European rural regions supporting and hindering the Sustainable Development Goals

European rural regions supporting and hindering the Sustainable Development Goals | 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 Analysis European rural regions supporting and hindering the Sustainable Development Goals Murray Scown, Kimberly Nicholas This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2941468/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Nov, 2024 Read the published version in Communications Earth & Environment → Version 1 posted You are reading this latest preprint version Abstract Transformation of rural land systems is essential if the European Union is to achieve its goal of fair and healthy food systems while becoming the first climate-neutral continent and halting biodiversity loss. Here we develop and apply a method to quantitatively assess the environmental and social sustainability of rural land systems in Europe, with regards to the EU Sustainable Development Goals and the Common Agricultural Policy. Using spatial hotspot analyses based on 24 indicators at the NUTS2 regional level, we identified two “brightspots” with good environmental and social performance (Nordics and Central Europe), and five “dragspots” hindering sustainability: the Balkans, the Lowlands, Northern Italy, Southern Italy and Malta, and Southern Spain. Existing subsidies over-reward large, intensive, unsustainable farms. A shift to low-intensity stewardship of high nature value farmland, and better integration of forests is necessary if rural systems are to transform to meet their social goals. Earth and environmental sciences/Environmental social sciences/Sustainability Scientific community and society/Agriculture Scientific community and society/Geography food systems land use multifunctionality leverage points SDGs spatial analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Land use and exploitation of animals and plants for consumption are the main drivers of global declines in nature 1 . To achieve the United Nations Sustainable Development Goals (SDGs) and stay within planetary boundaries, transforming rural land production systems while eliminating unsustainable consumption is essential 2,3 . The European Union (EU) aims to foster fair and healthy food systems and halt biodiversity loss as Europe becomes the first climate-neutral continent 4 . Around one-quarter of the 100 European Union SDG indicators (EU SDGs) are related to rural land systems 5 , but recent analyses show EU implementation of both SDGs and the Common Agricultural Policy (CAP) often counter established sustainability goals 5,6 . Making sustainability transformations requires both overcoming unsustainable instances of business as usual 7,8 , and identifying pathways for replacing the status quo with sustainable systems. Hotspot analysis builds on the principle of targeting limited resources at a particular place or process to maximize intervention effectiveness, like Meadows’ leverage points 9 . Hotspot analysis is often based on, for example, identifying outliers in the statistical distribution of a phenomenon (e.g., the top/bottom 10%), such as hotspots of change in land use extent and management intensity 10 ; or identifying sectors/groups/products with a disproportionately high contribution to some phenomenon, such as greenhouse gas emissions 11 . Advancing such approaches, spatial hotspot analysis incorporates both the statistical distribution and spatial patterns of a phenomenon, thus accounting for regional environmental conditions (e.g., climate, terrain) or national differences in governance and practices. While granular land use classifications (e.g., ref. 12 ) can be difficult to interpret and communicate at governance-relevant scales, spatial hotspot analysis enables identifying broader patterns which can be targeted by regional policies. Spatial hotspot analysis has been successfully applied in fields including ecosystem services 13 , welfare economics 14 , and human health 15 . The linked concept of “bright spots,” such as identifying “Lighthouse farms” 16 , refers to hotspots with the potential to point to pathways to transformation 17 . Here we use spatial hotspot analysis to identify spatial “brightspots” of sustainable rural land systems in Europe that can inspire us towards better pathways aligned with meeting SDGs, and spatial “dragspots” of unsustainable systems that hold us back. We developed indicators for environmentally and socially sustainable rural land systems in 283 NUTS2 regions. Specifically, we selected 15 environmental and 9 social indicators drawn from the overlap of the EU SDGs and the CAP. We then analysed the fraction of calories delivered to the food system from within each hotspot, and CAP subsidies paid to these regions. This enabled us to identify both the location and characteristics of sustainable rural land systems in Europe, their role in the food system, and their relationship to EU policy instruments. These results can inform the redirection of incentives away from unsustainable systems and towards successful practices. Results Summary of indices by country We calculated three sustainability indices for each NUTS2 region: ( 1 ) an environmental index based on 15 indicators, ( 2 ) a social index based on 9 indicators, and ( 3 ) an overall sustainability index, which combined these 24 indicators in an unweighted average (please see Methods for details). Indicators were range-standardised to between zero and one before calculating indices. All three indices were approximately normally distributed across the 283 NUTS2 regions (Extended Data Fig. 1 ). The environmental index had a substantially higher mean (0.63) than the social index (0.46), but values within the two spanned a similarly wide range (0.46 and 0.45, respectively; Extended Data Fig. 1 ). Due to its calculation as the average of the two, the overall sustainability index fell in between (mean 0.55, range 0.31) (Extended Data Fig. 1 ). We found that countries with high environmental scores tended to have low social ones, and vice versa. The four countries with the highest mean environmental index score (Croatia, Latvia, Slovenia, and Estonia) all had a mean social index in the lower half of countries (Extended Data Fig. 2 ). Similarly, the three highest countries for the mean social score (Germany, Luxembourg, and Iceland) were at the mid to lower end of environmental scores. Sweden had the highest mean score for the overall sustainability index (Extended Data Fig. 2 ), a reflection of being in the top six countries for both mean environmental and social scores. Malta had by far the lowest overall index score, due to having the lowest mean environmental score and the seventh-lowest mean social score (Extended Data Fig. 2 ). Countries whose NUTS2 scores ranged widely included Belgium for environmental, Switzerland and the United Kingdom for social, and Italy for both indices. Brightspots and dragspots Our analysis of 24 policy-aligned indicators of environmentally and socially sustainable rural land systems in Europe reveals: ( 1 ) two main brightspots and two main dragspots for the environmental subindex (Fig. 1 ); ( 2 ) one main brightspot and three dragspots for the social subindex (Fig. 2 ); and ( 3 ) two overall brightspots most in line with the EU SDGs and five overall dragspots most hindering achieving them (Fig. 3 ). In general, brightspots had at least two indicators with outstanding performance, along with solid performance across most indicators within the relevant index (Figs. 1 and 2 ). Good environmental performance was associated with less intensive methods (e.g., relatively low inputs of pesticides and irrigation, and low fertilizer application as indicated by P and N balance). Environmentally unsustainable food systems relied on intensive inputs in monocultural landscapes of little nature value, and particularly with high nutrient imbalances, most clearly exemplified in the Lowlands. Good social performance featured low energy use when accompanied by strong economic performance including high farm employment and income, supported by higher agricultural training, alongside low risk of poverty. Environmental sustainability index The main environmental brightspot areas were around in the Northern Baltic region (Sweden, Finland, Estonia, and Latvia), and in much of the Balkans region with parts of Slovakia (Fig. 1 ). The two areas were both characterised by very high (“good”) values for eight of the 15 environmental indicators: high groundwater quality, few artificial surfaces, low soil erosion, low gross N and P balances, low reliance on irrigation, low use of pesticides, and low emissions of greenhouse gases and ammonia. The Balkans environmental brightspot had generally lower forest area, less organic farming, and higher emissions of particulate matter from agriculture than the Northern Baltic environmental brightspot. However, the Balkans performed better in terms of Natura 2000 area and high nature value farmland. The main environmental dragspot areas were in The Netherlands, Belgium, and Denmark, along with western Germany and northern France (referred to as the “Northwest Continental” dragspot), and in Northern Italy (Fig. 1 ). Both environmental dragspot areas had very little organic farming, low forest area, low farmland birds, and little Natura 2000 area or high nature value farmland. Where they differed was that Northern Italy had much higher rates of soil erosion and greater dependence on irrigation, while the Northwest Continental dragspot performed very poorly in terms of gross N balance. [Fig. 1 ] Social sustainability index The main social brightspot was in Central Europe, including all of Germany, northern Austria, and western Czech Republic (Fig. 2 ). This brightspot performed very well in terms of low energy use in agriculture and low unemployment, as well as a low risk of poverty or social exclusion. This area also had much higher levels of agricultural training, more renewable energy use and production in agriculture, and higher farm incomes, compared to the dragspots, but still variable and in the mid-range when compared to all of Europe. The main social dragspots were in the Balkans and Cyprus, the Iberian Peninsula, and Southern Italy and Malta (Fig. 2 ). These areas all had extremely low levels of agricultural training, renewable energy use and production in agriculture, farm income, and GDP per capita. Rural purchasing power was also low in all of these dragspots, although not to the same degree as the previous indicators. Risk of poverty or social exclusion was generally lower in the Iberian dragspot than the other two, while the Balkans performed slightly better than Iberia and Southern Italy in terms of unemployment. Interestingly, all social dragspots also had relatively low rates of energy consumption in agriculture. [Fig. 2 ] Overall sustainability combining social and environmental The two brightspot areas for overall sustainability were the Nordics (Sweden and Finland) and Central Europe (including most of Germany, all of Austria, Luxembourg, parts of Switzerland, and small western portions of Slovakia and Czechia) (Fig. 3 ). The five dragspot areas for overall sustainability covered parts of: the Balkans (northern Romania and southeast Bulgaria, also including southeast Hungary), the Lowlands (the Netherlands and Belgium), Northern Italy, Southern Italy and Malta, and Southern Spain (Fig. 3 ). The Nordic brightspot was characterised by very high environmental index scores, with substantially lower scores for the social index, although still near the European average. By contrast, the Central Europe brightspot area had average environmental but very high social index scores. The Balkans, Southern Spain, and Southern Italy dragspots were all characterised by very low social index scores, but with good environmental conditions (with the exception of Malta), indicating that the poor social conditions dragged down the overall index in these regions. On the other hand, Northern Italy and the Lowlands performed poorly for the environment, and around average for social factors. In all but one case, to qualify as a bright (drag)spot, either the environmental or social index was extremely high (low) relative to the European average, while the other index was close to the average (Fig. 3 ). The Nordics were the exception, where environmental and social performance were both well above average. In some cases, good environmental performance despite poor social performance kept a region out of being an overall dragspot (e.g., Greece), and reasonable social performance