Database and operational classification system of ecosystem service – natural capital relationships
Deliverable D3.1 / WP3
February 2015
M. Pérez Soba1, P.A. Harrison2, A.C. Smith2, G. Simpson2, M. Uiterwijk1, L. Miguel Ayala1, F. Archaux3, T. Erős5, N. Fabrega Domenech6, Á.I. György5, R. Haines-Young4, S. Li2, E. Lommelen7, L. Meiresonne7, L. Mononen8, E. Stange9, F. Turkelboom7, C. Veerkamp10 and V. Wyllie de Echeverria2
1Alterra, Wageningen University and Research Centre, the Netherlands 2Environmental Change Institute, University of Oxford, UK 3IRSTEA, Domaine des Barres, 45290 Nogent-sur-Vernisson, France 4CEM, School of Geography, University of Nottingham, UK 5Centre for Ecological Research, Hungarian Academy of Sciences 6 European Commission, Joint Research Centre (JRC) 7Institute for Nature and Forest Research (INB0), Belgium 8 Finnish Environment Institute 9 Norwegian Institute for Nature Research (NINA) 10WUR/PBL, the Netherlands
From concepts to real-world applications www.openness-project.eu
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 2
Prepared under contract from the European Commission Contract n° 308428 Collaborative project FP7 Environment
Project acronym: OpenNESS Project full title: Operationalisation of natural capital and ecosystem services: from concepts to real-world applications Start of the project: 01 December 2012 Duration: 54 months Project coordinator: Finnish Environment Institute (SYKE) Project website http://www.openness-project.eu
Deliverable title: Database and operational classification system of ecosystem service - natural capital relationships Deliverable n°: D3.1 Nature of the deliverable: Report Dissemination level: Public
WP responsible: WP3 Lead beneficiary: Alterra
Citation: EU FP7 OpenNESS Project Deliverable 3.1, M. Pérez Soba, P.A. Harrison, A.C. Smith, G. Simpson, M. Uiterwijk, L. Miguel Ayala, F. Archaux, T. Erős, N. Fabrega, Á.I. György, R. Haines-Young, S. Li, E. Lommelen, L. Meiresonne, L. Mononen, E. Stange, F. Turkelboom, C. Veerkamp and V. Wyllie de Echeverria. Database and operational classification system of ecosystem service - natural capital relationships. European Commission FP7, 2015.
Due date of deliverable: Month n° 24 Actual submission date: Month n° 27
Deliverable status:
Version Status Date Reviewed by Author(s) 1.1 Final February 2014 M. Pérez Soba, P.A. Harrison, A.C. Smith et al.
The contents of this deliverable do not necessarily reflect the official opinions of the European Commission or other institutions of the European Union.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 3
Table of contents
Summary ...... 5 1 Introduction ...... 8 1.1 Links between natural capital and ecosystem services: current knowledge ...... 9 2 Methods ...... 10 2.1 Selection of ecosystem services ...... 12 2.2 Development of OpenNESS database tool ...... 14 2.3 Literature search and data entry ...... 23 2.4 Network diagrams ...... 24 3 Overview and synthesis of results ...... 25 3.1 Location and spatial scale ...... 25 3.2 Ecosystem type and condition...... 27 3.3 Linkages between natural capital and ecosystem services ...... 29 3.4 Human input and management ...... 38 3.5 Interactions with other ecosystem services ...... 41 3.6 Thresholds ...... 45 3.7 Indicators for ecosystem service assessment ...... 46 3.8 Policies ...... 47 4 Results for each ecosystem service ...... 47 4.1 Freshwater fishing ...... 47 4.2 Timber production ...... 50 4.3 Freshwater provision ...... 52 4.4 Food production (cultivated crops) ...... 55 4.5 Air quality regulation ...... 57 4.6 Atmospheric regulation (carbon sequestration) ...... 60 4.7 Mass flow regulation (erosion protection) ...... 63 4.8 Water flow regulation (flood protection) ...... 65 4.9 Water quality regulation ...... 68 4.10 Pollination ...... 71 4.11 Pest Regulation ...... 73 4.12 Recreation (species-based)...... 76 4.13 Aesthetic landscapes ...... 79 5 Discussion ...... 81
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 4
5.1 Coverage of regions and ecosystems ...... 82 5.2 Linkages between natural capital and ecosystem services ...... 82 5.3 The role of human intervention in protecting and enhancing ecosystem services ...... 85 6 Conclusions ...... 86 6.1 Outputs for OpenNESS ...... 86 6.2 Data gaps and recommendations for future work ...... 87 Annex 1: Supplementary literature search on ecosystem services and biophysical thresholds ...... 89 Annex 2: Analysis of variation of biotic attributes by location and spatial scale ...... 92 Annex 3: Indicators for ecosystem assessment ...... 97 Annex 4: Policies ...... 103 Annex 5: Short reports for each ecosystem service ...... 107 Freshwater fishing ...... 108 Timber production ...... 115 Freshwater provision ...... 122 Food production (cultivated crops) ...... 130 Air quality regulation ...... 136 Atmospheric Regulation ...... 145 Mass flow regulation (erosion protection)...... 157 Water flow regulation ...... 165 Water quality regulation ...... 175 Pollination...... 182 Pest regulation ...... 189 Recreation (species-based) ...... 198 Aesthetic landscapes ...... 206 Annex 6: The OpenNESS Database Tool ...... 215 Technical details ...... 215 Screenshots ...... 215 Domain Model ...... 224 References ...... 225
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 5
Summary
One of the objectives of OpenNESS is to investigate the ways in which natural capital (both biotic and abiotic components) underpins the delivery of ecosystem services. This report (deliverable under Task 3.1 in Work Package 3) approaches this objective by providing scientific evidence for the linkages between natural capital and ecosystem services based on a systematic search of peer-reviewed literature across four provisioning, seven regulation and maintenance, and two cultural ecosystem services. Data from 690 relevant journal articles published in English language was extracted into a spatial database structured according to a simple classification system which enabled analysis of the links between biotic (including biodiversity) and abiotic factors and associated ecosystem service providers for particular ecosystem types and geographical locations. The database also recorded any indicators measuring actual or potential ecosystem service delivery, as well as the impact of human activities and policies. Finally, it considered positive or negative interactions between the ecosystem services as reported in the papers, and the existence of any biophysical thresholds. A number of robust conclusions have emerged based on the analysis of the database results:
1. There is a general lack of literature explicitly addressing the relationship between natural capital attributes and ecosystem services.
2. Most of the biotic attributes identified in the review have a beneficial impact on ecosystem service delivery. Their contribution is related to three different clusters of attributes. The most commonly identified cluster relates to the physical amount of vegetation within an ecosystem, and includes habitat area, vegetation productivity, above- and below-ground biomass, stem density, species size/weight, growth rate, and successional stage. These tend to have beneficial impacts on a particular group of regulating services: atmospheric regulation (carbon storage), water flow regulation (flood protection), mass flow regulation (erosion prevention), water quality regulation (water purification) and air quality regulation. The second cluster focuses on the presence or abundance of particular species or functional groups. This is particularly important for the provision of freshwater fishing, timber, species-based recreation, pollination and pest regulation; a number of species-level traits (such as size or predation behaviour) are important for determining which are the most effective contributors to the ecosystem service. A third cluster, though less commonly discussed, comprises diversity-related indicators: species richness, species population diversity, functional richness, functional diversity, structural complexity and landscape diversity. Diversity is shown to be important for a wide range of services, including timber production, atmospheric regulation, pest regulation and pollination. It contributes to these services in two ways: niche complementarity, where efficiency is maximised because different organisms occupy different ecological niches; and the selection effect, where the presence of a wide range of different species improves the chance that one of them will be a high performer. Both of these mechanisms are shown to be important in different circumstances. Species richness and structural diversity also increase human enjoyment of species-based recreation and landscape aesthetics. There is also some mention of the role of diversity in ensuring resilience to environmental change.
3. Only a few biotic attributes had a negative impact on ecosystem service provision. These were: species mortality rate; the abundance of certain non-native species (e.g. invasive vegetation); the presence of forest plantations which can have a negative impact on freshwater supply; and a
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 6
limited number of examples where diversity had a negative effect because monocultures were found to provide a better service than polycultures.
4. Abiotic factors affect the delivery of ecosystem services, but their impact varies widely depending on the context. We found examples of both positive and negative effects as well as a large number of studies where the direction of the impact is unclear. Climatic factors are particularly relevant abiotic factors as they influence ecosystem dynamics, and exposure to climatic conditions outside a given range can limit the provision of ecosystem services. Temperature, precipitation and evaporation were cited frequently as important climatic factors for the services of freshwater provision, food provision, air quality regulation, mass flow regulation and water flow regulation. In particular, many articles assess how climate change might impair delivery of ecosystem services in the future. Other important abiotic factors include the impact of slope on mass flow regulation and the influence of soil on food production and freshwater provision.
5. Human activities are shown to have a range of positive and negative impacts on ecosystem service delivery, and many studies cite a mix of both. Overall, loss or damage of ecosystems through urban development, or over-exploitation of services (e.g. through intensive agriculture, deforestation or excessive eco-tourism), can alter the functioning of ecosystems and reduce the services they deliver. However, there are also many examples of ways in which protection, restoration and sustainable management of habitats can actively enhance ecosystem service delivery. Careful regulation and sustainable management offer opportunities to minimise the negative impacts of over-exploitation.
6. Ecosystem service delivery depends not only on the factors affecting a particular service, but also on the interactions with other services. It is crucial to assess bundles of services to understand the potential provision of individual services. Interactions between ecosystem services are mentioned in 40% of the articles reviewed, and 56% of these interactions are positive, highlighting the multiple benefits that ecosystems can provide. For example, forests can simultaneously provide many regulating services (atmospheric, water flow, mass flow, water quality) and cultural services (recreation and aesthetic landscapes). However, there are also some important negative interactions between services, especially between provisioning services and regulating or cultural services. These negative interactions are usually linked to human management activities that benefit one service but at the same time have negative impacts on another. Commonly cited examples include fertiliser application, which benefits food and timber production but has negative impacts on water quality regulation and freshwater fishing. Similarly, harvesting forests for timber has a negative impact on atmospheric regulation, water flow regulation and landscape aesthetics. Some management options may have short-term benefits but may result in adverse consequences in the long-term (such as a decline in pollinators and thus increased risks to food security due to intensive farming). Improved analysis of these interactions could help decision-makers to develop management strategies that exploit synergies and balance trade-offs more effectively.
The results of this review are useful in many ways in different venues:
1. Overall, the review emphasises the importance of conserving natural capital in order to continue to deliver robust and resilient ecosystem services in a world with increasing human demands. It also
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 7
provides valuable information on the contribution of different species, habitats and management techniques to the delivery of ecosystem services.
2. Within OpenNESS, this knowledge base can now be used to improve the design and selection of data for the ecosystem service mapping and modelling methods being developed in OpenNESS (Task 3.2), and the accompanying guidance (Task 3.4), both of which will be applied within the case studies (Work Package 5). Ultimately, a refined version of these tools, datasets and guidance will form part of the final OpenNESS contribution to the Oppla web portal (www.oppla.eu) for future application by the wider user community.
3. When focusing on the operationalisation of the concepts of natural capital and ecosystem services, this information base can help local decision-makers to identify opportunities for protecting and enhancing vital ecosystem services in their area, and for maximising positive interactions and minimising negative interactions between services, whilst national and regional policy-makers can design effective policies to facilitate these goals at a wider scale. Careful policy design is urgently needed to balance trade-offs between provisioning and regulating or cultural services that are linked to human decisions, and to protect ecosystems against future climate change and urban development. To preserve a balanced mix of ecosystem services as well as a healthy underlying level of biodiversity to sustain future services, it will be necessary to protect certain key habitats such as wetlands and old-growth forests from over-exploitation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 8
1 Introduction
Natural capital encompasses the elements of nature that directly or indirectly produce value for people, including ecosystems, species, freshwater, land, minerals, air and oceans, as well as natural processes and functions. Despite the importance of protecting the natural capital and ecosystem services on which we all depend, there is no consistent framework for including these concepts in policy-making. The OpenNESS project aims to find ways of integrating these concepts into land, water and urban management in Europe.
As part of that goal, OpenNESS (Work Package 3) aims to develop approaches for mapping and modelling the biophysical structure and associated ecological functions underpinning the delivery of ecosystem services, such as the way in which forests or grassland contribute to flood protection or carbon storage. This knowledge base can then be used to assess the effectiveness of different policy mechanisms and management practices for sustaining ecosystem service delivery whilst conserving biodiversity.
This will be achieved through the following sub-objectives: To analyse the contribution of natural capital stocks to ecosystem service flows by identifying the structural and functional factors that link them in different contexts (Task 3.1); To develop or improve a range of spatially-explicit methods for investigating the effects of multiple drivers on ecosystem services supply (Task 3.2); To develop methods for comparing ecosystem service supply with biodiversity conservation objectives to inform sustainable management practices and test the effectiveness of financial and governance instruments for conserving both ecosystem services and biodiversity (Task 3.3); and To create a set of guidelines for application of these methods within the OpenNESS case studies (Task 3.4).
This deliverable presents the method and results related to the first sub-objective (Task 3.1) on analysing the contribution of natural capital stocks to ecosystem service flows. The starting point of this analysis is a comprehensive literature review, based on peer-reviewed scientific articles published in English, and building on the database compiled by the BESAFE project (Biodiversity and Ecosystem Services: Arguments for our Future Environment1), which was focused on the linkages between biodiversity and ecosystem services. The results of this review are then presented in a series of network diagrams, tables and graphs, using a simplified topology that allows easy visualisation of the links between natural capital and ecosystem services. The findings from the review will be used to provide guidance to the case studies on the biotic and abiotic attributes which support the services potentially of interest within their case (Task 3.4).
In the next section, we summarise the current state of knowledge on the linkages between natural capital and ecosystem services, and describe how OpenNESS aims to extend that knowledge. We then present the methodology and results of our analysis, discuss the results and summarise our conclusions. At the end of the report are six Annexes providing full details of the information that has been summarised in the main document, and finally the reference section listing all the literature mentioned.
1 http://www.besafe-project.net/
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 9
1.1 Links between natural capital and ecosystem services: current knowledge
Ecosystems provide three types of service according to the CICES classification (Haines-Young and Potschin, 2013): provisioning (e.g. food), regulation and maintenance (e.g. water quality regulation, atmospheric regulation) and cultural (e.g. recreation). The importance of these services for human well-being is well known (e.g. Butler and Oluoch-Kosura, 2006; Costanza et al., 1997; Daily, 1997; de Groot et al., 2002). However, for most ecosystems, our understanding of the current status of the services they provide is poor (Millennium Ecosystem Assessment (MA), 2005; Harrison et al., 2010; UK NEA, 2011).
The role of biodiversity in underpinning the delivery of ecosystem services is well recognised (Diaz et al., 2006; MA, 2005), and our understanding of this relationship is increasing. Early work explored the contribution of habitats (Chan et al., 2006), individual species, and species-species and species- environment interactions (Balvanera et al., 2005) to ecosystem services. Recently there has been more emphasis on the role of functional diversity, which may be more significant than species richness in determining the properties of an ecosystem. Functional traits are properties of species, such as size, shape or behaviour, which determine their impact on ecosystem processes and their response to environmental change (De Bello et al., 2010; Diaz and Cabido, 2001). Analysis of these functional traits has been useful in identifying the way in which particular species contribute to ecosystem processes and services (e.g. Balvanera et al., 2006; Diaz et al., 2011; Fagan, et al., 2008; Gaston, 2000; Hooper et al., 2005; Lavorel and Grigulis, 2012; Lavorel, 2013; Luck et al., 2012).
As a result of these advances in research, it is now recognised that Ecosystem Service Providers can either be single species (such as a specialist pollinator), a functional group (such as parasitoid wasps for pest control) or an entire habitat (such as a forest for carbon storage) (Kremen, 2005; Luck et al., 2009). Building on this concept, a number of studies have started to address the crucial question of how loss of biodiversity will affect future provision of ecosystem services (Cardinale et al. 2006, 2012).
Despite these advances in research, our understanding of the role of biodiversity in ecosystem service provision is still limited (Balvanera et al., 2014; Mace et al., 2012; Schröter et al., 2014). Few studies cover a wide range of biotic attributes and ecosystem services, and very few use empirical evidence to investigate the role of biodiversity in providing ecosystem services (Mertz et al., 2007, Seppelt et al., 2011). As a result, the quantitative relationships between different aspects of biodiversity and ecosystem services are still poorly understood (Carpenter et al., 2009; Luck et al., 2009; de Bello et al., 2010, de Groot et al., 2010). There is even less evidence on the role of abiotic factors, and how they interact with biotic attributes to deliver different ecosystem services (Harrison et al., 2014).
