4.1 Interpretation and implications for EFA activation and implementation
This section explores the results of the spatial analysis with respect to countries and regions across Europe that have or have not activated each EFA. Table
5 provides an overview of which EFAs each MS has activated (note there are 32 as the UK and Belgium are split into 4 and 2 regions respectively) and Table
6 provides some example results for individual NUTS3 regions that are mentioned below.
Table 5
EFA Activation in the EU in 2017
Austria | | | | X | X | X | | | X | X | X | | |
Belgium—Flanders | X | X | X | X | X | X | X | X | X | X | X | | |
Belgium—Walonia | P | X | X | X | X | X | X | X | X | X | X | | X |
Bulgaria | P | | | X | | X | X | X | X | | X | | X |
Croatia | P | | X | X | X | X | | X | X | X | X | | X |
Cyprus | X | X | X | | | X | | | X | | | | |
Czech Republic | X | | | X | X | X | X | | X | | X | X | X |
Denmark | | | X | X | | X | | | | X | X | | |
Estonia | P | | | | X | X | | X | X | | X | | X |
Finland | | | | | | X | | | X | | X | | |
France | X | X | X | X | X | X | X | X | X | X | X | X | X |
Germany | X | X | X | X | X | X | X | X | X | O | X | X | X |
Greece | P | | X | | X | X | | | X | | | | |
Hungary | X | X | X | X | X | X | X | X | X | X | X | X | X |
Ireland | X | | X | X | X | X | | X | X | | X | | |
Italy | X | X | X | | X | X | X | X | X | X | X | X | X |
Latvia | P | | | X | | X | X | O | X | X | | | X |
Lithuania | P | | | X | X | X | X | | X | X | X | | |
Luxembourg | X | X | X | X | | X | X | X | X | X | X | | X |
Malta | | | | | | X | | | X | | | | |
Netherlands | P | | | X | X | | X | X | X | X | X | | X |
Poland | X | | X | X | X | X | X | X | X | X | X | | X |
Portugal | X | X | | | | X | | | X | | | O | |
Romania | X | | X | X | X | | X | X | X | X | X | X | X |
Slovakia | P | | X | X | | X | X | X | X | | X | X | X |
Slovenia | | | | X | | X | | | X | | | | |
Spain | X | X | | | | X | | | X | | | | |
Sweden | | | | X | | X | X | | X | | X | | |
UK—England | | | X | X | | X | | X | X | | | | |
UK—Northern Ireland | X | X | | | X | X | | X | X | | X | | |
UK—Scotland | | | X | X | FM | X | X | FM | X | | | | |
UK—Wales | X | | | | | X | | X | X | | X | | |
Table 6
Example region impact scores
Afforested areas |
Sydjylland (DK032) | Flood regulation | 28 | 81 | 23 | 65 | 16 | 46 |
Lincolnshire (UKF30) | Flood regulation | 28 | 81 | 11 | 32 | 10 | 29 |
Larisa (EL612) | Flood regulation | 28 | 81 | 14 | 40 | 7 | 19 |
Siauliu apskritis (LT006) | Flood regulation | 28 | 81 | 11 | 32 | 7 | 19 |
Eure-et-Loir (FR242) | Water provision | − 81 | − 28 | − 81 | − 28 | − 64 | − 22 |
Salzland (DEE0C) | Water provision | − 81 | − 28 | − 81 | − 28 | − 65 | − 23 |
Bekes (HU332) | Water provision | − 81 | − 28 | − 81 | − 28 | − 66 | − 23 |
Treviso (ITD34) | Water provision | − 86 | − 34 | − 86 | − 34 | − 54 | − 22 |
Lubelski (PL314) | Water provision | − 81 | − 28 | − 81 | − 28 | − 53 | − 19 |
Ialomita (RO315) | Water provision | − 81 | − 28 | − 81 | − 28 | − 67 | − 24 |
Valladolid (ES418) | Water provision | − 81 | − 28 | − 81 | − 28 | − 64 | − 23 |
Arr. Ieper (BE253) | Dilution | − 70 | − 50 | − 10 | − 7 | − 8 | − 5 |
Somme (FR223) | Dilution | − 72 | − 52 | − 10 | − 7 | − 7 | − 5 |
Emsland (DE949) | Dilution | − 80 | − 60 | − 8 | − 6 | − 6 | − 4 |