kept environmental dragspot from being overall ones (e.g., Denmark). [Fig. 3 ] Calories delivered to the food system We analysed the fraction of calories delivered to the food system produced in each overall brightspot and dragspot (please see Methods for details). Three of the five dragspots (Southern Italy, Lowlands, and Southern Spain) delivered, on average, a higher fraction of calories to the food system than did the two brightspots (Fig. 4 ). The Nordic brightspot delivered the lowest fraction of calories to the food system (Fig. 4 ), indicating that environmental and social successes here may be contributing to other goals besides maximizing food production. Of all brightspots and dragspots, the Central Europe brightspot (which also contained the largest number of NUTS2 regions) had the widest range of calorie fractions delivered to the food system (Fig. 4 ), with the Köln region (NUTS2 ID “DEA2”) being a particularly high outlier within this brightspot. The Southern Spain and Lowlands dragspots also had particularly wide ranges of calories delivered, compared to the other dragspots and the Nordic brightspot (Fig. 4 ). [Fig. 4 ] Subsidies paid to brightspots and dragspots To determine whether and how EU subsidies align with sustainability, we analysed the average CAP payment per hectare in each brightspot and dragspot—as well as in all other regions—according to their intended support scheme (please see Methods for details). We found that the dragspots of Northern Italy, Lowlands, and Southern Italy receive the highest level of support per hectare (Fig. 5 A). Southern Italy and the Lowlands also deliver the highest fractions of calories to the food system, on average (Fig. 4 ), indicating that large amounts of subsidies (intended to support non-market environmental and social values) are being paid to dragspots producing marketable food items but under-performing in terms of environmental and social outcomes. By contrast, the Northern Italy dragspot receives the highest average payments per hectare yet delivers a low fraction of calories to the food system. Future research should examine what is being produced in this region and in which ways to result in neither high-performing environmental no social outcomes, nor food, despite large amounts of subsidies. While all five dragspots had moderate to high subsidies per hectare, the two brightspots received very different amounts (Fig. 5 A). The Nordic brightspot, in particular, received less than €150 per hectare (in purchasing power parity), compared to €450 in the Central Europe brightspot, €350 in the lowest recipient dragspot (the Balkans), and just below €400 in all other regions. The Nordic brightspot also delivered the lowest fraction of calories to the food system (Fig. 4 ), indicating that this region is substantially different from other brightspots and dragspots in terms of the amounts of subsidies received and the amount of food produced. In terms of the distribution of subsidies between income support and environmental schemes (Fig. 5 B), we found the largest fractions of environmental support being paid to the two overall brightspots and to the Balkans dragspot, which is a reasonable use of such subsidies considering good environmental performance in these regions (Fig. 3 ). The largest fraction of income support was paid to the Lowlands dragspot, where its effectiveness is questionable considering the relatively poor environmental performance and average social conditions in this region (Fig. 3 ). The Southern Spain dragspot received the lowest fraction of environmental support despite having relatively high environmental performance (Fig. 3 ); conversely, the large fraction of income support (second only to the Lowlands) does not appear to be translating into social outcomes in this region (Fig. 3 ). Overall, the areas with the best environmental performance received proportionally more environmental subsidies, in line with a performance-based approach. However, already well-off and high-producing areas received high levels of income and general support, despite such support being unnecessary to achieve the CAP goals of securing farmer livelihoods, and already being compensated by the market for producing crops. [Fig. 5 ] Discussion We identify four unsustainable patterns in current European rural land systems: ( 1 ) a widespread trade-off between social and environmental sustainability; ( 2 ) the regions producing the most calories for food doing so at the cost of either environmental and/or social sustainability; ( 3 ) over-reliance on agricultural systems to deliver EU sustainability goals outside of urban areas, without governing forests; and ( 4 ) EU CAP subsidies often directed ineffectively. Intensive food production, intensive harm Intensive food production in Europe causes disproportionate environmental harm, while less agriculturally intensive areas are associated with better sustainability outcomes 18 . While Debonne et al. 18 described a dichotomy between productivist and post-productivist paradigms , this pattern of usage may be better explained by land use suitability . In a global context, European agriculture is rather steady, showing a rate of change in recent decades of < 0.5% on most outcome metrics 19 . However, trends within the continent show a bifurcation, with intensification occurring in highly suitable farming areas, while cropland is shrinking and inputs de-intensifying in more marginal areas 10 . This split pattern indicates that maximizing intensive production is prioritized wherever conditions allow. Overall, rural areas in Europe with the worst EU SDG performance feature intensive crop and animal agriculture. Most of Europe is at high risk of exceeding the planetary boundary for nitrogen, and increasing risk of exceeding the boundary for biosphere integrity 20 . Debonne et al. 18 saw that (productivist) megastables, with > 500 head of livestock, are concentrated in the Lowlands and Northern Italy, areas we also found to be dragspots. Our dragspots also overlap with Europe’s highest areas of nitrogen exceedance (the Lowlands) and excessive use of livestock antibiotics (Italy) 18 . Intensive farming methods in the Lowlands, characterized by increasing yields in croplands and permanent crops 12 , align with our dragspots, while areas with a share of organic agriculture five times above the European average, including central Sweden and parts of Switzerland 18 , align with our brightspots of sustainability. To keep global food production within planetary boundaries, a combination of dietary shifts to reduce animal consumption, prevention of food loss and waste, and adoption of efficient technology are necessary; no single measure is sufficient 2 . Forestry and sustainable rural land systems Intensification of agriculture is associated with lower performance on European SDGs, while intensification of forestry is not. Our two overall sustainability brightspots include areas with high harvested timber volumes and intensity 21 and with a strong increase in roundwood production since 1990 10 , with an intensification of wood production in southern Sweden and central Germany 12 . However, the EU SDGs have only one forest indicator for extent (“share of forest area”), and none for production or intensity. This gives heavy weight to agricultural sustainability to deliver overall rural land systems sustainability. The combination of variables we identified as supporting or hindering progress towards EU SDGs implies a range of both agricultural and broader social policies are necessary to achieve sustainable rural land systems. Implications for European SDGs and the Common Agricultural Policy Our results highlight brightspots and dragspots relative to European conditions and policy priorities. While global SDGs emphasise food provision in the first two goals, EU farm subsidies have been almost completely decoupled from production following overproduction in the 1970s and 80s (see brief history of CAP reform in ref. 6 ). Issues such as exceeding safe levels of nitrogen flows—for which the EU is a global risk 20 —predominate sustainability goals in Europe. We must note that even the “brightest” spot in Europe might not be a truly “sustainable” land system (e.g., may still exceed planetary boundaries). Our analysis shows which areas and what systems are better or worse, rather than saying any are sufficient. Our previous work has shown that farm income support under the CAP goes against the goals of securing living wages for farmers and social and environmental sustainability by over-subsidizing higher-income areas with poor environmental performance (see Fig. 2 in ref. 6 ). Making farm income support needs-based would address the issue of low-income areas, particularly Eastern Europe, receiving insufficient income support (see Fig. 3 in ref. 6 ), and would boost the Balkan dragspot. Since animal agriculture can be a major part of farmer income 22 , one possibility is for redirecting the CAP to incentivise reducing the size of the largest livestock herds. Using outcomes to align with policy goals From a systems perspective 23 , many previous analyses have assessed drivers (such as soil and climate) and management variables (such as fertilizer use and livestock density) to classify European rural land systems (e.g., refs. 10,21 ). In this analysis, we instead focused on sustainability outcomes relevant for rural land systems, using spatial hotspot analysis based on 15 environmental and 9 social indicators prioritized and monitored under EU policy at the intersection of both the EU SDGs and the CAP. Analyses using indicators directly derived from current policy goals can quantitatively assess progress towards sustainability and hold policymakers accountable for delivering it, although this approach is only as good as the data available for sustainability indicators. Our analysis of hotspots where sustainability is most supported and most hindered helps identify where to prioritize future research to link social and environmental drivers, management choices, and outcomes to develop a mechanistic understanding of how to transform the system towards sustainability. Methods Indicator selection We quantified brightspots and dragspots of European sustainable rural systems using a suite of environmental and social indicators relevant for sustainability, namely variables previously identified as aligned with the EU SDG indicators 5 for which data were available. In total we included 24 indicators in our spatial hotspot analysis: 15 environmental and 9 social (Table 1 ). This set was slightly reduced from the 29 variables aligning with EU SDG indicators 5 because we removed those that were highly correlated within a particular SDG, as well as those for which no data were available. We retained indicators that were correlated between SDGs to maximize the number of SDGs covered by at least one indicator in our analyses. For example, the risk of poverty or social exclusion (SDG 1) is highly correlated with rural purchasing power (SDG 10) but we retain the two indicators because of their relevance for different facets of sustainability (i.e., ending poverty and reducing inequalities). Thirteen pairwise comparisons of indicators were highly correlated (absolute value of Spearman’s rho > 0.7) within a particular SDG, for which we selected the most relevant indicator for the respective SDG (Supplementary Table 1). Table 1. Description of the 15 environmental and 9 social indicators used in our spatial hotspot analysis. Indicators are drawn from the overlap of the European SDGs and the Common Agricultural Policy (see ref. 5 ). Normative interpretations used for range-standardisation of each indicator are shown with a green upward arrow (indicating a higher value is desirable to meet the SDGs, i.e., “good”); orange downward arrow indicates a lower value is desirable to meet the SDGs. The number (and percentage) of NUTS2 regions analysed that had data is given for each indicator. Indicators where NUTS0 (country-level) data were used are indicated with an asterisk. Spatial data and coverage The spatial data structure we used was based on the two-digit regions of the Nomenclature of Territorial Units for Statistics (NUTS2; Eurostat, n.d.), of which there are approximately 260 throughout the (pre-Brexit) EU-28, plus additional regions in associated states (e.g., Iceland, Norway). We used the 2013 version of NUTS2 to align as closely as possible with the years of data used (2010–2017). Data were assembled from Eurostat, the Common Agricultural Policy (CAP) indicator tables, the European Environment Agency, the