 23 , as well as direct functional measurements such as enzymatic activities or microbial respiration associated with nutrient cycling and carbon storage, respectively. By capturing ecological attributes, biotic indicators offer a more comprehensive understanding of soil biodiversity, functioning, and overall soil health 24,25 . However, incorporating them in ecological assessments requires establishing a baseline reference status that allows to detect improvements or degradation, similar to the critical thresholds used for abiotic indicators. For a given indicator at a specific site, negligible deviations from values observed at comparable sites under non-degraded conditions indicate a high soil health status. In contrast, substantial deviations from these reference conditions may reflect soil degradation (Fig. 1). The challenge lies in defining an appropriate reference status for soils, which is inherently complex. Reference conditions are often implicitly associated with a natural state prior to major human disturbance 26 . However, in continents such as Europe, soils have been extensively shaped by long-term human activities, meaning that reference conditions may also encompass non-degraded soils under a range of land uses, including agriculture. This complexity complicates the definition of reference conditions at the continental scale, given the wide diversity of soil types, climates, and land-use histories. Here, we combine data on soil biodiversity and ecosystem functions from recent European monitoring programs (i.e., the LUCAS survey 10 ), along with environmental factors (e.g., climate, soil properties), and soil degradation processes (e.g., erosion, compaction), to develop an integrated assessment of soil health across the European Union. This approach complements the current official EU degradation assessments 20 . First, following the official EU framework, we classified all European soils as ‘healthy’, ‘degraded’ or ‘critically degraded’ , based on whether none, one or multiple degradation processes exceeded critical threshold documented in the literature 21 . The novelty of our approach lies in further refining the ‘healthy’ category into more nuanced categories, i.e., moderate, good or high soil health status, using biotic indicators of soil biodiversity and ecosystem functioning. To achieve this, we analysed LUCAS survey data from 881 sites across 25 European countries 10 , and selected 26 biotic indicators representing soil biota and ecosystem functioning (Table 1). These indicators encompassed the diversity of 19 functional groups from all domains of life (e.g., saprotrophic fungi, plant pathogenic protists, bacterivorous nematodes) and key ecosystem functions related to microbial activity, carbon cycling, and nutrient cycling. Random forest models were then used to predict each biotic indicator’s current value across the EU based on environmental conditions and soil degradation processes. To assess soil health status and soil restoration potential, we simulated a non-degradation reference scenario, defined as conditions in which degradation-related variables were set to zero or to their minimum observed value 26 (Supplementary Table 1), while all other predictors were kept unchanged. This was achieved by applying the trained model for each indicator. ‘Healthy’ soils were then classified into moderate, good or high status (Fig. 1) based on the magnitude of deviation in the biotic indicators between the current condition and the non-degradation reference. This deviation was estimated per land-use type and biogeographical region to account for local environmental specificities. The overall soil health was determined using ‘one out, all out’ principle, a widely used methodology in EU environmental management, where the worst status among the seven biotic indicator categories (Table 1) dictated the final classification. Complementary sensitivity analyses were conducted to assess the sensitivity of the soil health assessment to methodological choices. Results The assessed soil biotic indicators, including soil biodiversity groups and their associated functions, exhibited contrasting spatial patterns on a European scale, reflecting the variability in ecological characteristics (spatial predictions for each biotic indicator are available in Supplementary Data 1). This also implies that important trade-offs may arise when considering various biological dimensions of soil systems, particularly when considering restoration and degradation mitigation strategies. Overall, the biotic indicators considered were predominantly influenced by climate, nutrient availability, and soil physico-chemical factors (Supplementary Fig. 1). Additionally, indicators linked to biodiversity were also highly influenced by land-cover type. Interestingly, pH emerged as the sole predominant factor shaping bacterial diversity, while nutrients and soil texture were equally or even more critical than pH for ecosystem functions. Higher temperatures favoured the diversity of most soil functional groups, with some exceptions, such as ectomycorrhizal fungi and herbivorous protists, which exhibited higher diversity in colder regions, along with the ecological functions considered (Supplementary Fig. 2, 3). Nutrient availability strongly influenced most indicators, with several benefiting from increased nutrient levels (natural or otherwise), e.g., the diversity of most trophic groups was higher at lower soil C/N ratio. Likewise, as expected, most ecological function values increased with higher soil organic carbon and nitrogen contents (which are closely correlated Supplementary Fig. 4), given that most functional indicators were associated with carbon and nutrient cycling. (Supplementary Fig. 2, 3). Overall, acidic soils were generally detrimental to both soil biodiversity and functions, except for some groups, such as ectomycorrhizal and root endophytic fungi, and some soil enzymes. Importantly, within a kingdom, the response of functional groups to environmental predictors often diverged, emphasising the need of considering the functional aspects of soil taxa as biotic indicators to gain more insights into soil biodiversity patterns and their associated functions. The accumulation of soil degradation processes negatively impacted most biotic indicators, as evidenced by a decrease in their values when comparing the current state to the non-degradation reference (Supplementary Fig. 5). This highlights how soil degradation can trigger cascading negative system effects with economic and social consequences. Yet, la value of some indicators increased with accumulating soil degradation across land-cover types and biogeographical regions, including the diversity within pathogenic fungi and protists, and most groups of bacteria, basal respiration, microbial biomass, and enzymatic activities related to soil organic carbon degradation. Some biotic indicators exhibited a positive and strong response to specific soil degradation indicators (Supplementary Fig. 6, 7). For instance, plant feeder protists, heterotrophic bacteria, nematodes, and some enzymes, responded positively to heavy metals concentrations, in contrast to microbivorous and photoautotrophic protists or autotrophic bacteria. Among individual effects of soil degradation processes, erosion and heavy metal pollution had the most pronounced impact on biotic indicators (Supplementary Fig. 1). Additionally, although the model did not include an explicit variable to account for nutrient inputs and acidification (see Methods), biotic indicators showed strong responses to nitrogen, phosphorus, and pH (Supplementary Fig. 2, 3), suggesting a significant influence of these degradation processes on biotic indicators. Our results indicate that 62% of European soils are degraded, 31% exhibit a moderate health status, while only 6.5% and 0.5% show good and high health status, respectively (Fig. 2). Within the 62% of soils that were found to be degraded with at least one soil degradation process exceeding the critical limit (Supplementary Table 1), 22% correspond to soils that are critically degraded, i.e., with multiple soil degradation processes. Soil erosion was the soil degradation process affecting the largest area of European soils (33.4% of area), followed by nutrient excess or deficiency (22.3%), soil organic matter deficiency (17.8%) and pollution (10.5%) (Supplementary Fig. 8). Degraded soils were concentrated along the Atlantic coast, especially in the northwest of Europe and in the Mediterranean coasts, mostly in Italy, Greece, and South of Spain (Fig. 2A, 2B), and in a significant proportion of the Continental region, although they were more sparsely distributed. Both Atlantic and Mediterranean soils were mainly affected by nutrient excess or deficiency and erosion (Supplementary Fig. 9). From the multiple land covers considered, croplands and settlements had the highest proportion of degraded area (83.8%, 83.3%, and 79.1%, respectively; Fig. 2C), which means most of these land cover types are affected by at least one degradation process that requires the implementation of mitigation measures. Degradation of soils in croplands was mostly caused by erosion and nutrient excess or deficiency, with organic carbon deficiency being also important for annual crops (Supplementary Fig. 10). In settlements, sealing was the main degradation process threatening soils as expected. Across the EU soils, only 7% were healthy, with 0.5% classified as having ‘high health status’ and 6.5% as ‘good health status’. This indicates that for most European soils classified as ‘healthy’, at least one of the seven biotic indicator categories showed significant biodiversity or functioning declines compared to non-degradation reference. Soils with good status were predominantly located in forests of the Continental, Mediterranean, and Boreal regions. When considering only biodiversity-related indicators, 11% of soils were classified in good conditions (9% with good status and 2% with high status). In contrast, for ecosystem functioning indicators, this percentage rose to 23.6% (11.8% in good status and 11.8% in high status) (Supplementary Fig. 11). Note that these estimates are based on an approach that focuses solely on biodiversity and functioning losses (i.e., negative changes) when assessing deviations from the non-degradation reference (see Methods). When positive changes in biotic indicators, such as an increase in biodiversity, were also considered as potential signs of degradation (e.g., increases due to nutrient inputs), the proportion of soils in good condition (good or high status) dropped to 4.2% overall, and to 8.6% and 14% when considering only biodiversity or ecosystem functioning indicators, respectively. Despite the lower percentages under this alternative approach, the spatial distribution of soil health status categories remained largely consistent (Supplementary Fig. 12). When using a less conservative evaluation than the “one out, all out” principle, based on an additive aggregation across indicators (see Methods), the same soils were reclassified as 4% moderate, 22% good, and 11% high soil health (Supplementary Fig. 13). When accounting also for positive changes in biotic indicators (gains and losses) using this approach, the corresponding proportions were 16%, 17%, and 5%, respectively. It is important to note that, unlike the “one out, all out” rule, this approach relaxes the requirement that all biodiversity and ecosystem-function categories meet the same status, potentially masking deficits in individual components. We evaluated the restoration potential of degraded soils by comparing their current state to the non-degradation reference, with the aim of quantifying potential improvement under restoration and mitigation strategies. For this analysis, we excluded critically degraded soils not because these systems are of lower priority, but because restoration is more tractable where degradation is limited to a single process exceeding the degradation threshold value. Results indicated that Southeast Europe, particularly in Pannonian and Continental lands, but also in the eastern Mediterranean, showed the highest restoration potential for both biodiversity and ecosystem functioning (Fig. 3), especially in forests and annual crops (Supplementary Fig. 14). On average, soils affected by compaction and erosion showed slightly higher potential for biodiversity and ecosystem functioning gains. When also considering potential improvements through decreases in undesirable biodiversity (e.g., groups that proliferate under soil degradation) and ecosystem functioning, the restoration potential patterns remained similar but extended to the Atlantic coast of France and Ireland (Supplementary Fig. 15). Under this broader perspective, annual and permanent crops, rather than forests, exhibited the highest potential for change (Supplementary Fig. 16). Importantly, restoration potential was greater for biodiversity-related indicators than for ecosystem functions. Biodiversity gains were most pronounced in soils degraded by compaction or erosion, while ecosystem function improvements were slightly higher in soils affected by organic matter loss, compaction, nutrient depletion, or acidification. It is important to note that these patterns are indicative, given the strict and utopic nature of the restoration potential measure used here which assumes the reduction or removal of all soil degradation processes, not just those surpassing critical limits. Discussion Our study sheds light on the concerning state of soil health in the European Union by integrating soil biodiversity and functioning indicators into a comprehensive soil assessment. Only 7% of EU soils are classified as healthy, with good or high status, while the remaining 93% are either degraded (62%) or of moderate status (31%). These findings emphasize the urgent need for legislative action to support soil conservation and implement targeted restoration strategies to improve soil health in the EU. By providing spatially explicit insights into how soil biodiversity and functions respond to degradation, our framework helps refine soil health assessments and guide region-specific conservation and restoration efforts. Identifying both areas in need of urgent intervention and healthy soils that should be prioritized for protection enables more effective management strategies, such as including healthy soils in conservation areas 27 or promoting sustainable land-use practices. Trades-off in assessing soil health Evaluating soil health across multiple ecological dimensions provides a more comprehensive understanding of soil conditions, but can also be challenging due to the complex and sometimes contrasting responses of soil biodiversity and ecosystem functions 27 . Most of our indicators exhibited negative responses to soil degradation processes, validating their utility as broad-scale indicators for assessing soil degradation 28 . However, some biotic indicators showed positive responses to specific soil degradation processes. Soil degradation not only diminishes soil biodiversity but also simplifies soil food webs and alters community compositions, resulting in unequal impacts on the diversity and activity of various soil functional groups 4,29 . For instance, while the diversity of more sensitive groups declines due to the accumulation of degradation processes, the diversity of other groups, such as pathogens, can actually increase 30,31 . This may explain why, even when considering biodiversity and functioning ‘gains’ as a metric of degradation, the overall spatial patterns of soil health status remain largely unchanged. In many areas, biodiversity or ecosystem function losses coincide with positive changes in other biotic indicators in response to soil degradation. The variability in biotic indicator’s responses emphasises the importance of incorporating multiple dimensions of soils in soil health assessments. While certain aspects of the soil system may appear favourable, overall soil health can be compromised due to the degradation of other crucial components. Indeed, the low proportion of European soils classified as having a good or high status can be partly attributed to the limited overlap among the upper categories of each biotic indicator (Supplementary Fig. 11). This is a direct consequence of the “one out, all out” principle used in our assessment, which ensures that all components must be in good condition for soil to be classified as healthy. While restrictive, this approach effectively differentiate soil components and identifies specific aspects requiring improvement, ensuring sustainable use and an enhanced overall soil health 15,16 . Yet, looking independently to each component warrant cautious interpretation, acknowledging the multiple intrinsic and complex linkages that exist between biodiversity and ecosystem functions 32,33 . Using a less conservative, additive framework for soil health status classification, substantially shifted the distribution of soils across health categories (Supplementary Fig. 12), revealing higher proportions of soils in good status. Yet, this alternative aggregation allows soils to achieve high overall scores even when one or more components show deficits, which may have implications for ecosystem functioning 34 . While the “one out, all out” classification remains our primary approach, these results provide a complementary perspective that can inform policy and management decisions, highlighting areas where overall soil health is high or low despite localized weaknesses. Policy implications Effectively addressing soil degradation requires integrating quantitative target values into policies 9 . Our assessment identifies priority areas for conservation and restoration (Fig. 1- 3) and highlights the need to implement soil protection strategies beyond areas classified as degraded. The 31% of European soils classified as of moderate health status highly differ from what would be the ideal situation for the sustainable use of soils. This confirms that soil degradation processes, even if not surpassing the critical limits established by the EU, still show indications of deterioration when considering multiple soil biodiversity and ecosystem functioning indicators. For instance, forests are considered one of the healthiest land covers due to their low proportion of degraded soils. However, upon closer examination, many of these soils are only moderately healthy, indicating a potential vulnerability to degradation and the need for protection (Fig. 1). Furthermore, the implementation of soil conservation measures that go beyond the usual mitigation of impact in productive soils should also be considered 35 . In the EU, current conservation policies, such as the Habitat Directive 36 , do not explicitly account for soil biodiversity, leaving a significant gap in management strategies. Incorporating ecological soil indicators into the criteria for assessing habitat conservation status, e.g., within Natura 2000 sites, could pinpoint priority areas for targeted interventions 7,8 . Such interventions may include minimising soil disturbance, enhancing vegetation cover to mitigate erosion, restoring degraded soils, and implementing sustainable agricultural practices in sensitive habitat buffer zones 37,38 . Some land management practices can yield co-benefits—cover crops, for instance, not only reduce erosion but also enhance soil carbon storage and biodiversity 39 . Beyond localized management, addressing soil degradation also demands tackling broader drivers, including climate change, which intensifies processes like erosion through more frequent extreme events 38,40 . While our approach provides spatially explicit measures for potential changes in biodiversity and functioning, effective restoration planning must also consider socio-economical aspects 41 . For example, our results identified areas affected by erosion and compaction as key areas for biodiversity and functioning recovery, yet these processes are particularly difficult to mitigate in the short term, emphasizing the need for long-term, sustained restoration efforts 38 . We also emphasize that, although our analyses focused on degraded soils and excluded critically degraded soils, the latter remain a top priority and require urgent restoration strategies (Fig. 1). However, evaluating their restoration potential is more challenging and less realistic, as recovery trajectories become increasingly complex and uncertain when multiple interacting degradation processes are involved 17,41 . Advancing soil health assessments Our study leverages the most comprehensive biotic dataset currently available at the European scale, highlighting the importance of integrating biodiversity and ecosystem function indicators into soil health assessments. This integration is crucial for soil policy development 16 . Future monitoring efforts should further improve the representation of biodiversity and ecosystem functioning. For instance, biomass or abundance indicators, which can be directly linked to ecosystem functioning 42 , could provide complementary insights. Additionally, incorporating measurements of other key functions, such as water cycling, will enhance soil health assessments as new data become available. Our framework is designed to evolve alongside emerging data and methodologies. In the future, degradation thresholds tailored to specific pedo-climatic conditions may become available and refine the assessments 20 . Despite these anticipated advancements, our approach remains robust, as it is grounded in the best available data and aligns with current governance practices. By incorporating biotic indicators into official soil health evaluations, this study provides a foundation for more comprehensive assessments that can better inform policy and land management decisions. A spatially explicit and comprehensive assessment of soil health, as presented here, directly supports policy agendas in agriculture, forestry, restoration, and nature conservation. It provides essential tools for defining restoration and conservation targets at the European scale. While we proposed a policy indicator for EU-wide assessments, our methodology can be adapted to different geographic scales, enabling other regions to establish comparable baselines for soil health. Similarly, applying this approach at national and local scales, where higher-resolution and more data might be available, could provide more targeted guidance for management and restoration efforts. Methods Data on Soil Biodiversity and Functioning Soil biodiversity and functional data was obtained from the 2018 soil campaign of the Land Use/Land Cover Area Frame Survey (LUCAS) 10 . This dataset comprises data on soil functions and biodiversity from 881 sites in Europe covering most of the terrestrial land covers including annual crops, permanent crops, broadleaved forests, needle leaved