In particular, we need to understand the possible effects of natural capital loss on the delivery of ecosystem services. This is important in view of the possible existence of thresholds or tipping points at which natural capital loss severely compromises ecosystem service delivery. These have been demonstrated for some situations mainly focused on biodiversity variables (e.g. Scheffer, 2009), but we lack an understanding of how they might vary for different functions and services (Groffman et al., 2006) and under different abiotic conditions. Further evidence is essential so that services, and their underpinning natural capital, can be managed and monitored. A stronger evidence base could also support arguments for ecological restoration
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 10
(Rey Benayas et al., 2009; Bullock et al., 2011), and contribute to the management of protected and restored areas (Bastian, 2013) in order to meet the dual goal of conserving natural capital and biodiversity whilst optimising the delivery of ecosystem services (Palomo et al., 2014).
OpenNESS addresses this challenge and advances the state-of-the-art by evaluating, consolidating and expanding existing datasets on natural capital stocks (relevant biotic attributes and abiotic factors) and ecosystem service flows, including information on ecological thresholds. The work builds on the database generated by the BESAFE project (Harrison et al., 2014), which captures many journal articles from previous EU-funded projects, such as the FP6 RUBICODE project, to analyse linkages between biodiversity attributes and 11 ecosystem services. In OpenNESS, we have extended the conceptual framework used by BESAFE to enable analysis of additional factors including:
Geographical co-ordinates of the locations of the reported studies, to enable spatial analysis and mapping; Direction of relationships between abiotic factors, as well as biotic attributes, and ecosystem services; Interactions (synergies and trade-offs) between different ecosystem services; The impact of human interventions and policies; and Evidence for thresholds or tipping points where further natural capital loss would severely compromise ecosystem functioning and service delivery.
Using the database, we have analysed the relationships between natural capital and ecosystem services, applying a typology (or operational classification) that can reduce complexity and thereby help to put the ecosystem services concept into practice. This will be used to guide the development and testing of a set of methods for mapping and modelling spatially-explicit ecosystem services supply in further OpenNESS work (Task 3.2).
2 Methods
The OpenNESS analysis aimed to complement and supplement previous work by activities such as TEEB, the FP6 RUBICODE project and the FP7 BESAFE project. The starting point for the work was therefore a survey of all relevant completed and ongoing projects (including EU funded projects), databases and expert knowledge from the OpenNESS scientific community. This determined that the most efficient approach, to avoid duplicating previous work, was to extend and update the BESAFE database, which holds data on the relationship between biodiversity attributes and ecosystem services from approximately 50 peer-reviewed scientific journal articles for each of 11 ecosystem services, covering the period up to late 2012. We extended this work in eight ways:
1. Two additional ecosystem services were added - air quality regulation and food production (cultivated crops) - to address the main ecosystem services identified as important for the OpenNESS case studies. This brought the total number of services covered to 13. 2. Further biotic components were added based on the gaps identified in the BESAFE review. 3. The direction of relationships between abiotic components and ecosystem services was recorded to fully capture linkages between natural capital and service delivery. 4. Positive and negative interactions between ecosystem services were recorded.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 11
5. New information on thresholds or tipping points, indicators, the impact of human activity and policy were extracted from the articles. 6. A new database tool was designed, including these additional factors to enable more in-depth analysis. 7. Records were transferred from the BESAFE database and each article was re-analysed to extract data for the new fields. The content was updated by adding at least ten recent articles for most of the ecosystem services, to cover the period from late 2012 to mid-2014. 8. The database was extensively analysed, including the creation of network diagrams maps and summaries for each ecosystem service (in Annex 5).
Figure 1 shows the inter-linkages between the basic elements of our analysis using the ecosystem service cascade framework (De Groot et al., 2010; Potschin and Haines-Young, 2011), as adapted by Braat (2015).
Figure 1: Inter-linkages between the basic elements of our analysis of the literature review, using the ecosystem service cascade conceptual framework. We focused on the Ecosystem and the underpinned ecosystem services (framed by the blue box), considering the policies and management options developed by the Human System as responses to the benefits received from the ecosystem services and their social and economic value. Within the Ecosystem, we analysed the Natural Capital (biotic attributes and abiotic factors) where ecosystem service providers potentially deliver a bundle of ecosystem services that become actual services if they are used by humans. We also assessed the interactions between ecosystem services.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 12
In the following sections, we describe the methodology in more detail.
2.1 Selection of ecosystem services
Ecosystem services were selected to match those relevant to the OpenNESS case studies. It was not possible to cover all ecosystem services within the time and resources available, but 13 services were selected, covering 87% of those suggested as important for the case studies, excluding biodiversity which is considered as part of natural capital. Eleven of these services were already covered in the BESAFE database, and most of these journal articles were transferred to the new OpenNESS database tool. For the two additional services, a new literature search was carried out.
In total the report spans four provisioning services (freshwater fishing, timber production, freshwater supply, food production); seven regulation and maintenance services (air quality regulation, atmospheric regulation (carbon sequestration), mass flow regulation (erosion protection), water quality regulation (water purification), water flow regulation (flood protection), pollination and pest regulation); and two cultural services (species-based recreation and aesthetic landscapes). These are illustrated in Figure 2, which distinguishes the contribution of studies from OpenNESS and BESAFE to this review. Table 1 shows how these services relate to the CICES and Millennium Ecosystem Assessment classification schemes. Services relevant to the case studies that are not covered by the literature review are: bioenergy production (relevant to 4 case studies), cultural heritage (3), education (2), livestock (1) and local climate regulation (1). The category of “regulation and maintenance services” is sometimes abbreviated to “regulating services” in this report.
70
60 50 40 30 20
Number of studies 10 0
Figure 2: The number of studies included in this review per ecosystem service. This spans provisioning services (red), regulation and maintenance services (blue) and cultural services (green). Studies from the BESAFE review are indicated by the darker shades and new studies added during OpenNESS by the lighter shades. An article may include more than one study (see section 2.2).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 13
Table 1: Links between the ecosystem service classes analysed in the literature review and the Millennium Ecosystem Assessment (MA) and CICES classifications.
Ecosystem Service MA CICES group level Note Freshwater fishing Food (fodder) Biomass [Nutrition] Timber production Fibre, timber Biomass (fibres and other
materials from plants, algae and animals for direct use and processing) Freshwater provision (quantity) Freshwater Water (for drinking CICES also has Water purposes) [Nutrition] (for non-drinking purposes) Provisioning services [Materials] Food production (crops) Food (fodder) Biomass [Nutrition] Air quality regulation Air quality Dilution & sequestration regulation [Filtration of particulates and aerosols] Atmospheric regulation Climate Atmospheric composition (carbon sequestration) regulation and climate regulation
Mass flow regulation Erosion [Mediation of] mass flows Has some overlap (erosion protection) regulation with water regulation Water flow regulation Water [Mediation of] liquid flows TEEB splits into (flood protection) regulation Mediation of liquid flows and mediation of mass flows Water quality regulation Water Mediation [of waste, toxics Under this category (water purification) purification and and other nuisances] CICES also has water by biota and Mediation [of Mediation of treatment waste, toxics and other gaseous/air flows
Regulation and maintenanceservices nuisances] by ecosystems Pollination Pollination Lifecycle maintenance, habitat and gene pool protection Pest regulation Biological Pest and disease control control [Biological control mechanisms] Recreation (species-based) Recreation and Physical and experiential ecotourism interactions Aesthetic landscapes Aesthetic values Intellectual and
Cultural services representational interactions
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 14
2.2 Development of OpenNESS database tool
A new database tool was developed to summarise the results of the literature review. The aim was to capture the contribution that both biotic and abiotic aspects of natural capital make to the delivery of specified ecosystem services at different spatial and temporal scales. It is intended that the database will form part of ‘Oppla’, a shared web-platform with our twin project OPERAs.
The database includes the following information: the ecosystem service; the reference; the geographical co-ordinates of the location of the study (as a place name), to enable the spatial analysis and mapping of ecosystem services; interactions between different ecosystem services; the spatial scale; the temporal scale; type and condition of ecosystems, including whether they are actively managed; the ecosystem service provider (ESP); the important biotic attributes of the ESP, and direction of relationship; presence of abiotic factors which affect service delivery and their direction of influence; human input and management; indicators used to assess and quantify the ecosystem service; evidence for thresholds or tipping points; policies mentioned by the article; a rating of the overall strength of the evidence.
Further details on each of these data categories are presented below. Data were entered into a software tool consisting of a WinForms user interface linked to a Microsoft Access database. Full details of the tool, including screenshots, are presented in Annex 6. The main screen of the tool and the spatial location entry screen are shown in Figure 3 below.
Some journal articles were split into separate studies, which were entered separately into the database. This was done where articles addressed more than one ecosystem, location or ecosystem service provider, and the conclusions were significantly different in each case. Thus each journal article can have more than one database entry. For consistency, the individual database entries are referred to as “studies” throughout this document. The database contains 709 studies in total, derived from 690 journal articles. It is also possible for each database entry to have more than one study location – these are referred to as “study sites”. There are 903 study sites entered into the database.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 15
Figure 3: The main screen and the spatial location entry screen of the OpenNESS database tool.
2.2.1 Ecosystem service This records which of the 13 ecosystem services is being reviewed. The selection of the ecosystem services to be included in this review is discussed in Section 2.1.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 16
2.2.2 Reference This records the reviewer’s name and institution, and the article title, source, authors, year of publication and DOI. The DOI will be used for automatic linking to online resources through the Oppla web platform.
2.2.3 Interactions with other ecosystem services This stores the types of relationship between the main ecosystem service and any other ecosystem services that are mentioned in the article. The relationship can be positive, negative, both (positive and negative) or unclear (the article mentions the service being affected, but does not give an indication of the direction). Free text boxes are provided to add additional information on the nature of the relationship and to capture any quantitative information.
2.2.4 Spatial scales and locations: This records the spatial scale and location of the study. The scale is selected from a drop-down list of the six following options: Local (single land parcel, farm or sub-catchment); Sub-national (anything in between local and national, but not including them); National; Sub-continental (anything in between national and continental, but not including them); Continental (Asia, Africa, North America, South America, Antarctica, Australia or Europe); Global.
Location is identified using a search box in which the user can enter either a place name or geographical co- ordinates (see Figure 3). A pop-up map window (powered by a Microsoft application using Bing maps) then shows the location. The geographical location is registered as the bounding box shown in the map window. If the study covers more than one location or study site, additional locations can be added.
2.2.5 Temporal scale If the article mentions the temporal scale of the study (i.e. the period of observations, experiments or model simulations) it can be entered here as either: Snapshot; Seasonal; Annual; Long-term; Scenario analysis.
If information is available on the start and end of the period of observation, experiment or simulation, it can be entered as day, month and year.
2.2.6 Ecosystem types and conditions Any ecosystems mentioned in the article are recorded. There is a choice of 12 ecosystem types, based on level 2 of the ecosystem typology used in the MAES (Mapping and Assessment of Ecosystem and their
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 17
Services) Working Group2. This corresponds directly with the EUNIS habitat classification and SEBI 04 indicator on ecosystem coverage. It is relevant for EU policies and compatible with global ecosystem classifications. It is typological (enabling comparison between different parts of the European territory), keeps a (pan) European scale and takes into consideration regular mapping aspects (applying CORINE Land Cover (CLC) data for spatial delineation). Table 2 provides details of the ecosystem typology.
Table 2: Details of the ecosystem typology used in the database.
Major ecosystem category Ecosystem type for mapping and assessment (level 1) (level 2) Terrestrial Urban Cropland Grassland Woodland and forest Heathland and shrub Sparsely vegetated land Wetlands Fresh water Rivers and lakes Marine Marine inlets and transitional waters Coastal Shelf Open ocean
In addition, the condition of all ecosystems mentioned in the article can be recorded (if known) using a classification modified from the English Nature Special Site of Scientific Interest (SSSI) classification scheme: Condition not mentioned ( the ecosystem is mentioned, but no information is supplied about its ecological condition); Favourable (the ecosystem in question is being adequately conserved and is meeting its 'conservation objectives'. There may still be scope for enhancement, but there is no or very little ecological concern for this ecosystem); Unfavourable - Recovering (the ecosystem is not yet fully conserved, but there is evidence within the article that the ecological status is improving); Unfavourable - No Change (the ecosystem is not adequately conserved and the evidence within the article states that the ecological status is not changing); Unfavourable - Declining (the ecosystem is not adequately conserved and the evidence within the article suggests that the ecosystem condition is becoming progressively worse); Unfavourable - No Evidence (the ecosystem is not adequately conserved, but there is no evidence within the article as to the trajectory of change); Destroyed (the ecosystem existed at one point, but has been destroyed to the point that it would need to be re-created. The area is now considered as a different ecosystem type. This class includes both the Destroyed and Part-Destroyed classes of English Nature).
2 An Analytical framework for ecosystem assessments under Action 5 of the EU Biodiversity Strategy to 2020, Discussion paper, Final, April 2013, European Union, 2013 ISBN 978-92-79-29369-6 DOI 10.2779/12398
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 18
There is also a “Conservation Management” box which can be ticked if the article reports that any conservation action is either planned or taking place with respect to the ecosystem in question. Free text boxes are provided for additional information on the type (e.g. coniferous forest, extensive grassland for livestock grazing) and condition (e.g. any specific evidence to support the class selected) of each ecosystem.
2.2.7 Ecosystem Service Provider (ESP) classes The broad ecosystem services provider class is selected from a drop-down menu. ESPs are defined as the populations, functional groups or communities that contribute to service provision. The options are:
Single specific species population (A group of organisms, all of the same species, which occupies a particular area (geographic population), is genetically distinct (genetic population) or fluctuates synchronously (demographic population)); Two or more specific species populations; Single functional group (A collection of organisms with similar functional trait attributes in the study area, i.e. a feature of an organism, which has demonstrable links to the organism’s function. Sometimes referred to as a guild, especially when referring to animals); Two or more functional groups; Entire community or habitat (An association of interacting populations, usually defined by the nature of their interactions or by the place in which they live); Two or more communities or habitats; Dominant community (defined either qualitatively or quantitatively based on the article).
A free text box is provided to add more specific information on the ESP.
2.2.8 Biotic attributes/traits which affect ecosystem service delivery This holds information about all the biotic attributes and traits mentioned in the article that are important for the provision of the ecosystem service. Together with the list of abiotic factors (described in section 2.2.9 below), this describes the natural capital.
There are two descriptors for each attribute: Direction of the relationship: not mentioned, positive, negative, both (positive and negative) and unclear (the article mentions the relationship, but does not clearly indicate the direction). Free text box to add information on specific qualitative or quantitative information from the article.
SPECIES ATTRIBUTES: Presence of a specific species type (name of the species can be added in the free text box) Species abundance (number of individuals of a species expressed per unit area or volume of space. Synonymous with species population density) Species richness (number of different species represented in a set or collection of individuals) Species population diversity (the number, size, density, distribution and genetic variability of populations of a given species) Species size or weight (includes body size or weight, diameter at breast height – DBH – for trees, species/vegetation/tree height, basal area defined as the cross section area of the stem or stems of
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 19
a plant or of all plants in a stand, generally expressed as square units per unit area) (free text box can specify the type of measurement) Population growth rate (change in the number of individuals of a species in a population over time) Mortality rate (number of deaths of individuals per unit time) Natality rate (number of new individuals produced per unit time) Life span/longevity (duration of existence of an individual/expected average life span)
FUNCTIONAL GROUP ATTRIBUTES: Presence of a specific functional group type (the name of the functional group(s) can be recorded in the free text box) Abundance of a specific functional group Functional richness (the number of functional groups or trait attributes in the community) Functional diversity (range, actual values and relative abundance of functional trait attributes in a given community) Flower-visiting behavioural traits well suited to the system to provide pollination ecosystem services (free text box allows the behavioural type/preference/strategy to be entered) Predator behavioural traits well suited to the system to provide biocontrol ecosystem services (free text box allows the behavioural type/preference/strategy to be entered)
COMMUNITY/HABITAT ATTRIBUTES: Presence of a specific community/habitat type (the name of the habitat(s) or ecosystem(s) can be entered in the free text box) Community/habitat area (includes width or diameter, i.e. for buffer zones) Community/habitat structure (in terms of complexity - amount of structure or variation attributable to absolute abundance of individual structural component - and heterogeneity - kinds of structure or variation attributable to the relative abundance of different structural components) Primary productivity (rate at which plants and other photosynthetic organisms produce organic compounds in an ecosystem) Aboveground biomass (the total mass of aboveground living matter within a given area) Belowground biomass (the total mass of belowground living matter within a given area) Sapwood amount (including allocation of carbon to sapwood and sapwood area) Stem density (measured as the number of stems/specified area) Wood density (measured as the weight of a given volume of wood that has been air-dried) Successional stage (changes in the number of individuals of each species of a community by establishment of new species populations that may gradually replace the original inhabitants; categorised into early and late stages) Habitat/community/stand age (includes young and old-growth forests, even and uneven-aged forests, or can specify the age) Litter/crop residue quality (quality of plant litter with respect to decomposition: often defined by the C:N ratio, but ratios of C, N, lignin and polyphenols are other chemical properties and particle size and surface area to mass characteristics are physical properties) Leaf N content
OTHER ATTRIBUTES: Landscape diversity (diversity of landscapes and landscape features)
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 20
Other (if an attribute or trait is mentioned that is not in this list, it can be added here and described in the free text box)
2.2.9 Abiotic factors which affect ecosystem service delivery This allows the user to record information about all the abiotic factors mentioned in the article that are important for provision of the ecosystem service. The intention was to record only cases where an abiotic factor affects the ability of the ecosystem to deliver a service. For example, it is obvious that heavy rainfall can cause flooding, but this observation is not relevant for ecosystem service delivery. However, it is relevant to note whether the level of rainfall affects the ability of a forest or wetland to prevent flooding.