Emissions Database for Global Atmospheric Research (EDGAR), the 2012 CORINE land cover dataset, and published academic studies (Table 1 ). In order to fill annual gaps in Eurostat data and account for interannual variability, we took the average of the most recent years (maximum of five years) from the period 2010–2017. Most other data sources (e.g., the raster data from CORINE) were complete, so no averaging was performed. For indicators with no NUTS2 data available (e.g., farmland birds index), the finest-resolution values available were taken, either NUTS0 (country-level) or NUTS1. Additional processing was performed in the calculation of: 1) Natura 2000 area using the Natura 2000 end 2017 shapefile obtained from the European Environment Agency, which was rasterised to align with the CORINE data; and 2) High Nature Value (HNV) farmland processing described in Scown et al. (2020). All original data sources are publicly available and our final data are provided in the Supporting Information. [Table 1 ] We removed Turkey (TR), Liechtenstein (LI), North Macedonia (MK), and Montenegro (ME) from our analysis because they had no data for seven or more of our 24 indicators. The completeness of data for each indicator in our final set of 283 NUTS2 regions ranged from 100–54.4%, but 21 of the 24 indicators had data for more than 90% of NUTS2 regions (Table 1 ). Several EU overseas territories and islands for which some data were reported were also excluded (e.g., New Caledonia, Faroe Islands, Jersey and Guernsey). Index calculation We used a simple approach to dimension reduction of indicators based on normalization and index calculation before performing our hotspot analyses, in order to identify significant hotspots based on “good” performance across the suite of indicators. First, we normalized each indicator, with zero being considered the poorest performing NUTS2 for that indicator and one being the best performing region. This implies a normative interpretation of each indicator, which is specified in Table 1 ; for example, lower emissions of greenhouse gases, and higher production of renewable energy, are considered desirable. From the normalized indicators, we calculated separate environmental and social indices using the reduced set of environmental and social indicators, respectively. Each index was calculated as the unweighted average of all indicators. An overall sustainability index was then calculated as the unweighted average of the environmental (n = 15 indicators) and social (n = 9 indicators) indices, rather than using all 24 indicators combined. This was done to give equal weight to environmental and social sustainability, despite the differences in their constituent indicators due to data availability and the EU agricultural policy monitoring framework. However, this approach still implies that variance in each individual social indicator will have a greater effect on the overall index than variance in each environmental one. Missing data were ignored in the calculation of the indices, which means not all NUTS2 regions were evaluated based on all 24 indicators (see completeness in Table 1 ). In particular, rural Purchasing Power Standard is missing for predominantly urban or intermediate NUTS2 regions, therefore this indicator only plays a role in differentiating predominantly rural NUTS2 from each other and not from urban or intermediate regions. Brightspot and dragspot analyses We took a hybrid approach, combining spatial hotspot analyses with statistical analyses of the distribution of index scores. For the spatial analyses, we used the Getis-Ord Gi* statistic (described below), and for the statistical analyses we looked at whether adjacent NUTS2 regions were beyond one standard deviation from the index mean. Hotspots of “good” performance were our “brightspots”, and hotspots of “poor” performance were our “dragspots”. Getis-Ord Gi* hotpots We used the Getis-Ord Gi* statistic 24,25 implemented in GeoDa to quantify spatial hotspots for the environmental, social, and overall sustainability indices. The Getis-Ord Gi* analysis provides a z-score and p-value for each spatial unit (in our case NUTS2 regions) based on permutations. The z-score is calculated based on the local sum of a NUTS2 and its weighted neighbours, then compared to the expected local sum that would occur given a random distribution of all NUTS2 regions. The p-value is then calculated for the z-score as its probability at either end of the distribution, and is adjusted for multiple comparisons and spatial dependence. The neighbour matrix was created with the R package ‘spdep’ using Delaunay triangulation (spdep:: tri2nb) of NUTS2 centroids and thinning the graph using the Sphere of Influence function (spdep::soi.graph). This method was chosen over contiguity approaches because of the highly irregular distribution of NUTS2 regions, including isolated regions (e.g., Cyprus, Malta). Iceland’s neighbours were manually adjusted (Extended Data Fig. 3 ) and all others were deemed acceptable based on visual inspection. We used row-standardised neighbour weights, meaning that the weights of all neighbours of a particular NUTS2 are equal and sum to one, as opposed to binary weights where all neighbours have a weight of one and all non-neighbours a weight of zero. This is so the total weight of all neighbours for any NUTS2 equals one, regardless of the number of neighbours, which is preferred for highly irregular spatial units such as NUTS2. We employed the Gi* instead of the Gi statistic to include the value of the region along with its neighbour values in the calculation, which the latter does not do. We found only negligible differences in our results using Gi* compared to Gi and row-standardised compared to binary weights. We applied 99,999 permutations to the statistic calculation in GeoDa and controlled for the false discovery rate 26 . We did not rely solely on the Getis-Ord Gi* analyses because the results are dependent upon the underlying neighbourhood matrix used, and spurious results can emerge from highly irregular spatial data structures such as NUTS2 (please see details below and in Extended Data Figs. 3 and 4 ). In order for a NUTS2 region to be considered a brightspot (or dragspot) in our final analysis, it was required to meet one of the following two criteria: 1) Be a significant Getis-Ord Gi* bright(dragspot)spot AND have an index score above (below) the European median for that index AND be adjacent to at least one other bright(drag)spot NUTS2 region; OR 2) Be one of two or more adjacent NUTS2 regions with index scores more than one standard deviation above (or below) the European mean for that index. In other words, we excluded NUTS2 regions with below(above)-median index scores that the Getis-Ord Gi* statistic alone nonetheless identified as significant Gi* bright(drag)spots because their neighbours were significantly above average. Similarly, when multiple adjacent NUTS2 regions had index scores more than one standard deviation above (below) the mean but were not identified as significant Getis-Ord Gi* bright(drag)spots because of their neighbours, we manually corrected them to qualify as hotspots. This approach, however, still excludes isolated single-NUTS2 regions that might be considered a “lighthouse” in themselves. Several examples of regions with high or low index values that did not show up as significant Getis-Ord Gi* hotspots are worth mentioning here. Much of Sweden and Finland have very high values for the environmental and overall indices (Extended Data Fig. 4 a and e), yet not all of their NUTS2 regions show up as significant hotpots, likely due to having few neighbours overall and with lower scores (e.g., Norway; Extended Data Fig. 3 ). The opposite was observed for northern Norway and Cyprus, which have substantially lower environmental index scores than their neighbours (Extended Data Fig. 4 a), yet show up as significant hotspots (Extended Data Fig. 4 b)—again, likely due to the effects of few neighbours but this time with higher scores. Similar neighbourhood effects appear to happen around Berlin and Bucharest, whose regions show up as a social brightspot and dragspot, respectively (Extended Data Fig. 4 d), despite having lower and higher index scores, respectively, than their neighbouring regions (Extended Data Fig. 4 c). Thus, these results from the spatial hotspot analyses must be interpreted with caution, and this is why we took a hybrid approach combining the Getis-Ord Gi* and the (non-spatial) standard deviation of the index distributions for detailed analysis of hotspots. Analyses of calories, livestock, and subsidies In order to analyse the final overall index hotspots according to other factors related to sustainable rural systems, we compared our results to additional geospatial datasets. First, we analysed the fraction of calories delivered to the food system from each brightspot and dragspot based on the analysis of Cassidy et al. 27 . We used the data product DeliveredkcalFraction.tif from earthstat.org to calculate the average fraction of calories delivered within each NUTS2 region. Finally, we analysed the distribution of subsidies under different schemes of the CAP, within each brightspot and dragspot. Total payments in Euros (adjusted to purchasing power parity) for either 2014 (Denmark), 2016 (Bulgaria, Sweden, and the Czech Republic), or 2015 (all other countries), were grouped according to their purpose as income support payments, environmental payments, or other/unspecified payments (see Table S2 in ref. 6 ). Because the vast majority of CAP payments are based on agricultural land area, we calculated the average payment in Euros per hectare for each category of payment (income support, environmental, other/unspecified) in each hotspot, as well as for all other regions. Declarations Acknowledgements The authors wish to thank Loukas Christodoulou for helpful comments and Britta Ricker for mapping guidance and advice on earlier versions of the manuscript. We are grateful for the work of the Open Knowledge Foundation to initially compile the CAP data on farmsubsidies.org. This research was conceived under Swedish Research Council Grant 2014-5899/E0589901, but conducted on researchers’ free time after the grant ended. Data availability The full dataset and code used here is provided in the Supplementary Data. The shapefile contains the 283 NUTS2 regions analysed, the 24 variables used, the environment and social indices as well as the overall index calculated, the Gi* z-scores and p-values for each index, and the final hotspot regions. Field names and descriptions are found in Supplementary Table 2. Additional processing details (e.g., raster resolutions, projections) and code for deriving the data provided here from their raw original source can be obtained at https://github.com/murrayscown/EU-Agricultural-Systems-Database. All raw data are freely available from their original source, no additional requests for data were required in our processing and analysis. Author Contributions K.A.N. and M.S. conceived the research and wrote the manuscript; M.S. performed the analysis and created the visualizations; K.A.N. obtained funding to support the research. References Watson, R. et al. Summary for policymakers of the global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. IPBES Secretariat: Bonn, Germany 22–47 (2019). Springmann, M. et al. Options for keeping the food system within environmental limits. Nature 562 , 519–525 (2018). FAO. Transforming food and agriculture to achieve the SDGs: 20 interconnected actions to guide decision-makers . (Food and Agriculture Organization of the United Nations, 2018). EC. Communication from the Commission to the European Parliament, the European Council, the Council, the European Economic and Social Committee and the Committee of the Regions: The European Green Deal. COM(2019) 640 final. (European Commission, 2019). Scown, M. W. & Nicholas, K. A. European agricultural policy requires a stronger performance framework to achieve the Sustainable Development Goals. Global Sustainability 3 , 1–11 (2020). Scown, M. W., Brady, M. V. & Nicholas, K. A. Billions in Misspent EU Agricultural Subsidies Could Support the Sustainable Development Goals. One Earth 3 , 237–250 (2020). Stoddard, I. et al. Three decades of climate mitigation: why haven’t we bent the global emissions curve? Annu Rev Environ Resour 46 , 653–689 (2021). Seto, K. C. et al. Carbon lock-in: types, causes, and policy implications. Annu Rev Environ Resour 41 , 425–452 (2016). Meadows, D. H. Thinking in systems: A primer . (chelsea green publishing, 2008). Kuemmerle, T. et al. Hotspots of land use change in Europe. Environmental research letters 11 , 064020 (2016). Oberschelp, C., Pfister, S., Raptis, C. E. & Hellweg, S. Global emission hotspots of coal power generation. Nat Sustain 2 , 113–121 (2019). Levers, C. et al. Archetypical patterns and trajectories of land systems in Europe. Reg Environ Change 18 , 715–732 (2018). Cai, W., Gibbs, D., Zhang, L., Ferrier, G. & Cai, Y. Identifying hotspots and management of critical ecosystem services in rapidly urbanizing Yangtze River Delta Region, China. J Environ Manage 191 , 258–267 (2017). Wang, X. & Varady, D. P. Using hot-spot analysis to study the clustering of section 8 housing voucher families. Hous Stud 20 , 29–48 (2005). Zhang, H. & Tripathi, N. K. Geospatial hot spot analysis of lung cancer patients correlated to fine particulate matter (PM2. 