forests, grasslands, shrublands and human settlements. In short, soil samples were collected at each location by combining five subsamples taken in different directions around a pre-selected point and covering a depth of 20 cm. A total of 500 g of soil was stored on ice and transported to the JRC within 48 h of collection. The samples were then frozen and sent to the University of Tartu for DNA analyses. A separate sample was analysed by SGS Hungary for physical and chemical soil properties, and another in Germany for soil functions by the German Centre for Integrative Biodiversity Research (iDiv), in Leipzig, and the Helmholtz-Centre for Environmental Research (UFZ), in Halle (see below). Raw environmental DNA data (post sequencing) was obtained from the European Soil Data Centre (ESDAC) platform for Eukaryotes (18S) 43 , Bacteria (16S) and Fungi (ITS) 31 . Biodiversity data – Demultiplexed raw DNA sequence data for eukaryotes (18S) and prokaryotes (16S) were obtained from the Sequence Read Archive (SRA) (BioProject ID PRJNA952168 for 16S and ITS, and PRJNA985135 for 18S). For both datasets, forward and reverse reads were assessed for consistent orientation and corrected as necessary using SeqKit 44 and BBMap 45 tools. Reads re-sequenced due to low coverage in the 16S samples were merged with the original dataset to improve completeness. Primer sequences were verified for their starting position at the beginning of the reads. To maximise the detection of primer sequences that did not start immediately at the beginning, a general 5-base offset was applied across the dataset. Both forward and reverse primer sequences were then trimmed from each dataset using cutadapt 46 allowing a maximum of two mismatches per primer. During primer removal, reads shorter than 75 bp were discarded to avoid erroneous taxonomic assignments. Further, reads were end-trimmed with a minimum quality score threshold of 10 to ensure optimal merging of paired-end reads. For denoising and amplicon sequence variant (ASV) inference, the DADA2 algorithm 47 was implemented within the Qiime2 framework 48 . Input data from multiple sequencing efforts were categorised by sequencing run before error rate modelling, following DADA2 guidelines. A maximum expected error of 1.0 was used for quality filtering of reads prior to denoising. We implemented a minimum overlap of 12 bp for merging the denoised forward and reverse reads. Following chimaera removal, ASV tables from different sequencing efforts were combined for each dataset. For 18S data, ASVs clustered at 100% similarity were considered as species proxies following 43 . Taxonomic assignment was done using the assignTaxonomy function in DADA2, with the PR2 database 49 (Version 5.0) for 18S rRNA gene sequences and the SILVA database 50 (Version 138.1) for 16S rRNA gene sequences. Additionally, the 18S and 16S datasets were filtered for eukaryotic and bacterial ASVs, respectively. Due to the low representation of fewer than 2000 archaeal taxa, Archaea were excluded from the 16S dataset. Only ASVs present at a minimum of two sites, thus excluding singletons, were retained. ASV detected as potential contaminants based on their abundance in the controls were identified and removed with the contslayer function from the metabaR package 51 . At the end of the bioinformatic pipeline we obtained 90044 bacterial and 156298 eukaryotic ASVs. For fungi, OTU tables with taxonomic annotation were obtained from 31 . In short, PacBio amplicon data for fungal sequences (from the fungal ITS region) were demultiplexed using LotuS2 52 with default options. Paired-end reads were assembled using FLASH 1.2.1083 with default options (minimum overlap 10 bp) and unmerged reads were removed from the final datasets. Data were next processed as described in 53 . OTUs at 98% of similarity were obtained and taxonomic assignment was performed using BLAST + 2.11.091 by running MegaBLAST queries of representative OTU sequences against the updated UNITE 9.192 beta reference dataset. This resulted in 55647 fungal OTUs. OTU/ASV tables for the each kingdom were normalised to comparable sequencing depth (bacteria: 4200 reads; fungi: 1000 reads, protists: 400 reads, metazoans: 200 reads) using the Scaling with Ranked Subsampling (SRS) method 54 . Eukaryote data was separated into animals and protists, and the rest discarded for this study. Protists were classified into functional groups with standardised queries to retrieve the functional information from multiple data sources using the ontology-based data integration pipeline described 55 . From the query, protist ASVs were classified into photoautotroph, zooparasite, microbivore (including bacterivore, protistivore, and fungivore), and herbivore. When a protist ASV was associated with more than one functional group, it was included in each relevant category for ecological assessment. Due to the low taxonomic resolution of the 18S marker for animals, these were classified into key soil taxa, as in 43 , i.e. Arthropoda, Annelida, Rotifera, Tardigrada, and only nematodes were functionally annotated. ASV of nematodes with taxonomic annotation at the genus level were classified into bacterivore, fungivore, herbivore, predator or animal parasite using the NINJA platform 56 . The category of animal parasites was not retained for further analysis due to the low number of ASVs classified in this category. NINJA enables the classification of most nematode families, genera, and species found in soil into feeding types. This classification is based on 57 and additional information from the Nemaplex website (http://nemaplex.ucdavis.edu). Bacteria ASVs were functionally annotated using FAPROTAX 58 . The functional annotations of bacterial taxa were categorized into broad functional groups, including autotrophs, chemoheterotrophs, N-cycling bacteria, pathogens and parasites, and a more general decomposer group. FAPROTAX assigns ecological functions based on phenotype profiles supported by experimental evidence. All references supporting the functions can be found in 58 and the FAPROTAX database is available at: www.zoology.ubc.ca/louca/FAPROTAX. ASVs associated with autotrophy and phototrophy were categorised as autotrophic bacteria, including both oxygenic and anoxygenic phototrophic taxa, and all ASV from the phylum Chloroflexi. Oxygenic phototrophs perform photosynthesis using light energy to fix carbon, utilizing water (H 2 O) as an electron donor while producing oxygen as a byproduct. In contrast, anoxygenic phototrophs fix carbon without producing oxygen, utilising varied electron donors such as H 2 S, H 2 , and Fe 2+ . Chemoheterotrophs, which derive energy by oxidising organic compounds such as glucose, include aerobic chemoheterotrophs that operate in the presence of oxygen and anaerobic chemoheterotrophs that function in its absence. Bacterial ASVs assigned to any kind of chemoheterotrophy were classified under this group. N-cycling bacteria are involved in various steps of the nitrogen cycle, including nitrogen fixation (conversion of N 2 to NH 3 ), nitrification (oxidation of NH 3 to NO 2 - and NO 3 -), and denitrification (reduction of NO 2 - and NO 3 - to N 2 ). ASVs linked to such functions were categorised under N-cycling bacteria. Bacteria identified as pathogens and or parasites of plants, humans, animals, and other microbes were grouped as pathogens and parasites. ASVs that did not receive any of the above-mentioned functional assignment from FAPROTAX, were classified in the category of decomposer bacteria, due to the significance role that overall microbial diversity play on soil organic matter decomposition 59 . Functional annotations for fungi were obtained from 31 , and were retrieved using FungalTraits 60 . Similarly, the functional annotations of fungal taxa were categorized into multiple broad functional guilds. The classification encompassed various guilds of fungal plant pathogens, including those associated with roots, leaves, fruits, and seeds, as well as pathogens specific to algae, wood, and moss. OTUs were categorised as saprotrophs reflecting their decomposition roles and including saprotrophs with substrate preferences, e.g., litter, dung, wood. Furthermore, mycorrhizal associations were specifically differentiated based on their primary ecological interactions, between Arbuscular Mycorrhizal Fungi (AMF) and ectomycorrhizal fungi. Ecosystem functioning data – LUCAS samples transported to iDiv, Leipzig, Germany, were analysed for the measurement of potential basal respiration by O2-microcompensation 61 , and microbial biomass by substrate-induced respiration 62 . The enzymatic potential of acid phosphatase (phosphorus mineralization), β-glucosidase and cellulase (cellulose degradation), N-acetylglucosaminidase (chitin degradation), and xylosidase (hemicellulose degradation) were determined in the UFZ, in Halle. Determination of the activities of the hydrolytic enzymes was based on 4-methylumbelliferone (MUF)-coupled substrates (4-MUB-b-D-cellobioside, 4-MUB-b-D-xyloside, 4-MUB-N-acetyl-b-D-glucosaminide and 4-MUB-phosphate), at a pH of 5 at 25°C 63,64 . The method employed standardised conditions, mitigating variations in temperature and moisture to enable direct comparison of enzymatic potentials across diverse samples. Refer to 65 for more details on the methods used to measure potential respiration and microbial biomass and to 66 for methods used to measure enzymatic activities. Biotic indicators A Biotic Indicator is any living organism, its parts, its products (e.g. enzymes), or biological processes that can be used to assess the quality of the environment 28 . Building on the definition of soil health, our chosen indicators aimed to represent soil biodiversity and the associated ecosystem functions. Here, using the latest measured biotic data available at the European scale (see previous section), we defined 26 biotic indicators representing the diversity of soil organisms (19 biotic indicators) and the ecosystem function driven by soil organisms (7 biotic indicators) (Table 1). Indicators of soil biodiversity : To quantify soil biodiversity, we estimated the alpha diversity of the 19 assigned functional groups across the kingdoms Bacteria, Fungi, Protist and Metazoa. The effective number of OTUs, calculated as the exponential of the Shannon index (Hill numbers 67 ), was used as a measure of alpha diversity, ensuring better coverage across samples 68 . A functional approach allows for a more precise understanding of the relationships between habitat characteristics, disturbances, and ecosystem functions, providing more effective, reliable, and informative ecological indicators of system health 25 . Here, the diversity within a functional group serves as a measure of functional similarity 23 , as it reflects the extent to which species within a group share ecological roles while also accounting for the unique contributions of each species. Changes in functional similarity can influence ecosystem resilience and overall functioning, making it a relevant biotic indicator. An exception was made for most Metazoa, where taxonomic groups were used instead of functional groups due to the limited resolution of the marker (i.e., rotifers, tardigrades, arthropods, annelids). However, model performance for these taxonomic groups fell short of the established standard, leaving only the functional groups of nematodes as biotic indicators for the analyses (Supplementary Fig.18, Table 1). Indicators of ecosystem functioning : The measurements performed on soil samples (ecosystem functioning data) were used as biotic indicators serving as proxies for key soil ecosystem functions. Basal respiration and microbial biomass were interpreted as biotic indicators of soil microbial activity. While microbial activity itself is not an ecosystem function per se, microbial biomass and respiration reflect processes related to carbon storage, decomposition, and organic matter turnover, which are fundamental to soil ecosystem functioning. Enzymatic potentials were also used as biotic indicators of carbon and nutrient cycling, with xylosidase, β-glucosidase, and cellulase representing carbon cycling processes, while N-acetylglucosaminidase and acid phosphatase were associated with phosphorus and nitrogen cycling (referred to as nutrient cycling) 69,70 . Predictor variables and Data sources Environmental data – To model each biotic indicator across Europe, we used a comprehensive collection of environmental datasets from different repositories, including CHELSA (bioclim+ 71 ), the European Soil Data Centre (ESDAC) 72 , the European Digital Elevation Model (EU-DEM) 73,74 , and Land Cover databases 75,76 (the complete list of variables and sources is in Supplementary Table 2). By incorporating variables such as climate, soil physico-chemical properties, land cover, and topography, our analyses aimed to capture the complex interplay of factors that explain changes in the values of each biotic indicator at the continental scale. For model training, we obtained soil physical and chemical properties, as well as land cover, directly from the LUCAS survey, as they were measured in the same samples. For other variables, we extracted values from the aforementioned environmental datasets using the sample coordinates. To project data across Europe, we used the same datasets and maps available from the ESDAC platform for variables measured in the LUCAS survey. Soil degradation processes data – Building upon the EU Soil Observatory (EUSO) degradation assessment framework and data, we selected the following variables to represent soil degradation processes: soil erosion (measured as a combination of multiple indicators related to erosion processes by wind 77 , water 78 , tillage 38 or crop harvesting 79 ); concentration of heavy metals including copper 80 , mercury 81 , cadmium 82 , and zinc 83 , as indicators of soil pollution; soil compaction 84 ; and, sealing area for soil sealing (Copernicus). No specific variable was explicitly selected as an indicator of soil nutrient imbalances (e.g., phosphorus deficiency/excess, nitrogen surplus) or acidification (critical pH levels for crop production), as these factors were inherently accounted for by the environmental variables: phosphorus content, nitrogen content, and pH. Their contribution to soil degradation arises when these values exceed or fall below critical thresholds. This aspect was later addressed when predicting the non-degradation reference scenario (explained below). Modelling framework A stacked ensemble modelling framework was implemented with the aim to predict biotic indicators across the EU under both current conditions (current state) and a reference baseline (non-degradation reference). Each biotic indicator was modelled independently using an ensemble approach, and the resulting models were stacked to produce comprehensive predictions at the European scale (Fig. 4 in Extended Data). Data preparation – For each biotic indicator, a dataset was prepared containing the biotic indicator value along with the predictor values, measured in-situ or extracted, across the 881 samples (see Predictor variables and Data sources section). The dataset was pre-processed by removing missing values and one-hot encoding categorical variables where applicable. Missing values resulted either from the removal of a sample during the bioinformatic filtering of biodiversity data or from missing information for one of the predictors. The final dataset sizes, or the number of samples used to fit each model, were as follows: 827 for biotic indicators related to ecosystem functioning and to bacteria, 734 for biotic indicators related to fungi, and 597 for biotic indicators associated with protists or metazoans. Model training – The Random Forest algorithm was selected to model biotic indicator values in response to 25 environmental variables and 7 variables representing soil degradation processes. Random Forest constructs decision trees through recursive binary splitting, enabling it to capture complex, higher-order interactions among predictors without explicitly defining them. This capability makes it particularly effective for ecological modelling. To ensure robust predictions and account for spatial autocorrelation, a spatial block cross-validation approach was adopted 85 . Concretely, the dataset was divided into ~100×100 km spatial blocks for cross-validation. These blocks were randomly partitioned into five subsets (folds), with each subset serving as a validation set once while the remaining four were used for training. This process resulted in an ensemble of five Random Forest models per biotic indicator. Model parameters were optimized to improve predictive accuracy, using mean squared error (MSE) as the criterion. MSE quantifies the average squared difference between observed and predicted values, with lower values indicating better model performance. Key hyperparameters, including maximum tree depth and the number of features considered at each split, were fine-tuned using a holdout validation set. Predictive performance on the validation dataset was evaluated using R-squared (R²) and Spearman’s rank correlation coefficient (⍴) between observed and predicted values (Supplementary Fig. 17). The models were implemented in Python using scikit-learn 86 . Biotic indicators with both an R 2 lower than 0.5 and a Spearman’s ⍴ lower than 0.70 were removed from the analyses (Supplementary Fig. 18). These thresholds were chosen based on a visual assessment of overall model performance to exclude poorly predicted groups while retaining as many relevant groups as possible. Initially, models were fitted for 31 biotic indicators, but five were excluded due to poor model performance, resulting in 26 biotic indicators. Additionally, we applied the same modelling framework and subsequent analyses to model the Shannon diversity of the four soil kingdoms (Bacteria, Fungi, Protista, Metazoa) and one phyla (Nematoda) to capture broad diversity patterns at the kingdom or phylum level (Supplementary Data 1, Supplementary Figs. 2, 5, 6). This resulted in five additional models, which served as a reference but were not included in the soil health assessment. Predictions – The ensemble models per biotic indicator obtained after cross-validation were used to predict both the average and variance of each biotic indicator’s value across Europe under the two scenarios used to assess soil health: Current state: This scenario represents the estimated current state of soil biotic indicators across Europe, considering the current (or available) environmental and soil degradation processes conditions. Non-degradation reference: This scenario simulates a hypothetical condition with unaltered soil conditions, maintaining the environmental factors consistent with the current status while setting all soil degradation processes to zero or minimum values (see Supplementary Table 1 for specific values associated with each degradation process). To represent the reduction of soil nutrient imbalances and acidification, which were not explicitly represented by a variable in the model, phosphorus content values exceeding or falling below a critical threshold (i.e., phosphorus excess and deficiency, respectively) were set to the threshold value (Supplementary Table 1). Similarly, for acidification, pH values below 5 in croplands were adjusted to 5. However, nitrogen surplus and organic carbon deficiency could not be adjusted in the non-degradation scenario, as the model did not include a predictor for nitrogen surplus (only total nitrogen content, which is not equivalent) or for organic carbon deficiency. It is important to acknowledge that this scenario may not be entirely realistic, as soil degradation processes are rarely absent or minimal in real-world situations. However, in the absence of a non-degradation reference for soils in Europe, it provides an approximation of what a standardized reference state would be. We verified the reliability of our projections by assessing the extrapolation of the predictor variables used in the projections. We compared the projected values for the non-degradation reference with the current state, ensuring that the projections were not overly unrealistic in comparison to actual conditions observed in Europe (Supplementary Figure 19). Additionally, we checked the overlap between the training data and the projected environmental conditions using the Schoener's D metric of niche overlap. An 85% overlap in the 'environmental space' between calibration and projection suggests that the model adequately represents the conditions in both scenarios, supporting the reliability of the projections. The predictions were done at 1 km resolution. This resolution is relatively high for maps at the continental scale and is designed to summarize predicted average values based on the expected averages of climate, soil properties, and soil degradation processes at this spatial resolution. Model analyses – Relative importance of each predictor was estimated. While random forest models are robust to collinearity, variable importance can be affected when strongly correlated predictors are included, potentially underestimating their influence. However, correlations among predictors were generally low (Supplementary Fig. 4), and strong correlations (Pearson’s r > 0.65) were mostly within the same category (e.g., climate, nutrients, land cover), minimizing their impact on interpretation. Partial dependence plots were generated using built-in functions in scikit-learn. The uncertainty of the model was evaluated using the coefficient of variation (CV) in the predictions across the folds 87 . We estimated the uncertainty separately for biotic indicators related to soil biodiversity and ecosystem functions (Supplementary Fig. 20). To achieve this, we calculated the CV for each pixel by aggregating the CV values of individual biotic indicators. We then identified pixels with low confidence values corresponding to those above the 0.9 quantile in the summed CV distribution across all pixels and excluded them from further analyses. While ideally, uncertainty should account for the propagation of errors in input covariates, this remains a significant challenge in predictive ecological modelling. We acknowledge this limitation and recognize it as an important area for future research. However, our approach aligns with common practices in ecological modelling and provides a robust assessment given the available methodologies. Soil Health Indicator To assess soil health across Europe, soils were categorized into five soil health categories, representing a gradient from degraded to healthy. These categories included critical degraded, degraded, moderate status, good status, and high status (Fig. 1). Critical degraded soils represent the most degraded or unhealthy category, while high status soils are the least degraded or most healthy category. The classification process