Cremona (ITC4A) | Dilution | − 70 | − 50 | − 8 | − 6 | − 7 | − 5 |
Teruel (ES242) | Dilution | − 70 | − 50 | − 17 | − 12 | − 6 | − 4 |
Larisa (EL612) | Dilution | − 42 | − 22 | − 18 | − 9 | − 8 | − 4 |
Oost-Groningen (NL111) | Dilution | − 90 | − 70 | − 9 | − 7 | − 8 | − 6 |
Buffer strips |
Stredocesky kraj (CZ020) | Filtration | 8 | 75 | 1 | 7 | 0 | 4 |
Cádiz (ES612) | Dilution | 1 | 73 | 0 | 21 | 0 | 8 |
Tarragona (ES514) | Filtration | 5 | 76 | 2 | 33 | 0 | 6 |
West-Noord-Brabant (NL411) | Dilution | 9 | 76 | 1 | 11 | 1 | 7 |
Catch crops or green cover |
Campobasso (ITF22) | Dilution | − 2 | 35 | 0 | 8 | 0 | 5 |
Cádiz (ES612) | Dilution | − 2 | 28 | 0 | 8 | 0 | 3 |
Karditsa, Trikala (EL611) | Dilution | − 2 | 35 | − 1 | 15 | 0 | 7 |
Ditches |
Cádiz (ES612) | Dilution | − 20 | 75 | − 6 | 21 | − 2 | 8 |
Yambol (BG343) | Dilution | − 20 | 75 | − 3 | 11 | − 2 | 6 |
Kilkis (EL523) | Dilution | − 20 | 75 | − 8 | 32 | − 4 | 16 |
Sydjylland (DK032) | Flood regulation | 0 | 100 | 0 | 80 | 0 | 57 |
Lincolnshire (UKF30) | Flood regulation | 0 | 100 | 0 | 40 | 0 | 36 |
Fallow land |
Teleorman (RO317) | Dilution | − 22 | − 2 | − 1 | 0 | − 1 | 0 |
Zeeuwsch-Vlaanderen (NL341) | Dilution | − 25 | − 2 | − 3 | 0 | − 2 | 0 |
Somme (FR223) | Dilution | − 31 | − 2 | − 4 | 0 | − 3 | 0 |
Ascoli Piceno (ITE34) | Dilution | − 36 | − 2 | − 8 | − 1 | − 4 | 0 |
Karditsa, Trikala (EL611) | Dilution | − 25 | − 2 | − 10 | − 1 | − 5 | 0 |
Bekes (HU332) | Dilution | − 27 | − 2 | − 3 | 0 | − 3 | 0 |
Hedges or wooded strips |
Cádiz (ES612) | Dilution | 0 | 100 | 0 | 28 | 0 | 11 |
Kilkis (EL523) | Dilution | 0 | 100 | 0 | 42 | 0 | 21 |
Nitrogen fixing crops |
Stara Zaogora (BG344) | Dilution | − 26 | − 1 | − 5 | 0 | − 2 | 0 |
Somme (FR223) | Dilution | − 35 | − 2 | − 5 | 0 | − 4 | 0 |
Steinfurt (DEA37) | Dilution | − 36 | − 2 | − 4 | 0 | − 3 | 0 |
Larisa (EL612) | Dilution | − 23 | − 1 | − 10 | − 1 | − 5 | 0 |
Bekes (HU332) | Dilution | − 26 | − 1 | − 3 | 0 | − 3 | 0 |
Ascoli Piceno (ITE34) | Dilution | − 38 | − 2 | − 9 | 0 | − 5 | 0 |
Navarra (ES220) | Dilution | − 34 | − 2 | − 8 | 0 | − 3 | 0 |
Ponds |
Kilkis (EL523) | Dilution | 14 | 94 | 6 | 40 | 3 | 20 |
Zaragosa (ES243) | Dilution | 14 | 94 | 3 | 17 | 1 | 8 |
Yambol (BG343) | Dilution | 14 | 94 | 2 | 14 | 1 | 8 |
Afforested areas
The unweighted scores for afforested areas (Table
4) indicate that there are regions where the combination of attributes means that potential benefits and burdens are maximised (as indicated by the − 100 and + 100 scores). For dilution, the vulnerability (Score
VW) and area and vulnerability (Score
VAW) weighted scores are significantly lower, indicating that high vulnerability is not coinciding with the attributes for maximum impact and/or a high proportion of arable land in the region. This lack of coincidence is less so for flood regulation and water provision which have higher Score
VW and Score
VAW values. Thus it can be concluded that the implementation (or avoidance) of afforested areas has the greatest potential benefits with respect to flood regulation and burdens with respect to water provision, and lesser potential burdens with respect to dilution.