Similarly to the biotic attributes, the direction of the relationship (positive, negative, both or unclear) is recorded and there is a free text box for additional information. Abiotic factors are selected from the following list: Temperature Precipitation Evaporation Wind Snow Soil Geology Water availability Water quality Nutrient availability (soil minerals) Slope (angle, aspect) Other (if an abiotic factor is mentioned that is not in this list, it can be added here and described in the free text box)
2.2.10 Human input and management This records whether the article explicitly mentions any human inputs (in terms of either management or resource application) to the ecosystem, and whether these inputs are applied directly to the ecosystem or indirectly from neighbouring areas. A free text box is provided to give more specific details. Options include: Not mentioned (human management or input of resources are not mentioned in the article.) Direct human input (human inputs are explicitly mentioned in the article in terms of direct application of management or resources to the ecosystem. Also, there are no indirect human inputs mentioned.) Indirect human input (direct human inputs to the ecosystem are not mentioned in the article. However, management or resource application in neighbouring ecosystems/contexts is stated to be having an impact.) Both direct and indirect input (both direct and indirect inputs are explicitly mentioned in the article.) Unclear (human input is mentioned in the article, but it is unclear if it is direct or indirect.)
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 21
The intensity of human input or management is also recorded using the following options: None Intensive (if there is a significant use of capital and inputs relative to land area. To be classified as “Intensive” the article will mention that the main goal of the management is maximising production.) Extensive (if there is a low use of capital and inputs relative to land area. To be classified as “extensive” the article will mention that the land is not being managed to its utmost productive potential.) Intensive and extensive (if there is evidence of both intensive and extensive management within the article.) Unclear (if management is mentioned, but there is no indication as to the form this takes so that it is impossible to determine if intensive or extensive management is taking place.)
A free text box is provided to give more specific information on the intensity of human input or management.
Finally, there is a field to record the impact (in terms of direction) of management on the main ecosystem service and other ecosystem services mentioned in the article. A free text box is provided to give more specific details. Options include: Not mentioned Positive Negative Both (positive and negative) Unclear
2.2.11 Indicators Information on any indicators used for assessing the main ecosystem service can be recorded. A list of potential indicators for each service is generated, based on the JRC report ‘Indicators for mapping ecosystem services: a review’ (Egoh et al., 2012) (see Annex 3). Other indicators not included in this list can also be added. For each indicator, a check box is ticked if it is used in the article and the following information is recorded: • The assessment method used to calculate the indicator: o Primary data (the indicator is based on data from direct field observations or from public statistical databases) o Mechanistic model (the indicator is generated using a biophysical model of the ecosystem) o Correlative model (the indicator is estimated using a statistical relationship between primary data and environmental data) o Conceptual model (the indicator is based on an artificially defined rating or scoring system which reflects expert opinion) • Units or dimensions of the indicator, in a free text box • Information on any thresholds that are identified for the indicator, in a free text box • A comment box is provided for any other specific information
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 22
2.2.12 Thresholds This section records any information on thresholds that has not already been entered in the previous section for a particular indicator. There are three options: Other biophysical thresholds: biophysical thresholds not already included in the “Indicators” screen. These are defined as thresholds in ecological structures and/or functions that support ecosystem services, e.g. bumblebee density (nr/ha) or farmland bird richness (nr species/ha). Safe minimum standards: a scientifically grounded limit to avoid the risk of reaching a threshold, although not necessarily legally binding, e.g. maximum concentration of air pollutants recommended by WHO. Legal boundaries: a legally binding threshold, e.g. prescriptive limits on air pollutants.
Free text boxes are provided for each type of threshold to add any descriptive and quantitative information provided in the article.
2.2.13 Policies This section records any policies mentioned in the article. There is a list of 24 EU policies with check boxes and free text boxes to provide information on how the article refers to each relevant policy. These policies are selected from a list compiled within OpenNESS (Work Package 2). We tried to cover all the relevant sectors and at the same time keep the number manageable. More than one policy can be selected using check boxes: Biodiversity Strategy Green Infrastructure Strategy Birds Directive Habitat Directive Ambient Air Quality Directive Common Agriculture Policy Nitrates Directive Biocides Directive Plant Protection Products Regulation Common Fishery Policy (CFP) Rural Development Policy (2007-2013) Regulation on support for rural development by the European Agricultural Fund for Rural Development (EAFRD) Cohesion Policy (2014-2020) Water Framework Directive (WFD) Marine Strategy Framework Directive (MSFD) EU Floods Directive Renewable Energy Directive Forest Action Plan Thematic Strategy on the Urban Environment Environmental Impact Assessment Strategic Environmental Assessment EU Adaption Strategy
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 23
Soil Thematic Strategy (2007-2011) Thematic Strategy on the Sustainable Use of Natural Resources Other (details entered in free text box, including national policies).
2.2.14 Evidence rating Four questions are posed on the quality of the evidence within the article: Is the evidence qualitative, quantitative or both? Is the evidence based on a single or multiple observations? Is the evidence direct or indirect (i.e. through a surrogate)? Is the evidence based on empirical data, modelled information, or both?
A free text box is provided next to each question to explain the selection from the drop-down list.
2.2.15 Comments Finally, there is a free text box for any additional comments that are important, but which are not covered by other parts of the database.
2.3 Literature search and data entry
Following development of the data entry tool, the existing studies from the BESAFE database (around 30 to 50 studies for each service) were transferred into the OpenNESS database. This involved re-analysing each article in order to extract information on the role that natural capital (rather than only biodiversity) had on ecosystem services, as well as the new data attributes mentioned in the previous section.
For the two new ecosystem services - air quality regulation and food production – a new literature search had to be carried out. This consisted of three stages: (i) generation of keywords; (ii) a systematic search; using Web of Science; and (iii) extraction of the data into the database.
For the other services, as the BESAFE database only covered the period up to mid-2012, the search was updated to identify at least 10 new articles for each service, covering the period from mid-2012 to mid- 2014.
The search conformed to the methodology developed during the BESAFE project (Harrison et al., 2014). A pilot test showed that ‘ecosystem services’ is a relatively new term and, hence, only using this term in a literature search is likely to miss relevant articles. Thus, keywords specific to each ecosystem service were selected (e.g. “carbon storage”), accompanied by appropriate biodiversity terms which could be related to the given ecosystem service (e.g. “species richness”). Additional service-related terms were used if necessary to refine results when large numbers of articles were found for the initial search terms (e.g. “tree”, “forest”). The full list of search terms is presented in the short reports for each ecosystem service (Annex 5).
The objective was to find 50 relevant studies for each of the two new services, and at least 10 new studies for each of the existing services. Where the number of relevant results using the above methodology was too few, additional intelligent search approaches were utilised. These included: (i) searching the reference
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 24 lists of relevant articles for secondary references which may be of interest (termed snowballing); and (ii) searching articles that have cited the relevant articles (termed reverse snowballing).
The standard search did not identify many examples of thresholds above which ecosystem services deteriorate. Therefore an additional literature search was carried out using the terms “ecosystem AND service AND threshold”. The articles found were not entered into the database, but the supplementary review was used to inform the analysis of thresholds presented in the results section. Further details of this search and its results are given in Annex 1.
2.4 Network diagrams
Network diagrams were created for each ecosystem service, to explore the linkages between the biotic attributes and abiotic factors (natural capital) and the ecosystem service providers. Although the literature review distinguished between single and multiple ecosystem service providers (i.e. single species, multiple species; single functional group, multiple functional group; etc.) the network diagrams used a simpler typology based on just species, functional group or community/habitat. This was done to enable easier visual interpretation of the diagrams in order to create an operational classification for each service, and also because the analysis showed that the distinction between single and multiple ESPs did not affect the interpretation of the results. This is because the number of ESPs mentioned in an article is generally a function of the study design. Studies with multiple ESPs are usually comparing several species or habitats against each other, or simply choosing an arbitrary number of species to include in the experiment. The distinction between single and multiple ESPs does not therefore reflect the number of species or habitats required to provide the service.
The network diagrams demonstrate both the predominant direction of evidence for the relationship, which determines the colour of the lines, and the strength of the relationship (the proportion of studies for that ecosystem service which show that particular link), which determines the thickness of the lines. The direction of the relationship was calculated as shown in Table 3.
Table 3: Methodology to determine predominant direction of evidence for network diagrams.
Number of articles showing links that are: Positive- Predominant Colour of Positive Both positive Negative Unclear Negative direction line and negative ≥1 0 0 ≥0 ≥1 Positive Dark green ≥1 ≥1 ≥0 ≥1 Mostly positive Light green Positive-negative=0 AND/OR Unclear≥1 0 Neutral Grey ≥1 ≥1 ≥0 ≤1 Mostly negative Light red 0 0 ≥1 ≥0 ≤1 Negative Red 0 0 0 0 0 No relationship Not shown
The network diagrams were then created using the Pajek software (http://pajek.imfm.si/doku.php). Two examples are presented in the overview and synthesis of results (section 3), and the diagrams for each ecosystem service are presented in section 4.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 25
3 Overview and synthesis of results
This section presents an overview and synthesis of the results. Section 4 then presents separate summaries for each of the 13 ecosystem services.
3.1 Location and spatial scale
The majority of the studies focus on sites in Europe (33%) or North America (26%), with 14% in Asia and considerably fewer from Africa, Oceania and South America (Figure 4)Error! Reference source not found.. This may simply reflect the higher amount of research activity in Europe and North America, but could also be a consequence of the review focusing on journal articles written in English.
Location of studies reviewed
6% 7% 9% Africa 14% 5% Asia Europe North America South America 26% Oceania 33% Global
Figure 4: Location of the studies reviewed (% of study sites). Note that a study may cover more than one study site.
Figure 5Error! Reference source not found. shows how the geographical distribution of the studies varies for the different ecosystem services. Note that some studies cover a variety of locations, so that there are 903 study locations across the 709 studies entered into the database. The bias towards Europe and North America is particularly apparent for the service of air quality regulation, where 85% of study sites are in these locations, as well as timber production and aesthetic landscapes (73% of study sites). This probably reflects the higher air quality standards, the commercial importance of timber production, and the greater resources available for research in those regions. The only services where Europe and North America together account for less than 50% of the study sites are freshwater fishing, where 32% of study sites are in Asia (where aquaculture is an important industry); freshwater provision, where 42% of study sites are in Oceania (reflecting the severity of water shortages in Australia), and recreation, where the number of studies in Africa, Asia and Oceania indicate the importance of eco-tourism in these regions.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 26
Freshwater fishing Timber production Freshwater (quantity) Food production (crops) Air quality regulation Africa Atmospheric regulation Asia Mass flow regulation Europe Pollination North America Pest regulation South America Water flow regulation Oceania Water quality regulation Global Recreation (species-based) Aesthetic landscapes
0 20 40 60 80 100 120 Number of study sites
Figure 5: Location of studies by ecosystem service. Note that a study may have more than one study site.
The majority of articles reviewed focus on smaller scales, as shown in Figure 6. Most of the studies (385, i.e. 54% of the total) are at the local scale, covering either single or multiple local study sites, and 31% are at the sub-national scale. In contrast, just 6% of the studies are undertaken at a national scale, 1% at a continental scale and 7% at the global scale.
450 Multiple sites 400
350 43 Single site 300 250 200 31 342 150
Number of studies 100 185 50 2 41 3 10 50 0
Figure 6: Number of studies at each spatial scale across all ecosystem services.
The tendency to focus on smaller scales is particularly true for mass flow, air quality and pest regulation (where over 70% of studies are at the local scale), and food and timber production (50-60% are at the local
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 27 scale with the remainder being mainly at the sub-national scale). This reflects the typical format of the studies, which tend to be experimental investigations at particular sites, or occasionally (e.g. for pest regulation) laboratory studies. This may be because it can be difficult to measure ecosystem service flows at larger spatial scales. The major exception is for aesthetic landscapes, where over 80% of the studies are at the sub-national scale and only 2% are at the local scale, reflecting the focus on evaluating landscapes at a broader spatial scale (e.g. through surveys of visitor preferences in national parks or protected areas, or by questioning people from a range of towns or cities).
Most of the global studies are reviews, though there are some meta-analyses of global datasets (e.g. for the relationship between flood damage and forested area, Bradshaw et al. 2007). The lack of studies on a global scale may be simply due to the lower frequency of global studies in the published literature. However, it is probably also related to the decision to exclude articles flagged as “review” during the initial literature search (because the focus was on original studies), though many review-type articles were not screened out by this process.
Results related to the linkages between natural capital and ecosystem services (see Section 3.3) were analysed to see if they varied depending on the location and spatial scale. The analysis was limited by the small number of database entries for the larger spatial scales and for certain continents, and results were therefore not conclusive. Full details are given in Annex 2.
3.2 Ecosystem type and condition
Around 88% of articles refer to specific ecosystems, with 37% studying multiple ecosystems (e.g. comparing woodland and grassland). Figure 7 shows that the most commonly studied ecosystems are terrestrial: woodland and forest (30% of all the ecosystems studied), followed by cropland (20%) and grassland (14%). The least frequently cited ecosystems are those in marine environments, and sparsely vegetated land. This is partly a result of the choice of ecosystem services reviewed (e.g. the inclusion of freshwater fishing but exclusion of sea fishing). However it could also suggest that some ecosystems (e.g. sparsely vegetated land) are relatively unimportant for providing the selected ecosystem services, or that they have been poorly represented in the existing literature (this may be the case for marine ecosystems, see Liquete et al., 2013).
Figure 8 shows the split of ecosystem types by ecosystem service. It is interesting that most of the regulating and cultural services studied cover a wide variety of ecosystems, especially for aesthetic landscapes, recreation and water quality regulation, whereas some provisioning services tend to focus on one dominant ecosystem. There are obvious associations: food production, pollination and pest regulation studies focus on cropland; freshwater fishing studies focus mainly on rivers and lakes; timber production studies refer only to forests; and air quality studies tend to focus on urban areas. The importance of forests is clear: they feature in all 13 categories of ecosystem services, and are the main focus in studies of freshwater provision, atmospheric regulation (carbon storage) and water flow regulation (flood prevention). Wetlands feature strongly in studies of water flow and water quality regulation, and coastal environments are important for species-based recreation, including eco-tourism (e.g. whale and dolphin watching) and recreational fishing.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 28
Ecosystem types Woodland and forest 9% Grassland 3% Heathland and shrub 30% 9% Sparsely vegetated land Cropland Wetlands 8% Rivers and lakes Coastal Marine inlets and transitional waters 14% Shelf 20% Open ocean 5% 2% Urban
Figure 7: Types of ecosystems studied in the literature reviewed (% of total ecosystems mentioned). Note that a study can refer to more than one ecosystem.