consisted of two steps (Fig. 5 in Extended Data): 1) The first step involved separating degraded soils from healthy soils using the official procedure and standards from European institutions (EUSO); 2) In the second step, healthy soils were further classified into moderate, good, or high status using the approach developed in this study and detailed below. First step – First, soil were classified into degraded or healthy by applying the EU Soil Observatory (EUSO) degradation assessment framework and data (https://esdac.jrc.ec.europa.eu/esdacviewer/euso-dashboard/). In this approach, soils with at least one degradation process exceeding the defined critical threshold are classified as degraded. The critical thresholds for each degradation process indicator are summarized in Supplementary Table 1 and follow the same criteria established by the EUSO assessment. We used the available data from the EUSO to classify soils into degraded or healthy based on 19 soil degradation process indicators 20 . The data contains spatially explicit binary information indicating whether a given indicator for a soil degradation process exceed the critical threshold. If one or more indicators within a particular degradation process exceeded the critical threshold, it was considered degraded for that specific degradation process. While the EUSO framework classifies degradation on a gradient based on the number of co-occurring degradation processes, we simplified this into two categories: soils were classified as degraded if affected by at least one degradation process (i.e., at least one indicator for a given process exceeded the critical threshold) and as critically degraded if affected by multiple degradation processes. While striving to maintain consistency with the EUSO assessment, some modifications were made. Firstly, we excluded soil degradation processes that were redundant with our approach using biotic indicators and indirect, such as "potential threats to biological functions". Secondly, we eliminated processes that were only available for specific regions only, e.g., "salinization" in the Mediterranean region to avoid spatial biases in the assessment, or processes related to habitats not considered in our study, such as "peatland degradation”. Third, we also included acidification as a soil degradation process that was made available by European Environment Agency, but which only concerns croplands. In summary we kept the following soil degradation processes and respective indicators: soil erosion including multiple erosion processes by wind, water, tillage or crop harvesting, and post-fire recovery; soil pollution including excess of copper, mercury, cadmium, zinc, and arsenic; soil nutrients including phosphorus deficiency and excess, and nitrogen surplus; soil compaction; soil sealing; soil organic carbon deficiency; and acidification. Despite the differences, the percentage of degraded soils estimated in this study closely matched the EUSO’s published estimates. This suggests that excluding certain degradation processes, such as salinization, may decrease the number of degradation processes affecting an area but is unlikely to alter its classification as degraded. Second step – We further classified the remaining ‘healthy’ soils into moderate, good, or high status in the following way by comparing the deviation of biotic indicators in the current state scenario from the non-degradation reference scenario (Fig. 4 in Extended Data). Biotic indicators were grouped by category (Table 1): four categories for soil biodiversity indicators corresponding to the kingdoms Bacteria, Fungi, Protists, and Metazoa, and three categories for ecosystem functioning indicators corresponding to soil microbial activity, carbon cycling, and nutrients cycling. This categorization was done to balance the weight given to each category and reduce biases related to a specific kingdom or ecosystem function (e.g., there are more functional groups of fungi than bacteria, but we wanted to give equal weight to both kingdoms). Additionally, since functional groups within a kingdom may respond differently to environmental changes or degradation, grouping them allows for a broader understanding of predicted biodiversity or functioning changes within the category. For each category, the percentage of difference between the projected current state and the non-degradation reference was calculated to define the soil health status. We used a multivariate dissimilarity index (Temporal Beta-diversity Index, TBI), suggested by 88 that accounts for gains and losses of individual species or groups of species between two states. This index is also applicable to continuous variables such as our biotic indicator related to ecosystem functioning. The input for each category consisted of two matrices, one for each scenario (current and non-degradation), with sample IDs as rows and biotic indicators within that category as columns, with their respective predicted values. The TBI was then calculated for each biotic indicator category between the current and non-degradation status to assess overall changes in OTUs diversity within functional groups for a specific kingdom, or changes in functions values within a specific ecosystem function. We used the function TBI from the R package adespatial 89 , which additionally provides the associated probability of a site experiencing an "exceptional change", across the evaluated sites. Given the environmental differences existing across Europe that may lead to different order of magnitudes in the biotic indicator changes, we estimated the TBI and the associated probability for each land cover type within a biogeographical region (EEA, 2016: http://www.eea.europa.eu). We used the probability to define the soil helath status for each biotic indicator within each land cover and biogeographical region as follows: Moderate soil health status: The probability that the changes observed for a given biotic indicator between the status is ‘exceptional’ is relatively high (p > 0.67), indicating that the status of the biotic indicator in the site is very distant from the modelled "reference conditions". Good soil health status: The probability that the changes observed for a given biotic indicator between the status is ‘exceptional’ is intermediate (0.66 < p < 0.33), indicating that the site is between a moderate and a high status. High soil health status: The probability that the changes observed for a given biotic indicator between the status is ‘exceptional’ is relatively low (p < 0.33), indicating that the status of the biotic indicator in the site is very similar to the reference conditions. In addition, a sensitivity analysis was conducted to assess the effect of threshold choices by testing alternative, reasonable cut-points of 0.6/0.4 and 0.7/0.3 (Supplementary Fig. 13). We considered changes significant only when degradation processes led to a decrease in biodiversity or a loss of functions, following the official definition of soil health and degradation, which identifies biodiversity loss and ecosystem function decline as indicators of deterioration. However, since positive changes resulting from soil degradation processes may also indicate degradation, we conducted parallel analyses accounting for both positive and negative changes to ensure transparency in our approach (Supplementary Fig.12). At the end, each one of the seven categories of biotic indicators was classified into moderate, good, or high status. Maps of soil health status for each category are provided in Supplementary Fig. 21, 22. The overall soil health status classification for a soil area (1km 2 ) was determined by the element with the worst status out of all the categories of biotic indicators following the ‘one out, all out’ principle used in the Water Directive Framework 18 . To explore the sensitivity of this classification principle, we implemented an alternative additive approach (Supplementary Fig.13). For each soil area and each biotic indicator category, values were scored as 0 = moderate, 1 = good, and 2 = high. The scores were then summed across the seven indicator categories and standardized to a 0–2 range by dividing by the maximum possible total. Soils were subsequently classified into three classes: 0–1 = moderate, 1–1.5 = good, and >1.5 = high soil health. Soil restoration potential Within the framework presented here (Fig. 1), soil restoration refers specifically to the process of recovering soils that are degraded or critically degraded to a state where it is safe and suitable for the intended use. We only focused on degraded soils which are soils with severe loss of function and/or biological activity but with less cumulative impacts compared to critically degraded soils, to estimate the restoration potential. These soils represent more realistic and achievable targets for restoration actions in the short to medium term compared to critically degraded soils. Here, restoration potential was defined as the potential gain in soil biodiversity and functions if soil degradation processes were reduced to the minimum. For this, the TBI was also calculated by category of biotic indicators, but only for the degraded soils, and used as a metric of restoration potential. Higher values of TBI indicate that there is a higher potential of change for a given biotic indicator category if we were able to decrease or remove all soil degradation processes. To assess the overall restoration potential, TBI values were standardized for each category of biotic indicator and then summed, with weights assigned to give equal importance (50%) to soil biodiversity and ecosystem functioning. For the main analyses, we considered only positive changes from the current state to the non-degradation reference, as restoration strategies typically aim to enhance biodiversity and functioning rather than the opposite. However, since soil restoration may also seek to reduce undesirable biodiversity that proliferates under degradation, we conducted a side analysis accounting for overall changes (both gains and losses in biodiversity and functioning) to ensure a more objective assessment (Supplementary Fig. 15, 16). Declarations Data availability Data will be made available in a public repository after acceptance. Code availability Code will be made available in a public repository after acceptance. Acknowledgements We gratefully acknowledge the support of iDiv, which is funded by the German Research Foundation (DFG – FZT 118, 202548816), as well as by the DFG (Ei 862/29-1; Ei 862/31-1). We also acknowledge funding from the Horizon Europe EuropaBon project (grant agreement No. 101003553) to support the research. The LUCAS Survey is coordinated by Unit E4 of the Statistical Office of the European Union (EUROSTAT). The LUCAS Soil sample collection is supported by the Directorate-General Environment (DG-ENV), Directorate-General Agriculture and Rural Development (DG-AGRI), Directorate-General Climate Action (DG-CLIMA), and Directorate-General Eurostat (DG-ESTAT) of the European Commission. References Anthony, M. A., Bender, S. F. & van der Heijden, M. G. A. Enumerating soil biodiversity. Proc. Natl. Acad. Sci. 120 , e2304663120 (2023). Bardgett, R. D. & van der Putten, W. H. Belowground biodiversity and ecosystem functioning. Nature 515 , 505–511 (2014). Rillig, M. C. et al. The role of multiple global change factors in driving soil functions and microbial biodiversity. Science 366 , 886–890 (2019). Wall, D. H., Nielsen, U. N. & Six, J. Soil biodiversity and human health. Nature 528 , 69–76 (2015). Phillips, H. et al. Global change and their environmental stressors have a significant impact on soil biodiversity -- a meta-analysis. Preprint at https://doi.org/10.22541/au.167655684.49855023/v1 (2023). FAO, ITPS, GSBI, SCBD & EC. State of Knowledge of Soil Biodiversity: Status, Challenges and Potentialities, Report 2020. (Rome, 2020). Zeiss, R. et al. Challenges of and opportunities for protecting European soil biodiversity. Conserv. Biol. 36 , e13930 (2022). Köninger, J., Panagos, P., Jones, A., Briones, M. J. I. & Orgiazzi, A. In defence of soil biodiversity: Towards an inclusive protection in the European Union. Biol. Conserv. 268 , 109475 (2022). Paleari, S. Is the European Union protecting soil? A critical analysis of Community environmental policy and law. Land Use Policy 64 , 163–173 (2017). Orgiazzi, A. et al. LUCAS Soil Biodiversity and LUCAS Soil Pesticides, new tools for research and policy development. Eur. J. Soil Sci. 73 , (2022). Romero, F. et al. Soil health is associated with higher primary productivity across Europe. Nat. Ecol. Evol. 8 , 1847–1855 (2024). Eisenhauer, N. et al. The multidimensionality of soil macroecology. Glob. Ecol. Biogeogr. J. Macroecology 30 , 4–10 (2021). Lal, R. et al. Soils and sustainable development goals of the United Nations: An International Union of Soil Sciences perspective. Geoderma Reg. 25 , e00398 (2021). European Commission. Proposal for a Directive on Soil Monitoring and Resilience . (Online resource, https://environment.ec.europa.eu/publications/proposal-directive-soil-monitoring-and-resilience_en, 2023). Lehmann, J., Bossio, D. A., Kögel-Knabner, I. & Rillig, M. C. The concept and future prospects of soil health. Nat. Rev. Earth Environ. 1 , 544–553 (2020). Van Der Putten, W. H. et al. Soil biodiversity needs policy without borders. Science 379 , 32–34 (2023). IPBES, Montanarella, L., Scholes, R. & Brainich, A. The IPBES Assessment Report on Land Degradation and Restoration. https://zenodo.org/record/3237392 (2018) doi:10.5281/ZENODO.3237392. European Commission. Directorate General for the Environment. The EU Water Framework Directive. (Publications Office, LU, 2014). Bagnall, D. K. et al. A minimum suite of soil health indicators for North American agriculture. Soil Secur. 10 , 100084 (2023). Panagos, P. et al. How the EU Soil Observatory is providing solid science for healthy soils. Eur. J. Soil Sci. 75 , e13507 (2024). Prăvălie, R. et al. A unifying modelling of multiple land degradation pathways in Europe. Nat. Commun. 15 , 3862 (2024). Schoenholtz, S. H., Miegroet, H. V. & Burger, J. A. A review of chemical and physical properties as indicators of forest soil quality: challenges and opportunities. For. Ecol. Manag. 138 , 335–356 (2000). Eisenhauer, N., Hines, J., Maestre, F. T. & Rillig, M. C. Reconsidering functional redundancy in biodiversity research. Npj Biodivers. 2 , 1–4 (2023). Guerra, C. A. et al. Tracking, targeting, and conserving soil biodiversity. Science 371 , 239–241 (2021). Zwetsloot, M. J., Bongiorno, G., Barel, J. M., di Lonardo, D. P. & Creamer, R. E. A flexible selection tool for the inclusion of soil biology methods in the assessment of soil multifunctionality. Soil Biol. Biochem. 166 , 108514 (2022). Gibbons, P. et al. Rapidly quantifying reference conditions in modified landscapes. Biol. Conserv. 141 , 2483–2493 (2008). Guerra, C. A. et al. Global hotspots for soil nature conservation. Nature 610 , 693–698 (2022). Büchs, W. Biotic indicators for biodiversity and sustainable agriculture—introduction and background. Agric. Ecosyst. Environ. 98 , 1–16 (2003). Korthals, G. W., Bongers, T., Kammenga, J. E., Alexiev, A. D. & Lexmond, T. M. Long-term effects of copper and ph on the nematode community in an agroecosystem. Environ. Toxicol. Chem. 15 , 979–985 (1996). Qu, X. et al. Deforestation impacts soil biodiversity and ecosystem services worldwide. Proc. Natl. Acad. Sci. 121 , e2318475121 (2024). Labouyrie, M. et al. Patterns in soil microbial diversity across Europe. Nat. Commun. 14 , 3311 (2023). Hagan, J. G., Vanschoenwinkel, B. & Gamfeldt, L. We should not necessarily expect positive relationships between biodiversity and ecosystem functioning in observational field data. Ecol. Lett. 24 , 2537–2548 (2021). Hu, W. et al. Aridity-driven shift in biodiversity–soil multifunctionality relationships. Nat. Commun. 12 , 5350 (2021). Delgado-Baquerizo, M. et al. Multiple elements of soil biodiversity drive ecosystem functions across biomes. Nat. Ecol. Evol. 4 , 210–220 (2020). Panagos, P. et al. Soil Conservation in Europe: Wish or Reality? Land Degrad. Dev. 27 , 1547–1551 (2016). Mammola, S. et al. Towards a taxonomically unbiased European Union biodiversity strategy for 2030. Proc. R. Soc. B Biol. Sci. 287 , 20202166 (2020). Boardman, J. & Vandaele, K. Soil erosion and runoff: The need to rethink mitigation strategies for sustainable agricultural landscapes in western Europe. Soil Use Manag. 39 , 673–685 (2023). Borrelli, P. et al. Policy implications of multiple concurrent soil erosion processes in European farmland. Nat. Sustain. 6 , 103–112 (2022). Hartmann, M. & Six, J. Soil structure and microbiome functions in agroecosystems. Nat. Rev. Earth Environ. 4 , 4–18 (2022). Borrelli, P. et al. Land use and climate change impacts on global soil erosion by water (2015-2070). Proc. Natl. Acad. Sci. 117 , 21994–22001 (2020). Arneth, A. et al. Restoring Degraded Lands. Annu. Rev. Environ. Resour. 46 , 569–599 (2021). Potapov, A. M. et al. Global monitoring of soil animal communities using a common methodology. SOIL Org. 94 , 55–68 (2022). Köninger, J. et al. Ecosystem type drives soil eukaryotic diversity and composition in Europe. Glob. Change Biol. 29 , 5706–5719 (2023). Shen, W., Le, S., Li, Y. & Hu, F. SeqKit: A Cross-Platform and Ultrafast Toolkit for FASTA/Q File Manipulation. PLOS ONE 11 , e0163962 (2016). Bushnell, B. BBMap: A Fast, Accurate, Splice-Aware Aligner. https://escholarship.org/uc/item/1h3515gn (2014). Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17 , 10 (2011). Callahan, B. J. et al. DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods 13 , 581–583 (2016). Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37 , 852–857 (2019). Guillou, L. et al. The Protist Ribosomal Reference database (PR2): a catalog of unicellular eukaryote small sub-unit rRNA sequences with curated taxonomy. Nucleic Acids Res. 41 , D597-604 (2013). Quast, C. et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41 , D590–D596 (2013). Zinger, L. et al. metabaR: An r package for the evaluation and improvement of DNA metabarcoding data quality. Methods Ecol. Evol. 12 , 586–592 (2021). Özkurt, E. et al. LotuS2: an ultrafast and highly accurate tool for amplicon sequencing analysis. Microbiome 10 , 176 (2022). Tedersoo, L. et al. The Global Soil Mycobiome consortium dataset for boosting fungal diversity research. Fungal Divers. 111 , 573–588 (2021). Beule, L. & Karlovsky, P. Improved normalization of species count data in ecology by scaling with ranked subsampling (SRS): application to microbial communities. PeerJ 8 , e9593 (2020). Le Guillarme, N. & Thuiller, W. A practical approach to constructing a knowledge graph for soil ecological research. Eur. J. Soil Biol. 117 , 103497 (2023). Sieriebriennikov, B., Ferris, H. & De Goede, R. G. M. NINJA: An automated calculation system for nematode-based biological monitoring. Eur. J. Soil Biol. 61 , 90–93 (2014). Yeates, G. W., Bongers, T., De Goede, R. G., Freckman, D. W. & Georgieva, S. S. Feeding habits in soil nematode families and genera-an outline for soil ecologists. J. Nematol. 25 , 315–331 (1993). Louca, S., Parfrey, L. W. & Doebeli, M. Decoupling function and taxonomy in the global ocean microbiome. Science 353 , 1272–1277 (2016). Maron, P.-A. et al. High Microbial Diversity Promotes Soil Ecosystem Functioning. Appl. Environ. Microbiol. 84 , e02738-17 (2018). Põlme, S. et al. FungalTraits: a user-friendly traits database of fungi and fungus-like stramenopiles. Fungal Divers. 105 , 1–16 (2020). Scheu, S. Automated measurement of the respiratory response of soil microcompartments: Active microbial biomass in earthworm faeces. Soil Biol. Biochem. 24 , 1113–1118 (1992). Anderson, J. P. E. & Domsch, K. H. A physiological method for the quantitative measurement of microbial biomass in soils. Soil Biol. Biochem. 10 , 215–221 (1978). Breitkreuz, C. et al. Can We Estimate Functionality of Soil Microbial Communities from Structure-Derived Predictions? A Reality Test in Agricultural Soils. Microbiol. Spectr. 9 , e00278-21 (2021). Sinsabaugh, R. L. et al. Soil microbial activity in a Liquidambar plantation unresponsive to CO2-driven increases in primary production. Appl. Soil Ecol. 24 , 263–271 (2003). Smith, L. C. et al. Large‐scale drivers of relationships between soil microbial properties and organic carbon across Europe. Glob. Ecol. Biogeogr. 30 , 2070–2083 (2021). Sünnemann, M. et al. Climate change and cropland management compromise soil integrity and multifunctionality. Commun. Earth Environ. 4 , 394 (2023). Hill, M. O. Diversity and Evenness: A Unifying Notation and Its Consequences. Ecology 54 , 427–432 (1973). Calderón‐Sanou, I., Münkemüller, T., Boyer, F., Zinger, L. & Thuiller, W. From environmental DNA sequences to ecological conclusions: How strong is the influence of methodological choices? J. Biogeogr. 47 , 193–206 (2020). Ullah, S. et al. The responses of extracellular enzyme activities and microbial community composition under nitrogen addition in an upland soil. PLoS ONE 14 , e0223026 (2019). Neemisha & Sharma, S. Soil Enzymes and Their Role in Nutrient Cycling. in Structure and Functions of Pedosphere (eds Giri, B., Kapoor, R., Wu, Q.-S. & Varma, A.) 173–188 (Springer Nature, Singapore, 2022). doi:10.1007/978-981-16-8770-9_8. Brun, P., Zimmermann, N. E., Hari, C., Pellissier, L. & Karger, D. N. Global climate-related predictors at kilometer resolution for the past and future. Earth Syst. Sci. Data 14 , 5573–5603 (2022). Panagos, P. et al. European Soil Data Centre 2.0: Soil data and knowledge in support of the EU policies. Eur. J. Soil Sci. 73 , e13315 (2022). Riley, S. J., DeGloria, S. D. & Elliot, R. Index that quantifies topographic heterogeneity. Intermt. J. Sci. 5 , 23–27 (1999). Weiss, A. Topographic position and landforms analysis. in Poster presentation, ESRI user conference, San Diego, CA vol. 200 (2001). European Commission. Statistical Office of the European Union. Redesign Sample for Land Use/Cover Area Frame Survey (LUCAS) 2018: 2018 Edition. (Publications Office, LU, 2018). Dou, Y. et al. A new European land systems representation accounting for landscape characteristics. Landsc. Ecol. 36 , 2215–2234 (2021). Borrelli, P., Lugato, E., Montanarella, L. & Panagos, P. A New Assessment of Soil Loss Due to Wind Erosion in European Agricultural Soils Using a Quantitative Spatially Distributed Modelling Approach. Land Degrad. Dev. 28 , 335–344 (2017). Panagos, P. et al. A Soil Erosion Indicator for Supporting Agricultural, Environmental and Climate Policies in the European Union. Remote Sens. 12 , 1365 (2020). Panagos, P., Borrelli, P. & Poesen, J. Soil loss due to crop harvesting in the European Union: A first estimation of an underrated geomorphic process. Sci. Total Environ. 664 , 487–498 (2019). Ballabio, C. et al. Copper distribution in European topsoils: An assessment based on LUCAS soil survey. Sci. Total Environ. 