5) and industrial wind in Eastern Thailand. J Clean Prod 170 , 407–424 (2018). Valencia, V. et al. Learning from the future: mainstreaming disruptive solutions for the transition to sustainable food systems. Environmental Research Letters 17 , 051002 (2022). Bennett, E. M. et al. Bright spots: seeds of a good Anthropocene. Front Ecol Environ 14 , 441–448 (2016). Debonne, N. et al. The geography of megatrends affecting European agriculture. Global Environmental Change 75 , 102551 (2022). Dornelles, A. Z. et al. Transformation archetypes in global food systems. Sustain Sci 17 , 1827–1840 (2022). Gerten, D. et al. Feeding ten billion people is possible within four terrestrial planetary boundaries. Nat Sustain 3 , 200–208 (2020). Levers, C. et al. Drivers of forest harvesting intensity patterns in Europe. For Ecol Manage 315 , 160–172 (2014). Röös, E. et al. Moving beyond organic–A food system approach to assessing sustainable and resilient farming. Glob Food Sec 28 , 100487 (2021). Scown, M. W., Winkler, K. J. & Nicholas, K. A. Aligning research with policy and practice for sustainable agricultural land systems in Europe. Proc Natl Acad Sci U S A 116 , 4911–4916 (2019). Anselin, L. Local Spatial Autocorrelation (2): Other Local Spatial Autocorrelation Statistics. https://geodacenter.github.io/workbook/6b_local_adv/lab6b.html#getis-ord-statistics (2020). Getis, A. & Ord, J. K. The Analysis of Spatial Association by Use of Distance Statistics. Geogr Anal 24 , 189–206 (1992). Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal statistical society: series B (Methodological) 57 , 289–300 (1995). Cassidy, E. S., West, P. C., Gerber, J. S. & Foley, J. A. Redefining agricultural yields: from tonnes to people nourished per hectare. Environmental Research Letters 8 , 034015 (2013). Nicholas, K. A., Villemoes, F., Lehsten, E., Brady, M. V. & Scown, M. W. A harmonized and spatially-explicit dataset for the European Union’s €61 billion in Common Agricultural Policy payments to farmers for 2015. Preprint at https://doi.org/10.23644/uu.12706580.v1 (2020). Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryData.zip Dataset 1 ExtendedDataFigures.docx SupplementaryTables.docx Cite Share Download PDF Status: Published Journal Publication published 11 Nov, 2024 Read the published version in Communications Earth & Environment → 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-2941468","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Analysis","associatedPublications":[],"authors":[{"id":207780375,"identity":"e665ea6a-89b1-410d-81c7-962a063453fb","order_by":0,"name":"Murray Scown","email":"","orcid":"https://orcid.org/0000-0003-0663-7937","institution":"Utrecht University Faculty of Geosciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Murray","middleName":"","lastName":"Scown","suffix":""},{"id":207780376,"identity":"71972219-a14c-4b45-a509-3b18705c6111","order_by":1,"name":"Kimberly Nicholas","email":"data:image/png;base64,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","orcid":"","institution":"Lund University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Kimberly","middleName":"","lastName":"Nicholas","suffix":""}],"badges":[],"createdAt":"2023-05-16 08:45:56","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2941468/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2941468/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s43247-024-01736-6","type":"published","date":"2024-11-11T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":43441294,"identity":"ca44b21d-871e-47bc-8410-1330eb7a0c7c","added_by":"auto","created_at":"2023-09-20 23:18:16","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1028917,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCharacteristics of environmental brightspots (top two rows, green) and dragspots (bottom two rows, purple) of rural land systems in Europe.\u003c/strong\u003e Regions are specified based on results of Getis-Ord Gi* hotspot analysis of the environmental index comprised of the 15 indicators shown, and comparison against contiguous regions falling outside of one standard deviation either side of the mean index score. Indicators are range-standardised based on the full set of 283 NUTS2 regions analysed, and are scaled according to our normative interpretation so that one is always “best” and zero is “worst”. NUTS2 regions within designated brightspots or dragspots are coloured and country borders are shown in grey. Note that farmland birds data are missing for the Balkans brightspot. Please see Methods for full details.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/abf35425243d84ce423e6a01.png"},{"id":43444780,"identity":"6f75e4ab-94c2-4b90-8849-083083605c27","added_by":"auto","created_at":"2023-09-20 23:34:16","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":520586,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCharacteristics of social brightspots (top row, green) and dragspots (bottom three rows, purple) of rural land systems in Europe.\u003c/strong\u003e Regions are specified based on results of Getis-Ord Gi* hotspot analysis of the social index comprised of the 9 indicators shown, and comparison against contiguous regions falling outside of one standard deviation either side of the mean index score. Indicators are range-standardised based on the full set of 283 NUTS2 regions analysed, and are scaled according to our normative interpretation so that one is “best” and zero is “worst”. NUTS2 regions within designated brightspots or dragspots are coloured and country borders are shown in grey. Note that Cyprus is included under the “Balkans” coldspot and Malta is included under the “Southern Italy” coldspot. Please see Methods for full details.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/8e07eae4411544620fd81e6c.png"},{"id":43443017,"identity":"72bb38ee-30b7-4215-97ed-3c80ea864f5c","added_by":"auto","created_at":"2023-09-20 23:26:16","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":784190,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOverall brightspots (green) and dragspots (purple) of sustainable rural systems in Europe. \u003c/strong\u003eDark lines are national borders for those countries included, light lines within bright/dragspots are NUTS2 regional borders. Scores of the environmental and social indices within each bright/dragspot are shown as boxplots, along with the mean environmental and social score for all 283 NUTS2 analysed (grey diamond). Note that parts of Hungary are included under the “Balkans” dragspot and Malta is included under the “Southern Italy” dragspot. Please see Methods for full details.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/5b7e9f1117f3db2f1b015759.png"},{"id":43441290,"identity":"b5b9e2d9-085b-402a-8cfa-b2c5cee7d250","added_by":"auto","created_at":"2023-09-20 23:18:15","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":28182,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFraction of calories delivered to food system by overall sustainable rural brightspots (green) and dragspots (purple).\u003c/strong\u003e Boxplots are of NUTS2 regions within each brightspot/dragspot, calculated as the average fraction of calories delivered to the food system from agricultural land within each NUTS2 using the data of ref.. Please see Methods for full details.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/0f6c98d068bc28c33a533ffb.png"},{"id":43441291,"identity":"b2f02ce9-d1dd-4323-9a93-0ec98e4fea7c","added_by":"auto","created_at":"2023-09-20 23:18:15","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":256855,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSubsidies paid to overall rural sustainability brightspots and dragspots. \u003c/strong\u003e(A) Average subsidies in Euros per hectare (adjusted among countries to purchasing power parity) paid in 2015* to overall brightspots and dragspots; (B) those same subsidies shown as the proportional distribution among payment objectives (income, environmental, other/unspecified). Also shown in both panels between brightspots and dragspots is the subsidies (per hectare and fractional distribution) for all other regions that did not fall within a spatial hotspot. *Note: subsidies data are from ref., which contain the year 2015 for all countries except Denmark (2014) and Bulgaria, Sweden, and Czechia (2016).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/9dd335d26070b370553d9b39.png"},{"id":68806385,"identity":"1c0df46a-e750-4813-9836-d9315a7d9761","added_by":"auto","created_at":"2024-11-12 08:09:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3592563,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/9027b85d-9779-44f5-a148-3444895a9aa8.pdf"},{"id":43441289,"identity":"88d46325-d53a-44a9-ae5c-2eaf409abb69","added_by":"auto","created_at":"2023-09-20 23:18:15","extension":"zip","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2314053,"visible":true,"origin":"","legend":"\u003cp\u003eDataset 1\u003c/p\u003e","description":"","filename":"SupplementaryData.zip","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/ffe7199c75e513700bc3a06f.zip"},{"id":43443019,"identity":"e5203e03-e351-462e-aa8d-b56adc99766c","added_by":"auto","created_at":"2023-09-20 23:26:16","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":817967,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"ExtendedDataFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/52b65fd366089af6927885d8.docx"},{"id":43441296,"identity":"c2a94e8e-fd55-4f6f-913a-9b9abd043bd3","added_by":"auto","created_at":"2023-09-20 23:18:16","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":23235,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTables.docx","url":"https://assets-eu.researchsquare.com/files/rs-2941468/v1/3d3b662ad0b06cce6c6ff00f.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"European rural regions supporting and hindering the Sustainable Development Goals","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLand use and exploitation of animals and plants for consumption are the main drivers of global declines in nature\u003csup\u003e1\u003c/sup\u003e. To achieve the United Nations Sustainable Development Goals (SDGs) and stay within planetary boundaries, transforming rural land production systems while eliminating unsustainable consumption is essential\u003csup\u003e2,3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe European Union (EU) aims to foster fair and healthy food systems and halt biodiversity loss as Europe becomes the first climate-neutral continent\u003csup\u003e4\u003c/sup\u003e. Around one-quarter of the 100 European Union SDG indicators (EU SDGs) are related to rural land systems\u003csup\u003e5\u003c/sup\u003e, but recent analyses show EU implementation of both SDGs and the Common Agricultural Policy (CAP) often counter established sustainability goals\u003csup\u003e5,6\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMaking sustainability transformations requires both overcoming unsustainable instances of business as usual\u003csup\u003e7,8\u003c/sup\u003e, and identifying pathways for replacing the status quo with sustainable systems.\u0026nbsp;Hotspot analysis builds on the principle of targeting limited resources at a particular place or process to maximize intervention effectiveness, like Meadows\u0026rsquo; leverage points\u003csup\u003e9\u003c/sup\u003e. Hotspot analysis is often based on, for example, identifying outliers in the statistical distribution of a phenomenon\u0026nbsp;(e.g., the top/bottom 10%), such as hotspots of change in land use extent and management intensity\u003csup\u003e10\u003c/sup\u003e; or identifying sectors/groups/products with a disproportionately high contribution to some phenomenon, such as greenhouse gas emissions\u003csup\u003e11\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eAdvancing such approaches, spatial hotspot analysis incorporates both the statistical distribution and spatial patterns of a phenomenon, thus accounting for regional environmental conditions (e.g., climate, terrain) or national differences in governance and practices. While granular land use classifications (e.g., ref.