The areas with the greatest potential benefits for flood regulation are those where the vulnerability to flooding is high, annual rainfall is high and there is a high proportion of arable land (thus high potential to implement this land use as an EFA). This includes central Denmark, eastern England, northwest France, central-northern Italy, central Greece, north-east Germany, southern Poland and Lithuania (see Fig.
1). There are some minor differences between the worst to best case scenarios. The areas with the greatest potential burdens for water provision are those where the region is vulnerable (i.e. currently water stressed and predicted to become warmer and drier) and paradoxically where rainfall is high (as there is greater scope for a reduction in water provision). This includes eastern-England, Denmark, northern France, central Spain, central Germany, Poland, eastern Italy, Hungary and southeast Romania (see Fig.
2). There are some major differences between the worst and best case scenarios, with the best case scenario eliminating any significant burdens, whilst the worst case scenario significantly extends the number of regions where burdens could occur. There are other more localised and management factors, which have not been specified in this spatial analysis, which affect the benefits and burdens (these unspecified factors contribute the difference between the worst and best case scenarios). These are that older woodlands (20+ years) tend to have a greater effect on flood regulation and water provision (Farley et al.
2005), with coniferous woodland having a slightly greater effect than broadleaved woodland (Sahin and Hall
1996).
The areas with the greatest potential burdens for dilution are those where vulnerability to dilution is high (i.e. where the current chemical condition of water is low, and the region is predicted to become drier resulting in less volume of water) and where the risk of acid and nutrient deposition is high and the buffering capacity of the soil is low. This includes include northern France and Belgium, eastern Spain, south and eastern Netherlands, western Germany, north and eastern Italy, Hungary and eastern Greece (see Fig.
3). There are some minor differences between the worst to best case scenarios. Other more geographically localised and/or management factors that influence the burdens and were unspecified are woodland type (with coniferous and eucalypts having a greater burden than broadleaved woodland), and whether the woodland is commercially harvested or not, with commercially harvested woodlands having a slightly lower burden (Allen and Chapman
2001; Bastrup-Birk and Gundersen
2004).
There are eight MSs that have not activated the ‘afforested areas’ or the ‘trees in groups and field copses’ EFAs. Two of these have been identified as having significant potential for benefits for flood regulation (based on the combination of the performance of the EFA, the vulnerability and the area of arable land). These are (with example regions) Denmark (e.g. Sydjylland—DK032) and UK—England (e.g. Lincolnshire—UKF30). A further nine MSs have activated ‘trees in groups and field copses’ only, and two of these have been identified as having significant potential for benefits for flood regulation should there be more uptake of afforested areas. These are Greece (e.g. Larisa—EL612) and Lithuania (e.g. Siauliu apskritis—LT006).