636 , 282–298 (2018). Ballabio, C. et al. A spatial assessment of mercury content in the European Union topsoil. Sci. Total Environ. 769 , 144755 (2021). Ballabio, C., Jones, A. & Panagos, P. Cadmium in topsoils of the European Union – An analysis based on LUCAS topsoil database. Sci. Total Environ. 912 , 168710 (2024). Van Eynde, E., Fendrich, A. N., Ballabio, C. & Panagos, P. Spatial assessment of topsoil zinc concentrations in Europe. Sci. Total Environ. 892 , 164512 (2023). European Commission. Joint Research Centre. Institute for Environment and Sustainability. Threats to Soil Quality in Europe. (Publications Office, LU, 2008). Roberts, D. R. et al. Cross‐validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure. Ecography 40 , 913–929 (2017). Pedregosa, F. et al. Scikit-learn: Machine Learning in Python. https://doi.org/10.48550/ARXIV.1201.0490 (2012) doi:10.48550/ARXIV.1201.0490. Thuiller, W., Lafourcade, B., Engler, R. & Araújo, M. B. BIOMOD – a platform for ensemble forecasting of species distributions. Ecography 32 , 369–373 (2009). Legendre, P. A temporal beta-diversity index to identify sites that have changed in exceptional ways in space–time surveys. Ecol. Evol. 9 , 3500–3514 (2019). Dray, S. et al. adespatial: Multivariate Multiscale Spatial Analysis. (2021). Table Table 1. Biotic indicators used for the assessment of the soil health and sources on the methods employed to identify or measure each indicator. Category Biotic indicator Description Source/References Soil biodiversity Metazoa Nematode bacterivores Diversity of nematodes feeding on bacteria NINJA Sieriebrienniko, Ferris & de Goede (2014) Nematode fungivores Diversity of nematodes feeding on fungal hyphae Nematode herbivores Diversity of nematodes feeding on plant roots Nematode predator Diversity of nematodes feeding on other animals or nematodes Fungi Animal parasitic fungi Diversity of fungi that parasite animals FungalTraits Põlme et al. (2020) Arbuscular mycorrhizal fungi Diversity of arbuscular mycorrhizal fungi Ectomycorrhizal fungi Diversity of ectomycorrhizal fungi Mycoparasitic fungi Diversity of fungi that parasite other fungi Plant pathogenic fungi Diversity of fungi that parasite plants Root endophytic fungi Diversity of root endophytes Saprotrophic fungi Diversity of saprotrophic fungi Protist Photoautotrophic protist Diversity of phototrophic and autotrophic protists Multiple sources Microbivorous protist Diversity of protists feeding on bacteria, fungi and/or other protists Plant pathogenic protist Diversity of protists that parasite plants Bacteria Chemoheterotrophic bacteria Diversity of chemoheterotrophs, which derive energy by oxidising organic compounds such as glucose FAPROTAX Louca, Parfrey, & Doebeli (2016) Pathogenic and parasitic bacteria Diversity of bacteria that parasite animals and plants Autotrophic bacteria Diversity of phototrophic and autotrophic bacteria (includes all Chloroflexi) N-cycling bacteria Diversity of bacteria involved in nitrification, denitrification and N-fixation Decomposer bacteria Diversity of all other bacteria found in soils (and not classified in previous groups) Ecosystem functions Soil microbial activity Basal respiration Potential basal respiration measured in lab with O2-microcompensation Scheu (1992) Microbial biomass Microbial biomass by substrate-induced respiration Anderson & Domsch (1978) Carbon cycling Xylosidase Activity potential, i.e. substrate turnover rate at pH 5 and 25°C Breitkreuz et al. (2021) Sinsabaugh et al. 2003 Cellulase Beta-glucosidase Nutrient cycling N-acetyl-glucosaminidase Acid phosphatase Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryInformationEXTENDEDDATAV2.docx EXTENDED DATA SupplementaryInformationV2.docx Supplementary Information Supplementarydatamaps.pdf Supplementary DATA: Maps Cite Share Download PDF Status: Posted Version 1 posted 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-8660880","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":580658525,"identity":"37af2612-ad1c-4c35-b2ce-9000cbd721d1","order_by":0,"name":"Irene Calderón-Sanou","email":"data:image/png;base64,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","orcid":"","institution":"UMR EcoFoG (AgroParisTech, CIRAD, CNRS, INRAE, Université des Antilles, Université de la Guyane), Kourou, France","correspondingAuthor":true,"prefix":"","firstName":"Irene","middleName":"","lastName":"Calderón-Sanou","suffix":""},{"id":580658526,"identity":"1beca8ea-809f-42eb-8687-2fec58318357","order_by":1,"name":"Cristiano Ballabio","email":"","orcid":"https://orcid.org/0000-0001-7452-9271","institution":"Joint Research Centre - European Comission","correspondingAuthor":false,"prefix":"","firstName":"Cristiano","middleName":"","lastName":"Ballabio","suffix":""},{"id":580658527,"identity":"1ea3e8df-213c-4b47-a588-78397ca46054","order_by":2,"name":"Claudia Breitkreuz","email":"","orcid":"","institution":"Julius Kühn Institute (JKI) / Helmholtz-Centre for Environmental Research","correspondingAuthor":false,"prefix":"","firstName":"Claudia","middleName":"","lastName":"Breitkreuz","suffix":""},{"id":580658528,"identity":"418de54b-3e28-47a5-8ebd-fd14cecb10a5","order_by":3,"name":"Nico Eisenhauer","email":"","orcid":"https://orcid.org/0000-0002-0371-6720","institution":"Leipzig University","correspondingAuthor":false,"prefix":"","firstName":"Nico","middleName":"","lastName":"Eisenhauer","suffix":""},{"id":580658529,"identity":"abe2a19a-e2de-491e-89a1-1ce7d5579771","order_by":4,"name":"Arwyn Jones","email":"","orcid":"","institution":"European Commission, Joint Research Centre","correspondingAuthor":false,"prefix":"","firstName":"Arwyn","middleName":"","lastName":"Jones","suffix":""},{"id":580658530,"identity":"babce0d1-96f8-4108-967f-dc78770b4e5a","order_by":5,"name":"Julia Köninger","email":"","orcid":"","institution":"Departamento de Ecología y Biología Animal","correspondingAuthor":false,"prefix":"","firstName":"Julia","middleName":"","lastName":"Köninger","suffix":""},{"id":580658531,"identity":"74a6d226-de1a-425d-afc2-aab61c91926d","order_by":6,"name":"Kirsten Küsel","email":"","orcid":"https://orcid.org/0000-0002-5396-0975","institution":"Aquatic Geomicrobiology, Institute of Biodiversity, Ecology and Evolution, Friedrich Schiller University Jena, Jena, Germany","correspondingAuthor":false,"prefix":"","firstName":"Kirsten","middleName":"","lastName":"Küsel","suffix":""},{"id":580658532,"identity":"304991b1-3957-4b0d-9447-a729f21bac7d","order_by":7,"name":"Alberto Orgiazzi","email":"","orcid":"","institution":"European Commission, Joint Research Centre, Ispra, Italy","correspondingAuthor":false,"prefix":"","firstName":"Alberto","middleName":"","lastName":"Orgiazzi","suffix":""},{"id":580658533,"identity":"ce9f9eb8-0909-49d2-b451-d1ff477efbf8","order_by":8,"name":"Panos Panagos","email":"","orcid":"https://orcid.org/0000-0003-1484-2738","institution":"European Commission, Joint Research Centre","correspondingAuthor":false,"prefix":"","firstName":"Panos","middleName":"","lastName":"Panagos","suffix":""},{"id":580658534,"identity":"a8e1b7cf-8fab-41e8-bb60-d1a3b5d92d2e","order_by":9,"name":"Thomas Reitz","email":"","orcid":"https://orcid.org/0000-0001-6581-6316","institution":"Helmholtz Centre for Environmental Research","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Reitz","suffix":""},{"id":580658535,"identity":"f4ce9cf9-75f5-4640-959f-4b7eff3fdad3","order_by":10,"name":"Sara Si-Moussi","email":"","orcid":"","institution":"Univ. Grenoble Alpes, Univ. Savoie Mont Blanc, CNRS, LECA","correspondingAuthor":false,"prefix":"","firstName":"Sara","middleName":"","lastName":"Si-Moussi","suffix":""},{"id":580658536,"identity":"7c029042-1b84-4799-a1f6-989cb58c3d80","order_by":11,"name":"Bala Singavarapu","email":"","orcid":"","institution":"Institute of Biodiversity, Friedrich Schiller University Jena","correspondingAuthor":false,"prefix":"","firstName":"Bala","middleName":"","lastName":"Singavarapu","suffix":""},{"id":580658537,"identity":"3f301911-20f3-4952-afb7-3dac8ea0ef1a","order_by":12,"name":"Carlos A. Guerra","email":"","orcid":"","institution":"Universidade de Coimbra, Departamento de Geografia","correspondingAuthor":false,"prefix":"","firstName":"Carlos","middleName":"A.","lastName":"Guerra","suffix":""}],"badges":[],"createdAt":"2026-01-21 14:18:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8660880/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8660880/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101280518,"identity":"f82cead5-2bc8-44eb-8660-05a13d28bbdb","added_by":"auto","created_at":"2026-01-28 05:04:21","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":178046,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic representation of the soil health assessment.\u003c/strong\u003e In green, a gradient of healthy soils is shown from “high soil health status”, referring to soils with high biological diversity and activity, and structure, to “moderate soil health status”, referring to still non-degraded soils with moderate levels of biological and functional activity, both estimated when comparing to the same reference system (e.g., a combination of biogeographical region and land-use type). In dark orange and red, the soil degradation gradient is shown from “critically degraded soils”, referring to soils with high levels of degradation and cumulative impacts resulting from severe or continued disturbances to “degraded soils”, referring to soils with severe loss of function and/or biological activity but with less cumulative impacts. Black dashed arrows indicate different strategies for the improvement of soil health and the potential for healthy soils to be sustainably used (e.g., for agricultural production) or to be spared for nature conservation of soil organisms.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8660880/v1/8317299d4ad22cac18cc43a7.png"},{"id":101280519,"identity":"2c1ef27c-4868-4186-81d8-3ceb817b0346","added_by":"auto","created_at":"2026-01-28 05:04:21","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":330928,"visible":true,"origin":"","legend":"\u003cp\u003eSoil health assessment across Europe (A) including the percentage of area in each category of soil health status by biogeographical region (B) and by land cover type (C). Grey areas indicate countries that were not sampled in the LUCAS survey or with missing data for the soil degradation processes evaluated, and regions of low model confidence that were excluded.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8660880/v1/9488f385ee2903158c5cd038.png"},{"id":101297638,"identity":"4db4e535-481f-47f9-af5b-ddccc4f8c636","added_by":"auto","created_at":"2026-01-28 09:28:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":479558,"visible":true,"origin":"","legend":"\u003cp\u003eSoil restoration potential in Europe (A) and the gain potential in soil biodiversity and ecosystem functioning across soils suffering from different degradation factors (B). The gain potential represents the average value across biotic indicators in each category or between the 2 categories (for the total), which were previously standardised per biotic indicator by dividing by the maximum value to ensure comparability across the indicators. The boxes span the interquartile range (IQR), with the vertical black line indicating the median and whiskers extend 1.5 × IQR from the hinge. Grey areas in A indicate countries that were not sampled in the LUCAS survey or with missing data for the soil degradation processes evaluated, and regions of low model confidence that were excluded. Soc_deficit: Organic Carbon deficit.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8660880/v1/ccb5d886135051039fbe267b.png"},{"id":103506208,"identity":"589d2eff-47b4-40c0-8ffd-4d1819f85e9f","added_by":"auto","created_at":"2026-02-26 13:34:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1849369,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8660880/v1/ac4b2bee-81ac-4029-b7d6-441bdc0b8f4a.pdf"},{"id":101297701,"identity":"f6007d32-b62e-4188-acb0-c472a9bdb9f8","added_by":"auto","created_at":"2026-01-28 09:28:40","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":763528,"visible":true,"origin":"","legend":"EXTENDED DATA","description":"","filename":"SupplementaryInformationEXTENDEDDATAV2.docx","url":"https://assets-eu.researchsquare.com/files/rs-8660880/v1/16a4a7589ffcef0c8c4969b7.docx"},{"id":101280522,"identity":"76790f63-e526-4ac7-b68c-43394c52563d","added_by":"auto","created_at":"2026-01-28 05:04:21","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":13129594,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformationV2.docx","url":"https://assets-eu.researchsquare.com/files/rs-8660880/v1/c175e0f09a512c184500d5d4.docx"},{"id":101280523,"identity":"8c2bdd45-6f1c-4d98-a4d9-5a1242516973","added_by":"auto","created_at":"2026-01-28 05:04:22","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":45088070,"visible":true,"origin":"","legend":"Supplementary DATA: Maps","description":"","filename":"Supplementarydatamaps.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8660880/v1/c4120293470108a024a830ee.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"An Integrated Assessment of European Soil Health and Restoration potential","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSoils are essential components of terrestrial ecosystems, hosting 59% of Earth's species\u003csup\u003e1\u003c/sup\u003e and providing critical functions such as nutrient cycling, carbon sequestration, water purification and food production\u003csup\u003e2\u003c/sup\u003e. However, soil biodiversity and its ecosystem functions face ongoing threats from a diverse range of degradation processes including land-use and climate change, pollution, overexploitation, and urbanisation\u003csup\u003e3–6\u003c/sup\u003e. Despite the significant consequences of inaction and a growing recognition of its importance, soils continue to be underestimated and frequently ignored in current nature conservation and environmental policies, failing to achieve policy relevance comparable to air or water\u003csup\u003e7,8\u003c/sup\u003e. This is particularly true for the European Union (EU), where the proposal for a framework directive to protect soils in 2007 was blocked by a minority of member states\u003csup\u003e9\u003c/sup\u003e, and new soil protection legislation is now under debate as part of the EU Green Deal. A key barrier to soil legislation is the perceived lack of actionable data to inform policymaking\u003csup\u003e8\u003c/sup\u003e. Although national-scale data remain scarce, ongoing efforts at the European level to standardize data collection have created the opportunity to establish comprehensive soil health assessments that integrate soil biodiversity and functioning\u003csup\u003e10,11\u003c/sup\u003e. These assessments can be directly translated into decision support tools, guiding policymakers in the development of effective strategies for the protection and sustainable use of soils.\u003c/p\u003e\n\u003cp\u003eAn effective assessment of soils requires an integrative approach that considers multiple ecological dimensions\u003csup\u003e12\u003c/sup\u003e and provides insights into the consequences of environmental impacts\u003csup\u003e13\u003c/sup\u003e. The concept of soil health refers to the physical, chemical, and biological condition of the soil determining its capacity to function as a vital living system and to provide ecosystem services\u003csup\u003e14\u003c/sup\u003e. It aims at evaluating and enhancing the sustainable use of soils\u003csup\u003e15,16\u003c/sup\u003e and supports the design of legislation that fosters healthy soils\u003csup\u003e7,8,10\u003c/sup\u003e. A healthy soil thus becomes a sociopolitical priority, standing in contrast to soil degradation, which is defined as\u0026nbsp;the diminishing capacity of the soil to provide ecosystem goods and services\u003csup\u003e17\u003c/sup\u003e.\u0026nbsp;Soil health can be assessed using a classification system along a gradient from ‘degraded’ to ‘healthy’, referred to here as soil health status. Such classification offers an operational system for integrating soil health into complementary policies (Fig. 1). For instance, identifying degraded soils can be directly used to guide restoration priorities by offering differentiated paths for improvement. Furthermore, distinguishing soils with a moderate health status from those with good status allows for prioritizing mitigation efforts or establishing specific conservation measures, as soils with moderate status may be more vulnerable to degradation. By identifying soils in good condition, nature conservation strategies may be prioritised, not only through protected areas but also through legislation that promotes sustainable land use, such as in agriculture or managed forests. While similar operational frameworks have been successfully implemented for other compartments (e.g., water ecosystems through the Water Directive framework\u003csup\u003e18\u003c/sup\u003e), challenges arise for soils due to the absence of standardised measured ecological criteria and difficulty establishing reference conditions accounting for soil variation across habitats and biogeographical regions.\u003c/p\u003e\n\u003cp\u003eSoil health indicators must encompass physical, chemical, and biological components. Notably, the European Commission and international organizations (e.g., FAO, UNESCO) increasingly recognize soil biodiversity as essential for maintaining ecosystem functioning and soil health\u003csup\u003e6,11,19\u003c/sup\u003e. However, biological aspects are frequently missing or indirectly addressed in assessments\u003csup\u003e15\u003c/sup\u003e, limiting a comprehensive understanding of soil functionality. Current EU soil assessments predominantly classify soils based on soil degradation processes such as erosion, nutrient depletion, carbon losses, and pollution with heavy metal loads\u003csup\u003e20,21\u003c/sup\u003e. These are measured through physical and chemical measurements, such as soil pH, changes in carbon content, nutrient surpluses, elemental analyses, and erosion rates\u003csup\u003e22\u003c/sup\u003e, along with critical thresholds that serve as baselines above or below which each degradation process indicator reduces the soil’s capacity to provide goods and services\u003csup\u003e20\u003c/sup\u003e. While these indicators effectively identify degraded soils, they fail to capture the status of the rest of soils considered ‘healthy’, that is, soils that have not yet crossed degradation thresholds, nor they provide a path for ecological restoration strategies. Biotic indicators, in contrast, can be directly linked to soil biodiversity and their functional roles to support ecosystem functioning. These include diversity metrics, which reflect both the intrinsic value of biodiversity and the distinct contributions of coexisting species to ecosystem functioning\u003csup\u003e23\u003c/sup\u003e, as well as direct functional measurements such as enzymatic activities or microbial respiration associated with nutrient cycling and carbon storage, respectively. By capturing ecological attributes, biotic indicators offer a more comprehensive understanding of soil biodiversity, functioning, and overall soil health\u003csup\u003e24,25\u003c/sup\u003e. However, incorporating them in ecological assessments requires establishing a baseline reference status that allows to detect improvements or degradation, similar to the critical thresholds used for abiotic indicators. For a given indicator at a specific site, negligible deviations from values observed at comparable sites under non-degraded conditions indicate a high soil health status. In contrast, substantial deviations from these reference conditions may reflect soil degradation (Fig. 1). The challenge lies in defining an appropriate reference status for soils, which is inherently complex. Reference conditions are often implicitly associated with a natural state prior to major human disturbance\u003csup\u003e26\u003c/sup\u003e. However, in continents such as Europe, soils have been extensively shaped by long-term human activities, meaning that reference conditions may also encompass non-degraded soils under a range of land uses, including agriculture. This complexity complicates the definition of reference conditions at the continental scale, given the wide diversity of soil types, climates, and land-use histories.\u003c/p\u003e\n\u003cp\u003eHere, we combine data on soil biodiversity and ecosystem functions from recent European monitoring programs (i.e., the LUCAS survey\u003csup\u003e10\u003c/sup\u003e), along with environmental factors (e.g., climate, soil properties), and soil degradation processes (e.g., erosion, compaction), to develop an integrated assessment of soil health across the European Union. This approach complements the current official EU degradation assessments\u003csup\u003e20\u003c/sup\u003e. First, following the official EU framework, we classified all European soils as ‘healthy’, ‘degraded’ or ‘critically degraded’ , based on whether none, one or multiple degradation processes exceeded critical threshold documented in the literature\u003csup\u003e21\u003c/sup\u003e. The novelty of our approach lies in further refining the ‘healthy’ category into more nuanced categories, i.e., moderate, good or high soil health status, using biotic indicators of soil biodiversity and ecosystem functioning. To achieve this, we analysed LUCAS survey data from 881 sites across 25 European countries\u003csup\u003e10\u003c/sup\u003e, and selected 26 biotic indicators representing soil biota and ecosystem functioning (Table 1). These indicators encompassed the diversity of 19 functional groups from all domains of life (e.g., saprotrophic fungi, plant pathogenic protists, bacterivorous nematodes) and key ecosystem functions related to microbial activity, carbon cycling, and nutrient cycling. Random forest models were then used to predict each biotic indicator’s current value across the EU based on environmental conditions and soil degradation processes. To assess soil health status and soil restoration potential, we simulated a non-degradation reference scenario, defined as conditions in which degradation-related variables were set to zero or to their minimum observed value\u003csup\u003e26\u003c/sup\u003e (Supplementary Table 1), while all other predictors were kept unchanged. This was achieved by applying the trained model for each indicator. ‘Healthy’ soils were then classified into moderate, good or high status (Fig. 1) based on the magnitude of deviation in the biotic indicators between the current condition and the non-degradation reference. This deviation was estimated per land-use type and biogeographical region to account for local environmental specificities. The overall soil health was determined using ‘one out, all out’ principle, a widely used methodology in EU environmental management, where the worst status among the seven biotic indicator categories (Table 1) dictated the final classification. Complementary sensitivity analyses were conducted to assess the sensitivity of the soil health assessment to methodological choices.