\u003csup\u003e12\u003c/sup\u003e) can be difficult to interpret and communicate at governance-relevant scales, spatial hotspot analysis enables identifying broader patterns which can be targeted by regional policies. Spatial hotspot analysis has been successfully applied in fields including\u0026nbsp;ecosystem services\u003csup\u003e13\u003c/sup\u003e, welfare economics\u003csup\u003e14\u003c/sup\u003e, and human health\u003csup\u003e15\u003c/sup\u003e.\u0026nbsp;The linked concept of \u0026ldquo;bright spots,\u0026rdquo; such as identifying \u0026ldquo;Lighthouse farms\u0026rdquo;\u003csup\u003e16\u003c/sup\u003e, refers to hotspots with the potential to point to pathways to transformation\u003csup\u003e17\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHere we use spatial hotspot analysis to identify spatial \u0026ldquo;brightspots\u0026rdquo; of sustainable rural land systems in Europe that can inspire us towards better pathways aligned with meeting SDGs, and spatial \u0026ldquo;dragspots\u0026rdquo; of unsustainable systems that hold us back. We developed indicators for environmentally and socially sustainable rural land systems in 283 NUTS2 regions. Specifically, we selected 15 environmental and 9 social indicators drawn from the overlap of the EU SDGs and the CAP. We then analysed the fraction of calories delivered to the food system from within each hotspot, and CAP subsidies paid to these regions. This enabled us to identify both the location and characteristics of sustainable rural land systems in Europe, their role in the food system, and their relationship to EU policy instruments. These results can inform the redirection of incentives away from unsustainable systems and towards successful practices.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eSummary of indices by country\u003c/h2\u003e \u003cp\u003eWe calculated three sustainability indices for each NUTS2 region: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) an environmental index based on 15 indicators, (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) a social index based on 9 indicators, and (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) an overall sustainability index, which combined these 24 indicators in an unweighted average (please see Methods for details). Indicators were range-standardised to between zero and one before calculating indices. All three indices were approximately normally distributed across the 283 NUTS2 regions (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The environmental index had a substantially higher mean (0.63) than the social index (0.46), but values within the two spanned a similarly wide range (0.46 and 0.45, respectively; Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Due to its calculation as the average of the two, the overall sustainability index fell in between (mean 0.55, range 0.31) (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe found that countries with high environmental scores tended to have low social ones, and vice versa. The four countries with the highest mean environmental index score (Croatia, Latvia, Slovenia, and Estonia) all had a mean social index in the lower half of countries (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Similarly, the three highest countries for the mean social score (Germany, Luxembourg, and Iceland) were at the mid to lower end of environmental scores. Sweden had the highest mean score for the overall sustainability index (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), a reflection of being in the top six countries for both mean environmental and social scores. Malta had by far the lowest overall index score, due to having the lowest mean environmental score and the seventh-lowest mean social score (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Countries whose NUTS2 scores ranged widely included Belgium for environmental, Switzerland and the United Kingdom for social, and Italy for both indices.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBrightspots and dragspots\u003c/h3\u003e\n\u003cp\u003eOur analysis of 24 policy-aligned indicators of environmentally and socially sustainable rural land systems in Europe reveals: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) two main brightspots and two main dragspots for the environmental subindex (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e); (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) one main brightspot and three dragspots for the social subindex (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e); and (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) two overall brightspots most in line with the EU SDGs and five overall dragspots most hindering achieving them (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In general, brightspots had at least two indicators with outstanding performance, along with solid performance across most indicators within the relevant index (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eGood environmental performance was associated with less intensive methods (e.g., relatively low inputs of pesticides and irrigation, and low fertilizer application as indicated by P and N balance). Environmentally unsustainable food systems relied on intensive inputs in monocultural landscapes of little nature value, and particularly with high nutrient imbalances, most clearly exemplified in the Lowlands. Good social performance featured low energy use when accompanied by strong economic performance including high farm employment and income, supported by higher agricultural training, alongside low risk of poverty.\u003c/p\u003e\n\u003ch3\u003eEnvironmental sustainability index\u003c/h3\u003e\n\u003cp\u003eThe main environmental brightspot areas were around in the Northern Baltic region (Sweden, Finland, Estonia, and Latvia), and in much of the Balkans region with parts of Slovakia (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The two areas were both characterised by very high (\u0026ldquo;good\u0026rdquo;) values for eight of the 15 environmental indicators: high groundwater quality, few artificial surfaces, low soil erosion, low gross N and P balances, low reliance on irrigation, low use of pesticides, and low emissions of greenhouse gases and ammonia. The Balkans environmental brightspot had generally lower forest area, less organic farming, and higher emissions of particulate matter from agriculture than the Northern Baltic environmental brightspot. However, the Balkans performed better in terms of Natura 2000 area and high nature value farmland.\u003c/p\u003e \u003cp\u003eThe main environmental dragspot areas were in The Netherlands, Belgium, and Denmark, along with western Germany and northern France (referred to as the \u0026ldquo;Northwest Continental\u0026rdquo; dragspot), and in Northern Italy (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Both environmental dragspot areas had very little organic farming, low forest area, low farmland birds, and little Natura 2000 area or high nature value farmland. Where they differed was that Northern Italy had much higher rates of soil erosion and greater dependence on irrigation, while the Northwest Continental dragspot performed very poorly in terms of gross N balance.\u003c/p\u003e \u003cp\u003e[Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e]\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eSocial sustainability index\u003c/h2\u003e \u003cp\u003eThe main social brightspot was in Central Europe, including all of Germany, northern Austria, and western Czech Republic (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). This brightspot performed very well in terms of low energy use in agriculture and low unemployment, as well as a low risk of poverty or social exclusion. This area also had much higher levels of agricultural training, more renewable energy use and production in agriculture, and higher farm incomes, compared to the dragspots, but still variable and in the mid-range when compared to all of Europe.\u003c/p\u003e \u003cp\u003eThe main social dragspots were in the Balkans and Cyprus, the Iberian Peninsula, and Southern Italy and Malta (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These areas all had extremely low levels of agricultural training, renewable energy use and production in agriculture, farm income, and GDP per capita. Rural purchasing power was also low in all of these dragspots, although not to the same degree as the previous indicators. Risk of poverty or social exclusion was generally lower in the Iberian dragspot than the other two, while the Balkans performed slightly better than Iberia and Southern Italy in terms of unemployment. Interestingly, all social dragspots also had relatively low rates of energy consumption in agriculture.\u003c/p\u003e \u003cp\u003e[Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e]\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eOverall sustainability combining social and environmental\u003c/h2\u003e \u003cp\u003eThe two brightspot areas for overall sustainability were the Nordics (Sweden and Finland) and Central Europe (including most of Germany, all of Austria, Luxembourg, parts of Switzerland, and small western portions of Slovakia and Czechia) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The five dragspot areas for overall sustainability covered parts of: the Balkans (northern Romania and southeast Bulgaria, also including southeast Hungary), the Lowlands (the Netherlands and Belgium), Northern Italy, Southern Italy and Malta, and Southern Spain (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe Nordic brightspot was characterised by very high environmental index scores, with substantially lower scores for the social index, although still near the European average. By contrast, the Central Europe brightspot area had average environmental but very high social index scores. The Balkans, Southern Spain, and Southern Italy dragspots were all characterised by very low social index scores, but with good environmental conditions (with the exception of Malta), indicating that the poor social conditions dragged down the overall index in these regions. On the other hand, Northern Italy and the Lowlands performed poorly for the environment, and around average for social factors.\u003c/p\u003e \u003cp\u003eIn all but one case, to qualify as a bright (drag)spot, either the environmental or social index was extremely high (low) relative to the European average, while the other index was close to the average (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The Nordics were the exception, where environmental and social performance were both well above average. In some cases, good environmental performance despite poor social performance kept a region out of being an overall dragspot (e.g., Greece), and reasonable social performance kept environmental dragspot from being overall ones (e.g., Denmark).