It could be argued that these MSs should consider activating the ‘afforested areas’ EFA. However, there are also burdens from afforested areas on water provision and dilution. In many respects, this is a trade-off as the consequence of forests retaining water (to minimise flooding) is that this can reduce water provision and dilution downstream in the catchment. However, this trade-off depends on the circumstances of each region. With regard to water provision, France, Germany, Hungary, Italy, Poland, Romania and Spain have all activated ‘afforested areas’ and have areas where there could be significant burdens on water provision (e.g. Eure-et-Loir—FR242, Salzland—DEE0C, Bekes—HU332, Treviso—ITD34, Lubelski—PL314, Ialomita—RO315 and Valladolid—ES418). Denmark and England, which could gain benefits for flood regulation by activating this EFA, also have areas that would experience significant burdens on water provision. Thus if activated, the potential trade-offs would need to be considered on a region by region basis. Potential dilution issues have been identified for Belgium, France, Germany, Italy, Spain which have activated ‘afforested areas’ and Greece and the Netherlands which have activated ‘trees in groups and field copses’ (e.g. Arr. Ieper—BE253, Somme—FR223, Emsland—DE949, Cremona—ITC4A, Teruel—ES242, Larisa—EL612 and Oost-Groningen—NL111). So again it may be advisable for these MSs to avoid implementing ‘afforested areas’ in regions most vulnerable to dilution issues, and Greece may also need to consider the trade-off between flood regulation benefits and dilution burdens on a region by region basis.
Ditches
The unweighted scores for ditches (Table
4) indicate that there are regions where the combination of attributes results in high potential benefit (+ 75) or moderate burden (− 20) for dilution, and the benefit is maximised (+ 100) for flood regulation. For dilution, the Score
VW and Score
VAW values are significantly lower indicating that high vulnerability and/or high proportion of arable land in the region are not coinciding with the attributes for high benefits or burdens, but they are still moderate and low for the best and worst case scenarios. For flood regulation, there is a greater coincidence with a maximum Score
VW value (+ 100) and a slightly lower Score
VAW value (+ 59). There is consequently scope for moderate to high benefits in relation to flood regulation, moderate benefits for dilution and low burdens for dilution.
The areas where ditches have a high benefit for flood regulation are simply those where the vulnerability to flooding is high and there is a high proportion of arable land. This includes central Denmark, eastern England, northwest France, central-northern Italy, central Greece, northeast Germany, southern Poland and Lithuania (see Fig.
8). There are some minor differences between the worst and best case scenarios, with the worst case scenario resulting in no benefits for flood regulation. Areas where ditches have benefits and burdens for dilution relate to the dilution vulnerability in the region and the proportion of arable land. The greatest benefits are in northern France and Belgium, eastern and southwest Spain, south and eastern Netherlands, western Germany, northeast and southern Italy, Hungary, southern Bulgaria and eastern Greece (see Fig.
9). There are some minor differences between the worst and best case scenarios, with the worst case scenario resulting in burdens in northern France, southeast England, eastern and southwest Spain, Germany, Czech Republic, Hungary, southern Romania, Bulgaria and Greece. The worst to best case scenarios are influenced by more geographically localised and/or management factors which have not been specified in this spatial analysis. These factors include disposal of cut weeds (whether they are removed or not), presence of low-grade weirs/small dams in ditch (with sediments removed periodically), the general in-ditch flow rate, the presence of in-ditch vegetation, dredging of ditch sediments, disposal of dredged sediments, time of dredging, whether high sulphate soil additions are used, intermittent periods of ditch drying and whether livestock have access to the ditch bank (Kröger et al.
2014; Needelman et al.
2007; Shore et al.
2015; Smith
2009).