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe assessed soil biotic indicators, including soil biodiversity groups and their associated functions, exhibited contrasting spatial patterns on a European scale, reflecting the variability in ecological characteristics (spatial predictions for each biotic indicator are available in Supplementary Data 1). This also implies that important trade-offs may arise when considering various biological dimensions of soil systems, particularly when considering restoration and degradation mitigation strategies. Overall, the biotic indicators considered were predominantly influenced by climate, nutrient availability, and soil physico-chemical factors (Supplementary Fig. 1). Additionally, indicators linked to biodiversity were also highly influenced by land-cover type. Interestingly, pH emerged as the sole predominant factor shaping bacterial diversity, while nutrients and soil texture were equally or even more critical than pH for ecosystem functions. Higher temperatures favoured the diversity of most soil functional groups, with some exceptions, such as ectomycorrhizal fungi and herbivorous protists, which exhibited higher diversity in colder regions, along with the ecological functions considered (Supplementary Fig. 2, 3). Nutrient availability strongly influenced most indicators, with several benefiting from increased nutrient levels (natural or otherwise), e.g., the diversity of most trophic groups was higher at lower soil C/N ratio. Likewise, as expected, most ecological function values increased with higher soil organic carbon and nitrogen contents (which are closely correlated Supplementary Fig. 4), given that most functional indicators were associated with carbon and nutrient cycling. (Supplementary Fig. 2, 3). Overall, acidic soils were generally detrimental to both soil biodiversity and functions, except for some groups, such as ectomycorrhizal and root endophytic fungi, and some soil enzymes. Importantly, within a kingdom, the response of functional groups to environmental predictors often diverged, emphasising the need of considering the functional aspects of soil taxa as biotic indicators to gain more insights into soil biodiversity patterns and their associated functions.\u003c/p\u003e\n\u003cp\u003eThe accumulation of soil degradation processes negatively impacted most biotic indicators, as evidenced by a decrease in their values when comparing the current state to the non-degradation reference (Supplementary Fig. 5). This highlights how soil degradation can trigger cascading negative system effects with economic and social consequences. Yet, la value of some indicators increased with accumulating soil degradation across land-cover types and biogeographical regions, including the diversity within pathogenic fungi and protists, and most groups of bacteria, basal respiration, microbial biomass, and enzymatic activities related to soil organic carbon degradation. Some biotic indicators exhibited a positive and strong response to specific soil degradation indicators (Supplementary Fig. 6, 7). For instance, plant feeder protists, heterotrophic bacteria, nematodes, and some enzymes, responded positively to heavy metals concentrations, in contrast to microbivorous and photoautotrophic protists or autotrophic bacteria. Among individual effects of soil degradation processes, erosion and heavy metal pollution had the most pronounced impact on biotic indicators (Supplementary Fig. 1). Additionally, although the model did not include an explicit variable to account for nutrient inputs and acidification (see Methods), biotic indicators showed strong responses to nitrogen, phosphorus, and pH (Supplementary Fig. 2, 3), suggesting a significant influence of these degradation processes on biotic indicators.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur results indicate that 62% of European soils are degraded, 31% exhibit a moderate health status, while only 6.5% and 0.5% show good and high health status, respectively (Fig. 2). Within the 62% of soils that were found to be degraded with at least one soil degradation process exceeding the critical limit (Supplementary Table 1), 22% correspond to soils that are critically degraded, i.e., with multiple soil degradation processes. Soil erosion was the soil degradation process affecting the largest area of European soils (33.4% of area), followed by nutrient excess or deficiency (22.3%), soil organic matter deficiency (17.8%) and pollution (10.5%) (Supplementary Fig. 8). Degraded soils were concentrated along the Atlantic coast, especially in the northwest of Europe and in the Mediterranean coasts, mostly in Italy, Greece, and South of Spain (Fig. 2A, 2B), and in a significant proportion of the Continental region, although they were more sparsely distributed. Both Atlantic and Mediterranean soils were mainly affected by nutrient excess or deficiency and erosion (Supplementary Fig. 9). From the multiple land covers considered, croplands and settlements had the highest proportion of degraded area (83.8%, 83.3%, and 79.1%, respectively; Fig. 2C), which means most of these land cover types are affected by at least one degradation process that requires the implementation of mitigation measures. Degradation of soils in croplands was mostly caused by erosion and nutrient excess or deficiency, with organic carbon deficiency being also important for annual crops (Supplementary Fig. 10). In settlements, sealing was the main degradation process threatening soils as expected.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAcross the EU soils, only 7% were healthy, with 0.5% classified as having ‘high health status’ and 6.5% as ‘good health status’. This indicates that for most European soils classified as ‘healthy’, at least one of the seven biotic indicator categories showed significant biodiversity or functioning declines compared to non-degradation reference. Soils with good status were predominantly located in forests of the Continental, Mediterranean, and Boreal regions. When considering only biodiversity-related indicators, 11% of soils were classified in good conditions (9% with good status and 2% with high status). In contrast, for ecosystem functioning indicators, this percentage rose to 23.6% (11.8% in good status and 11.8% in high status) (Supplementary Fig. 11).\u0026nbsp;Note that these estimates are based on an approach that focuses solely on biodiversity and functioning losses (i.e., negative changes) when assessing deviations from the non-degradation reference (see Methods). When positive changes in biotic indicators, such as an increase in biodiversity, were also considered as potential signs of degradation (e.g., increases due to nutrient inputs), the proportion of soils in good condition (good or high status) dropped to 4.2% overall, and to 8.6% and 14% when considering only biodiversity or ecosystem functioning indicators, respectively. Despite the lower percentages under this alternative approach, the spatial distribution of soil health status categories remained largely consistent (Supplementary Fig. 12). When using a less conservative evaluation than the “one out, all out” principle, based on an additive aggregation across indicators (see Methods), the same soils were reclassified as 4% moderate, 22% good, and 11% high soil health (Supplementary Fig. 13). When accounting also for positive changes in biotic indicators (gains and losses) using this approach, the corresponding proportions were 16%, 17%, and 5%, respectively. It is important to note that, unlike the “one out, all out” rule, this approach relaxes the requirement that all biodiversity and ecosystem-function categories meet the same status, potentially masking deficits in individual components.\u003c/p\u003e\n\u003cp\u003eWe evaluated the restoration potential of degraded soils by comparing their current state to the non-degradation reference, with the aim of quantifying potential improvement under restoration and mitigation strategies. For this analysis, we excluded critically degraded soils not because these systems are of lower priority, but because restoration is more tractable where degradation is limited to a single process exceeding the degradation threshold value. Results indicated that Southeast Europe, particularly in Pannonian and Continental lands, but also in the eastern Mediterranean, showed the highest restoration potential for both biodiversity and ecosystem functioning (Fig. 3), especially in forests and annual crops (Supplementary Fig. 14). On average, soils affected by compaction and erosion showed slightly higher potential for biodiversity and ecosystem functioning gains. When also considering potential improvements through decreases in undesirable biodiversity (e.g., groups that proliferate under soil degradation) and ecosystem functioning, the restoration potential patterns remained similar but extended to the Atlantic coast of France and Ireland (Supplementary Fig. 15). Under this broader perspective, annual and permanent crops, rather than forests, exhibited the highest potential for change (Supplementary Fig. 16). Importantly, restoration potential was greater for biodiversity-related indicators than for ecosystem functions. Biodiversity gains were most pronounced in soils degraded by compaction or erosion, while ecosystem function improvements were slightly higher in soils affected by organic matter loss, compaction, nutrient depletion, or acidification. It is important to note that these patterns are indicative, given the strict and utopic nature of the restoration potential measure used here which assumes the reduction or removal of all soil degradation processes, not just those surpassing critical limits.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur study sheds light on the concerning state of soil health in the European Union by integrating soil biodiversity and functioning indicators into a comprehensive soil assessment. Only 7% of EU soils are classified as healthy, with good or high status, while the remaining 93% are either degraded (62%) or of moderate status (31%). These findings emphasize the urgent need for legislative action to support soil conservation and implement targeted restoration strategies to improve soil health in the EU. By providing spatially explicit insights into how soil biodiversity and functions respond to degradation, our framework helps refine soil health assessments and guide region-specific conservation and restoration efforts. Identifying both areas in need of urgent intervention and healthy soils that should be prioritized for protection enables more effective management strategies, such as including healthy soils in conservation areas\u003csup\u003e27\u003c/sup\u003e or promoting sustainable land-use practices.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrades-off in assessing soil health\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEvaluating soil health across multiple ecological dimensions provides a more comprehensive understanding of soil conditions, but can also be challenging due to the complex and sometimes contrasting responses of soil biodiversity and ecosystem functions\u003csup\u003e27\u003c/sup\u003e. Most of our indicators exhibited negative responses to soil degradation processes, validating their utility as broad-scale indicators for assessing soil degradation\u003csup\u003e28\u003c/sup\u003e. However, some biotic indicators showed positive responses to specific soil degradation processes. Soil degradation not only diminishes soil biodiversity but also simplifies soil food webs and alters community compositions, resulting in unequal impacts on the diversity and activity of various soil functional groups\u003csup\u003e4,29\u003c/sup\u003e. For instance, while the diversity of more sensitive groups declines due to the accumulation of degradation processes, the diversity of other groups, such as pathogens, can actually increase\u003csup\u003e30,31\u003c/sup\u003e. This may explain why, even when considering biodiversity and functioning \u0026lsquo;gains\u0026rsquo; as a metric of degradation, the overall spatial patterns of soil health status remain largely unchanged. In many areas, biodiversity or ecosystem function losses coincide with positive changes in other biotic indicators in response to soil degradation. The variability in biotic indicator\u0026rsquo;s responses emphasises the importance of incorporating multiple dimensions of soils in soil health assessments. While certain aspects of the soil system may appear favourable, overall soil health can be compromised due to the degradation of other crucial components. Indeed, the low proportion of European soils classified as having a good or high status can be partly attributed to the limited overlap among the upper categories of each biotic indicator (Supplementary Fig. 11). This is a direct consequence of the \u0026ldquo;one out, all out\u0026rdquo; principle used in our assessment, which ensures that all components must be in good condition for soil to be classified as healthy.\u0026nbsp;While restrictive, this approach effectively differentiate soil components and identifies specific aspects requiring improvement, ensuring sustainable use and an enhanced overall soil health\u003csup\u003e15,16\u003c/sup\u003e. Yet, looking independently to each component warrant cautious interpretation, acknowledging the multiple intrinsic and complex linkages that exist between biodiversity and ecosystem functions\u003csup\u003e32,33\u003c/sup\u003e. Using a less conservative, additive framework for soil health status classification, substantially shifted the distribution of soils across health categories (Supplementary Fig. 12), revealing higher proportions of soils in good status. Yet, this alternative aggregation allows soils to achieve high overall scores even when one or more components show deficits, which may have implications for ecosystem functioning\u003csup\u003e34\u003c/sup\u003e. While the \u0026ldquo;one out, all out\u0026rdquo; \u0026nbsp;classification remains our primary approach, these results provide a complementary perspective that can inform policy and management decisions, highlighting areas where overall soil health is high or low despite localized weaknesses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePolicy implications\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEffectively addressing soil degradation requires integrating quantitative target values into policies\u003csup\u003e9\u003c/sup\u003e. Our assessment identifies priority areas for conservation and restoration (Fig. 1- 3) and highlights the need to implement soil protection strategies beyond areas classified as degraded. The 31% of European soils classified as of moderate health status highly differ from what would be the ideal situation for the sustainable use of soils. This confirms that soil degradation processes, even if not surpassing the critical limits established by the EU, still show indications of deterioration when considering multiple soil biodiversity and ecosystem functioning indicators. For instance, forests are considered one of the healthiest land covers due to their low proportion of degraded soils. However, upon closer examination, many of these soils are only moderately healthy, indicating a potential vulnerability to degradation and the need for protection (Fig. 1).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFurthermore, the implementation of soil conservation measures that go beyond the usual mitigation of impact in productive soils should also be considered\u003csup\u003e35\u003c/sup\u003e. In the EU, current conservation policies, such as the Habitat Directive\u003csup\u003e36\u003c/sup\u003e, do not explicitly account for soil biodiversity, leaving a significant gap in management strategies. Incorporating ecological soil indicators into the criteria for assessing habitat conservation status, e.g., within Natura 2000 sites, could pinpoint priority areas for targeted interventions\u003csup\u003e7,8\u003c/sup\u003e. Such interventions may include minimising soil disturbance, enhancing vegetation cover to mitigate erosion, restoring degraded soils, and implementing sustainable agricultural practices in sensitive habitat buffer zones\u003csup\u003e37,38\u003c/sup\u003e. Some land management practices can yield co-benefits\u0026mdash;cover crops, for instance, not only reduce erosion but also enhance soil carbon storage and biodiversity\u003csup\u003e39\u003c/sup\u003e. Beyond localized management, addressing soil degradation also demands tackling broader drivers, including climate change, which intensifies processes like erosion through more frequent extreme events\u003csup\u003e38,40\u003c/sup\u003e. While our approach provides spatially explicit measures for potential changes in biodiversity and functioning, effective restoration planning must also consider socio-economical aspects\u003csup\u003e41\u003c/sup\u003e. For example, our results identified areas affected by erosion and compaction as key areas for biodiversity and functioning recovery, yet these processes are particularly difficult to mitigate in the short term, emphasizing the need for long-term, sustained restoration efforts\u003csup\u003e38\u003c/sup\u003e. We also emphasize that, although our analyses focused on degraded soils and excluded critically degraded soils, the latter remain a top priority and require urgent restoration strategies (Fig. 1). However, evaluating their restoration potential is more challenging and less realistic, as recovery trajectories become increasingly complex and uncertain when multiple interacting degradation processes are involved\u003csup\u003e17,41\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdvancing soil health assessments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur study leverages the most comprehensive biotic dataset currently available at the European scale, highlighting the importance of integrating biodiversity and ecosystem function indicators into soil health assessments. This integration is crucial for soil policy development\u003csup\u003e16\u003c/sup\u003e. Future monitoring efforts should further improve the representation of biodiversity and ecosystem functioning. For instance, biomass or abundance indicators, which can be directly linked to ecosystem functioning\u003csup\u003e42\u003c/sup\u003e, could provide complementary insights.\u0026nbsp;Additionally, incorporating measurements of other key functions, such as water cycling, will enhance soil health assessments as new data become available. Our framework is designed to evolve alongside emerging data and methodologies. In the future, degradation thresholds tailored to specific pedo-climatic conditions may become available and refine the assessments\u003csup\u003e20\u003c/sup\u003e. Despite these anticipated advancements, our approach remains robust, as it is grounded in the best available data and aligns with current governance practices. By incorporating biotic indicators into official soil health evaluations, this study provides a foundation for more comprehensive assessments that can better inform policy and land management decisions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA spatially explicit and comprehensive assessment of soil health, as presented here, directly supports policy agendas in agriculture, forestry, restoration, and nature conservation. It provides essential tools for defining restoration and conservation targets at the European scale. While we proposed a policy indicator for EU-wide assessments, our methodology can be adapted to different geographic scales, enabling other regions to establish comparable baselines for soil health. Similarly, applying this approach at national and local scales, where higher-resolution and more data might be available, could provide more targeted guidance for management and restoration efforts.