\u003c/p\u003e \u003cp\u003e[Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e]\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCalories delivered to the food system\u003c/h3\u003e\n\u003cp\u003eWe analysed the fraction of calories delivered to the food system produced in each overall brightspot and dragspot (please see Methods for details). Three of the five dragspots (Southern Italy, Lowlands, and Southern Spain) delivered, on average, a higher fraction of calories to the food system than did the two brightspots (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The Nordic brightspot delivered the lowest fraction of calories to the food system (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), indicating that environmental and social successes here may be contributing to other goals besides maximizing food production. Of all brightspots and dragspots, the Central Europe brightspot (which also contained the largest number of NUTS2 regions) had the widest range of calorie fractions delivered to the food system (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), with the K\u0026ouml;ln region (NUTS2 ID \u0026ldquo;DEA2\u0026rdquo;) being a particularly high outlier within this brightspot. The Southern Spain and Lowlands dragspots also had particularly wide ranges of calories delivered, compared to the other dragspots and the Nordic brightspot (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e[Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e]\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eSubsidies paid to brightspots and dragspots\u003c/h2\u003e \u003cp\u003eTo determine whether and how EU subsidies align with sustainability, we analysed the average CAP payment per hectare in each brightspot and dragspot\u0026mdash;as well as in all other regions\u0026mdash;according to their intended support scheme (please see Methods for details). We found that the dragspots of Northern Italy, Lowlands, and Southern Italy receive the highest level of support per hectare (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Southern Italy and the Lowlands also deliver the highest fractions of calories to the food system, on average (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), indicating that large amounts of subsidies (intended to support non-market environmental and social values) are being paid to dragspots producing marketable food items but under-performing in terms of environmental and social outcomes. By contrast, the Northern Italy dragspot receives the highest average payments per hectare yet delivers a low fraction of calories to the food system. Future research should examine what is being produced in this region and in which ways to result in neither high-performing environmental no social outcomes, nor food, despite large amounts of subsidies.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhile all five dragspots had moderate to high subsidies per hectare, the two brightspots received very different amounts (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). The Nordic brightspot, in particular, received less than \u0026euro;150 per hectare (in purchasing power parity), compared to \u0026euro;450 in the Central Europe brightspot, \u0026euro;350 in the lowest recipient dragspot (the Balkans), and just below \u0026euro;400 in all other regions. The Nordic brightspot also delivered the lowest fraction of calories to the food system (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), indicating that this region is substantially different from other brightspots and dragspots in terms of the amounts of subsidies received and the amount of food produced.\u003c/p\u003e \u003cp\u003eIn terms of the distribution of subsidies between income support and environmental schemes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB), we found the largest fractions of environmental support being paid to the two overall brightspots and to the Balkans dragspot, which is a reasonable use of such subsidies considering good environmental performance in these regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The largest fraction of income support was paid to the Lowlands dragspot, where its effectiveness is questionable considering the relatively poor environmental performance and average social conditions in this region (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The Southern Spain dragspot received the lowest fraction of environmental support despite having relatively high environmental performance (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e); conversely, the large fraction of income support (second only to the Lowlands) does not appear to be translating into social outcomes in this region (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Overall, the areas with the best environmental performance received proportionally more environmental subsidies, in line with a performance-based approach. However, already well-off and high-producing areas received high levels of income and general support, despite such support being unnecessary to achieve the CAP goals of securing farmer livelihoods, and already being compensated by the market for producing crops.\u003c/p\u003e \u003cp\u003e[Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e]\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe identify four unsustainable patterns in current European rural land systems: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) a widespread trade-off between social and environmental sustainability; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) the regions producing the most calories for food doing so at the cost of either environmental and/or social sustainability; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) over-reliance on agricultural systems to deliver EU sustainability goals outside of urban areas, without governing forests; and (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) EU CAP subsidies often directed ineffectively.\u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eIntensive food production, intensive harm\u003c/h2\u003e \u003cp\u003eIntensive food production in Europe causes disproportionate environmental harm, while less agriculturally intensive areas are associated with better sustainability outcomes\u003csup\u003e18\u003c/sup\u003e. While Debonne et al.\u003csup\u003e18\u003c/sup\u003e described a dichotomy between productivist and post-productivist \u003cem\u003eparadigms\u003c/em\u003e, this pattern of usage may be better explained by land use \u003cem\u003esuitability\u003c/em\u003e. In a global context, European agriculture is rather steady, showing a rate of change in recent decades of \u0026lt; 0.5% on most outcome metrics\u003csup\u003e19\u003c/sup\u003e. However, trends within the continent show a bifurcation, with intensification occurring in highly suitable farming areas, while cropland is shrinking and inputs de-intensifying in more marginal areas\u003csup\u003e10\u003c/sup\u003e. This split pattern indicates that maximizing intensive production is prioritized wherever conditions allow.\u003c/p\u003e \u003cp\u003eOverall, rural areas in Europe with the worst EU SDG performance feature intensive crop and animal agriculture. Most of Europe is at high risk of exceeding the planetary boundary for nitrogen, and increasing risk of exceeding the boundary for biosphere integrity\u003csup\u003e20\u003c/sup\u003e. Debonne et al.\u003csup\u003e18\u003c/sup\u003e saw that (productivist) megastables, with \u0026gt; 500 head of livestock, are concentrated in the Lowlands and Northern Italy, areas we also found to be dragspots. Our dragspots also overlap with Europe’s highest areas of nitrogen exceedance (the Lowlands) and excessive use of livestock antibiotics (Italy)\u003csup\u003e18\u003c/sup\u003e. Intensive farming methods in the Lowlands, characterized by increasing yields in croplands and permanent crops\u003csup\u003e12\u003c/sup\u003e, align with our dragspots, while areas with a share of organic agriculture five times above the European average, including central Sweden and parts of Switzerland\u003csup\u003e18\u003c/sup\u003e, align with our brightspots of sustainability. To keep global food production within planetary boundaries, a combination of dietary shifts to reduce animal consumption, prevention of food loss and waste, and adoption of efficient technology are necessary; no single measure is sufficient\u003csup\u003e2\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eForestry and sustainable rural land systems\u003c/h2\u003e \u003cp\u003eIntensification of agriculture is associated with lower performance on European SDGs, while intensification of forestry is not. Our two overall sustainability brightspots include areas with high harvested timber volumes and intensity\u003csup\u003e21\u003c/sup\u003e and with a strong increase in roundwood production since 1990\u003csup\u003e10\u003c/sup\u003e, with an intensification of wood production in southern Sweden and central Germany\u003csup\u003e12\u003c/sup\u003e. However, the EU SDGs have only one forest indicator for extent (“share of forest area”), and none for production or intensity. This gives heavy weight to agricultural sustainability to deliver overall rural land systems sustainability. The combination of variables we identified as supporting or hindering progress towards EU SDGs implies a range of both agricultural and broader social policies are necessary to achieve sustainable rural land systems.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eImplications for European SDGs and the Common Agricultural Policy\u003c/h2\u003e \u003cp\u003eOur results highlight brightspots and dragspots relative to European conditions and policy priorities. While global SDGs emphasise food provision in the first two goals, EU farm subsidies have been almost completely decoupled from production following overproduction in the 1970s and 80s (see brief history of CAP reform in ref.\u003csup\u003e6\u003c/sup\u003e). Issues such as exceeding safe levels of nitrogen flows—for which the EU is a global risk\u003csup\u003e20\u003c/sup\u003e—predominate sustainability goals in Europe. We must note that even the “brightest” spot in Europe might not be a truly “sustainable” land system (e.g., may still exceed planetary boundaries). Our analysis shows which areas and what systems are better or worse, rather than saying any are sufficient.\u003c/p\u003e \u003cp\u003eOur previous work has shown that farm income support under the CAP goes against the goals of securing living wages for farmers and social and environmental sustainability by over-subsidizing higher-income areas with poor environmental performance (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e in ref.\u003csup\u003e6\u003c/sup\u003e). Making farm income support needs-based would address the issue of low-income areas, particularly Eastern Europe, receiving insufficient income support (see Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e in ref.\u003csup\u003e6\u003c/sup\u003e), and would boost the Balkan dragspot. Since animal agriculture can be a major part of farmer income\u003csup\u003e22\u003c/sup\u003e, one possibility is for redirecting the CAP to incentivise reducing the size of the largest livestock herds.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eUsing outcomes to align with policy goals\u003c/h2\u003e \u003cp\u003eFrom a systems perspective\u003csup\u003e23\u003c/sup\u003e, many previous analyses have assessed drivers (such as soil and climate) and management variables (such as fertilizer use and livestock density) to classify European rural land systems (e.g., refs.\u003csup\u003e10,21\u003c/sup\u003e). In this analysis, we instead focused on sustainability \u003cem\u003eoutcomes\u003c/em\u003e relevant for rural land systems, using spatial hotspot analysis based on 15 environmental and 9 social indicators prioritized and monitored under EU policy at the intersection of both the EU SDGs and the CAP. Analyses using indicators directly derived from current policy goals can quantitatively assess progress towards sustainability and hold policymakers accountable for delivering it, although this approach is only as good as the data available for sustainability indicators. Our analysis of hotspots where sustainability is most supported and most hindered helps identify where to prioritize future research to link social and environmental drivers, management choices, and outcomes to develop a mechanistic understanding of how to transform the system towards sustainability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e "},{"header":"Methods","content":"\u003ch2\u003eIndicator selection\u003c/h2\u003e\n\u003cp\u003eWe quantified brightspots and dragspots of European sustainable rural systems using a suite of environmental and social indicators relevant for sustainability, namely variables previously identified as aligned with the EU SDG indicators\u003csup\u003e5\u003c/sup\u003e for which data were available. In total we included 24 indicators in our spatial hotspot analysis: 15 environmental and 9 social (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). This set was slightly reduced from the 29 variables aligning with EU SDG indicators\u003csup\u003e5\u003c/sup\u003e because we removed those that were highly correlated \u003cem\u003ewithin\u003c/em\u003e a particular SDG, as well as those for which no data were available. We retained indicators that were correlated \u003cem\u003ebetween\u003c/em\u003e SDGs to maximize the number of SDGs covered by at least one indicator in our analyses. For example, the risk of poverty or social exclusion (SDG 1) is highly correlated with rural purchasing power (SDG 10) but we retain the two indicators because of their relevance for different facets of sustainability (i.e., ending poverty and reducing inequalities). Thirteen pairwise comparisons of indicators were highly correlated (absolute value of Spearman\u0026rsquo;s rho\u0026thinsp;\u0026gt;\u0026thinsp;0.7) within a particular SDG, for which we selected the most relevant indicator for the respective SDG (Supplementary Table 1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1.