Freshwater fishing
Timber production Woodland and forest Freshwater (quantity) Grassland Food production (crops) Heathland and shrub Air quality regulation Sparsely vegetated land Atmospheric regulation Cropland Mass flow regulation Wetlands Water flow regulation Rivers and lakes Water quality regulation
Pollination
Pest regulation
Recreation (species-based)
Aesthetic landscapes
0 50 100 Number of studies
Figure 8: Types of ecosystems studied in the literature reviewed, by ecosystem service. Note that a study can refer to more than one ecosystem.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 29
It is not possible to quantitatively analyse whether the contribution of biotic attributes to the ecosystem service varies by ecosystem, because, for studies where multiple ecosystems are described, attributes were not recorded separately for each ecosystem. Timber production is the only service where no articles reviewed refer to multiple ecosystems, but in this case (unsurprisingly) “woodland and forest” is the only ecosystem discussed and so cross-comparison is not possible. However, detailed information on the contribution of different biotic attributes for each ecosystem is recorded in the comments within the database, and this information has been used to perform a qualitative comparison as part of Section 3.3.2 on biotic attributes.
Information on ecosystem condition is limited, being present in only 34% of the articles reviewed. In fact, information on condition is rarely stated explicitly in the articles, but it can sometimes be inferred, for example if the area is described as being protected, or if the article states that forests have been degraded through logging. It should be noted that this information was recorded to aid interpretation of the impact of ecosystem condition on ecosystem services, and not to evaluate ecosystem conditions in general: that would require a more extensive review. However, some trends can be seen. Food production is associated in most cases with cropland (for obvious reasons), which appears to be in a worsening condition, with seven articles out of 50 citing an unfavourable and declining condition due to factors related to over- exploitation, such as soil erosion or lack of water availability, compared to only one article stating a favourable condition. The negative impact of agricultural activities on the condition of other ecosystems, mainly wetlands (due to conversion for agriculture or pollution from fertilisers), is also noted in several studies. The condition of rivers/lakes used for freshwater fishing also appears to be predominantly declining. For timber production, however, condition is cited in only 18% of studies, but appears to be predominantly favourable (five studies) or unfavourable but recovering (four studies).
3.3 Linkages between natural capital and ecosystem services
This section analyses the links between natural capital and the delivery of ecosystem services. Natural capital consists of both biotic and abiotic components of nature that directly or indirectly produce value for people. Biotic components include plants, animals, fungi, bacteria and dead organic material; abiotic components include water, soil and rock. These components, and the interactions between them, provide ecosystem services. In addition, the condition of the natural capital and the delivery of the services are influenced by abiotic factors such as temperature, wind, evaporation rate, slope and nutrient supply.
In this section we first describe the range of ecosystem service providers identified in the review, which are the species populations, functional groups, communities or habitats (with the latter including both biotic and abiotic components) that contribute to providing a service. We then analyse the contribution of biotic attributes to ecosystem service provision – not just the contribution of individual biotic components such as particular species of plants and animals, but also the diversity of species, and specific characteristics such as species size, community age and animal behaviour. Next we look at the influence of abiotic factors on service delivery. Finally, we combine all components to analyse the direction and strength of evidence linking natural capital with ecosystem services using network diagrams to show the links between ecosystem service providers, biotic attributes and abiotic factors in an operational classification.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 30
3.3.1 Ecosystem Service Providers (ESPs) The most common ESP is the entire community or habitat, which features in 46% of the studies reviewed (Figure 9Error! Reference source not found.). Most of these studies focus on a single community or habitat, though 14% cover multiple habitats – typically comparing a variety of habitats such as woodland and grassland, or primary forest and secondary forest. Specific species populations are the next most commonly cited ESP, accounting for 36% of the articles reviewed, with over half of these studying multiple species. Functional groups (single and multiple) receive less attention, accounting for only 16% of the articles reviewed, although evidence for the importance of functional groups is growing in more recent literature.
It should be noted that Figure 9Error! Reference source not found. does not indicate the relative overall importance of different ESPs, as the ESP is partly a function of the scientific method chosen to investigate the ecosystem service (i.e. whether the researchers choose to investigate the role of one or more species, functional groups or entire communities). Also, the ESP is rarely explicitly identified as such in the journal articles, because much of the literature does not specifically investigate ecosystem services. More often, the ESP was inferred by the reviewer from the information given in the article.
Entire community or habitat
22% Two or more communities or habitats 32% Dominant community
Single functional group
14% Two or more functional groups Single specific species 6% 14% population Two or more specific species 10% 2% populations
Figure 9: A breakdown of the Ecosystem Service Providers (ESPs) studied in the reviewed literature (% of studies). The ESPs for each ecosystem service are shown in Figure 10. It is immediately obvious that for studies investigating provision of food, fish and timber, together with species-based recreation, the ESP is generally one or more specific species. Interestingly, studies of freshwater fishing and timber production focus mainly on two or more specific species (in 63% and 71% of studies respectively), whereas for food production the ESP is more often one specific species such as maize or corn. In contrast, freshwater is predominantly (in 64% of studies) provided by two or more communities/habitats e.g. forest and pasture.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 31
Freshwater fishing Timber production Entire community or habitat Freshwater provision Food production (crops) Two or more communities or habitats Air quality regulation Atmospheric regulation Dominant community Mass flow regulation Water flow regulation Single functional group Water quality regulation Pollination Two or more functional groups Pest regulation Recreation (species) Single specific species population Aesthetic landscapes Two or more specific species 0 20 40 60 80 populations Number of studies
Figure 10: Number of studies showing a linkage between a specific ESP and ecosystem service.
For the regulation and maintenance services, pollination and pest regulation are typically provided by one or more functional groups (such as wasps, bees or pest predators in general), whereas for atmospheric, mass flow, water flow and water quality regulation the ESP is predominantly a single community or habitat such as a forest. For air quality regulation, the ESP is generally categorised as one or more functional groups, e.g. urban trees and/or shrubs, or urban vegetation in general. Finally, for aesthetic landscapes the ESP is almost always one or more entire habitats.
3.3.2 Biotic attributes of natural capital contributing to ecosystem services The literature review found that the selected ecosystem services are influenced by a wide range of biotic attributes (Figure 11). The most commonly cited attribute is community/habitat area (30% of studies). This reflects the large number of studies where the most important attribute is simply the size of the area of the ecosystem, such as lakes for fishing, forests for controlling flood risk, wetlands for purifying water, tree cover for improving air quality, natural habitats for encouraging pest predators, or flower meadows for supporting pollinating insects.
Perhaps surprisingly, the second most important attribute is community/habitat structure (26% of studies). This term covers a number of different aspects of structure, but one of the most commonly cited is structural diversity, which is important in a number of different ways: more diverse habitats provide better food and shelter for pollinating insects and pest predators; structural diversity enhances the aesthetic appeal of landscapes; and structural complexity tends to improve water flow, water quality and atmospheric regulation.
The third most cited attribute is a classic measure of biodiversity: species richness (20% of studies). This is capable of enhancing ecosystem services in two ways: through the ‘selection effect’, where the presence of a greater number of species increases the chances that some of them will be good providers of a particular ecosystem service, and through the ‘niche complementarity’ effect, where a mix of species with different characteristics enables greater use of the available resources. The review identifies several examples of
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 32 niche complementarity, such as greater carbon storage in forests with a range of tree and shrub canopy heights and root depths. Functional diversity and richness are investigated less often, but are found to be important in 6% and 5% of cases, respectively.
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Successional stage Primary productivity Aboveground biomass Belowground biomass Stem density Litter/crop residue quality Landscape diversity Species richness Functional richness Functional diversity Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Population growth rate Life span/longevity Natality rate Mortality rate Wood density Sapwood amount Leaf N content Flower-visiting behavioural traits (pollination) Predator behavioural traits (biocontrol) Other
0 50 100 150 200 250 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 11: Frequency and direction of links between biotic attributes and ecosystem services.
Other important attributes include the abundance of a specific species (18%) and the presence of a specific community/habitat (17%) or species (17%). Additional biotic attributes, not included in the original list, are identified in 11% of studies and are categorised as “Other”. These include a very wide variety of attributes, such as species evenness, leaf-area index (important for air quality regulation) and root density.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 33
In most cases (68%), biotic attributes are found to improve ecosystem services provision, with 8% of studies citing both positive and negative impacts and 15% being unclear. However, in 10% of cases there is a negative influence, particularly (for obvious reasons) for mortality rate, and also for the abundance of non- native species such as commercial honeybees which compete with native bees, introduced fish which predate native fish, or invasive vegetation which chokes rivers and increases flood risk. There are also examples of the reduction of water supply by forest plantations, through interception of rainfall and transpiration, so that forest area, age and stem density may all have a negative impact on freshwater provision.
Table 4 shows how the biotic attributes vary between ecosystem services. The importance of community/habitat area and structure is clear, especially for the regulation and maintenance services, but also for freshwater provision and aesthetic landscapes. The table also shows the importance of species richness for a wide range of services, especially for timber production where 98% of articles report a link: this includes 26 articles showing a positive link, usually where polycultures or mixed forests were more productive than monocultures, and eight with a negative link, typically where competition from other species reduced the productivity of commercial timber species.
Five services show a strong dependence on the abundance of particular species: these are freshwater fishing, timber production, pollination, pest regulation and species-based recreation. For pollination and pest regulation, the abundance and diversity of functional groups (typically groups of pollinating or predatory insects) is also important. For fishing, and to a lesser extent species-based recreation, the size or weight of species is also important.
Atmospheric regulation shows a strong dependence on above- and below-ground biomass (as this is directly correlated with carbon storage), and also with community/stand age and successional stage, because older forests tend to have larger trees and thus store more carbon. Community/stand age is also important for the services of water flow regulation, water quality regulation and freshwater provision, though for freshwater supply ten studies show a negative link with the service and only one link is positive.
For air quality regulation, 52% of links are classified as “other”: this mainly refers to the attribute of leaf area index, which is critical for air quality regulation but which was not included in the pre-defined database list as this was extended from the BESAFE database which did not include this service.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 34
Table 4: Percentage of studies showing a linkage between a specific biotic attribute and ecosystem service. Greatest percentages highlighted in darker shades of grey. Note that food production is omitted from this table because the search did not focus on biotic attributes.
Community attributes Functional attributes Species attributes
Presence of a specificPresence community/habitat type Flower Abundance of a specific functional group functional specific a of Abundance Predator Predator - Presence of a specificPresence group functional visiting (pollination) behavioural traits Presence of a specificPresence species type Community/habitat/stand age behavioural traits (biocontrol) Community/habitat structure Species population diversity population Species Litter/crop residue quality Community/habitat area Population rate growth Aboveground Aboveground Belowground biomass Primary productivityPrimary Functional diversity Functional Landscape diversity Species size/weight Functional richness Functional Life span/longevity Life Species abundance Successional stage Sapwood amount Sapwood Species richness Species Leaf N content Leaf Mortality rate Wood density Wood Stem density Natality rate biomass Other
Freshwater fishing 2 8 4 6 3 2 12 6 27 24 3 2 11
Timber production 3 2 1 1 3 41 2 4 2 8 11 1 1 1 1 4 1
Freshwater provision 1 29 7 19 1 1 1 3 7 1 1 4 1 1 2 3 3
Air quality 2 23 5 1 2 7 1 2 6 2 1 6 3 15 3 4 6 1 2 26
Atmospheric 9 15 17 22 7 11 38 26 6 19 2 13 6 7 15 4 6 13 7 3 1 10 7 8 1 1 4
Mass flow 23 1 7 2 2 1 1 3 2 4 3
Water flow 5 52 21 13 3 1 2 1 5 3 5 4 5 1 4 1
Water quality 28 41 14 8 2 3 6 5 1 4 8 1 1 5 3 12 7 4 9 2 4 1 5
Pollination 11 27 1 7 15 14 9 1 20 15 30 9 3 3 18 10
Pest regulation 32 26 33 1 2 2 6 7 3 14 11 9 14 19 5 17 1 5 4 5 4 5 6 1 15 12
Recreation 3 2 13 1 3 1 23 22 14 12 2 4 4
Aesthetic 15 6 48 1 1 9 4 2 9
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 35
3.3.3 Abiotic factors affecting ecosystem service provision A range of abiotic factors are found to influence ecosystem service provision, with precipitation, soil and temperature being the most frequently cited, but with mixed positive and negative impacts (Figure 12). To take an obvious example, precipitation generally enhances the service of freshwater provision, but may reduce the ability of ecosystems to provide flood protection if the ground becomes saturated. It should be noted that the intention was to record the way in which abiotic factors affect the ability of the ecosystem to provide the service, and not simply the direct impact of the abiotic factor, such as the increased risk of flooding due to heavy rain.
There is considerable variation between ecosystem services (Figure 13, Table 5). Freshwater provision appears to be particularly highly influenced by abiotic factors, with soil, precipitation and evaporation mentioned in over 70% of journal articles. Food production is also highly dependent on abiotic factors including soil, water availability, precipitation and temperature. Air quality regulation is affected by wind, temperature and precipitation. In contrast, pest regulation, species-based recreation and aesthetic landscapes seem to be influenced by a much smaller range of abiotic factors.
Temperature Precipitation Snow Water availability Evaporation Wind Nutrient availability Water quality Soil Geology Slope Other
0 20 40 60 80 100 120 140 160 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 12: Frequency and direction of linkages between abiotic factors and the 13 ecosystem services reviewed.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 36
Freshwater fishing Timber production Precipitation Freshwater (quantity) Snow Food production (crops) Water availability Air quality regulation Water quality Atmospheric regulation Wind Mass flow regulation Evaporation Water flow regulation Temperature Water quality regulation Slope Pollination Geology Pest regulation Soil Recreation (species-based) Nutrient availability Aesthetic landscapes Other 0 50 100 150 200 250 Number of studies
Figure 13: The importance of abiotic factors for each ecosystem service.
Table 5: Percentage of studies showing a linkage between a specific abiotic factor and ecosystem service. Greatest percentages are highlighted in darker shades of grey. Nutrient availability Nutrient Water availability availability Water Water quality quality Water Temperature Temperature Precipitation Evaporation Evaporation Geology Other Other Snow Slope Wind Soil
Freshwater fishing 6 2 2 6 2 20 20 4 12
Timber production 7 7 5 36 2 5
Freshwater provision 49 95 16 11 13 71 5 9 76 16 47 31 Food production (crops) 26 32 2 40 6 10 16 64 34
Air quality regulation 54 14 12 52 24 2 4 34
Atmospheric regulation 15 23 5 2 5 8 28 3 13
Mass flow regulation 30 2 8 2 10 2 18 8
Water flow regulation 7 33 3 7 3 2 18 10 12 10
Water quality regulation 7 3 2 10 10 5 7 15
Pollination 16 8 14 2 2 16
Pest regulation 13 5 2
Recreation (species-based) 8 3 2 3 2 16
Aesthetic landscapes 15 5 3 8 11
Total number of studies 111 139 12 50 50 62 30 30 131 21 59 111
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 37
3.3.4 Network diagrams showing the links between natural capital and ecosystem services The network diagrams summarise the typology (classification scheme) that we have developed to describe the relationships between different components of natural capital and ecosystem services. For each service, they show the main ecosystem service providers and the way in which biotic attributes and abiotic factors contribute to (or detract from) provision of the service. As described in section 2.4, the colour of the links corresponds to the direction of the interaction (positive, negative or neutral/unclear), and the thickness of the lines represents the strength of the evidence for the links (based on the percentage of studies supporting the link).
The diagrams, which are presented and described fully in section 4, highlight the degree of complexity in the interactions between natural capital and ecosystem service provision. The provision of most of the regulation and maintenance services appears to be quite complex, with studies demonstrating the influence of a large spread of biotic and abiotic attributes. Figure 14 shows one example, for the service of atmospheric regulation. In contrast, the attributes influencing some of the cultural and provisioning services, especially the service of aesthetic landscapes, appear to be simpler (Figure 15).Figure 14
Positive Mostly positive Neutral Mostly negative Negative
Figure 14: Network of links to ESPs, biotic attributes and abiotic factors for the service of atmospheric regulation. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 38
Positive Mostly positive Neutral Mostly negative Negative
Figure 15: Network of links to ESPs, biotic attributes and abiotic factors for the service of aesthetic landscapes. AF = abiotic factors. Line thickness indicates the number of studies showing a given link. As discussed in sections 3.3.2 and 3.3.3 above, the diagrams illustrate that biotic attributes tend to enhance ecosystem service provision, shown by the dominance of green lines for biotic attributes, whereas the effect of abiotic factors is variable, depending very much on the ecosystem service and study concerned.
3.4 Human input and management
Over half of the journal articles reviewed mention some form of human management or human impact on the ecosystem service. This covers a wide range of activities: intensive crop production; extensive grazing; management of timber plantations; management of nature reserves and fishing lakes; planting of trees to improve air quality, or grass to stabilise soil; provision of suitable habitat for pest predators and pollinators; and loss or damage to ecosystems through development. There is a roughly an equal split between positive and negative impacts, with a large number of studies recording both together (Figure 16).