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eData on Soil Biodiversity and Functioning\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSoil biodiversity and functional data was obtained from the 2018 soil campaign of the Land Use/Land Cover Area Frame Survey (LUCAS)\u003csup\u003e10\u003c/sup\u003e. This dataset comprises data on soil functions and biodiversity from 881 sites in Europe covering most of the terrestrial land covers including annual crops, permanent crops, broadleaved forests, needle leaved forests, grasslands, shrublands and human settlements. In short, soil samples were collected at each location by combining five subsamples taken in different directions around a pre-selected point and covering a depth of 20 cm. A total of 500 g of soil was stored on ice and transported to the JRC within 48\u0026thinsp;h of collection. The samples were then frozen and sent to the University of Tartu for DNA analyses. A separate sample was analysed by SGS Hungary for physical and chemical soil properties, and another in Germany for soil functions by the German Centre for Integrative Biodiversity Research (iDiv), in Leipzig, and the Helmholtz-Centre for Environmental Research (UFZ), in Halle (see below). Raw environmental DNA data (post sequencing) was obtained from the European Soil Data Centre (ESDAC) platform for Eukaryotes (18S)\u003csup\u003e43\u003c/sup\u003e, Bacteria (16S) and Fungi (ITS)\u003csup\u003e31\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBiodiversity data \u0026ndash;\u003c/em\u003eDemultiplexed raw DNA sequence data for eukaryotes (18S) and prokaryotes (16S) were obtained from the Sequence Read Archive (SRA) (BioProject ID PRJNA952168 for 16S and ITS, and PRJNA985135 for 18S). For both datasets, forward and reverse reads were assessed for consistent orientation and corrected as necessary using SeqKit\u003csup\u003e44\u003c/sup\u003e and BBMap\u003csup\u003e45\u003c/sup\u003e tools. Reads re-sequenced due to low coverage in the 16S samples were merged with the original dataset to improve completeness. Primer sequences were verified for their starting position at the beginning of the reads. To maximise the detection of primer sequences that did not start immediately at the beginning, a general 5-base offset was applied across the dataset. Both forward and reverse primer sequences were then trimmed from each dataset using cutadapt\u003csup\u003e46\u003c/sup\u003e allowing a maximum of two mismatches per primer. During primer removal, reads shorter than 75 bp were discarded to avoid erroneous taxonomic assignments. Further, reads were end-trimmed with a minimum quality score threshold of 10 to ensure optimal merging of paired-end reads. For denoising and amplicon sequence variant (ASV) inference, the DADA2 algorithm\u003csup\u003e47\u003c/sup\u003e was implemented within the Qiime2 framework\u003csup\u003e48\u003c/sup\u003e. Input data from multiple sequencing efforts were categorised by sequencing run before error rate modelling, following DADA2 guidelines. A maximum expected error of 1.0 was used for quality filtering of reads prior to denoising. We implemented a minimum overlap of 12 bp for merging the denoised forward and reverse reads. Following chimaera removal, ASV tables from different sequencing efforts were combined for each dataset. For 18S data, ASVs clustered at 100% similarity were considered as species proxies following\u003csup\u003e43\u003c/sup\u003e. Taxonomic assignment was done using the \u003cem\u003eassignTaxonomy\u003c/em\u003e function in DADA2, with the PR2 database\u003csup\u003e49\u003c/sup\u003e (Version 5.0) for 18S rRNA gene sequences and the SILVA database\u003csup\u003e50\u003c/sup\u003e (Version 138.1) for 16S rRNA gene sequences. Additionally, the 18S and 16S datasets were filtered for eukaryotic and bacterial ASVs, respectively. Due to the low representation of fewer than 2000 archaeal taxa, Archaea were excluded from the 16S dataset. Only ASVs present at a minimum of two sites, thus excluding singletons, were retained. ASV detected as potential contaminants based on their abundance in the controls were identified and removed with the \u003cem\u003econtslayer\u003c/em\u003e function from the metabaR package\u003csup\u003e51\u003c/sup\u003e. At the end of the bioinformatic pipeline we obtained 90044 bacterial and 156298 eukaryotic ASVs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor fungi, OTU tables with taxonomic annotation were obtained from\u003csup\u003e31\u003c/sup\u003e. In short, PacBio amplicon data for fungal sequences (from the fungal ITS region) were demultiplexed using LotuS2\u003csup\u003e52\u003c/sup\u003e with default options. Paired-end reads were assembled using FLASH 1.2.1083 with default options (minimum overlap 10\u0026thinsp;bp) and unmerged reads were removed from the final datasets. Data were next processed as described in\u0026nbsp;\u003csup\u003e53\u003c/sup\u003e. OTUs at 98% of similarity were obtained and taxonomic assignment was performed using BLAST\u0026thinsp;+\u0026thinsp;2.11.091 by running MegaBLAST queries of representative OTU sequences against the updated UNITE 9.192 beta reference dataset. This resulted in 55647 fungal OTUs. OTU/ASV tables for the each kingdom were normalised to comparable sequencing depth (bacteria: 4200 reads; fungi: 1000 reads, protists: 400 reads, metazoans: 200 reads) using the Scaling with Ranked Subsampling (SRS) method\u003csup\u003e54\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eEukaryote data was separated into animals and protists, and the rest discarded for this study. Protists were classified into functional groups with standardised queries to retrieve the functional information from multiple data sources using the ontology-based data integration pipeline described\u003csup\u003e55\u003c/sup\u003e. From the query, protist ASVs were classified into photoautotroph, zooparasite, microbivore (including bacterivore, protistivore, and fungivore), and herbivore. When a protist ASV was associated with more than one functional group, it was included in each relevant category for ecological assessment.\u003c/p\u003e\n\u003cp\u003eDue to the low taxonomic resolution of the 18S marker for animals, these were classified into key soil taxa, as in\u003csup\u003e43\u003c/sup\u003e, i.e. Arthropoda, Annelida, Rotifera, Tardigrada, and only nematodes were functionally annotated. ASV of nematodes with taxonomic annotation at the genus level were classified into bacterivore, fungivore, herbivore, predator or animal parasite using the NINJA platform\u003csup\u003e56\u003c/sup\u003e. The category of animal parasites was not retained for further analysis due to the low number of ASVs classified in this category. NINJA enables the classification of most nematode families, genera, and species found in soil into feeding types. This classification is based on\u0026nbsp;\u003csup\u003e57\u003c/sup\u003e and additional information from the Nemaplex website (http://nemaplex.ucdavis.edu).\u003c/p\u003e\n\u003cp\u003eBacteria ASVs were functionally annotated using FAPROTAX\u003csup\u003e58\u003c/sup\u003e. The functional annotations of bacterial taxa were categorized into broad functional groups, including autotrophs, chemoheterotrophs, N-cycling bacteria, pathogens and parasites, and a more general decomposer group. FAPROTAX assigns ecological functions based on phenotype profiles supported by experimental evidence. All references supporting the functions can be found in\u0026nbsp;\u003csup\u003e58\u003c/sup\u003e and the FAPROTAX database is available at: www.zoology.ubc.ca/louca/FAPROTAX. ASVs associated with autotrophy and phototrophy were categorised as autotrophic bacteria, including both oxygenic and anoxygenic phototrophic taxa, and all ASV from the phylum Chloroflexi. Oxygenic phototrophs perform photosynthesis using light energy to fix carbon, utilizing water (H\u003csub\u003e2\u003c/sub\u003eO) as an electron donor while producing oxygen as a byproduct. In contrast, anoxygenic phototrophs fix carbon without producing oxygen, utilising varied electron donors such as H\u003csub\u003e2\u003c/sub\u003eS, H\u003csub\u003e2\u003c/sub\u003e, and Fe\u003csup\u003e2+\u003c/sup\u003e. Chemoheterotrophs, which derive energy by oxidising organic compounds such as glucose, include aerobic chemoheterotrophs that operate in the presence of oxygen and anaerobic chemoheterotrophs that function in its absence. Bacterial ASVs assigned to any kind of chemoheterotrophy were classified under this group. N-cycling bacteria are involved in various steps of the nitrogen cycle, including nitrogen fixation (conversion of N\u003csub\u003e2\u003c/sub\u003e to NH\u003csub\u003e3\u003c/sub\u003e), nitrification (oxidation of NH\u003csub\u003e3\u003c/sub\u003e to NO\u003csub\u003e2\u003c/sub\u003e- and NO\u003csub\u003e3\u003c/sub\u003e-), and denitrification (reduction of NO\u003csub\u003e2\u003c/sub\u003e- and NO\u003csub\u003e3\u003c/sub\u003e- to N\u003csub\u003e2\u003c/sub\u003e). ASVs linked to such functions were categorised under N-cycling bacteria. Bacteria identified as pathogens and or parasites of plants, humans, animals, and other microbes were grouped as pathogens and parasites. ASVs that did not receive any of the above-mentioned functional assignment from FAPROTAX, were classified in the category of decomposer bacteria, due to the significance role that overall microbial diversity play on soil organic matter decomposition\u003csup\u003e59\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFunctional annotations for fungi were obtained from\u003csup\u003e31\u003c/sup\u003e, and were retrieved using FungalTraits\u003csup\u003e60\u003c/sup\u003e. Similarly, the functional annotations of fungal taxa were categorized into multiple broad functional guilds. The classification encompassed various guilds of fungal plant pathogens, including those associated with roots, leaves, fruits, and seeds, as well as pathogens specific to algae, wood, and moss. OTUs were categorised as saprotrophs reflecting their decomposition roles and including saprotrophs with substrate preferences, e.g., litter, dung, wood. Furthermore, mycorrhizal associations were specifically differentiated based on their primary ecological interactions, between Arbuscular Mycorrhizal Fungi (AMF) and ectomycorrhizal fungi.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEcosystem functioning data \u0026ndash;\u003c/em\u003eLUCAS samples transported to iDiv, Leipzig, Germany, were analysed for the measurement of potential basal respiration by O2-microcompensation\u003csup\u003e61\u003c/sup\u003e, and microbial biomass by substrate-induced respiration\u003csup\u003e62\u003c/sup\u003e. The enzymatic potential of acid phosphatase (phosphorus mineralization), \u0026beta;-glucosidase and cellulase (cellulose degradation), N-acetylglucosaminidase (chitin degradation), and xylosidase (hemicellulose degradation) were determined in the UFZ, in Halle. Determination of the activities of the hydrolytic enzymes was based on 4-methylumbelliferone (MUF)-coupled substrates (4-MUB-b-D-cellobioside, 4-MUB-b-D-xyloside, 4-MUB-N-acetyl-b-D-glucosaminide and 4-MUB-phosphate), at a pH of 5 at 25\u0026deg;C\u003csup\u003e63,64\u003c/sup\u003e. The method employed standardised conditions, mitigating variations in temperature and moisture to enable direct comparison of enzymatic potentials across diverse samples. Refer to\u0026nbsp;\u003csup\u003e65\u003c/sup\u003e for more details on the methods used to measure potential respiration and microbial biomass and to\u0026nbsp;\u003csup\u003e66\u003c/sup\u003e for methods used to measure enzymatic activities.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBiotic indicators\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA Biotic Indicator is any living organism, its parts, its products (e.g. enzymes), or biological processes that can be used to assess the quality of the environment\u003csup\u003e28\u003c/sup\u003e. Building on the definition of soil health, our chosen indicators aimed to represent soil biodiversity and the associated ecosystem functions. Here, using the latest measured biotic data available at the European scale (see previous section), we defined 26 biotic indicators representing the diversity of soil organisms (19 biotic indicators) and the ecosystem function driven by soil organisms (7 biotic indicators) (Table 1).\u0026nbsp;\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cem\u003eIndicators of soil biodiversity\u003c/em\u003e: To quantify soil biodiversity, we estimated the alpha diversity of the 19 assigned functional groups across the kingdoms Bacteria, Fungi, Protist and Metazoa. The effective number of OTUs, calculated as the exponential of the Shannon index (Hill numbers\u003csup\u003e67\u003c/sup\u003e), was used as a measure of alpha diversity, ensuring better coverage across samples\u003csup\u003e68\u003c/sup\u003e. A functional approach allows for a more precise understanding of the relationships between habitat characteristics, disturbances, and ecosystem functions, providing more effective, reliable, and informative ecological indicators of system health\u003csup\u003e25\u003c/sup\u003e. Here, the diversity within a functional group serves as a measure of\u0026nbsp;functional similarity\u003csup\u003e23\u003c/sup\u003e, as it reflects the extent to which species within a group share ecological roles while also accounting for the unique contributions of each species. Changes in functional similarity can influence ecosystem resilience and overall functioning, making it a relevant biotic indicator. An exception was made for most Metazoa, where taxonomic groups were used instead of functional groups due to the limited resolution of the marker (i.e., rotifers, tardigrades, arthropods, annelids). However, model performance for these taxonomic groups fell short of the established standard, leaving only the functional groups of nematodes as biotic indicators for the analyses (Supplementary Fig.18, Table 1).\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eIndicators of ecosystem functioning\u003c/em\u003e: The measurements performed on soil samples (ecosystem functioning data) were used as biotic indicators serving as proxies for key soil ecosystem functions. Basal respiration and microbial biomass were interpreted as biotic indicators of soil microbial activity. While microbial activity itself is not an ecosystem function per se, microbial biomass and respiration reflect processes related to\u0026nbsp;carbon storage, decomposition, and organic matter turnover, which are fundamental to soil ecosystem functioning. Enzymatic potentials were also used as biotic indicators of carbon and nutrient cycling, with xylosidase, \u0026beta;-glucosidase, and cellulase representing carbon cycling processes, while N-acetylglucosaminidase and acid phosphatase were associated with phosphorus and nitrogen cycling (referred to as nutrient cycling)\u003csup\u003e69,70\u003c/sup\u003e.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003ePredictor variables and Data sources\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEnvironmental data \u0026ndash;\u003c/em\u003e To model each biotic indicator across Europe, we used a comprehensive collection of environmental datasets from different repositories, including CHELSA (bioclim+\u003csup\u003e71\u003c/sup\u003e), the European Soil Data Centre (ESDAC)\u003csup\u003e72\u003c/sup\u003e, the European Digital Elevation Model (EU-DEM)\u003csup\u003e73,74\u003c/sup\u003e, and Land Cover databases\u003csup\u003e75,76\u003c/sup\u003e (the complete list of variables and sources is in Supplementary Table 2). By incorporating variables such as climate, soil physico-chemical properties, land cover, and topography, our analyses aimed to capture the complex interplay of factors that explain changes in the values of each biotic indicator at the continental scale. For model training, we obtained soil physical and chemical properties, as well as land cover, directly from the LUCAS survey, as they were measured in the same samples. For other variables, we extracted values from the aforementioned environmental datasets using the sample coordinates. To project data across Europe, we used the same datasets and maps available from the ESDAC platform for variables measured in the LUCAS survey.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSoil degradation processes data \u0026ndash;\u0026nbsp;\u003c/em\u003eBuilding upon the EU Soil Observatory (EUSO) degradation assessment framework and data, we selected the following variables to represent soil degradation processes: soil erosion (measured as a combination of multiple indicators related to erosion processes by wind\u003csup\u003e77\u003c/sup\u003e, water\u003csup\u003e78\u003c/sup\u003e, tillage\u003csup\u003e38\u003c/sup\u003e or crop harvesting\u003csup\u003e79\u003c/sup\u003e); concentration of heavy metals including copper\u003csup\u003e80\u003c/sup\u003e, mercury\u003csup\u003e81\u003c/sup\u003e, cadmium\u003csup\u003e82\u003c/sup\u003e, and zinc\u003csup\u003e83\u003c/sup\u003e, as indicators of soil pollution; soil compaction\u0026nbsp;\u003csup\u003e84\u003c/sup\u003e; and, sealing area for soil sealing (Copernicus). No specific variable was explicitly selected as an indicator of soil nutrient imbalances (e.g., phosphorus deficiency/excess, nitrogen surplus) or acidification (critical pH levels for crop production), as these factors were inherently accounted for by the environmental variables: phosphorus content, nitrogen content, and pH. Their contribution to soil degradation arises when these values exceed or fall below critical thresholds. This aspect was later addressed when predicting the non-degradation reference scenario (explained below).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eModelling framework\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA stacked ensemble modelling framework was implemented with the aim to predict biotic indicators across the EU under both current conditions (current state) and a reference baseline (non-degradation reference). Each biotic indicator was modelled independently using an ensemble approach, and the resulting models were stacked to produce comprehensive predictions at the European scale (Fig. 4 in Extended Data).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData preparation\u003c/em\u003e \u0026ndash; For each biotic indicator, a dataset was prepared containing the biotic indicator value along with the predictor values, measured \u003cem\u003ein-situ\u0026nbsp;\u003c/em\u003eor extracted, across the 881 samples (see \u003cem\u003ePredictor variables and Data sources\u003c/em\u003e section). The dataset was pre-processed by removing missing values and one-hot encoding categorical variables where applicable. Missing values resulted either from the removal of a sample during the bioinformatic filtering of biodiversity data or from missing information for one of the predictors. The final dataset sizes, or the number of samples used to fit each model, were as follows: 827 for biotic indicators related to ecosystem functioning and to bacteria, 734 for biotic indicators related to fungi, and 597 for biotic indicators associated with protists or metazoans.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eModel training\u003c/em\u003e \u0026ndash; The Random Forest algorithm was selected to model biotic indicator values in response to 25 environmental variables and 7 variables representing soil degradation processes. Random Forest constructs decision trees through recursive binary splitting, enabling it to capture complex, higher-order interactions among predictors without explicitly defining them. This capability makes it particularly effective for ecological modelling.\u003c/p\u003e\n\u003cp\u003eTo ensure robust predictions and account for spatial autocorrelation, a spatial block cross-validation approach was adopted\u003csup\u003e85\u003c/sup\u003e. Concretely, the dataset was divided into ~100\u0026times;100 km spatial blocks for cross-validation. These blocks were randomly partitioned into five subsets (folds), with each subset serving as a validation set once while the remaining four were used for training. This process resulted in an ensemble of five Random Forest models per biotic indicator.\u003c/p\u003e\n\u003cp\u003eModel parameters were optimized to improve predictive accuracy, using mean squared error (MSE) as the criterion. MSE quantifies the average squared difference between observed and predicted values, with lower values indicating better model performance. Key hyperparameters, including maximum tree depth and the number of features considered at each split, were fine-tuned using a holdout validation set. Predictive performance on the validation dataset was evaluated using R-squared (R\u0026sup2;) and Spearman\u0026rsquo;s rank correlation coefficient (⍴) between observed and predicted values (Supplementary Fig. 17). The models were implemented in Python using scikit-learn\u003csup\u003e86\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eBiotic indicators with both an R\u003csup\u003e2\u003c/sup\u003e lower than 0.5 and a Spearman\u0026rsquo;s\u0026nbsp;⍴\u0026nbsp;lower than 0.70 were removed from the analyses (Supplementary Fig. 18). These thresholds were chosen based on a visual assessment of overall model performance to exclude poorly predicted groups while retaining as many relevant groups as possible. Initially, models were fitted for 31 biotic indicators, but five were excluded due to poor model performance, resulting in 26 biotic indicators. Additionally, we applied the same modelling framework and subsequent analyses to model the Shannon diversity of the four soil kingdoms (Bacteria, Fungi, Protista, Metazoa) and one phyla (Nematoda) to capture broad diversity patterns at the kingdom or phylum level (Supplementary Data 1, Supplementary Figs. 2, 5, 6). This resulted in five additional models, which served as a reference but were not included in the soil health assessment.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePredictions\u0026nbsp;\u003c/em\u003e\u0026ndash; The ensemble models per biotic indicator obtained after cross-validation were used to predict both the average and variance of each biotic indicator\u0026rsquo;s value across Europe under the two scenarios used to assess soil health:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eCurrent state:\u003c/strong\u003e This scenario represents the estimated current state of soil biotic indicators across Europe, considering the current (or available) environmental and soil degradation processes conditions.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eNon-degradation\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;reference:\u003c/strong\u003e This scenario simulates a hypothetical condition with unaltered soil conditions, maintaining the environmental factors consistent with the current status while setting all soil degradation processes to zero or minimum values (see Supplementary Table 1 for specific values associated with each degradation process). To represent the reduction of soil nutrient imbalances and acidification, which were not explicitly represented by a variable in the model, phosphorus content values exceeding or falling below a critical threshold (i.e., phosphorus excess and deficiency, respectively) were set to the threshold value (Supplementary Table 1). Similarly, for acidification, pH values below 5 in croplands were adjusted to 5. However, nitrogen surplus and organic carbon deficiency could not be adjusted in the non-degradation scenario, as the model did not include a predictor for nitrogen surplus (only total nitrogen content, which is not equivalent) or for organic carbon deficiency. It is important to acknowledge that this scenario may not be entirely realistic, as soil degradation processes are rarely absent or minimal in real-world situations. However, in the absence of a non-degradation reference for soils in Europe, it provides an approximation of what a standardized reference state would be.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eWe verified the reliability of our projections by assessing the extrapolation of the predictor variables used in the projections. We compared the projected values for the non-degradation reference with the current state, ensuring that the projections were not overly unrealistic in comparison to actual conditions observed in Europe (Supplementary Figure 19). Additionally, we checked the overlap between the training data and the projected environmental conditions using the Schoener\u0026apos;s D metric of niche overlap. An 85% overlap in the \u0026apos;environmental space\u0026apos; between calibration and projection suggests that the model adequately represents the conditions in both scenarios, supporting the reliability of the projections. The predictions were done at 1 km resolution. This resolution is relatively high for maps at the continental scale and is designed to summarize predicted average values based on the expected averages of climate, soil properties, and soil degradation processes at this spatial resolution.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eModel analyses\u003c/em\u003e \u0026ndash; Relative importance of each predictor was estimated. While random forest models are robust to collinearity, variable importance can be affected when strongly correlated predictors are included, potentially underestimating their influence. However, correlations among predictors were generally low (Supplementary Fig. 4), and strong correlations (Pearson\u0026rsquo;s \u003cem\u003er\u0026nbsp;\u003c/em\u003e\u0026gt; 0.65) were mostly within the same category (e.g., climate, nutrients, land cover), minimizing their impact on interpretation. Partial dependence plots were generated using built-in functions in scikit-learn.