\u003c/strong\u003e \u003cstrong\u003eDescription of the 15 environmental and 9 social indicators used in our spatial hotspot analysis.\u003c/strong\u003e Indicators are drawn from the overlap of the European SDGs and the Common Agricultural Policy (see ref.\u003csup\u003e5\u003c/sup\u003e). Normative interpretations used for range-standardisation of each indicator are shown with a green upward arrow (indicating a higher value is desirable to meet the SDGs, i.e., \u0026ldquo;good\u0026rdquo;); orange downward arrow indicates a lower value is desirable to meet the SDGs. The number (and percentage) of NUTS2 regions analysed that had data is given for each indicator. Indicators where NUTS0 (country-level) data were used are indicated with an asterisk.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"https://myfiles.space/user_files/95224_ce634422aaf2e7a6/95224_custom_files/img1695221962.png\"\u003e\u003c/p\u003e\n\u003ch2\u003e\u003cimg src=\"https://myfiles.space/user_files/95224_ce634422aaf2e7a6/95224_custom_files/img1695222015.png\"\u003e\u003cbr\u003e\u003c/h2\u003e\n\u003ch2\u003eSpatial data and coverage\u003c/h2\u003e\n\u003cp\u003eThe spatial data structure we used was based on the two-digit regions of the Nomenclature of Territorial Units for Statistics (NUTS2; Eurostat, n.d.), of which there are approximately 260 throughout the (pre-Brexit) EU-28, plus additional regions in associated states (e.g., Iceland, Norway). We used the 2013 version of NUTS2 to align as closely as possible with the years of data used (2010\u0026ndash;2017). Data were assembled from Eurostat, the Common Agricultural Policy (CAP) indicator tables, the European Environment Agency, the Emissions Database for Global Atmospheric Research (EDGAR), the 2012 CORINE land cover dataset, and published academic studies (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). In order to fill annual gaps in Eurostat data and account for interannual variability, we took the average of the most recent years (maximum of five years) from the period 2010\u0026ndash;2017. Most other data sources (e.g., the raster data from CORINE) were complete, so no averaging was performed. For indicators with no NUTS2 data available (e.g., farmland birds index), the finest-resolution values available were taken, either NUTS0 (country-level) or NUTS1. Additional processing was performed in the calculation of: 1) Natura 2000 area using the Natura 2000 end 2017 shapefile obtained from the European Environment Agency, which was rasterised to align with the CORINE data; and 2) High Nature Value (HNV) farmland processing described in Scown et al. (2020). All original data sources are publicly available and our final data are provided in the Supporting Information.\u003c/p\u003e\n\u003cp\u003e[Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e]\u003c/p\u003e\n\u003cp\u003eWe removed Turkey (TR), Liechtenstein (LI), North Macedonia (MK), and Montenegro (ME) from our analysis because they had no data for seven or more of our 24 indicators. The completeness of data for each indicator in our final set of 283 NUTS2 regions ranged from 100\u0026ndash;54.4%, but 21 of the 24 indicators had data for more than 90% of NUTS2 regions (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Several EU overseas territories and islands for which some data were reported were also excluded (e.g., New Caledonia, Faroe Islands, Jersey and Guernsey).\u003c/p\u003e\n\u003ch2\u003eIndex calculation\u003c/h2\u003e\n\u003cp\u003eWe used a simple approach to dimension reduction of indicators based on normalization and index calculation before performing our hotspot analyses, in order to identify significant hotspots based on \u0026ldquo;good\u0026rdquo; performance across the suite of indicators. First, we normalized each indicator, with zero being considered the poorest performing NUTS2 for that indicator and one being the best performing region. This implies a normative interpretation of each indicator, which is specified in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e; for example, lower emissions of greenhouse gases, and higher production of renewable energy, are considered desirable. From the normalized indicators, we calculated separate environmental and social indices using the reduced set of environmental and social indicators, respectively. Each index was calculated as the unweighted average of all indicators. An overall sustainability index was then calculated as the unweighted average of the environmental (n\u0026thinsp;=\u0026thinsp;15 indicators) and social (n\u0026thinsp;=\u0026thinsp;9 indicators) indices, rather than using all 24 indicators combined. This was done to give equal weight to environmental and social sustainability, despite the differences in their constituent indicators due to data availability and the EU agricultural policy monitoring framework. However, this approach still implies that variance in each individual social indicator will have a greater effect on the overall index than variance in each environmental one. Missing data were ignored in the calculation of the indices, which means not all NUTS2 regions were evaluated based on all 24 indicators (see completeness in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). In particular, rural Purchasing Power Standard is missing for predominantly urban or intermediate NUTS2 regions, therefore this indicator only plays a role in differentiating predominantly rural NUTS2 from each other and not from urban or intermediate regions.\u003c/p\u003e\n\u003ch2\u003eBrightspot and dragspot analyses\u003c/h2\u003e\n\u003cp\u003eWe took a hybrid approach, combining spatial hotspot analyses with statistical analyses of the distribution of index scores. For the spatial analyses, we used the Getis-Ord Gi* statistic (described below), and for the statistical analyses we looked at whether adjacent NUTS2 regions were beyond one standard deviation from the index mean. Hotspots of \u0026ldquo;good\u0026rdquo; performance were our \u0026ldquo;brightspots\u0026rdquo;, and hotspots of \u0026ldquo;poor\u0026rdquo; performance were our \u0026ldquo;dragspots\u0026rdquo;.\u003c/p\u003e\n\u003ch2\u003eGetis-Ord Gi* hotpots\u003c/h2\u003e\n\u003cp\u003eWe used the Getis-Ord Gi* statistic\u003csup\u003e24,25\u003c/sup\u003e implemented in GeoDa to quantify spatial hotspots for the environmental, social, and overall sustainability indices. The Getis-Ord Gi* analysis provides a z-score and p-value for each spatial unit (in our case NUTS2 regions) based on permutations. The z-score is calculated based on the local sum of a NUTS2 and its weighted neighbours, then compared to the expected local sum that would occur given a random distribution of all NUTS2 regions. The p-value is then calculated for the z-score as its probability at either end of the distribution, and is adjusted for multiple comparisons and spatial dependence.\u003c/p\u003e\n\u003cp\u003eThe neighbour matrix was created with the R package \u0026lsquo;spdep\u0026rsquo; using Delaunay triangulation (spdep:: tri2nb) of NUTS2 centroids and thinning the graph using the Sphere of Influence function (spdep::soi.graph). This method was chosen over contiguity approaches because of the highly irregular distribution of NUTS2 regions, including isolated regions (e.g., Cyprus, Malta). Iceland\u0026rsquo;s neighbours were manually adjusted (Extended Data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e) and all others were deemed acceptable based on visual inspection. We used row-standardised neighbour weights, meaning that the weights of all neighbours of a particular NUTS2 are equal and sum to one, as opposed to binary weights where all neighbours have a weight of one and all non-neighbours a weight of zero. This is so the total weight of all neighbours for any NUTS2 equals one, regardless of the number of neighbours, which is preferred for highly irregular spatial units such as NUTS2.\u003c/p\u003e\n\u003cp\u003eWe employed the Gi* instead of the Gi statistic to include the value of the region along with its neighbour values in the calculation, which the latter does not do. We found only negligible differences in our results using Gi* compared to Gi and row-standardised compared to binary weights. We applied 99,999 permutations to the statistic calculation in GeoDa and controlled for the false discovery rate\u003csup\u003e26\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eWe did not rely solely on the Getis-Ord Gi* analyses because the results are dependent upon the underlying neighbourhood matrix used, and spurious results can emerge from highly irregular spatial data structures such as NUTS2 (please see details below and in Extended Data Figs. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e). In order for a NUTS2 region to be considered a brightspot (or dragspot) in our final analysis, it was required to meet one of the following two criteria:\u003c/p\u003e\n\u003cp\u003e1) Be a significant Getis-Ord Gi* bright(dragspot)spot AND have an index score above (below) the European median for that index AND be adjacent to at least one other bright(drag)spot NUTS2 region;\u003c/p\u003e\n\u003ch2\u003eOR\u003c/h2\u003e\n\u003cp\u003e2) Be one of two or more adjacent NUTS2 regions with index scores more than one standard deviation above (or below) the European mean for that index.\u003c/p\u003e\n\u003cp\u003eIn other words, we excluded NUTS2 regions with below(above)-median index scores that the Getis-Ord Gi* statistic alone nonetheless identified as significant Gi* bright(drag)spots because their neighbours were significantly above average. Similarly, when multiple adjacent NUTS2 regions had index scores more than one standard deviation above (below) the mean but were not identified as significant Getis-Ord Gi* bright(drag)spots because of their neighbours, we manually corrected them to qualify as hotspots. This approach, however, still excludes isolated single-NUTS2 regions that might be considered a \u0026ldquo;lighthouse\u0026rdquo; in themselves.\u003c/p\u003e\n\u003cp\u003eSeveral examples of regions with high or low index values that did not show up as significant Getis-Ord Gi* hotspots are worth mentioning here. Much of Sweden and Finland have very high values for the environmental and overall indices (Extended Data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea and e), yet not all of their NUTS2 regions show up as significant hotpots, likely due to having few neighbours overall and with lower scores (e.g., Norway; Extended Data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). The opposite was observed for northern Norway and Cyprus, which have substantially lower environmental index scores than their neighbours (Extended Data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea), yet show up as significant hotspots (Extended Data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eb)\u0026mdash;again, likely due to the effects of few neighbours but this time with higher scores. Similar neighbourhood effects appear to happen around Berlin and Bucharest, whose regions show up as a social brightspot and dragspot, respectively (Extended Data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ed), despite having lower and higher index scores, respectively, than their neighbouring regions (Extended Data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ec). Thus, these results from the spatial hotspot analyses must be interpreted with caution, and this is why we took a hybrid approach combining the Getis-Ord Gi* and the (non-spatial) standard deviation of the index distributions for detailed analysis of hotspots.\u003c/p\u003e\n\u003ch2\u003eAnalyses of calories, livestock, and subsidies\u003c/h2\u003e\n\u003cp\u003eIn order to analyse the final overall index hotspots according to other factors related to sustainable rural systems, we compared our results to additional geospatial datasets. First, we analysed the fraction of calories delivered to the food system from each brightspot and dragspot based on the analysis of Cassidy et al.\u003csup\u003e27\u003c/sup\u003e. We used the data product \u003cem\u003eDeliveredkcalFraction.tif\u003c/em\u003e from earthstat.org to calculate the average fraction of calories delivered within each NUTS2 region. Finally, we analysed the distribution of subsidies under different schemes of the CAP, within each brightspot and dragspot. Total payments in Euros (adjusted to purchasing power parity) for either 2014 (Denmark), 2016 (Bulgaria, Sweden, and the Czech Republic), or 2015 (all other countries), were grouped according to their purpose as income support payments, environmental payments, or other/unspecified payments (see Table \u003cspan class=\"InternalRef\"\u003eS2\u003c/span\u003e in ref.