Although 76% of studies concern direct input to the ecosystem, there are also examples of indirect effects, where human influence on surrounding areas affects provision of the service. This is most common for the services of freshwater provision (where land use in the catchment affects water supply) and pollination (due to the presence or absence of habitats for pollinators near to crops) (Figure 17).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 39
Human input and management
Indirect Both (Direct & Indirect)
Input Direct Unclear Intensive Both (Intensive & Extensive) Extensive Unclear Positive Both (Positive & Negative) Negative Impact Intensity Unclear 0 100 200 300 400 Number of studies
Figure 16: Number of studies showing different types of human input and management, management intensity and impact for all ecosystem services reviewed.
Freshwater fishing Timber production Freshwater provision (quantity) Food production (cultivated crops) Air quality regulation Atmospheric regulation (carbon sequestration) Mass flow regulation (erosion protection) Water flow regulation (flood protection) Water quality regulation Pollination Pest regulation Recreation (species-based) Aesthetic landscapes 0 10 20 30 40 50 60 Number of studies
Direct Both (Direct & Indirect) Indirect Unclear
Figure 17: Number of studies showing direct or indirect human input and management for each ecosystem service reviewed.
Figure 18 illustrates the split between intensive and extensive inputs. Food production and freshwater fishing are dominated by intensive forms of management, including crop fertilisation and fish stocking programs. Water quality regulation is also dominated by the impacts of intensive practices such as deforestation (negative) or wetland construction (positive). In contrast, the majority of articles on air quality regulation refer to input as both intensive and extensive. For example, the practice of planting trees in urban areas is extensive, but also intensive due to their subsequent irrigation needs (Jim and Chen,
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 40
2008); whereas the construction of green roofs in Chicago could be intensive, extensive or semi-intensive depending on which type was planted (Yang et al., 2008).
Freshwater fishing Timber production Freshwater (quantity) Food production (cultivated crops) Air quality regulation Atmospheric regulation (carbon sequestration) Mass flow regulation (erosion protection) Water flow regulation (flood protection) Water quality regulation Pollination Pest regulation Recreation (species-based) Aesthetic landscapes 0 10 20 30 40 50 60 Number of studies
Extensive Both (Intensive & Extensive) Intensive Unclear
Figure 18: Number of studies showing extensive or intensive human input and management for each ecosystem service reviewed.
Figure 19 shows the split of impacts due to human input and management by ecosystem service. Most services record a mix of positive and negative impacts, and often both together. This reflects a typical pattern observed in many articles that record damage to ecosystem services from unsustainable development, such as loss of carbon storage due to clear-felling of forests, together with positive impacts from beneficial interventions such as reforestation or designation of protected areas. Another common approach is for articles to compare two alternative forms of management with positive and negative impacts, such as the impact on pollinators or pest predators of intensive agriculture vs. organic agriculture.
Food production provides interesting examples of the conflicts and trade-offs often encountered between and within ecosystem services. Although human input can increase food yield through beneficial interventions such as irrigation to reduce drought risk (Hoffman et al., 2009), over-extraction of groundwater can also endanger food production and other services such as freshwater provision (Nainggolan et al., 2013). Other services also cite numerous negative impacts from human input and management associated with agricultural activities, such as the impact of forest clearance for agriculture on water flow regulation (flood protection), atmospheric regulation (carbon storage), pollination and pest control; the impact of tillage on mass flow regulation (soil erosion); and the impact of fertiliser run-off on water quality.
It is also interesting to compare the services of species-based recreation and aesthetic landscapes. For recreation, a large number of positive impacts are recorded which consist largely of conservation actions to protect wildlife and encourage eco-tourism. However, a few negative impacts are also recorded where tourist behaviour or illegal hunting can deplete wildlife populations. For aesthetic landscapes, the majority
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 41 of impacts are negative because in most cases people express a preference for unmanaged landscapes. However, some studies find that positive impacts can also arise because farmers and certain ethnic groups prefer managed landscapes.
An exception to the generally observed even mix of both positive and negative impacts is for air quality management, where human impact is mainly recorded as beneficial (in 78% of studies) due to the planting of trees to improve air quality (though one comment notes that the removal of urban trees due to development could be classified as a negative impact). However, in some cases there can be both positive and negative impacts as avenues of trees along busy roads can inhibit pollution dispersal, and certain tree species can worsen air quality due to emission of ozone-forming volatile organic compounds.
Freshwater fishing Timber production Freshwater provision (quantity) Food production (cultivated crops) Air quality regulation Atmospheric regulation (carbon sequestration) Mass flow regulation (erosion protection) Water flow regulation (flood protection) Water quality regulation Pollination Pest regulation Recreation (species-based) Aesthetic landscapes
0 10 20 30 40 50 60 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 19: Impact (direction) of human input and management for each ecosystem service reviewed.
3.5 Interactions with other ecosystem services
Interactions between different ecosystem services were mentioned in 40% of articles, though they were not always investigated explicitly. Over half of the studies on freshwater provision, atmospheric and water flow regulation cite interactions with other services, compared with less than 25% of those on species- based recreation, water quality regulation and timber production.
Interactions between different ecosystem services are shown in Error! Reference source not found.Table 6. The most frequent interaction cited in the literature, with 27 counts, is that between mass flow and water flow regulation. All of the other frequently cited interactions (>20 counts) occur between the regulating and provisioning services, for example between food production and atmospheric regulation (24 counts); water flow regulation and freshwater provision (26 counts); and food production and pollination (22 counts). Interestingly, no specific interactions were identified between the two cultural services, though
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 42 species-rich landscapes were appreciated by both services, and crowding (which could be seen as “too much” recreation) was cited as a negative impact for aesthetic landscapes.
Table 6: The number of studies mentioning interactions between ecosystem services. Most frequent interactions are highlighted in darker shades of grey. Green indicates provisioning services; blue indicates regulation and maintenance services; and red indicates cultural services.
Food production
Freshwater fishing 7
Timber production 2 1
Freshwater provision 7 1 18
Air quality regulation 3 4
Atmospheric regulation 24 19 22 16
Mass flow regulation 18 1 2 5 6 11
Pollination 22 1 1
Pest regulation 21 1 3 1 7
Water flow regulation 20 7 13 26 13 10 27 2 1
Water quality regulation 15 6 6 9 2 9 4 1 3 18
Recreation 3 14 2 7 2 2 5 8 3
Aesthetic landscapes 11 2 8 1 18 5 1 11
Other 3 1 12 22 12 6 4 10 6 1 2 Food production Freshwater fishing Timber production Fresh Air quality regulation Atmospheric regulation flow Mass Pollination Pest regulation Waterflow regulation Waterquality regulation Recreation landscapes Aesthetic
water
regulation
The majority of interactions (56%) identified between ecosystem services were documented as being positive, with an additional 8% citing both positive and negative interactions. These are shown in Table 7, and are most common between different regulating services. The greatest number of positive links (25) is between mass flow and water flow regulation, as both services are provided by measures such as afforestation, with less runoff and erosion occurring in vegetated areas. There are also strong positive links (18) between food production and pollination, as many crops (e.g. fruit, cacao, oilseed and coffee) are clearly dependent on pollination, and between food production and pest regulation (16 links). There is a strong synergy between water quality and water flow regulation, both of which are typically enhanced by wetlands, forests and other vegetation. Finally, planting trees to improve air quality also provides many benefits to other services: atmospheric regulation, water flow regulation and aesthetic landscapes. It is also
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 43 worth noting the large number of positive interactions (21) for air quality regulation under the “other” category: this refers mainly to the benefits of urban trees and green roofs for microclimate regulation.
Table 7: The number of studies mentioning positive interactions between ecosystem services. Includes studies citing “both positive and negative interactions”. Most commonly occurring positive interactions have been highlighted in darker shades of green. Green indicates provisioning services; blue indicates regulation and maintenance services; and red indicates cultural services.
Food production
Freshwater fishing 6
Timber production 1
Freshwater provision 2 1 2
Air quality regulation 3 4
Atmospheric regulation 14 8 8 16
Mass flow regulation 5 1 1 2 6 9
Pollination 18 0 0 0 0
Pest regulation 16 0 0 1 0 5
Water flow regulation 8 2 6 11 13 7 25 1 1
Water quality regulation 8 5 3 4 2 2 4 1 3 17
Recreation 3 11 0 1 7 2 0 2 5 7 2
Aesthetic landscapes 5 1 1 0 18 4 1 0 0 9 0 0
Other 1 1 2 21 6 4 3 5 4 1 Food production Freshwater fishing Timber production Fresh Air quality regulation Atmospheric regulation regulation flow Mass Pollination Pest regulation Waterflow regulation Waterquality regulation Recreation landscapes Aesthetic
water
Negative interactions between ecosystem services are cited in only 16% of studies, plus the 8% that cite both positive and negative interactions. Table 8 shows that most of these antagonistic interactions occur between provisioning and regulating services, particularly for food and timber provision. Articles on water flow regulation and atmospheric regulation, for example, often mention the negative impact of deforestation driven by conversion of forest to farmland, or clear-felling of forest for timber. Food production also adversely affects mass flow regulation, as cultivation makes land more vulnerable to soil erosion. Similarly, the literature on atmospheric regulation often cites a negative interaction with timber production, as regular harvesting of timber reduces carbon storage.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 44
However, there are also major negative interactions between two provisioning services: timber production and freshwater provision. Timber plantations may be very intensive users of limited water supplies in areas where water is scarce. In other examples, conversion of land for timber production competes with cloud forest habitats which are beneficial to water supplies (Hamilton, 1995); and the change from broad-leaved forest to coniferous plantation causes reductions in water yield (Komatsu et al., 2008).
Table 8: The number of studies mentioning negative interactions between ecosystem services. Includes studies citing “both positive and negative interactions”. Most commonly occurring negative interactions have been highlighted in darker shades of red. Green indicates provisioning services; blue indicates regulation and maintenance services; and red indicates cultural services.
Food production
Freshwater fishing 2
Timber production
Freshwater provision 2 11
Air quality regulation 0
Atmospheric regulation 9 8 4 0
Mass flow regulation 10 1 1 0 0 0
Pollination 2 0 0 0 0
Pest regulation 9 0 1 0 0 2
Water flow regulation 12 5 7 7 0 2 1 1 0
Water quality regulation 8 1 3 0 0 4 0 0 1 2
Recreation 2 5 0 0 0 0 0 0 1 1 1
Aesthetic landscapes 7 0 0 0 0 0 0 0 2 0 0
Other 1 1 1 2 1 4 2 Food production Freshwater fishing Timber production Fresh Air quality regulation Atmospheric regulation regulation flow Mass Pollination Pest regulation Waterflow regulation Waterquality regulation Recreation landscapes Aesthetic
water
There are some interesting examples of trade-offs amongst the 8% of interactions with both positive and negative effects. For example, one study linking food production and atmospheric regulation in Spain finds that olive and almond farming sequesters carbon, whereas vineyards are net emitters of CO2 (Nainggolan et al., 2013). Similarly, for the effect of pest regulation on food production, Bisseleua et al. (2013) find that although a higher Shade Index increased beneficial predators and reduced pest populations and damage, it also reduced cocoa yield.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 45
3.6 Thresholds
In total only 68 instances of thresholds were identified in the main literature review. This includes safe minimum standards (10) and legal boundaries (5), mainly cited for air quality regulation, but the majority are categorised as biophysical thresholds, with nine being attached to specific ecosystem service indicators, and the remaining 44 being classed as “other biophysical thresholds”. These are often unique thresholds for each study where changes in behaviour were observed, such as the percentage of vegetation or forest cover that can safely be removed before a decrease in water flow regulation (flood protection) occurs. Thresholds may also correspond to a classification scheme, e.g. habitats classified as either high existing value or marginal existing value (Jackson et al., 2013).
Figure 20 highlights that studies related to air and water quality regulation, freshwater supply and food production cite thresholds most frequently. In contrast, none of the articles discussing aesthetic landscapes or timber production indicate the existence of any thresholds.
Freshwater provision Food production (cultivated crops) Air quality regulation Atmospheric regulation (carbon sequestration) Mass flow regulation (erosion protection) Water flow regulation (flood protection) Water quality regulation Pollination Pest regulation Recreation (species-based)
0 2 4 6 8 10 12 14 Number of studies Legal Boundaries Safe Minimum Standards Other Biophysical
Figure 20: The frequency of thresholds discussed in the literature for the 13 ecosystem services.
A wide range of thresholds for freshwater provision were identified, including the level of deforestation in the Amazon after which significant changes in cloud cover are predicted (Nosetto et al., 2005); the amount of rainfall required before recharge occurs (Petheram et al., 2002), and two studies describing thresholds for the reduction of runoff due to afforestation (e.g. Farley et al., 2005). In contrast, for water flow regulation, several studies note that the attenuation of peak stream flow starts to decline after the removal of 20-30% of forest cover in the catchment area (e.g. Bathurst et al., 2011; Lin and Wei, 2008). Some studies on water flow regulation found that the service declined for larger flood events (e.g. the presence of salt marsh vegetation was found to have little effect on flood attenuation above an inundation depth of 1.3m, Möller et al., 2007; forest cover has no effect for extreme events, e.g. Bruijnzeel, 2004; Moore and Wondzell, 2005). However, Green & Alila (2012) dispute this view (see report on water flow regulation in Annex 5).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 46
Thresholds for water quality regulation include the minimum level of richness, evenness and diversity required in a microbial community for denitrification (Martin et al., 1999); a saturation effect between species richness and ecosystem functioning (Cardinale et al., 2011); a threshold for chronic sediment delivery to streams (Rashin et al., 2006); and a biomass threshold for seagrass (Moore, 2004). Food production thresholds include the effect of short-term exposure to high temperature on wheat grain yields (Ferris et al., 1998); the increase of crop yield with soil thickness until a certain depth and decrease thereafter (Power et al., 1981); and a significant reduction in wheat production if the groundwater table depth falls below 0.5 m (Chen and Zong, 1999). For atmospheric regulation, it was found that although carbon storage increases with species richness, it can saturate at a low number of species, with storage increasing at a slower rate thereafter (Chen, 2006).
Air quality regulation thresholds included safe minimum standards and legal boundaries for pollutants including ozone, particulate matter, nitrous oxides and carbon monoxide. Examples included international programs (WHO standards), national programs (the US Ambient Air Quality Standard), and local standards (e.g. the California Ambient Air Quality Standard).
Other examples include a biophysical threshold for pollination describing the point when bee visits no longer increase pollination (Greenleaf and Kremen, 2006). Finally, a study concerning recreational angling identified a noise threshold from boats after which fish stress was exhibited (Lewin et al., 2006).
As the main literature review did not specifically search for the word “threshold”, a supplementary review was carried out (see Annex 1 for details). This identified only 28 original journal articles that attempted to quantify or otherwise assess biophysical thresholds that relate to ecosystem services. Of these, only around half provided a quantitative value for a threshold, and not all of these are related to a specific ecosystem service. Only one article was identified in both the main and supplementary literature reviews: the study by Letourneau et al. (2012), which found a steep decline in pest parasitism at thresholds of 38% or 51% crop cover (for two different species). Most of the other thresholds found in the supplementary review also related to land cover, including the impact of vegetation cover on soil productivity (Gao et al., 2011); the impact of removal of forest patches in Madagascar on pollination and seed dispersal (Bodin et al., 2006) and loss of tree cover in Australia on pest control (Fischer et al., 2010); the effect of landscape complexity in Argentina on a range of services including food production and water flow regulation (Laterra et al., 2012); the provision of pollinator habitat (Keitt, 2009); the impact of seagrass cover on fish habitat and atmospheric regulation (Wagner et al, 2012); and the impact of “biologically vital areas” on a range of regulating and cultural services (Szulczewska et al., 2014). However, there are also one or two examples of thresholds related to temperature and salinity, such as the impact of drought on oyster populations in estuaries (Petes et al., 2012) and the impact of water temperatures on trout (Wardynsky et al., 2013).
3.7 Indicators for ecosystem service assessment
The majority of the ecosystem service literature reviewed (83%) used assessment methods which were based on primary data, such as direct field observations or public statistical databases. Mechanistic, correlative and conceptual models were used very rarely.
The frequency of the different indicators used for each ecosystem service is presented in Annex 3. It is worth noting that the initial list of potential indicators, derived from the JRC review on indicators for
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 47 mapping ecosystem services (Egoh et al., 2012), included a number of indicators that were never used or cited in the literature, and 37% of the studies reviewed used indicators that were not included in the JRC review. This information could be used to considerably improve the list of indicators for future work on ecosystem services. Full details, including the list of newly identified indicators, are given in Annex 3.