There are 14 MSs that have not activated the ‘ditches’ EFA. Two of these have been identified as having significant potential for benefits for dilution. These are Spain (e.g. Cádiz—ES612) and Bulgaria (e.g. Yambol—BG343). Consequently, activation of this EFA in these regions may be beneficial. In the case of Bulgaria, the ‘ditches’ EFA was available in 2015–2016, but has been deactivated in 2017. The reasons for this are unknown, but it should perhaps be reconsidered for activation in the future. A number of MSs also have regions where low burdens in relation to dilution might result from ditches, such as Kilkis (EL523) in Greece. However, it should be noted that the worst case values, as explained above, would only be realised under specific local and/or management factors.
Two MSs that have not activated the ‘ditches’ EFA have been identified as having significant potential for benefits for flood regulation. These are Denmark (e.g. Sydjylland—DK032) and UK—England (e.g. Lincolnshire—UKF30).
4.2 Limitations and wider perspectives
This study has brought together the outputs from two continental-scale projects, both of which synthesised a substantial amount of scientific data and information. In some respects, it has been an exercise in the conversion of scientific knowledge into a format that can aid pragmatic decision making. This conversion process inherently involves aggregation and simplification and the application of novel processes and techniques. Consequently, it is important to acknowledge some of the weaknesses in the approach and any limitations on the findings of this study.
Firstly, it is important to acknowledge that the assessments have been made based on a knowledge base of existing evidence of the potential impact of EFAs, and this is not complete for all ecosystem services across all the EFAs. Consequently, there may be benefits and burdens for some ecosystem services from some of the EFAs explored in this study, which have not been highlighted in this analysis. For example, afforested areas are the only EFA that has been examined with respect to impacts on water provision. Other EFAs, such as agroforestry, ditches, hedges, ponds and short rotation coppice, may also have an impact. However, the knowledge base (Tzilivakis et al.
2015b) that underpins the indicator framework (Tzilivakis et al.
2016) is based on the weight of evidence that was available at the time, and there was a lack of scientific evidence specifically exploring the impact of these EFAs in relation to water provision. Similarly, only afforested areas and ditches have been examined with respect to impacts on flood regulation. Many other EFAs could also potentially affect hydrological processes and thus flood regulation, but again evidence associating (and quantifying) the effect of specific EFAs was lacking. This is not a new phenomenon, as all environmental assessments are made based on the most established and relevant scientific understanding available at the time. Scientific knowledge is evolving and growing all the time; consequently, efforts are ongoing to plug gaps in the knowledge base that underpins the EFA calculator, and as the science improves, the assessments made within the tool should reflect this.
It is important to acknowledge that the spatial analysis presented herein is a relatively broad-brush approach and that any benefits or burdens identified for a region do not mean this will definitely occur should those EFAs be implemented on any farm within the region. Specific benefits and burdens will depend on the details of their implementation on each farm within each region, e.g. buffer strips being located in the right places to be most effective in a catchment. The analysis presented herein does account for this to some extent by presenting the worst and best case scenarios (which account for unspecified parameters), but it is likely there will be greater variability.
In relation to the last point, it is also important to acknowledge that no account is taken of the performance of an EFA in a region in relation to a baseline. The analysis is simply highlighting the potential benefits and burdens should the EFA be implemented on 5% of the arable land within the region. The EFA policy only requires that an area of the farm that is equivalent to 5% of arable land area is declared as EFA. This may not necessarily require the creation of new EFAs, as EFAs covering 5% may already exist. The baseline situation could be accounted for within the spatial analysis by adjusting the area weighting using data on existing EFA declaration or using data such as the distribution of semi-natural vegetation in agricultural land (as done by Angileri et al.
2017). However, in this instance, the study is examining potential future vulnerabilities and the need to increase capacity within a region to cope with climate change. Therefore, the results presented herein should be interpreted as a need to increase certain EFAs in order gain the benefits (or avoiding certain EFAs to decrease burdens).
It is also important to reiterate that the indicator framework used (Tzilivakis et al.