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe uncertainty of the model was evaluated using the coefficient of variation (CV) in the predictions across the folds\u003csup\u003e87\u003c/sup\u003e. We estimated the uncertainty separately for biotic indicators related to soil biodiversity and ecosystem functions (Supplementary Fig. 20). To achieve this, we calculated the CV for each pixel by aggregating the CV values of individual biotic indicators. We then identified pixels with low confidence values corresponding to those above the 0.9 quantile in the summed CV distribution across all pixels and excluded them from further analyses. While ideally, uncertainty should account for the propagation of errors in input covariates, this remains a significant challenge in predictive ecological modelling. We acknowledge this limitation and recognize it as an important area for future research. However, our approach aligns with common practices in ecological modelling and provides a robust assessment given the available methodologies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSoil Health Indicator\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo assess soil health across Europe, soils were categorized into five soil health categories, representing a gradient from degraded to healthy. These categories included critical degraded, degraded, moderate status, good status, and high status (Fig. 1). Critical degraded soils represent the most degraded or unhealthy category, while high status soils are the least degraded or most healthy category. The classification process consisted of two steps (Fig. 5 in Extended Data): 1) The first step involved separating degraded soils from healthy soils using the official procedure and standards from European institutions (EUSO); 2) In the second step, healthy soils were further classified into moderate, good, or high status using the approach developed in this study and detailed below.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFirst step \u0026ndash;\u003c/em\u003e First, soil were classified into degraded or healthy by applying the EU Soil Observatory (EUSO) degradation assessment framework and data (https://esdac.jrc.ec.europa.eu/esdacviewer/euso-dashboard/). In this approach, soils with at least one degradation process exceeding the defined critical threshold are classified as degraded. The critical thresholds for each degradation process indicator are summarized in Supplementary Table 1 and follow the same criteria established by the EUSO assessment. We used the available data from the EUSO to classify soils into degraded or healthy based on 19 soil degradation process indicators\u003csup\u003e20\u003c/sup\u003e. The data contains spatially explicit binary information indicating whether a given indicator for a soil degradation process exceed the critical threshold. If one or more indicators within a particular degradation process exceeded the critical threshold, it was considered degraded for that specific degradation process. While the EUSO framework classifies degradation on a gradient based on the number of co-occurring degradation processes, we simplified this into two categories: soils were classified as degraded if affected by at least one degradation process (i.e., at least one indicator for a given process exceeded the critical threshold) and as critically degraded if affected by multiple degradation processes. While striving to maintain consistency with the EUSO assessment, some modifications were made. Firstly, we excluded soil degradation processes that were redundant with our approach using biotic indicators and indirect, such as \u0026quot;potential threats to biological functions\u0026quot;. Secondly, we eliminated processes that were only available for specific regions only, e.g., \u0026quot;salinization\u0026quot; in the Mediterranean region to avoid spatial biases in the assessment, or processes related to habitats not considered in our study, such as \u0026quot;peatland degradation\u0026rdquo;. Third, we also included acidification as a soil degradation process that was made available by European Environment Agency, but which only concerns croplands. In summary we kept the following soil degradation processes and respective indicators: soil erosion including multiple erosion processes by wind, water, tillage or crop harvesting, and post-fire recovery; soil pollution including excess of copper, mercury, cadmium, zinc, and arsenic; soil nutrients including phosphorus deficiency and excess, and nitrogen surplus; soil compaction; soil sealing; soil organic carbon deficiency; and acidification. Despite the differences, the percentage of degraded soils estimated in this study closely matched the EUSO\u0026rsquo;s published estimates. This suggests that excluding certain degradation processes, such as salinization, may decrease the number of degradation processes affecting an area but is unlikely to alter its classification as degraded.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSecond step \u0026ndash;\u003c/em\u003e We further classified the remaining \u0026lsquo;healthy\u0026rsquo; soils into moderate, good, or high status in the following way by comparing the deviation of biotic indicators in the current state scenario from the non-degradation reference scenario (Fig. 4 in Extended Data). Biotic indicators were grouped by category (Table 1): four categories for soil biodiversity indicators corresponding to the kingdoms Bacteria, Fungi, Protists, and Metazoa, and three categories for ecosystem functioning indicators corresponding to soil microbial activity, carbon cycling, and nutrients cycling. This categorization was done to balance the weight given to each category and reduce biases related to a specific kingdom or ecosystem function (e.g., there are more functional groups of fungi than bacteria, but we wanted to give equal weight to both kingdoms). Additionally, since functional groups within a kingdom may respond differently to environmental changes or degradation, grouping them allows for a broader understanding of predicted biodiversity or functioning changes within the category. For each category, the percentage of difference between the projected current state and the non-degradation reference was calculated to define the soil health status. We used a multivariate dissimilarity index (Temporal Beta-diversity Index, TBI), suggested by\u0026nbsp;\u003csup\u003e88\u003c/sup\u003e that accounts for gains and losses of individual species or groups of species between two states. This index is also applicable to continuous variables such as our biotic indicator related to ecosystem functioning. The input for each category consisted of two matrices, one for each scenario (current and non-degradation), with sample IDs as rows and biotic indicators within that category as columns, with their respective predicted values. The TBI was then calculated for each biotic indicator category between the current and non-degradation status to assess overall changes in OTUs diversity within functional groups for a specific kingdom, or changes in functions values within a specific ecosystem function. We used the function \u003cem\u003eTBI\u003c/em\u003e from the R package adespatial\u003csup\u003e89\u003c/sup\u003e, which additionally provides the associated probability of a site experiencing an \u0026quot;exceptional change\u0026quot;, across the evaluated sites. Given the environmental differences existing across Europe that may lead to different order of magnitudes in the biotic indicator changes, we estimated the TBI and the associated probability for each land cover type within a biogeographical region (EEA, 2016: http://www.eea.europa.eu). We used the probability to define the soil helath status for each biotic indicator within each land cover and biogeographical region as follows:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eModerate soil health status: The probability that the changes observed for a given biotic indicator between the status is \u0026lsquo;exceptional\u0026rsquo; is relatively high (p \u0026gt; 0.67), indicating that the status of the biotic indicator in the site is very distant from the modelled \u0026quot;reference conditions\u0026quot;.\u003c/li\u003e\n \u003cli\u003eGood soil health status: The probability that the changes observed for a given biotic indicator between the status is \u0026lsquo;exceptional\u0026rsquo; is intermediate (0.66 \u0026lt; p \u0026lt; 0.33), indicating that the site is between a moderate and a high status.\u003c/li\u003e\n \u003cli\u003eHigh soil health status: The probability that the changes observed for a given biotic indicator between the status is \u0026lsquo;exceptional\u0026rsquo; is relatively low (p \u0026lt; 0.33), indicating that the status of the biotic indicator in the site is very similar to the reference conditions.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eIn addition, a sensitivity analysis was conducted to assess the effect of threshold choices by testing alternative, reasonable cut-points of 0.6/0.4 and 0.7/0.3 (Supplementary Fig. 13). We considered changes significant only when degradation processes led to a decrease in biodiversity or a loss of functions, following the official definition of soil health and degradation, which identifies biodiversity loss and ecosystem function decline as indicators of deterioration. However, since positive changes resulting from soil degradation processes may also indicate degradation, we conducted parallel analyses accounting for both positive and negative changes to ensure transparency in our approach (Supplementary Fig.12). At the end, each one of the seven categories of biotic indicators was classified into moderate, good, or high status. Maps of soil health status for each category are provided in Supplementary Fig. 21, 22. The overall soil health status classification for a soil area (1km\u003csup\u003e2\u003c/sup\u003e) was determined by the element with the worst status out of all the categories of biotic indicators following the \u0026lsquo;one out, all out\u0026rsquo; principle used in the Water Directive Framework\u003csup\u003e18\u003c/sup\u003e. To explore the sensitivity of this classification principle, we implemented an alternative additive approach (Supplementary Fig.13). For each soil area and each biotic indicator category, values were scored as 0 = moderate, 1 = good, and 2 = high. The scores were then summed across the seven indicator categories and standardized to a 0\u0026ndash;2 range by dividing by the maximum possible total. Soils were subsequently classified into three classes: 0\u0026ndash;1 = moderate, 1\u0026ndash;1.5 = good, and \u0026gt;1.5 = high soil health.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSoil restoration potential\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWithin the framework presented here (Fig. 1), soil restoration refers specifically to the process of recovering soils that are degraded or critically degraded to a state where it is safe and suitable for the intended use. We only focused on degraded soils which are soils with severe loss of function and/or biological activity but with less cumulative impacts compared to critically degraded soils, to estimate the restoration potential. These soils represent more realistic and achievable targets for restoration actions in the short to medium term compared to critically degraded soils. Here, restoration potential was defined as the potential gain in soil biodiversity and functions if soil degradation processes were reduced to the minimum. For this, the TBI was also calculated by category of biotic indicators, but only for the degraded soils, and used as a metric of restoration potential. Higher values of TBI indicate that there is a higher potential of change for a given biotic indicator category if we were able to decrease or remove all soil degradation processes. To assess the overall restoration potential, TBI values were standardized for each category of biotic indicator and then summed, with weights assigned to give equal importance (50%) to soil biodiversity and ecosystem functioning. For the main analyses, we considered only positive changes from the current state to the non-degradation reference, as restoration strategies typically aim to enhance biodiversity and functioning rather than the opposite. However, since soil restoration may also seek to reduce undesirable biodiversity that proliferates under degradation, we conducted a side analysis accounting for overall changes (both gains and losses in biodiversity and functioning) to ensure a more objective assessment (Supplementary Fig. 15, 16).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData will be made available in a public repository after acceptance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCode will be made available in a public repository after acceptance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge the support of iDiv, which is funded by the German Research Foundation (DFG – FZT 118, 202548816), as well as by the DFG (Ei 862/29-1; Ei 862/31-1). We also acknowledge funding from the Horizon Europe EuropaBon project (grant agreement No. 101003553) to support the research. The LUCAS Survey is coordinated by Unit E4 of the Statistical Office of the European Union (EUROSTAT). The LUCAS Soil sample collection is supported by the Directorate-General Environment (DG-ENV), Directorate-General Agriculture and Rural Development (DG-AGRI), Directorate-General Climate Action (DG-CLIMA), and Directorate-General Eurostat (DG-ESTAT) of the European Commission.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAnthony, M. A., Bender, S. F. \u0026amp; van der Heijden, M. G. A. Enumerating soil biodiversity. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e120\u003c/strong\u003e, e2304663120 (2023).\u003c/li\u003e\n\u003cli\u003eBardgett, R. D. \u0026amp; van der Putten, W. H. Belowground biodiversity and ecosystem functioning. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e515\u003c/strong\u003e, 505\u0026ndash;511 (2014).\u003c/li\u003e\n\u003cli\u003eRillig, M. C. \u003cem\u003eet al.\u003c/em\u003e The role of multiple global change factors in driving soil functions and microbial biodiversity. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e366\u003c/strong\u003e, 886\u0026ndash;890 (2019).\u003c/li\u003e\n\u003cli\u003eWall, D. H., Nielsen, U. N. \u0026amp; Six, J. Soil biodiversity and human health. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e528\u003c/strong\u003e, 69\u0026ndash;76 (2015).\u003c/li\u003e\n\u003cli\u003ePhillips, H. \u003cem\u003eet al.\u003c/em\u003e Global change and their environmental stressors have a significant impact on soil biodiversity -- a meta-analysis. Preprint at https://doi.org/10.22541/au.167655684.49855023/v1 (2023).\u003c/li\u003e\n\u003cli\u003eFAO, ITPS, GSBI, SCBD \u0026amp; EC. \u003cem\u003eState of Knowledge of Soil Biodiversity: Status, Challenges and Potentialities, Report 2020.\u003c/em\u003e (Rome, 2020).\u003c/li\u003e\n\u003cli\u003eZeiss, R. \u003cem\u003eet al.\u003c/em\u003e Challenges of and opportunities for protecting European soil biodiversity. \u003cem\u003eConserv. Biol.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, e13930 (2022).\u003c/li\u003e\n\u003cli\u003eK\u0026ouml;ninger, J., Panagos, P., Jones, A., Briones, M. J. I. \u0026amp; Orgiazzi, A. In defence of soil biodiversity: Towards an inclusive protection in the European Union. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e268\u003c/strong\u003e, 109475 (2022).\u003c/li\u003e\n\u003cli\u003ePaleari, S. Is the European Union protecting soil? A critical analysis of Community environmental policy and law. \u003cem\u003eLand Use Policy\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 163\u0026ndash;173 (2017).\u003c/li\u003e\n\u003cli\u003eOrgiazzi, A. \u003cem\u003eet al.\u003c/em\u003e LUCAS Soil Biodiversity and LUCAS Soil Pesticides, new tools for research and policy development. \u003cem\u003eEur. J. Soil Sci.\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, (2022).\u003c/li\u003e\n\u003cli\u003eRomero, F. \u003cem\u003eet al.\u003c/em\u003e Soil health is associated with higher primary productivity across Europe. \u003cem\u003eNat. Ecol. Evol.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1847\u0026ndash;1855 (2024).\u003c/li\u003e\n\u003cli\u003eEisenhauer, N. \u003cem\u003eet al.\u003c/em\u003e The multidimensionality of soil macroecology. \u003cem\u003eGlob. Ecol. Biogeogr. J. Macroecology\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 4\u0026ndash;10 (2021).\u003c/li\u003e\n\u003cli\u003eLal, R. \u003cem\u003eet al.\u003c/em\u003e Soils and sustainable development goals of the United Nations: An International Union of Soil Sciences perspective. \u003cem\u003eGeoderma Reg.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, e00398 (2021).\u003c/li\u003e\n\u003cli\u003eEuropean Commission. \u003cem\u003eProposal for a Directive on Soil Monitoring and Resilience\u003c/em\u003e. (Online resource, https://environment.ec.europa.eu/publications/proposal-directive-soil-monitoring-and-resilience_en, 2023).\u003c/li\u003e\n\u003cli\u003eLehmann, J., Bossio, D. A., K\u0026ouml;gel-Knabner, I. \u0026amp; Rillig, M. C. The concept and future prospects of soil health. \u003cem\u003eNat. Rev. Earth Environ.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 544\u0026ndash;553 (2020).\u003c/li\u003e\n\u003cli\u003eVan Der Putten, W. H. \u003cem\u003eet al.\u003c/em\u003e Soil biodiversity needs policy without borders. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e379\u003c/strong\u003e, 32\u0026ndash;34 (2023).\u003c/li\u003e\n\u003cli\u003eIPBES, Montanarella, L., Scholes, R. \u0026amp; Brainich, A. \u003cem\u003eThe IPBES Assessment Report on Land Degradation and Restoration.\u003c/em\u003e https://zenodo.org/record/3237392 (2018) doi:10.5281/ZENODO.3237392.\u003c/li\u003e\n\u003cli\u003eEuropean Commission. Directorate General for the Environment. \u003cem\u003eThe EU Water Framework Directive.\u003c/em\u003e (Publications Office, LU, 2014).\u003c/li\u003e\n\u003cli\u003eBagnall, D. K. \u003cem\u003eet al.\u003c/em\u003e A minimum suite of soil health indicators for North American agriculture. \u003cem\u003eSoil Secur.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 100084 (2023).\u003c/li\u003e\n\u003cli\u003ePanagos, P. \u003cem\u003eet al.\u003c/em\u003e How the EU Soil Observatory is providing solid science for healthy soils. \u003cem\u003eEur. J. Soil Sci.\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, e13507 (2024).\u003c/li\u003e\n\u003cli\u003ePrăvălie, R. \u003cem\u003eet al.\u003c/em\u003e A unifying modelling of multiple land degradation pathways in Europe. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 3862 (2024).\u003c/li\u003e\n\u003cli\u003eSchoenholtz, S. H., Miegroet, H. V. \u0026amp; Burger, J. A. A review of chemical and physical properties as indicators of forest soil quality: challenges and opportunities. \u003cem\u003eFor. Ecol. Manag.\u003c/em\u003e \u003cstrong\u003e138\u003c/strong\u003e, 335\u0026ndash;356 (2000).\u003c/li\u003e\n\u003cli\u003eEisenhauer, N., Hines, J., Maestre, F. T. \u0026amp; Rillig, M. C. Reconsidering functional redundancy in biodiversity research. \u003cem\u003eNpj Biodivers.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 1\u0026ndash;4 (2023).\u003c/li\u003e\n\u003cli\u003eGuerra, C. A. \u003cem\u003eet al.\u003c/em\u003e Tracking, targeting, and conserving soil biodiversity. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e371\u003c/strong\u003e, 239\u0026ndash;241 (2021).\u003c/li\u003e\n\u003cli\u003eZwetsloot, M. J., Bongiorno, G., Barel, J. M., di Lonardo, D. P. \u0026amp; Creamer, R. E. A flexible selection tool for the inclusion of soil biology methods in the assessment of soil multifunctionality. \u003cem\u003eSoil Biol. Biochem.\u003c/em\u003e \u003cstrong\u003e166\u003c/strong\u003e, 108514 (2022).\u003c/li\u003e\n\u003cli\u003eGibbons, P. \u003cem\u003eet al.\u003c/em\u003e Rapidly quantifying reference conditions in modified landscapes. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e141\u003c/strong\u003e, 2483\u0026ndash;2493 (2008).\u003c/li\u003e\n\u003cli\u003eGuerra, C. A. \u003cem\u003eet al.\u003c/em\u003e Global hotspots for soil nature conservation. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e610\u003c/strong\u003e, 693\u0026ndash;698 (2022).\u003c/li\u003e\n\u003cli\u003eB\u0026uuml;chs, W. Biotic indicators for biodiversity and sustainable agriculture\u0026mdash;introduction and background. \u003cem\u003eAgric. Ecosyst. Environ.\u003c/em\u003e \u003cstrong\u003e98\u003c/strong\u003e, 1\u0026ndash;16 (2003).\u003c/li\u003e\n\u003cli\u003eKorthals, G. W., Bongers, T., Kammenga, J. E., Alexiev, A. D. \u0026amp; Lexmond, T. M. Long-term effects of copper and ph on the nematode community in an agroecosystem. \u003cem\u003eEnviron. Toxicol. Chem.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 979\u0026ndash;985 (1996).\u003c/li\u003e\n\u003cli\u003eQu, X. \u003cem\u003eet al.\u003c/em\u003e Deforestation impacts soil biodiversity and ecosystem services worldwide. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, e2318475121 (2024).\u003c/li\u003e\n\u003cli\u003eLabouyrie, M. \u003cem\u003eet al.\u003c/em\u003e Patterns in soil microbial diversity across Europe. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 3311 (2023).\u003c/li\u003e\n\u003cli\u003eHagan, J. G., Vanschoenwinkel, B. \u0026amp; Gamfeldt, L. We should not necessarily expect positive relationships between biodiversity and ecosystem functioning in observational field data. \u003cem\u003eEcol. Lett.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 2537\u0026ndash;2548 (2021).\u003c/li\u003e\n\u003cli\u003eHu, W. \u003cem\u003eet al.\u003c/em\u003e Aridity-driven shift in biodiversity\u0026ndash;soil multifunctionality relationships. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 5350 (2021).\u003c/li\u003e\n\u003cli\u003eDelgado-Baquerizo, M. \u003cem\u003eet al.\u003c/em\u003e Multiple elements of soil biodiversity drive ecosystem functions across biomes. \u003cem\u003eNat. Ecol. Evol.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 210\u0026ndash;220 (2020).\u003c/li\u003e\n\u003cli\u003ePanagos, P. \u003cem\u003eet al.\u003c/em\u003e Soil Conservation in Europe: Wish or Reality? \u003cem\u003eLand Degrad. Dev.\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 1547\u0026ndash;1551 (2016).\u003c/li\u003e\n\u003cli\u003eMammola, S. \u003cem\u003eet al.\u003c/em\u003e Towards a taxonomically unbiased European Union biodiversity strategy for 2030. \u003cem\u003eProc. R. Soc. B Biol. Sci.\u003c/em\u003e \u003cstrong\u003e287\u003c/strong\u003e, 20202166 (2020).