3.8 Policies
Only 13% of the studies reviewed cite a particular policy. Some of these cite multiple policies, with 121 examples identified in total. However, policy is discussed more often than this implies, with many articles describing existing national/local level policies, or discussing policy recommendations in general terms. Of the 24 EU policies included in the database, only the following nine are mentioned in the literature, and these are rarely cited:
Ambient Air Quality Directive Biodiversity Strategy Birds Directive Common Agriculture Policy Environmental Impact Assessment Habitat Directive Nitrates Directive Soil Thematic Strategy Water Framework Directive (WFD)
Policies are cited most frequently for water flow regulation (21 counts, 5 of which are the Water Framework Directive), followed by air quality regulation, atmospheric regulation, food production, pollination, species-based recreation and freshwater provision. Full details are given in Annex 4.
4 Results for each ecosystem service
In this section we use network diagrams to present a summary of the results for each of the 13 ecosystem services in terms of the operational classification scheme (typology) that we have developed. The network diagrams show the links between ecosystem service provider, biotic attributes and abiotic factors for each ecosystem service. We also briefly discuss interactions between services. Full results for each service are presented in Annex 5.
4.1 Freshwater fishing
Around 30% of the studies on freshwater fishing focus on aquaculture ponds, many of which are in Asia, and 37% cover rivers, lakes or reservoirs used for recreational fishing, many of which are in North America. The remaining articles are wider reviews or laboratory studies. The main indicator used to measure the service is the weight of fish caught per unit area of the water body per day or year.
The main provider identified for the service of freshwater fishing is specific populations of commercially important fish species, such as carp, tilapia or trout, though eight studies also refer to the entire community, meaning the community of fish within a particular habitat such as a fishing lake (Figure 21).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 48
A wide range of biotic attributes are discussed for both classes of ESP. Species-level attributes are the most important, with species abundance (stocking rate), species size/weight and population growth rate all having a predominantly positive impact on freshwater fishing. Larger fish were preferred by fishermen, and were also found to produce larger yields due to their higher survival rate (Li, 1999). Species richness, primary productivity and community/habitat area are also found to have a positive influence, with a number of studies finding higher productivity and yield in polycultures compared to monocultures. Mortality rate was the only biotic attribute found to have a purely negative impact. However, there was a trade-off between stocking density and yield, because over-stocking reduces fish size and eventually leads to increased mortality (e.g. Lorenzen, 1995). Species abundance can also have a negative impact in a few cases due to predation: for example, non-indigenous sea lamprey (Petromyzon marinus) caused a large decrease in populations of commercially important fish in Lake Superior (Lawrie, 1978).
Positive Mostly positive Neutral Mostly negative Negative
Figure 21: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of freshwater fishing. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
A range of abiotic factors are discussed. Nutrient availability and water quality are the most important for the provision of freshwater fishing. The direction of impact varies between the studies.
Human management includes intensive activities such as stocking ponds, feeding and rearing fish, especially for the 30% of studies covering aquaculture. However, 20% of studies discuss more extensive management for recreational fisheries where fish feed mainly from natural sources. Management activities such as feeding and stocking, as expected, have a positive impact on fish catch (provided that overstocking is avoided). However, for natural fisheries, human activity can cause a negative impact from over-fishing or habitat degradation. No specific thresholds were identified, though several articles mention that fish catch would be expected to decline above a certain stocking density.
Figure 22 shows the interactions with other ecosystem services that were identified in the freshwater fishing review (not including interactions with fishing that were identified in the other ecosystem service
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 49 reviews). Not surprisingly, the main interaction is a positive link with species-based recreation, i.e. the sport of fishing. There are also negative interactions with water flow regulation, as the construction of dams, drainage channels or irrigation systems can damage fish habitat and disrupt fish migration.
POSITIVE NEGATIVE
BOTH
Figure 22: Interactions between freshwater fishing and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 50
4.2 Timber production
The studies reviewed are split roughly equally between natural forests and timber plantations, and focus on forests containing commercially important species such as conifers and eucalyptus. There is a wide geographical spread, though the studies on plantations were located mainly in tropical regions. Timber production was measured using various indicators such as growth of the trees (height, diameter or circumference) or the production of wood (m³ or tonnes).
The service of timber production is mainly provided by one or more specific species populations (e.g. Scots Pine and Silver Birch), although some studies are classified as dealing with functional groups (e.g. mixed conifer forest) (Figure 23).
Positive Mostly positive Neutral Mostly negative Negative
Figure 23: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of timber production. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
The impact of biotic attributes seems to be predominantly positive, with species richness being cited the most often. Most studies (26) found evidence that plantation species are more productive in mixtures than in monocultures (e.g. Erskine et al., 2006), but there is some conflicting evidence, with eight studies finding monocultures to be more productive (e.g. Nguyen et al., 2012). Other factors with a positive impact on timber production include functional diversity, species abundance, stem density and community/habitat structure. For example, Donoso et al. (2007) found that forests with mixed canopy heights are more productive due to better use of the available light. The only negative influence was cited by Vila et al. (2003) who found that pine forests are less productive at a later successional stage.
For the abiotic factors, the most commonly mentioned is soil, though other factors such as precipitation and temperature are also found to have a positive impact in a small number of cases. Additional abiotic factors include radiation (Liang et al., 2007) and insecticide (Riedel et al., 2013), both of which have a negative impact on timber production.
Perhaps surprisingly, only two of the articles on timber production mention interactions with other ecosystem services (Figure 24). One of these mentions the positive impact of trees on water flow regulation (flood protection), through enhanced evapotranspiration, and the other mentions the benefits of
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 51 plantations for wildlife habitat and non-timber forest products. In contrast, articles on other ecosystem services mention numerous interactions with timber production, including both positive and negative interactions with atmospheric regulation and water flow regulation, and mainly negative interactions with freshwater provision (see section 3.5).
POSITIVE NEGATIVE
BOTH
Figure 24: Interactions between timber production and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 52
4.3 Freshwater provision
There is little literature that explicitly investigates freshwater provision as an ecosystem service; instead, the reviewer generally inferred the impact of natural capital from studies of the effect of land cover change in a catchment area (especially change in forest cover) on proxy indicators including stream flows, water levels or runoff. Because of this focus on land use studies, the ecosystem service provider is typically classified as the entire community or habitat, although some studies focus on specific species or functional groups (Figure 25). Studies mainly focus on areas where freshwater supply is limited or threatened, such as Australia and South Africa.
Positive Mostly positive Neutral Mostly negative Negative Figure 25: Network diagram showing the linkages between ESPs, biotic attributes and abiotic factors for the service of freshwater (potable water) provision. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
In contrast to the other ecosystem services, biotic attributes have a relatively low impact and in many cases the impact is unclear or negative. A number of articles describe a negative impact from forest plantations, as trees can intercept rainwater and absorb groundwater, thus reducing the supply available for humans downstream. The two most commonly cited attributes with a negative impact are community/habitat area and stand age, as older/larger trees use more water (e.g. Nosetto et al., 2005). Similarly, higher stem density and higher sapwood area can increase water use (Kagawa et al., 2009), and harvesting and thinning are found to significantly increase runoff and therefore increase provision in many studies (e.g. Petheram et al., 2002; Sahin and Hall, 1996). There is a distinction between plantations and natural forests: many of the 20 articles citing reduced water yield concerned plantations of species such as pine and eucalyptus, and some studies showed that water yield decreased when old growth forest was replaced with pine plantations (Rowe and Pearce, 1994; Komatsu, 2008). Five studies showed that natural forests can have beneficial impacts on water supply. Of these, four were cloud forests (e.g. Gomez-Peralta et al. 2005, Brauman et al. 2010) and one was old forest in the Western Ghats in India (Singh and Mishra, 2012). Grasslands are also found to increase water provision compared to bare soil (Holdsworth and Mark, 1990).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 53
Abiotic factors appear to have an important influence on this service, with precipitation and snow adding freshwater to the system, and hence improving provision, whereas evaporation (mainly evapo- transpiration by trees) leads to a loss of moisture and sometimes an increase of salt content in soils (e.g. Ruprecht and Stoneman, 1993) and thus reduces freshwater provision. The direction for the remaining abiotic attributes, including soil, temperature and slope, is unclear. Soil is often mentioned (for example in terms of soil moisture content), but the direction of impact is generally unclear. However, with 41 records categorised as unclear and just one as positive, the green colour of the thick line is somewhat misleading. One significant soil-related impact is reported by Buytaert et al. (2007), who find that soil compaction arising from forestry operations has a negative impact on water supply by reducing infiltration.
Human input clearly affects freshwater provision, through land use management. As described above, the review found that afforestation can decrease freshwater supply, but that the choice of species is important: plantations of species such as pine and eucalyptus often cause a significant decrease in stream flow while native old growth forests can have a positive impact. Management techniques can also have an influence, with thinning of plantations increasing the water yield. Several studies found thresholds concerning the proportion of a catchment that could be forested before a reduction in water yield is observed, or the amount of rainfall necessary before surplus runoff is generated. However, the afforestation thresholds may be related to the difficulty of measuring small changes in stream flow rather than a true biophysical threshold.
Figure 26 shows that a number of interactions between freshwater provision and other ecosystem services were identified, with the clearest being a negative interaction with timber production, due to the negative impact of plantations on water levels as discussed above. The relationship with atmospheric regulation (carbon storage) is less clear, with a mix of positive and negative interactions reflecting the fact that some old growth forest and cloud forests can be beneficial for water supply, and that vegetated land can be better than bare soil for both these services. There are also links to water quality, as poor water quality reduces the supply of fresh water. In some cases native forests can enhance both these services, but in other cases forests can improve water quality but decrease water quantity as noted above.
The interactions with forest ecosystems are reflected in the coverage of policies in the literature, which mentions not only a number of national and local policies to improve water availability and quality, such as Integrated Water Resources Management and subsidies for water conservation, but also policies related to biodiversity and carbon storage in forests.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 54
POSITIVE NEGATIVE
BOTH
Figure 26: Interactions between freshwater provision and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 55
4.4 Food production (cultivated crops)
The studies reviewed on food production cover a wide geographical range but over half are in North America or Europe. The most frequently cited indicator is ‘crop yield’, usually in tons/ha but sometimes in $/ha, though a range of other indicators are also used including water productivity (kg food/m3 water) and food security (average daily production of calories/household member). The service of food production is provided mainly by one or more specific species populations, for example a mixture of crops, or a single crop such as corn or sugar beet (Figure 27). One study focuses on functional groups (perennials and annuals).
The review for food production is more limited than for the other services as it did not specifically search for biotic attributes. For this reason, the network diagram shows few biotic attributes. However, the review does emphasise the importance of abiotic factors for food provision. Unsurprisingly, nutrient availability has a positive effect, and many articles discuss problems arising from lack of precipitation, reduced snowmelt and salinity of groundwater. Temperature, evaporation and soil are also mentioned. Additional abiotic factors include atmospheric CO2 concentration. Thresholds were identified for temperature, organic matter content in the soil, soil thickness, groundwater levels and soil salinity, above or below which crop yield decreased.
Positive Mostly positive Neutral Mostly negative Negative
Figure 27: Network diagram showing the linkages between ESPs and abiotic factors for the service of food production (cultivated crops). AF = abiotic factors. Line thickness indicates the number of studies showing a given link. Few biotic attributes are shown because the review focused on abiotic factors.
All but one of the 50 articles discuss some form of human management, either agricultural activities designed to increase crop yield (e.g. drainage, irrigation, mechanical cultivation and application of fertilisers and pesticide) or non-agricultural activities such as dam construction, that have an indirect impact on food production. In most cases the management is direct, though four articles mentioned indirect effects such as the loss of wetlands to agriculture (Zhao et al., 2004) or the impact of water management at a regional scale (Misra, 2013). Where the intensity could be determined, it was generally classed as intensive (19 out of 23 studies). Impacts are mixed: six articles cited the benefits of irrigation or fertilisation for increasing crop yield, but 11 articles mentioned negative impacts such as loss of soil fertility, soil erosion, damage to wetlands from agriculture, or the impacts of dam construction on farmland.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 56
The complexity of these impacts is reflected in the wide range of interactions between food production and other ecosystem services, as shown in Figure 28.
POSITIVE NEGATIVE
BOTH
Figure 28: Interactions between food production and other ecosystem services (unclear interactions omitted).
As water availability is a crucial factor for agriculture, it is not surprising that there are interactions with water flow regulation, freshwater provision and water quality. These can be positive or negative: several articles mention the vulnerability of farmland to water-logging, flooding, sea-level rise and saline intrusion, so that improvements in water flow regulation and water quality regulation can enhance food production, but agriculture also has negative impacts on water quality due to pollution with excess fertilisers. There are also interactions with atmospheric regulation: the articles mainly refer to positive impacts on crop yield from the CO2 fertilisation effect, or to the potential of agricultural soils to sequester carbon with the right
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 57 management techniques, though one article refers to the adverse climate impact of deforestation for agriculture. Finally, nine articles address the problem of soil erosion, which is shown to reduce crop yield.
There is considerable potential for management techniques to exploit the synergies and reduce the negative interactions associated with food production, and this is partly reflected in the number of European level policies mentioned in the review for this service, including the Biodiversity Strategy, Habitats Directive, Common Agriculture Policy, Nitrates Directive, Water Framework Directive and Environmental Impact Assessment. However, these policies are not applicable to non-European countries.
4.5 Air quality regulation
Urban trees are the primary vegetation capable of having an impact on air quality. They do this in two ways: firstly by physically intercepting airborne pollutants on the leaf surface or absorbing them through the leaf pores (stomata), and secondly (indirectly) by providing shade and shelter so that buildings require less energy for heating and cooling, so that less pollution is generated. The literature included in this review focuses exclusively on the first method.
All but seven of the studies reviewed are located in Europe or North America, and they are mainly local studies concerned with assessing the impacts of trees, green roofs or other urban vegetation on dry deposition of pollutants in one or more cities. The studies use a mix of primary data and modelling techniques, and the main indicators are air purification rates or deposition velocity (both typically measured in weight of pollutant removed per unit area per unit time), or changes in pollution concentrations (in weight of pollutant per cubic metre of air). Percentage tree cover is almost always an important parameter in the models.
The main provider of the service is urban vegetation, which can be classified as an entire community (urban vegetation), a specific functional group (urban trees) or a mix of specific species, depending on the context. Examples of all these ESPs were identified in the review, with a degree of overlap, and a range of species-, community- and functional group-level biotic attributes are identified as contributing to the service (Figure 29).
Two of the main attributes are leaf area index and canopy cover, both of which are classified under “Other biotic” as they were not in the original list of biotic attributes. Other key attributes are community/habitat area, typically referring to benefits of tree cover for pollution removal (e.g. Jim and Chen, 2008); functional diversity, which is found to enable pollution removal to be performed across varying conditions (Manes et al., 2012); and species size/weight with larger trees having a larger leaf area enabling them to trap airborne pollutants (Brack, 2002). All attributes except mortality have a predominantly beneficial impact, but there are some negative impacts. Several studies assess the impact of isoprene and other biogenic volatile organic compounds (BVOCs), which are emitted at far higher rates by certain tree species and can stimulate ozone production, though all these studies conclude that the net impact of urban trees on air quality is beneficial (e.g., Morani et al., 2011). Five studies present evidence that trees planted along busy roads can create a canyon effect that prevents atmospheric mixing of polluted air and lowers air quality (e.g. Salmond et al., 2013).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 58
Positive Mostly positive Neutral Mostly negative Negative
Figure 29: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of air quality regulation. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
Abiotic factors, particularly those involved in the mechanics of atmospheric mixing or air flow, have highly complex and non-linear relationships with the contribution of urban trees to air quality regulation. Relationships described in the studies reviewed are highly context dependent. For example, factors related to air movement (i.e. wind direction, wind speed and boundary layer height) can either increase deposition rates on plant surfaces or increase re-suspension of particulates, depending on the factor’s magnitude and on the spatial and temporal context (e.g. the interactions with topography and physical structure of the vegetation; Freiman et al., 2006; Nowak et al., 2006). Similarly, high temperatures can decrease the uptake of pollutants by plants (e.g. Alonso et al., 2011) and can also reduce air quality by increasing the rate of ground level ozone production from BVOCs (Salmond et al., 2011). However, temperature can have an almost purely positive effect on air quality regulation over ranges where it increases plant growth and metabolism. Simplified generalisations about the relationship between abiotic factors and air quality regulation are therefore largely impossible.