2016), and thus the analysis presented, is one that identifies the relative performance of an EFA and does not quantify actual benefits or burdens for ecosystem services. Quantification of ecosystem services requires more sophisticated techniques, many of which do not exist for such broad-scale analyses (Baveye
2017; Maes et al.
2016). Efforts are being made to develop such techniques, such as those that have emerged from the Quantification of Ecological Services for Sustainable Agriculture project (QuESSA
2017). Some of these have been incorporated into the latest version of the EFA calculator, but these are only available for a limited number of ecosystem services, including pollination, pest control, soil erosion, carbon and aesthetics. As such techniques evolve and develop, a similar analysis to the one presented herein could be undertaken using these techniques, and could liberate additional information that could aid decision making and the selection and implementation of EFAs that maximise benefits and minimise burdens for ecosystem services under current and future climatic scenarios.
This study has explored the potential benefits and burdens of EFAs on water-related ecosystem services vulnerable to climate change. It is important to remember that the EFAs examined in this study (and those not examined) also have other benefits and burdens which will influence decisions with regard to their promotion (via policy) and uptake on the ground. There can be numerous effects and impacts to consider, for example climate regulation (greenhouse gas emissions, including above and below ground carbon), soil erosion, pollination, pest control, aesthetic and other cultural services and biodiversity. Some of these could be explored using the spatial analysis presented herein, but many will depend on more localised variables and management factors for which georeferenced spatial data do not exist. Should this data for these factors become available, then these could be analysed in the same way, extending the assessment to include more ecosystem services and EFAs, and thus providing a more holistic perspective.
It is important that other benefits and burdens are taken into account in the decision making processes associated with their promotion, uptake and implementation. This will involve identifying where there are synergies and trade-offs to be taken into account. The spatial analysis presented in this study can support these decision making processes by identifying regions where the greatest benefits and burdens might occur in the event of climate change projections being realised. For example, afforested areas have the potential to provide benefits for flood regulation and burdens for water provision and dilution. Under the right circumstances, they can also provide benefits for numerous wildlife species, climate regulation (via carbon sequestration), pollination and aesthetic services. The synergies and trade-offs between these benefits and burdens will need to be considered within each region when considering the implementation of EFAs. As such, the spatial analysis presented herein can contribute towards understanding such synergies and trade-offs in the context of future vulnerabilities and threats from climate change.
This study has explored the potential impact on ecosystem services from a number of land uses and landscape features in the context of Europe and its land use policies in the light of projected changes in climate in the future. Other continents and countries around the world face similar challenges (Chang and Bonnette
2016; Fu et al.
2017; Malinga et al.
2015; Overbeck et al.
2015; Rodríguez-Echeverry et al.
2018; Tolessa et al.
2017), so there is a need for tools and techniques to aid policy assessment and development. For example, Tolessa et al. (
2017) utilised GIS techniques to estimate changes in ecosystem services over 40-year period as a consequence of land use change in the central highlands of Ethiopia. This study highlighted the impact of Ethiopian land use policies during this period and thus can be used to adapt policies towards sustaining important ecosystem services. Similarly, Rodríguez-Echeverry et al. (
2018) adopted a spatial approach to assess the impact of land use change ecosystem services from Chilean temperate forests and identified that losses in habitat and diversity are having detrimental effects on soil erosion and water supply. Fu et al. (
2017) also used a spatial approach to assess the effects of land use and climate change on ecosystem services in central Asia’s arid regions and propose a number of policy responses to sustain ecosystem services. All these studies attempt to take into account land use, ecosystem service provision and vulnerability to climate change. An important element amongst them all is the utilisation and tailoring of appropriate measures and indicators in the regions being studied. The study presented herein demonstrates how an indicator framework can be developed based on the latest knowledge of the effects and impacts of land use and then applied in the context of climate change vulnerability using spatial techniques. This ‘generic’ concept could be applied elsewhere in the world and thus help spatially target mitigation and adaptation policy interventions based on current knowledge and understanding in any given region.