\u003csup\u003e6\u003c/sup\u003e). Because the vast majority of CAP payments are based on agricultural land area, we calculated the average payment in Euros per hectare for each category of payment (income support, environmental, other/unspecified) in each hotspot, as well as for all other regions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eThe authors wish to thank Loukas Christodoulou for helpful comments and Britta Ricker for mapping guidance and advice on earlier versions of the manuscript. We are grateful for the work of the Open Knowledge Foundation to initially compile the CAP data on farmsubsidies.org. This research was conceived under Swedish Research Council Grant 2014-5899/E0589901, but conducted on researchers\u0026rsquo; free time after the grant ended.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData availability\u003c/p\u003e\n\u003cp\u003eThe full dataset and code used here is provided in the Supplementary Data. The shapefile contains the 283 NUTS2 regions analysed, the 24 variables used, the environment and social indices as well as the overall index calculated, the Gi* z-scores and p-values for each index, and the final hotspot regions. Field names and descriptions are found in Supplementary Table 2. Additional processing details (e.g., raster resolutions, projections) and code for deriving the data provided here from their raw original source can be obtained at https://github.com/murrayscown/EU-Agricultural-Systems-Database. All raw data are freely available from their original source, no additional requests for data were required in our processing and analysis.\u003c/p\u003e\n\u003cp\u003eAuthor Contributions\u003c/p\u003e\n\u003cp\u003eK.A.N. and M.S. conceived the research and wrote the manuscript; M.S. performed the analysis and created the visualizations; K.A.N. obtained funding to support the research.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eWatson, R. \u003cem\u003eet al.\u003c/em\u003e Summary for policymakers of the global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. \u003cem\u003eIPBES Secretariat: Bonn, Germany\u003c/em\u003e 22\u0026ndash;47 (2019).\u003c/li\u003e\n \u003cli\u003eSpringmann, M. \u003cem\u003eet al.\u003c/em\u003e Options for keeping the food system within environmental limits. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e562\u003c/strong\u003e, 519\u0026ndash;525 (2018).\u003c/li\u003e\n \u003cli\u003eFAO. \u003cem\u003eTransforming food and agriculture to achieve the SDGs: 20 interconnected actions to guide decision-makers\u003c/em\u003e. (Food and Agriculture Organization of the United Nations, 2018).\u003c/li\u003e\n \u003cli\u003eEC. \u003cem\u003eCommunication from the Commission to the European Parliament, the European Council, the Council, the European Economic and Social Committee and the Committee of the Regions: The European Green Deal. COM(2019) 640 final.\u003c/em\u003e (European Commission, 2019).\u003c/li\u003e\n \u003cli\u003eScown, M. W. \u0026amp; Nicholas, K. A. European agricultural policy requires a stronger performance framework to achieve the Sustainable Development Goals. \u003cem\u003eGlobal Sustainability\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 1\u0026ndash;11 (2020).\u003c/li\u003e\n \u003cli\u003eScown, M. W., Brady, M. V. \u0026amp; Nicholas, K. A. Billions in Misspent EU Agricultural Subsidies Could Support the Sustainable Development Goals. \u003cem\u003eOne Earth\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 237\u0026ndash;250 (2020).\u003c/li\u003e\n \u003cli\u003eStoddard, I. \u003cem\u003eet al.\u003c/em\u003e Three decades of climate mitigation: why haven\u0026rsquo;t we bent the global emissions curve? \u003cem\u003eAnnu Rev Environ Resour\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 653\u0026ndash;689 (2021).\u003c/li\u003e\n \u003cli\u003eSeto, K. C. \u003cem\u003eet al.\u003c/em\u003e Carbon lock-in: types, causes, and policy implications. \u003cem\u003eAnnu Rev Environ Resour\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 425\u0026ndash;452 (2016).\u003c/li\u003e\n \u003cli\u003eMeadows, D. H. \u003cem\u003eThinking in systems: A primer\u003c/em\u003e. (chelsea green publishing, 2008).\u003c/li\u003e\n \u003cli\u003eKuemmerle, T. \u003cem\u003eet al.\u003c/em\u003e Hotspots of land use change in Europe. \u003cem\u003eEnvironmental research letters\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 064020 (2016).\u003c/li\u003e\n \u003cli\u003eOberschelp, C., Pfister, S., Raptis, C. E. \u0026amp; Hellweg, S. Global emission hotspots of coal power generation. \u003cem\u003eNat Sustain\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 113\u0026ndash;121 (2019).\u003c/li\u003e\n \u003cli\u003eLevers, C. \u003cem\u003eet al.\u003c/em\u003e Archetypical patterns and trajectories of land systems in Europe. \u003cem\u003eReg Environ Change\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 715\u0026ndash;732 (2018).\u003c/li\u003e\n \u003cli\u003eCai, W., Gibbs, D., Zhang, L., Ferrier, G. \u0026amp; Cai, Y. Identifying hotspots and management of critical ecosystem services in rapidly urbanizing Yangtze River Delta Region, China. \u003cem\u003eJ Environ Manage\u003c/em\u003e \u003cstrong\u003e191\u003c/strong\u003e, 258\u0026ndash;267 (2017).\u003c/li\u003e\n \u003cli\u003eWang, X. \u0026amp; Varady, D. P. Using hot-spot analysis to study the clustering of section 8 housing voucher families. \u003cem\u003eHous Stud\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 29\u0026ndash;48 (2005).\u003c/li\u003e\n \u003cli\u003eZhang, H. \u0026amp; Tripathi, N. K. Geospatial hot spot analysis of lung cancer patients correlated to fine particulate matter (PM2. 5) and industrial wind in Eastern Thailand. \u003cem\u003eJ Clean Prod\u003c/em\u003e \u003cstrong\u003e170\u003c/strong\u003e, 407\u0026ndash;424 (2018).\u003c/li\u003e\n \u003cli\u003eValencia, V. \u003cem\u003eet al.\u003c/em\u003e Learning from the future: mainstreaming disruptive solutions for the transition to sustainable food systems. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 051002 (2022).\u003c/li\u003e\n \u003cli\u003eBennett, E. M. \u003cem\u003eet al.\u003c/em\u003e Bright spots: seeds of a good Anthropocene. \u003cem\u003eFront Ecol Environ\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 441\u0026ndash;448 (2016).\u003c/li\u003e\n \u003cli\u003eDebonne, N. \u003cem\u003eet al.\u003c/em\u003e The geography of megatrends affecting European agriculture. \u003cem\u003eGlobal Environmental Change\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, 102551 (2022).\u003c/li\u003e\n \u003cli\u003eDornelles, A. Z. \u003cem\u003eet al.\u003c/em\u003e Transformation archetypes in global food systems. \u003cem\u003eSustain Sci\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1827\u0026ndash;1840 (2022).\u003c/li\u003e\n \u003cli\u003eGerten, D. \u003cem\u003eet al.\u003c/em\u003e Feeding ten billion people is possible within four terrestrial planetary boundaries. \u003cem\u003eNat Sustain\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 200\u0026ndash;208 (2020).\u003c/li\u003e\n \u003cli\u003eLevers, C. \u003cem\u003eet al.\u003c/em\u003e Drivers of forest harvesting intensity patterns in Europe. \u003cem\u003eFor Ecol Manage\u003c/em\u003e \u003cstrong\u003e315\u003c/strong\u003e, 160\u0026ndash;172 (2014).\u003c/li\u003e\n \u003cli\u003eR\u0026ouml;\u0026ouml;s, E. \u003cem\u003eet al.\u003c/em\u003e Moving beyond organic\u0026ndash;A food system approach to assessing sustainable and resilient farming. \u003cem\u003eGlob Food Sec\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 100487 (2021).\u003c/li\u003e\n \u003cli\u003eScown, M. W., Winkler, K. J. \u0026amp; Nicholas, K. A. Aligning research with policy and practice for sustainable agricultural land systems in Europe. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e116\u003c/strong\u003e, 4911\u0026ndash;4916 (2019).\u003c/li\u003e\n \u003cli\u003eAnselin, L. Local Spatial Autocorrelation (2): Other Local Spatial Autocorrelation Statistics. https://geodacenter.github.io/workbook/6b_local_adv/lab6b.html#getis-ord-statistics (2020).\u003c/li\u003e\n \u003cli\u003eGetis, A. \u0026amp; Ord, J. K. The Analysis of Spatial Association by Use of Distance Statistics. \u003cem\u003eGeogr Anal\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 189\u0026ndash;206 (1992).\u003c/li\u003e\n \u003cli\u003eBenjamini, Y. \u0026amp; Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. \u003cem\u003eJournal of the Royal statistical society: series B (Methodological)\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 289\u0026ndash;300 (1995).\u003c/li\u003e\n \u003cli\u003eCassidy, E. S., West, P. C., Gerber, J. S. \u0026amp; Foley, J. A. Redefining agricultural yields: from tonnes to people nourished per hectare. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 034015 (2013).\u003c/li\u003e\n \u003cli\u003eNicholas, K. A., Villemoes, F., Lehsten, E., Brady, M. V. \u0026amp; Scown, M. W. A harmonized and spatially-explicit dataset for the European Union\u0026rsquo;s \u0026euro;61 billion in Common Agricultural Policy payments to farmers for 2015. Preprint at https://doi.org/10.23644/uu.12706580.v1 (2020).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"food systems, land use, multifunctionality, leverage points, SDGs, spatial analysis","lastPublishedDoi":"10.21203/rs.3.rs-2941468/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2941468/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eTransformation of rural land systems is essential if the European Union is to achieve its goal of fair and healthy food systems while becoming the first climate-neutral continent and halting biodiversity loss. Here we develop and apply a method to quantitatively assess the environmental and social sustainability of rural land systems in Europe, with regards to the EU Sustainable Development Goals and the Common Agricultural Policy. Using spatial hotspot analyses based on 24 indicators at the NUTS2 regional level, we identified two \u0026ldquo;brightspots\u0026rdquo; with good environmental and social performance (Nordics and Central Europe), and five \u0026ldquo;dragspots\u0026rdquo; hindering sustainability: the Balkans, the Lowlands, Northern Italy, Southern Italy and Malta, and Southern Spain. Existing subsidies over-reward large, intensive, unsustainable farms. A shift to low-intensity stewardship of high nature value farmland, and better integration of forests is necessary if rural systems are to transform to meet their social goals.\u003c/p\u003e","manuscriptTitle":"European rural regions supporting and hindering the Sustainable Development Goals","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-09-20 23:18:10","doi":"10.21203/rs.3.rs-2941468/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"communications-earth-and-environment","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"commsenv","sideBox":"Learn more about [Communications Earth and Environment](https://www.nature.com/commsenv/)","snPcode":"","submissionUrl":"","title":"Communications Earth \u0026 Environment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Communications Series","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"0bfbcfb7-2207-46cd-b2b1-530c10002329","owner":[],"postedDate":"September 20th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":22200002,"name":"Earth and environmental sciences/Environmental social sciences/Sustainability"},{"id":22200003,"name":"Scientific community and society/Agriculture"},{"id":22200004,"name":"Scientific community and society/Geography"}],"tags":[],"updatedAt":"2024-11-12T08:09:52+00:00","versionOfRecord":{"articleIdentity":"rs-2941468","link":"https://doi.org/10.1038/s43247-024-01736-6","journal":{"identity":"communications-earth-and-environment","isVorOnly":false,"title":"Communications Earth \u0026 Environment"},"publishedOn":"2024-11-11 05:00:00","publishedOnDateReadable":"November 11th, 2024"},"versionCreatedAt":"2023-09-20 23:18:10","video":"","vorDoi":"10.1038/s43247-024-01736-6","vorDoiUrl":"https://doi.org/10.1038/s43247-024-01736-6","workflowStages":[]},"version":"v1","identity":"rs-2941468","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2941468","identity":"rs-2941468","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}