Urban environments are highly managed ecosystems, and therefore human management is integral to the air quality regulation provided by urban trees. Urban trees and shrubs are managed either directly through planting and maintenance (watering, pruning, feeding), or indirectly by protecting surrounding woodland areas from development. Where these management activities increase the extent of urban vegetation, air quality regulation generally increases. The only exceptions to this were cases where planting efforts either utilised species with higher BVOC emission rates or resulted in tree configurations that prevented adequate mixing of polluted air.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 59
Thresholds are mentioned only in the context of air pollution standards above which there is presumed to be a negative health impact. Vegetation removal of airborne pollutants is generally regarded as an asymptotic function of vegetation cover, so it seems unlikely that a threshold value of vegetation cover that is necessary to sustain this service could be identified.
POSITIVE NEGATIVE
BOTH
Figure 30: Interactions between air quality regulation and other ecosystem services (unclear interactions omitted).
Overall, the evidence for substantial air quality benefits from urban vegetation was found to be weak, with 38 of the 50 studies relying largely on modelling and plant chamber experiments, and little empirical evidence in a real urban setting. Moreover, both modelled and empirical data generally indicate that the
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 60 proportion of airborne pollutants removed by urban vegetation is low (Setala et al., 2013). However, as shown in Figure 30, virtually all of the studies reviewed emphasise the positive contributions urban trees and other vegetation made to other ecosystem services. The most frequently cited interaction is with microclimate regulation (21 studies), which is classified as “other”, but studies also identified benefits for aesthetic landscapes (18 studies), atmospheric regulation (14 studies), mass flow regulation (6 studies), water flow regulation (12 studies), recreation (7 studies) and timber production (3 studies). The only negative interaction is with pest regulation, as trees can provide habitat for insect pests.
4.6 Atmospheric regulation (carbon sequestration)
Most of the studies on atmospheric regulation are experimental measurements of vegetation biomass at a particular local site – often sampling a group of plots in a forest, or comparing two different habitats such as forest and farmland, or logged forest and intact forest. The estimates of biomass are then used to estimate carbon storage in tons per hectare, or carbon sequestration in tons/hectare/year. All continents are represented, though 81% of the studies are in North America, Europe or Asia.
Most of the studies assess the service at the level of the entire community or habitat, which can include not just trees and shrubs but also grass, understory plants, dead wood, leaf litter and soil carbon. However, some studies focus on specific species or functional groups (Figure 31Figure 37).
Positive Mostly positive Neutral Mostly negative Negative
Figure 31: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of atmospheric regulation (carbon storage). AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 61
The literature shows that this service is affected by a large range of attributes, both biotic and abiotic, and the influence is mainly positive. The main determinant of carbon storage is simply the amount of biomass, so key attributes are community (forest) area, above- and below-ground biomass, stand age, primary productivity, growth rate and species size/weight. For example, Kirby and Potvin (2007) find that trees with diameter at breast height (DBH) over 10cm account for 90% of the aboveground carbon stocks in the forest area studied. A number of studies investigate the impact of species richness, functional richness, functional diversity and structural diversity, finding that this has a positive impact in many studies, but that sometimes a less diverse mix could store more carbon if it consists of large tree species. Chen (2006) reports that carbon storage increases with species richness but that it may saturate at a low number of species, after which it increases more slowly: this is the only example of a threshold found in the review. Many of the more recent articles highlight an interesting debate over the role of niche complementarity versus the selection effect, which is discussed in detail in the report on atmospheric regulation in Annex 5. Mortality rate is the only attribute to negatively affect carbon storage, for example as a result of wildfire (e.g. Haugaasen et al., 2003), pests such as bark beetle (Seidl et al., 2008), or grazing (Klumpp et al., 2009).
The relationships between abiotic factors and atmospheric regulation are less clear, with the review finding that these are highly dependent on the ecosystem and location considered. Factors include water availability, precipitation, evaporation, temperature and soil (including the effect of pH; Keeton et al., 2010; and soil moisture; Yurova and Lankreiger, 2007). Drought and high temperatures, both exacerbated by climate change, are often cited as having a negative impact on this service (e.g. Beier et al., 2009, Law et al., 2003), and wildfire occurrence is an additional (often related) abiotic factor (e.g. Wardle et al., 2012).
Human impacts are mainly negative, through logging and over-harvesting of forest products, grazing, forest fires and urban development. However, there are also examples of positive impacts where there is management to promote biodiversity and ecosystem services, such as on organic coffee farms (Hager, 2012) or no-till farming to enhance soil carbon (Mishra et al., 2010), or where forest or grassland that has been damaged by logging or overgrazing is now being protected or restored. Fourteen articles mention policies to promote carbon storage, including REDD and the Clean Development Mechanism as well as some national and local policies such as logging bans or forestry management.
Figure 32 shows a wide range of mainly positive interactions between atmospheric regulation and other ecosystem services, reflecting the multiple benefits of forests for services such as water quality regulation, mass flow regulation, recreation and aesthetic landscapes. The interactions with timber production are complex, and can be positive or negative depending on whether the timber is produced sustainably. There are also negative interactions with food production due to deforestation to clear land for farming, although there are also benefits for crop yield from farming techniques that improve soil carbon storage.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 62
POSITIVE NEGATIVE
BOTH
Figure 32: Interactions between atmospheric regulation and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 63
4.7 Mass flow regulation (erosion protection)
For mass flow regulation, 42% of the studies reviewed are from Europe, often from mountainous or Mediterranean areas where soil erosion is a common problem. Studies typically measure soil erosion empirically, in units such as kg/m2/year. Most of the studies focus on the role of vegetation cover in stabilising soil. They generally refer to the impact of the entire community/habitat, but some examine specific species or functional groups. The overall impact of biotic attributes is positive, with no predominantly negative or unclear relationships (Figure 33).
Positive Mostly positive Neutral Mostly negative Negative
Figure 33: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of mass flow regulation (erosion protection). AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
The review found that the presence of some form of vegetation, generally a specific habitat (e.g. grassland), but sometimes a functional group (trees or shrubs) or specific species (e.g. Stipa grasses), is the most important factor in providing this service. For example, grassland is found to reduce soil erosion when compared to cropland (Lorencová et al., 2013) and forest has lower erosion rates than both cropland and grassland (Wei et al., 2010). Older and taller vegetation tends to have a more stabilising effect, so community structure, age, above- and below-ground biomass and successional stage all have positive impacts. Three articles also consider the role of soil-engineering animals: earthworms/termites, dung beetles and prairie dogs.
For the abiotic factors, precipitation and steeper slopes are the main factors that increase erosion (e.g. Bini et al., 2006; Zuazo et al., 2004). However, in drier climates low rainfall can be associated with greater erosion, such as in a study of Mediterranean shrubland by Kosmas et al. (1997).
Nine articles gave examples of negative impacts of human management, including forest clearing or logging, grazing, tillage, burning, and the introduction of invasive species, although one article cited a
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 64 positive impact from restoration of native forests. Despite the widespread nature of the problem, there was little mention of policies to address soil erosion, with just one mention of the Common Agricultural Policy and one of the Soil Thematic Strategy, plus a few local or national policies.
Mass flow regulation has synergies with many other ecosystem services, especially water flow regulation (Figure 34). Policies such as planting vegetation on steep slopes to reduce soil erosion, or increasing the organic content of the soil to improve infiltration, will also help with flood protection and carbon storage, as well as reducing loss of soil from fields which is beneficial for farming.
POSITIVE NEGATIVE
BOTH
Figure 34: Interactions between mass flow regulation and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 65
4.8 Water flow regulation (flood protection)
Ecosystems can provide a flood protection service in several ways: (i) Forests and other vegetation intercept precipitation, increasing the rate at which it evaporates before reaching the ground; (ii) Plants suck up water from the ground and release it into the air through transpiration; (iii) Plant roots can increase the rate at which water infiltrates into the ground; (iv) Wetland areas, when not completely saturated, can act as a sink for surface runoff; (v) Coastal vegetation can reduce the speed and height of waves, providing protection against floods during storms and tsunamis.
Flood protection is not always specifically assessed in the literature, but it is possible to use proxy indicators, usually reductions in peak stream flow or (for coastal areas) reductions in wave height. The literature includes examples of all the above mechanisms, but most of the studies found investigate the role of forests in reducing surface runoff, typically through “paired catchment” studies that compare empirical measurements of stream flow in similar-sized forested and unforested catchments. For this reason, it was generally inferred that water flow regulation was provided by the entire community or habitat, with forest being the most commonly discussed, though some studies examine the role of individual species (Figure 35).
The studies are located mainly in Europe (26 studies, of which 13 are in the UK), North America (9 studies) and Asia (10 studies, of which 4 are in China). Almost half of the studies use long-term data sets, to capture stream flow under a range of rainfall conditions.
Positive Mostly positive Neutral Mostly negative Negative
Figure 35: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of water flow regulation (flood protection). AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 66
Community/habitat area is the most commonly identified biotic attribute, with most articles observing that forests tend to reduce peak run-off (by intercepting precipitation, absorbing groundwater through transpiration and improving the infiltration capacity of the soil). As larger trees tend to intercept and absorb more water, stand age and community structure are also cited a number of times, as are above-and below-ground biomass, successional stage, species size/weight and growth rate. Two studies show a positive impact of litter/crop residue quality on rainwater infiltration rates, but in two further studies the impact is found to be unclear. Several studies also examine the role of wetlands, which can provide a sink for floodwater provided that they are not already saturated, and coastal vegetation such as mangroves, which can provide an important physical barrier against high waves generated during storms or tsunamis.
In three studies, however, species abundance is cited as having a negative impact on flood protection as a result of invasive species reducing river channel capacity and trapping sediment. These include mangrove (Kandelia candel) (Lee and Shih, 2004), willow (Erskine and Webb, 2003) and tamarisk (Zavaleta, 2000).
Abiotic factors are often cited in the literature. The impact of precipitation is debated, with recent articles challenging the earlier view that forests provide diminishing levels of protection for higher rainfall events. Several articles illustrate ways in which climate change can reduce the ability of ecosystems to provide flood protection: these include the damaging impact of rising sea levels on coastal wetlands (Wamsley et al., 2010); the degradation of coral reefs due to rising temperatures (Ferrario et al., 2014); and damage to coastal marsh vegetation due to more frequent storms (Möller et al., 2007). Steeper slopes are associated with a more rapid downward movement of water (Nedkov and Burkhard, 2012) and a larger volume of runoff (Lana-Renault et al., 2014).
Human impacts are predominantly negative (17 studies), due to deforestation, soil compaction, or over- grazing of marshes and grassland. Even after abandonment of farmland, long-term negative impacts on soil structure can persist. However, six studies cite positive impacts from restoration of forests, mangroves, wetlands, coral reefs or riparian ecosystems. Three examples of thresholds were found, including an estimate that the flood protection service would start to decline after 20-30% of a catchment area was deforested (Schnorbus and Alila, 2013).
Multiple interactions with other ecosystem services were found (Figure 36). There are many positive synergies, relating to the benefits of forests, wetlands and coastal mangroves for water quality regulation, atmospheric regulation, landscape aesthetics, mass flow regulation, recreation and pollination. In urban areas, green roofs can also contribute to air quality regulation and cooling. However, there can also be trade-offs: raising the water level in wetlands would have a beneficial effect on wildlife and biodiversity but would reduce the flood storage capacity, for example. There is often a negative correlation with the service of freshwater supply, because the absorption of water by forests tends to reduce stream low-flows as well as peak flows, which can exacerbate water shortages in regions where water supply is limited. Several of the studies provided empirical evidence for this, as the catchment studies often measured both peak flows and low flows. However, some authors found that forests can both reduce peak flows and enhance low flows, perhaps through improving infiltration of surface runoff, thus providing a long-term store of groundwater to be released over time. There are also negative interactions with the services of timber and food production, due to forest clearance for timber or agriculture, though some studies cite the establishment of timber plantations on former grassland as a positive benefit in terms of reduction of flood risk.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 67
Perhaps reflecting the many interactions with other services, a variety of different policies were mentioned in the literature, including the Common Agriculture Policy, the Water Framework Directive and the Habitat Directive (each mentioned four times).
POSITIVE NEGATIVE
BOTH
Figure 36: Interactions between water flow regulation (flood protection) and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 68
4.9 Water quality regulation
Most of the 60 studies reviewed for the service of water quality regulation examine the impact of entire communities or habitats (with forests and wetlands being most commonly cited), though 13 studies focus on specific species populations (e.g. common reed) and one assessed the role of two functional groups (submerged macrophytes and zooplankton) (Figure 37 ).
The articles reviewed include large-scale land use studies such as the impact of deforestation on water quality in rivers and lakes; smaller scale experimental studies of the impact of vegetation type on water quality in wetlands; and studies of the impact of riparian buffer zones along streams and rivers. Almost half of the studies (27/60) were located in North America, with 13 in Asia and only eight in Europe. The main indicators were direct measurements of water quality, typically concentrations of various forms of nitrogen and phosphorous and/or suspended sediments, and measurements of nutrient removal rates.
Positive Mostly positive Neutral Mostly negative Negative
Figure 37: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of water quality regulation. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
The review identifies a number of ways in which ecosystems such as forests, wetlands and grassland can improve water quality: (i) Permanent vegetation reduces soil erosion compared to bare ground or farmland; (ii) Vegetation and marshes can trap sediment before it reaches water courses; (iii) Vegetation can absorb and adsorb excess nutrients and other impurities; (iv) Soils can host de-nitrifying bacteria that break down nitrates from fertiliser runoff into harmless nitrogen gas; (v) Vegetation roots can improve infiltration, allowing more impurities to be filtered out by the soil and preventing pollution of adjacent streams and lakes).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 69
Because of the role of vegetation in preventing erosion, physically trapping sediment and absorbing pollution, biotic attributes related to the amount of vegetation are found to have a positive impact. By far the most commonly cited attributes are community/habitat area (41 studies) and presence of a specific community/habitat (27 studies), but other attributes include above- and below-ground biomass, primary productivity, stand age, stem density and species size or weight. Several studies refer to the abundance of highly effective species, such as California bulrush, poplar, willow or seagrass, or functional groups such as mangroves. Ten studies also find an impact from various types of diversity, including species richness, species population diversity, functional richness and functional diversity. The impacts are predominantly positive and seem to be related to the ability of more diverse mixtures to be more productive, and therefore take up more nutrients, due to niche complementarity (i.e. exploitation of a wider range of resources) (Fisher et al., 2009; Cardinale, 2011). However, in two studies the impact is unclear, with Cardinale et al. (2011) stating that there is no evidence that polycultures out-perform the most efficient monocultures. Similarly, Weisner and Thiere (2010) found that wetlands dominated by a less diverse mix of tall, emergent vegetation are more efficient at nitrogen removal. These two studies therefore support the selection effect rather than the niche complementarity effect.
Abiotic factors cited in the literature include the effects of temperature, slope, precipitation, and soil. The relationship with water quality regulation is often unclear or mixed (both positive and negative), and varies between studies. For example, Tomimatsu et al. (2014) find that higher temperatures in summer speed up nitrogen removal in wetlands due to higher plant growth rates, but Rodrigo et al. (2013) find that warmer weather stimulates algal blooms.
A range of human impacts were noted, including negative impacts from deforestation, degradation of wetlands, urbanisation and pollution from agricultural fertilisers, pesticides and sewage. However, some articles cite benefits from restoration of wetlands, construction of artificial wetlands, afforestation, protection of forests and wetlands, and planting of riparian buffer strips.
Twelve articles refer to some kind of threshold, including minimum safe standards for water quality, or percentage land use in catchment areas. Negative impacts on water quality are seen beyond thresholds including 40-60% removal of indigenous vegetation, 10% impervious surfaces, 50% of land use in a catchment for agriculture, and 20% urbanisation. One study finds that minimum levels of richness, evenness and diversity are required for a microbial community to perform denitrification. However, Cardinale et al. (2011) found a saturation effect whereby ecosystem function ceased to improve after a certain level of species richness.
Links to other ecosystem services are mentioned in 19 studies, comprising a varied mix of both positive and negative interactions (Figure 38). For example, fertilisers from timber and crop production are a cause of water pollution, but cleaner water supplies are also beneficial for irrigation of food crops such as rice, and riparian buffer strips can be a source of sustainably harvested timber. Similarly, carbon is sequestered in forests and in wetland soils, but wetlands can also be a source of methane and nitrous oxide, two powerful greenhouse gases. However, forests and wetlands also have mainly positive benefits for fishing, mass flow regulation, water flow regulation and aesthetic landscapes.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 70
POSITIVE NEGATIVE
BOTH
Figure 38: Interactions between water quality regulation and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 71
4.10 Pollination
The studies spanned a wide range of locations from all over the world, but most were conducted in Europe (36) or North America (19). Most of the articles on pollination refer either to the functional group of pollinating insects, or to specific species populations (e.g. honey bees, bumble bees, hoverflies) (Figure 39). It is difficult to measure pollination effectiveness directly, so a range of proxy indicators were used, including crop yield, fruit or seed set, the number of pollinating insects, the percentage of natural land cover, or distance of agricultural fields to (semi)natural habitats.