4.3 Conclusions
There is little doubt that climate change and its consequent effects and impacts pose a great challenge to society, thus there is a need for robust mitigation and adaptation strategies and policies. Understanding the threats and vulnerabilities is a key aspect to tackle this challenge. Another key aspect is to determine the actions that will minimise any negative effects and impacts. This applies to all industry sectors, including rural land management, and applies to both policy makers and managers on the ground. It is important that relevant stakeholders have access to reliable information, data and analysis that will aid decision making on what are often complex topics. This often involves processing a substantial amount of complex data and information and presenting and interpreting it in a form to support decision making processes.
This study has used the continent of Europe as example to explore the challenges involved in assessing the benefits and burdens of land use policies on water-related ecosystem services that are vulnerable to climate change. Many of these are common to other regions of the world, for example water provision/supply in California, USA (Byrd et al.
2015), Chile (Rodríguez-Echeverry et al.
2018) and Ethiopia (Tolessa et al.
2017); water regulation and flooding in China (Ouyang et al.
2016) and the USA (Blumstein and Thompson
2015); and water quality in China (Gao et al.
2017) and Japan (Fan and Shibata
2015). Consequently, the study presented in this paper has global relevance. Many of the land uses and landscape features explored in this study exist in other regions of the world. The effects and impacts will vary with location, as they do in Europe, but under similar circumstances, similar benefits and burdens may arise. Where circumstances are different, the approach presented in this paper of combining pertinent indicators, spatial data and vulnerability assessments could be applied to other regions and continents to aid the development of mitigation and adaptation strategies and policies for land use and ecosystem services.
This study combined the outputs from two continental-scale projects to provide a means to identify European regions where EFAs may help increase the capacity of water-related ecosystem services in areas where they are vulnerable due to climate change. Six EFAs have been identified which have not been activated in some MSs, but which have the potential to provide benefits. These include afforested areas for flood regulation in Denmark, England, Greece and Lithuania; buffer strips for dilution and filtration in the Czech Republic, Netherlands and Spain; catch crops or green cover for dilution in Greece, Italy and Spain; ditches for dilution in Bulgaria and Spain and for flood regulation in Denmark and England; trees in a line (hedges) for dilution in Greece and Spain; and ponds for dilution in Bulgaria, Greece and Spain. It is recommended that these MSs consider activating these EFAs in the future. Eleven Member States have also been identified which have regions where some EFAs should perhaps be avoided, due to potential burdens on vulnerable ecosystem services. These include afforested areas for water provision in some regions in France, Germany, Hungary, Italy, Poland, Romania and Spain, and for dilution in some regions in Belgium, France, Germany, Italy, Spain, Greece and the Netherlands; fallow land for dilution in some regions in the Netherlands, Romania, France, Italy, Greece and Hungary; and nitrogen fixing crops for dilution in some regions in Bulgaria, France, Germany, Greece, Hungary, Italy and Spain.
Current European policy does not require that new EFAs be created to meet the 5% arable area threshold for farms, as existing land use and landscape features can be declared as EFAs. This is still a positive policy as it aims to maintain these areas for the future. However, given the challenges that lie ahead with respect to climate change, there is a growing need to increase, rather than just maintain, the capacity of landscapes to perform the ecosystem services that society relies on, such as water provision, flood regulation and the water purifying services of filtration and dilution. As such, EFAs and/or other land management policies need to encourage this increase in capacity, especially in the most vulnerable regions.
It will also be important to ensure that there are demonstrable benefits from the implementation of EFAs or other interventions. The impact of land use policies on ecosystem services should be monitored to ensure expected outcomes are realised, and when they are not realised evidence on the reasons for this should be gathered. This will generate new scientific data, evidence and understanding that can be used to improve the knowledge base, improve the tools available to aid future decision making and consequently make progress towards more sustainable and resilient land use that delivers the ecosystem services society requires.