\u003c/li\u003e\n\u003cli\u003eBoardman, J. \u0026amp; Vandaele, K. Soil erosion and runoff: The need to rethink mitigation strategies for sustainable agricultural landscapes in western Europe. \u003cem\u003eSoil Use Manag.\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 673\u0026ndash;685 (2023).\u003c/li\u003e\n\u003cli\u003eBorrelli, P. \u003cem\u003eet al.\u003c/em\u003e Policy implications of multiple concurrent soil erosion processes in European farmland. \u003cem\u003eNat. Sustain.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 103\u0026ndash;112 (2022).\u003c/li\u003e\n\u003cli\u003eHartmann, M. \u0026amp; Six, J. Soil structure and microbiome functions in agroecosystems. \u003cem\u003eNat. Rev. Earth Environ.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 4\u0026ndash;18 (2022).\u003c/li\u003e\n\u003cli\u003eBorrelli, P. \u003cem\u003eet al.\u003c/em\u003e Land use and climate change impacts on global soil erosion by water (2015-2070). \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 21994\u0026ndash;22001 (2020).\u003c/li\u003e\n\u003cli\u003eArneth, A. \u003cem\u003eet al.\u003c/em\u003e Restoring Degraded Lands. \u003cem\u003eAnnu. Rev. Environ. Resour.\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 569\u0026ndash;599 (2021).\u003c/li\u003e\n\u003cli\u003ePotapov, A. M. \u003cem\u003eet al.\u003c/em\u003e Global monitoring of soil animal communities using a common methodology. \u003cem\u003eSOIL Org.\u003c/em\u003e \u003cstrong\u003e94\u003c/strong\u003e, 55\u0026ndash;68 (2022).\u003c/li\u003e\n\u003cli\u003eK\u0026ouml;ninger, J. \u003cem\u003eet al.\u003c/em\u003e Ecosystem type drives soil eukaryotic diversity and composition in Europe. \u003cem\u003eGlob. Change Biol.\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 5706\u0026ndash;5719 (2023).\u003c/li\u003e\n\u003cli\u003eShen, W., Le, S., Li, Y. \u0026amp; Hu, F. SeqKit: A Cross-Platform and Ultrafast Toolkit for FASTA/Q File Manipulation. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e0163962 (2016).\u003c/li\u003e\n\u003cli\u003eBushnell, B. BBMap: A Fast, Accurate, Splice-Aware Aligner. https://escholarship.org/uc/item/1h3515gn (2014).\u003c/li\u003e\n\u003cli\u003eMartin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. \u003cem\u003eEMBnet.journal\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 10 (2011).\u003c/li\u003e\n\u003cli\u003eCallahan, B. J. \u003cem\u003eet al.\u003c/em\u003e DADA2: High-resolution sample inference from Illumina amplicon data. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 581\u0026ndash;583 (2016).\u003c/li\u003e\n\u003cli\u003eBolyen, E. \u003cem\u003eet al.\u003c/em\u003e Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 852\u0026ndash;857 (2019).\u003c/li\u003e\n\u003cli\u003eGuillou, L. \u003cem\u003eet al.\u003c/em\u003e The Protist Ribosomal Reference database (PR2): a catalog of unicellular eukaryote small sub-unit rRNA sequences with curated taxonomy. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, D597-604 (2013).\u003c/li\u003e\n\u003cli\u003eQuast, C. \u003cem\u003eet al.\u003c/em\u003e The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, D590\u0026ndash;D596 (2013).\u003c/li\u003e\n\u003cli\u003eZinger, L. \u003cem\u003eet al.\u003c/em\u003e metabaR: An r package for the evaluation and improvement of DNA metabarcoding data quality. \u003cem\u003eMethods Ecol. Evol.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 586\u0026ndash;592 (2021).\u003c/li\u003e\n\u003cli\u003e\u0026Ouml;zkurt, E. \u003cem\u003eet al.\u003c/em\u003e LotuS2: an ultrafast and highly accurate tool for amplicon sequencing analysis. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 176 (2022).\u003c/li\u003e\n\u003cli\u003eTedersoo, L. \u003cem\u003eet al.\u003c/em\u003e The Global Soil Mycobiome consortium dataset for boosting fungal diversity research. \u003cem\u003eFungal Divers.\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 573\u0026ndash;588 (2021).\u003c/li\u003e\n\u003cli\u003eBeule, L. \u0026amp; Karlovsky, P. Improved normalization of species count data in ecology by scaling with ranked subsampling (SRS): application to microbial communities. \u003cem\u003ePeerJ\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, e9593 (2020).\u003c/li\u003e\n\u003cli\u003eLe Guillarme, N. \u0026amp; Thuiller, W. A practical approach to constructing a knowledge graph for soil ecological research. \u003cem\u003eEur. J. Soil Biol.\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 103497 (2023).\u003c/li\u003e\n\u003cli\u003eSieriebriennikov, B., Ferris, H. \u0026amp; De Goede, R. G. M. NINJA: An automated calculation system for nematode-based biological monitoring. \u003cem\u003eEur. J. Soil Biol.\u003c/em\u003e \u003cstrong\u003e61\u003c/strong\u003e, 90\u0026ndash;93 (2014).\u003c/li\u003e\n\u003cli\u003eYeates, G. W., Bongers, T., De Goede, R. G., Freckman, D. W. \u0026amp; Georgieva, S. S. Feeding habits in soil nematode families and genera-an outline for soil ecologists. \u003cem\u003eJ. Nematol.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 315\u0026ndash;331 (1993).\u003c/li\u003e\n\u003cli\u003eLouca, S., Parfrey, L. W. \u0026amp; Doebeli, M. Decoupling function and taxonomy in the global ocean microbiome. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e353\u003c/strong\u003e, 1272\u0026ndash;1277 (2016).\u003c/li\u003e\n\u003cli\u003eMaron, P.-A. \u003cem\u003eet al.\u003c/em\u003e High Microbial Diversity Promotes Soil Ecosystem Functioning. \u003cem\u003eAppl. Environ. Microbiol.\u003c/em\u003e \u003cstrong\u003e84\u003c/strong\u003e, e02738-17 (2018).\u003c/li\u003e\n\u003cli\u003eP\u0026otilde;lme, S. \u003cem\u003eet al.\u003c/em\u003e FungalTraits: a user-friendly traits database of fungi and fungus-like stramenopiles. \u003cem\u003eFungal Divers.\u003c/em\u003e \u003cstrong\u003e105\u003c/strong\u003e, 1\u0026ndash;16 (2020).\u003c/li\u003e\n\u003cli\u003eScheu, S. Automated measurement of the respiratory response of soil microcompartments: Active microbial biomass in earthworm faeces. \u003cem\u003eSoil Biol. Biochem.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 1113\u0026ndash;1118 (1992).\u003c/li\u003e\n\u003cli\u003eAnderson, J. P. E. \u0026amp; Domsch, K. H. A physiological method for the quantitative measurement of microbial biomass in soils. \u003cem\u003eSoil Biol. Biochem.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 215\u0026ndash;221 (1978).\u003c/li\u003e\n\u003cli\u003eBreitkreuz, C. \u003cem\u003eet al.\u003c/em\u003e Can We Estimate Functionality of Soil Microbial Communities from Structure-Derived Predictions? A Reality Test in Agricultural Soils. \u003cem\u003eMicrobiol. Spectr.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e00278-21 (2021).\u003c/li\u003e\n\u003cli\u003eSinsabaugh, R. L. \u003cem\u003eet al.\u003c/em\u003e Soil microbial activity in a \u003cem\u003eLiquidambar\u003c/em\u003e plantation unresponsive to CO2-driven increases in primary production. \u003cem\u003eAppl. Soil Ecol.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 263\u0026ndash;271 (2003).\u003c/li\u003e\n\u003cli\u003eSmith, L. C. \u003cem\u003eet al.\u003c/em\u003e Large‐scale drivers of relationships between soil microbial properties and organic carbon across Europe. \u003cem\u003eGlob. Ecol. Biogeogr.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 2070\u0026ndash;2083 (2021).\u003c/li\u003e\n\u003cli\u003eS\u0026uuml;nnemann, M. \u003cem\u003eet al.\u003c/em\u003e Climate change and cropland management compromise soil integrity and multifunctionality. \u003cem\u003eCommun. Earth Environ.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 394 (2023).\u003c/li\u003e\n\u003cli\u003eHill, M. O. Diversity and Evenness: A Unifying Notation and Its Consequences. \u003cem\u003eEcology\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 427\u0026ndash;432 (1973).\u003c/li\u003e\n\u003cli\u003eCalder\u0026oacute;n‐Sanou, I., M\u0026uuml;nkem\u0026uuml;ller, T., Boyer, F., Zinger, L. \u0026amp; Thuiller, W. From environmental DNA sequences to ecological conclusions: How strong is the influence of methodological choices? \u003cem\u003eJ. Biogeogr.\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, 193\u0026ndash;206 (2020).\u003c/li\u003e\n\u003cli\u003eUllah, S. \u003cem\u003eet al.\u003c/em\u003e The responses of extracellular enzyme activities and microbial community composition under nitrogen addition in an upland soil. \u003cem\u003ePLoS ONE\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, e0223026 (2019).\u003c/li\u003e\n\u003cli\u003eNeemisha \u0026amp; Sharma, S. Soil Enzymes and Their Role in Nutrient Cycling. in \u003cem\u003eStructure and Functions of Pedosphere\u003c/em\u003e (eds Giri, B., Kapoor, R., Wu, Q.-S. \u0026amp; Varma, A.) 173\u0026ndash;188 (Springer Nature, Singapore, 2022). doi:10.1007/978-981-16-8770-9_8.\u003c/li\u003e\n\u003cli\u003eBrun, P., Zimmermann, N. E., Hari, C., Pellissier, L. \u0026amp; Karger, D. N. Global climate-related predictors at kilometer resolution for the past and future. \u003cem\u003eEarth Syst. Sci. Data\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 5573\u0026ndash;5603 (2022).\u003c/li\u003e\n\u003cli\u003ePanagos, P. \u003cem\u003eet al.\u003c/em\u003e European Soil Data Centre 2.0: Soil data and knowledge in support of the EU policies. \u003cem\u003eEur. J. Soil Sci.\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, e13315 (2022).\u003c/li\u003e\n\u003cli\u003eRiley, S. J., DeGloria, S. D. \u0026amp; Elliot, R. Index that quantifies topographic heterogeneity. \u003cem\u003eIntermt. J. Sci.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 23\u0026ndash;27 (1999).\u003c/li\u003e\n\u003cli\u003eWeiss, A. Topographic position and landforms analysis. in \u003cem\u003ePoster presentation, ESRI user conference, San Diego, CA\u003c/em\u003e vol. 200 (2001).\u003c/li\u003e\n\u003cli\u003eEuropean Commission. Statistical Office of the European Union. \u003cem\u003eRedesign Sample for Land Use/Cover Area Frame Survey (LUCAS) 2018: 2018 Edition.\u003c/em\u003e (Publications Office, LU, 2018).\u003c/li\u003e\n\u003cli\u003eDou, Y. \u003cem\u003eet al.\u003c/em\u003e A new European land systems representation accounting for landscape characteristics. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 2215\u0026ndash;2234 (2021).\u003c/li\u003e\n\u003cli\u003eBorrelli, P., Lugato, E., Montanarella, L. \u0026amp; Panagos, P. A New Assessment of Soil Loss Due to Wind Erosion in European Agricultural Soils Using a Quantitative Spatially Distributed Modelling Approach. \u003cem\u003eLand Degrad. Dev.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 335\u0026ndash;344 (2017).\u003c/li\u003e\n\u003cli\u003ePanagos, P. \u003cem\u003eet al.\u003c/em\u003e A Soil Erosion Indicator for Supporting Agricultural, Environmental and Climate Policies in the European Union. \u003cem\u003eRemote Sens.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1365 (2020).\u003c/li\u003e\n\u003cli\u003ePanagos, P., Borrelli, P. \u0026amp; Poesen, J. Soil loss due to crop harvesting in the European Union: A first estimation of an underrated geomorphic process. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cstrong\u003e664\u003c/strong\u003e, 487\u0026ndash;498 (2019).\u003c/li\u003e\n\u003cli\u003eBallabio, C. \u003cem\u003eet al.\u003c/em\u003e Copper distribution in European topsoils: An assessment based on LUCAS soil survey. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cstrong\u003e636\u003c/strong\u003e, 282\u0026ndash;298 (2018).\u003c/li\u003e\n\u003cli\u003eBallabio, C. \u003cem\u003eet al.\u003c/em\u003e A spatial assessment of mercury content in the European Union topsoil. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cstrong\u003e769\u003c/strong\u003e, 144755 (2021).\u003c/li\u003e\n\u003cli\u003eBallabio, C., Jones, A. \u0026amp; Panagos, P. Cadmium in topsoils of the European Union \u0026ndash; An analysis based on LUCAS topsoil database. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cstrong\u003e912\u003c/strong\u003e, 168710 (2024).\u003c/li\u003e\n\u003cli\u003eVan Eynde, E., Fendrich, A. N., Ballabio, C. \u0026amp; Panagos, P. Spatial assessment of topsoil zinc concentrations in Europe. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cstrong\u003e892\u003c/strong\u003e, 164512 (2023).\u003c/li\u003e\n\u003cli\u003eEuropean Commission. Joint Research Centre. Institute for Environment and Sustainability. \u003cem\u003eThreats to Soil Quality in Europe.\u003c/em\u003e (Publications Office, LU, 2008).\u003c/li\u003e\n\u003cli\u003eRoberts, D. R. \u003cem\u003eet al.\u003c/em\u003e Cross‐validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure. \u003cem\u003eEcography\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 913\u0026ndash;929 (2017).\u003c/li\u003e\n\u003cli\u003ePedregosa, F. \u003cem\u003eet al.\u003c/em\u003e Scikit-learn: Machine Learning in Python. https://doi.org/10.48550/ARXIV.1201.0490 (2012) doi:10.48550/ARXIV.1201.0490.\u003c/li\u003e\n\u003cli\u003eThuiller, W., Lafourcade, B., Engler, R. \u0026amp; Ara\u0026uacute;jo, M. B. BIOMOD \u0026ndash; a platform for ensemble forecasting of species distributions. \u003cem\u003eEcography\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 369\u0026ndash;373 (2009).\u003c/li\u003e\n\u003cli\u003eLegendre, P. A temporal beta-diversity index to identify sites that have changed in exceptional ways in space\u0026ndash;time surveys. \u003cem\u003eEcol. Evol.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 3500\u0026ndash;3514 (2019).\u003c/li\u003e\n\u003cli\u003eDray, S. \u003cem\u003eet al.\u003c/em\u003e adespatial: Multivariate Multiscale Spatial Analysis. (2021).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u003c/strong\u003e Biotic indicators used for the assessment of the soil health and sources on the methods employed to identify or measure each indicator.\u003c/p\u003e\n\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"664\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 30px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCategory\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBiotic indicator\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDescription\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSource/References\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"19\" style=\"width: 30px;\"\u003e\n \u003cp\u003eSoil biodiversity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" style=\"width: 96px;\"\u003e\n \u003cp\u003eMetazoa\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eNematode bacterivores\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of nematodes feeding on bacteria\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" style=\"width: 88px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNINJA\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eSieriebrienniko, Ferris \u0026amp; de Goede (2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eNematode fungivores\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of nematodes feeding on fungal hyphae\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eNematode herbivores\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of nematodes feeding on plant roots\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eNematode predator\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of nematodes feeding on other animals or nematodes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"7\" style=\"width: 96px;\"\u003e\n \u003cp\u003eFungi\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eAnimal parasitic fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of fungi that parasite animals\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"7\" style=\"width: 88px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFungalTraits\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eP\u0026otilde;lme \u003cem\u003eet al.\u0026nbsp;\u003c/em\u003e(2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eArbuscular mycorrhizal fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of arbuscular mycorrhizal fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eEctomycorrhizal fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of ectomycorrhizal fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eMycoparasitic fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of fungi that parasite other fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003ePlant pathogenic fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of fungi that parasite plants\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eRoot endophytic fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of root endophytes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eSaprotrophic fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of saprotrophic fungi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" style=\"width: 96px;\"\u003e\n \u003cp\u003eProtist\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003ePhotoautotrophic protist\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of phototrophic and autotrophic protists\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" style=\"width: 88px;\"\u003e\n \u003cp\u003eMultiple sources\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eMicrobivorous protist\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of protists feeding on bacteria, fungi and/or other protists\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003ePlant pathogenic protist\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of protists that parasite plants\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" style=\"width: 96px;\"\u003e\n \u003cp\u003eBacteria\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eChemoheterotrophic bacteria\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of chemoheterotrophs, which derive energy by oxidising organic compounds such as glucose\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"5\" style=\"width: 88px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFAPROTAX\u003c/strong\u003e Louca, Parfrey, \u0026amp; Doebeli (2016)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003ePathogenic and parasitic bacteria\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of bacteria that parasite animals and plants\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eAutotrophic bacteria\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of phototrophic and autotrophic bacteria (includes all Chloroflexi)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eN-cycling bacteria\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of bacteria involved in nitrification, denitrification and N-fixation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eDecomposer bacteria\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eDiversity of all other bacteria found in soils (and not classified in previous groups)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"7\" style=\"width: 30px;\"\u003e\n \u003cp\u003eEcosystem functions\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 96px;\"\u003e\n \u003cp\u003eSoil microbial activity\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eBasal respiration\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003ePotential basal respiration measured in lab with O2-microcompensation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003eScheu (1992)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eMicrobial biomass\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 300px;\"\u003e\n \u003cp\u003eMicrobial biomass by substrate-induced respiration\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003eAnderson \u0026amp; Domsch (1978)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" style=\"width: 96px;\"\u003e\n \u003cp\u003eCarbon cycling\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eXylosidase\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"5\" style=\"width: 300px;\"\u003e\n \u003cp\u003eActivity potential, i.e. substrate turnover rate at pH 5 and 25\u0026deg;C\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"5\" style=\"width: 88px;\"\u003e\n \u003cp\u003eBreitkreuz \u003cem\u003eet al.\u003c/em\u003e (2021)\u003c/p\u003e\n \u003cp\u003eSinsabaugh \u003cem\u003eet al.\u003c/em\u003e 2003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eCellulase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eBeta-glucosidase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" style=\"width: 96px;\"\u003e\n \u003cp\u003eNutrient cycling\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eN-acetyl-glucosaminidase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 150px;\"\u003e\n \u003cp\u003eAcid phosphatase\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"soil biodiversity, soil functioning, biotic indicators, soil degradation, machine learning","lastPublishedDoi":"10.21203/rs.3.rs-8660880/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8660880/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSoils host a significant proportion of biodiversity on Earth providing ecosystem functions vital to human well-being, making it imperative to include them and their ecological features when addressing sustainability goals. We performed a comprehensive assessment of soil health across Europe by explicitly integrating biotic and abiotic indicators alongside soil degradation processes. We further identified areas with high restoration potential, quantifying the potential positive changes achievable when mitigating degradation processes. Our results show that 93% of soils in Europe are either degraded (62%) or in a moderate state (31%), with only 7% having a good (6.5%) or high (0.5%) health status. We found Southeast Europe to have the highest restoration potential, particularly for forests and annual crops. By providing spatially explicit indicators of soil health and restoration potential, our approach offers valuable guidance to support sustainable soil management and inform policies aimed at enhancing soil health across Europe.\u003c/p\u003e","manuscriptTitle":"An Integrated Assessment of European Soil Health and Restoration potential","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-28 05:04:16","doi":"10.21203/rs.3.rs-8660880/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a851a550-4278-41ed-967c-9415427bdfee","owner":[],"postedDate":"January 28th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":61771153,"name":"Biological sciences/Ecology/Ecological modelling"},{"id":61771154,"name":"Biological sciences/Ecology/Ecosystem ecology"},{"id":61771155,"name":"Earth and environmental sciences/Ecology/Macroecology"}],"tags":[],"updatedAt":"2026-02-23T19:50:39+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-28 05:04:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8660880","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8660880","identity":"rs-8660880","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}