The most commonly cited biotic attribute is the abundance of a specific pollinator species (30 counts), with the abundance of a functional group (of pollinators) also being important (20 counts). However, the presence of suitable habitat for pollinators is also of vital importance, with community or habitat area being cited 11 times and community/habitat structure 27 times. Many articles mentioned that a diverse, natural habitat with a variety of flowering plants was needed to support populations of pollinators. Pollinating services and the diversity of pollinators tended to decline with increasing distance from natural habitat (e.g. Carvalheiro et al., 2010). Behavioural traits such as foraging distance, flight range, pollinator size, and bee tongue length (Bommarco et al., 2011) are also important in determining which pollinators can access certain flowers.
Positive Mostly positive Neutral Mostly negative Negative Figure 39: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of pollination. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
The beneficial impact of diversity and richness appears to be more important for pollination than for any of the other ecosystem services studied here. Functional richness is mentioned 14 times, functional diversity 9 times, species richness 15 times and species population diversity 9 times. This can refer either to the diversity of the pollinators, or to the diversity of the plant species in the habitats needed to sustain the pollinators. The impact of pollinator diversity is mainly positive, with various studies finding that more diverse populations of pollinators increased seed production (e.g. Albrecht et al., 2007), coffee fruit set (e.g. Vergara and Badano, 2009) and pollination efficiency (Hoehn et al., 2008; Balvanera et al., 2005). This is generally because different species visit different plants (Winfree et al., 2008) or visit different areas and at different times (Hoehn et al., 2008), so that a more diverse community provides a more complete
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 72 pollination service. Many articles also discuss the need for plant species richness and functional diversity in the surrounding habitat, in order to support populations of pollinators (e.g. Holzschuh et al., 2011). In fact, the strong relationship between plant diversity and pollinator diversity is demonstrated by Batary et al. (2010) who find that the richness of insect-pollinated plant species is directly correlated with bee species richness in three different European countries. The relationship works both ways, with Fontaine et al. (2006) showing that after two years, plant communities pollinated by more functionally diverse pollinator assemblages contained about 50% more species than those pollinated by less diverse assemblages. However, there are also examples of negative impacts on pollination, associated with the introduction of honey bees which compete with native bees (Shavit et al., 2009; Badano and Vergara, 2011).
Abiotic factors such as temperature and wind speed are mentioned in a number of journal articles, but the direction of impact on pollination is usually unclear. Light intensity was found to have a positive influence (e.g. Munyuli, 2012) and there is also a negative correlation between the distance to natural habitats and pollinator abundance and diversity (e.g. Klein et al., 2003).
Human input was found to have a negative impact on pollination in 16 studies, due to loss or degradation of natural habitat and the use of pesticides and fertilisers. However, positive impacts are found from the effect of managed beehives in agricultural areas, or the conservation and protection of natural or semi- natural habitats in National Parks or urban gardens. Less intensive farming (e.g. extensive, organic farming) supports pollinator abundance and diversity and thus benefits pollination services. Several policies were mentioned in connection with this, including agri-environment schemes and organic farming policy. Only one threshold was found: a point in ecosystems where additional bee visits no longer increases production/productivity.
The review found 17 positive interactions with food production, for obvious reasons, but surprisingly there were no clear links to other ecosystem services including pest regulation (Figure 40). However, links between these two services were identified by the pest regulation review (see below).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 73
POSITIVE NEGATIVE
BOTH
Figure 40: Interactions between pollination and other ecosystem services.
4.11 Pest Regulation
Pest regulation is mainly provided by functional groups, e.g. spiders, wasps or beetles, although 17 studies investigate specific species populations (both single and multiple) and six studies focus on an entire habitat/community (typically farmland) (Figure 41). The most commonly cited indicator is pest density, though many studies use the population or richness of pest predators as a proxy. Some articles also cite crop yield, pest damage or the cost of pest control.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 74
This service is affected by a wide range of biotic attributes, with the linkages found to be predominantly positive. The most commonly cited attributes are community/habitat presence, area and structure, because many articles focus on the importance of natural or semi-natural habitats for supporting populations of pest predators. The studies find that pest predation is positively influenced by complex habitats (e.g. Bianchi et al., 2006); by crop lands interspersed with and/or surrounded by semi-natural habitat (e.g. Letourneau et al., 2012); by good connectivity between patches (e.g. Boccaccio and Petacchi, 2009); and by diverse plant communities (e.g. Drapela et al., 2008). Habitat management can therefore influence predator density through modifications such as thicker ground cover (Colloff et al., 2013) or creation of semi-natural field edges (Krauss et al., 2011).
Other important attributes include the presence and abundance of a specific functional group (i.e. predators), and species abundance. A number of studies found that species richness, functional richness and functional diversity are important, though while several find that land use management can enhance predator diversity, fewer demonstrate that predator diversity reduces pest activity. Those that do attribute this to niche complementarity, with different predators attacking different prey sizes, life stages, population densities and behaviour (e.g. flying vs. ground dwelling), but other studies find no effect of diversity. Predator behavioural traits are also cited, such as the ability to disperse over long distances (e.g. Öberg, 2007) or the ability to form aggregations during dormancy so that they can hatch en masse and attack prey (Iperti, 1999). These linkages are predominantly positive.
A small range of abiotic factors are discussed in the literature, with temperature and precipitation being the most common, although the effect is variable.
Positive Mostly positive Neutral Mostly negative Negative
Figure 41: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of pest regulation. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
The review found a mix of positive and negative impacts from human input and management, with examples of negative impacts mainly concerning habitat loss, and positive impacts including habitat improvement such as planting wildflower patches, mulching, thick groundcover between orchard rows, and
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 75 less intensive rotations. One study identified a threshold related to the percentage of land covered by annual crops, finding that pest parasitism rates dropped precipitously when annual crop cover exceeded species-specific thresholds of 38% and 51% (Letourneau et al., 2012). However, only one study discusses the role of policy: Woodcock et al. (2014) discuss the potential for agri-environment schemes to increase the robustness of ecosystem services provided by pest control and pollination by conserving and creating semi-natural habitats.
Interactions with several other ecosystem services were identified: pollination, food production, recreation and four classified as ‘other’ (Figure 42).
POSITIVE NEGATIVE
BOTH
Figure 42: Interactions between pest regulation and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 76
The positive interactions with food production and pollination are obvious, as pest regulation reduces damage to food crops, and the creation of semi-natural habitats to boost pest predators will also have benefits for pollinating insects (e.g. bees, hoverflies). Other services include eco-tourism opportunities (recreation), and the benefits of shade trees for climate adaptation. However, there can also be some trade-offs: for example, shade trees can reduce crop yield; ground cover between crop rows can compete with crops for nutrients and water and predators themselves can sometimes damage crops.
4.12 Recreation (species-based)
Species-based recreation is generally based in national parks or nature reserves, and includes activities such as wildlife viewing, hunting or fishing. Wildlife viewing is often assessed based on tourist surveys, while hunting and fishing may be assessed using evidence such as catch per unit effort or exploitation rates. A very wide variety of indicators were used, some related to humans, such as visitor numbers, distance to protected area, cost per trip, viewing preferences or willingness to pay for an experience, and some related to the ecosystem itself, such as area of protected land, bird species richness or fish abundance.
Not surprisingly, species-based recreation is found to be predominantly provided by specific species populations, though in 18 studies the focus is an entire community/habitat (e.g. coral reefs, nature reserves or hunting areas) (Figure 43).
Positive Mostly positive Neutral Mostly negative Negative
Figure 43: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of species-based recreation. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
Biotic attributes are found to affect this service in a predominantly positive way, with the only negative impacts arising from mortality rate. The most important biotic attributes are the presence and abundance of specific species. These include charismatic species such as whales and dolphins for marine eco-tourism, or large mammals such as lions, tigers and elephants for land-based eco-tourism, as well as mammals such as deer for hunting, and fish such as salmon and trout for recreational fishing. Species size or weight can also be significant, with visitors, fishermen and hunters often expressing a preference for larger species such as sharks and lions. Species richness and diversity are also valued. For example, Lindsey et al. (2007) find that tourists in South Africa consider functional group diversity (in this case, the variety of large
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 77 mammals) to be the most important feature of their wildlife viewing experience, and Ruiz-Frau et al. (2013) find that marine biodiversity is important for scuba divers.
A number of abiotic factors are cited in the literature. Weather-related factors such as precipitation and temperature are often cited, especially for fishing (e.g. Smallwood et al., 2006). These have mixed effects, with extreme conditions found to negatively affect recreation (Cooke and Suski, 2005).
As most of the studies were located in actively managed or protected areas such as national parks, all of the studies were classified as having some kind of human management. This typically had both positive and negative effects: the main impact was generally positive, due to protection of the wildlife in the area, but there could also be negative impacts from visitors such as disturbance to wildlife. As a result, many of the studies discussed methods of minimising these impacts, such as regulating fish catch, restricting hunting or limiting visitor numbers. Good governance, including the active participation of local communities, was shown to be important to maximise conservation benefits, enforce regulations and minimise problems such as poaching. There were several mentions of local or national regulations applicable to hunting, fishing and national park management, but the EU Habitats Directive was only mentioned twice.
Few interactions with other ecosystem services were identified (Figure 44). There are positive links to food production, freshwater fishing (where recreational species also have a commercial value), pest regulation (through hunting of nuisance animals such as feral pigs, and the protection of pest predators such as bats and dragonflies) and pollination (again by bats). Figure 26
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 78
POSITIVE NEGATIVE
BOTH
Figure 44: Interactions between species-based recreation and other ecosystem services (unclear interactions omitted).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 79
4.13 Aesthetic landscapes
Almost all studies consist of surveys of individual preferences for different landscapes, often using photographs, and almost all are located in Europe or North America. The most common indicators are scores or rankings for perceived aesthetic value or scenic beauty. In all studies, the provider is the entire habitat (community is less relevant for this service), such as a woodland, ornamental park or mixed rural landscape (Figure 45).
The effect of biotic attributes is predominantly positive, with community-level attributes most often cited. The presence of a particular habitat is cited 15 times, with forests and water features being most often mentioned. There is evidence that people value urban trees and green space (e.g. Kaplan, 2007). Proximity of forests and parks has a significant positive effect on housing prices and households are willing to pay to enlarge nature areas. Habitat structure is the most frequently cited attribute, appearing in 48 of the 61 studies, with the term ‘structure’ being interpreted as covering a broad range of characteristics including landscape diversity and complexity, vegetation density, naturalness and uniqueness. (For this ecosystem service, the attribute of ‘landscape diversity’ is included under ‘community/habitat structure’ and is not classified as a separate attribute). Many studies find a preference for wilder, more complex, more natural landscapes (e.g. Acar and Sakici, 2008; Heyman, 2012; Daniel et al., 2012). However, the review finds important differences between cultural groups. Farmers and certain ethnic groups may prefer a more open, managed landscape whereas most user groups tend to prefer a more natural landscape. Some studies also find that people in developed countries tend to appreciate natural landscapes, but some groups originating from other countries (e.g. black Americans or Islamic immigrants) appear to prefer less dense vegetation and more open landscapes, with man-made elements.
Positive Mostly positive Neutral Mostly negative Negative
Figure 45: Network diagram showing linkages between ESPs, biotic attributes and abiotic factors for the service of aesthetic landscapes. AF = abiotic factors. Line thickness indicates the number of studies showing a given link.
Species-level attributes are mentioned in 13 studies. Most people appreciate the presence of a rich variety of species, and the presence of rare or remarkable species, such as sea turtles in Barbados (Schuhmann et
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 80 al., 2013). Other biotic attributes cited in the literature include subjective terms such as a feeling of ‘naturalness’ or presence of ancestral spirits.
A range of abiotic factors also appear to affect landscape aesthetics including slope, water availability, geology (relief), and water quality. Five studies cite positive effects from water availability (i.e. the presence of water features), and three found that poor water quality has a negative effect on aesthetic value. There was also a general appreciation of mountainous areas and varied topography (slope and geology). Other abiotic factors include the travel distance to reach the area, the weather, and the presence of hiking trails.
Human intervention in the form of deforestation, intensive agriculture and built infrastructure (roads, car parks and large buildings) is generally ranked as a negative impact (16 studies). However, positive appreciation for human influences on the landscape was found in six studies, including traditional agriculture, cultural buildings and hiking trails for access.
Few clear interactions with other ecosystem services were identified in this review (Figure 46). There are mixed links with food production, depending on the reaction of different groups of people to different types of agricultural landscape. Figure 26 Timber production is positively correlated to aesthetic value once, in a study where forestry was considered to be part of local cultural heritage, though there are seven unclear correlations (not shown). Freshwater fishing is appreciated by some groups, but the general opinion about it is unclear.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 81
POSITIVE NEGATIVE
BOTH
Figure 46: Interactions between aesthetic landscapes and other ecosystem services (unclear interactions omitted).
5 Discussion
This review of almost 700 journal articles represents a thorough investigation of the current state of the literature on the links between natural capital and ecosystem services. The review has collected a large amount of useful data, but has also revealed significant gaps in the literature.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 82
5.1 Coverage of regions and ecosystems
There seems to be a significant bias in the geographical coverage, with 59% of the study sites identified in the literature review being located in Europe and North America. Only 7% are in Africa, 14% in Asia, 9% in Oceania and 5% in South America. However, this might be in part related to the focus on journal articles written in English. Given the importance of natural capital in underpinning essential ecosystem services worldwide, this seems to indicate a need for a wider geographical spread of research effort. Although there is no indication of a major variation in the influence of biotic attributes across continents, there are certainly examples that illustrate the importance of local factors when evaluating ecosystem service provision. For example, forest plantations seem to have a predominantly negative effect on freshwater provision in hot, dry climates such as Australia, but cloud forests in South America have a beneficial impact on water supply.
There was a reasonably broad spread of ecosystems, dominated by forests (30%), cropland (20%) and grassland (14%), followed by urban (9%), rivers and lakes (9%) and wetlands (8%). Though some ecosystems were discussed very rarely, this does not necessarily indicate a research gap: it may be partly due to a lower level of ecosystem service provision by some categories such as sparsely vegetated land, and partly to the choice of services for this review, which accounts for the low number of studies addressing marine ecosystems.
There was very little information on the impact of the condition of ecosystems on their ability to provide ecosystem services, despite the obvious threat posed by the widespread degradation of natural capital identified in the Millennium Ecosystem Assessment (MA, 2005). In particular, information on thresholds where further natural capital loss would severely compromise ecosystem functioning and service delivery is very limited, and even the additional search using the term “threshold” failed to find a significant body of research in this area. The thresholds identified by the review focused mainly on a limited number of services such as food production, air quality regulation, freshwater provision and water quality regulation, and thresholds relating to other services including timber production, atmospheric regulation, freshwater fishing, aesthetic landscapes, pest regulation and pollination are rarely examined. However, the limited information found clearly demonstrates the potential importance of this issue. For example, Nosetto et al. (2005) suggest that deforestation above a threshold of 20% would begin to significantly affect cloud cover in the Amazon Basin, thus reducing the provision of freshwater. This is clearly an area where more research is needed.
5.2 Linkages between natural capital and ecosystem services
The review shows that the literature directly addressing linkages between natural capital, ecosystem function and ecosystem services is quite limited. Relatively few articles were found that explicitly identify ecosystem service providers and the ways in which biotic and abiotic attributes affect ecosystem service provision. Indeed, as the concept of “ecosystem services” is relatively new, many articles did not use terms such as “ecosystem service provider” or “biotic attribute”, and the reviewer used their own judgement to identify the provider and to infer links to attributes. This lack of ecosystem service related terminology in the literature also led to the possibility of alternative interpretations by different reviewers. For example, for an article examining three different species of pollinating insects, one reviewer might classify the ecosystem service provider as being “multiple species” and another might classify it as representing the
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 83
“functional group” of pollinators. To minimise these effects, a quality control check was undertaken on the final database by a single researcher. However, it is still important not to over-interpret apparent differences between ecosystem services in the database.
Despite these data gaps, and the possibility of minor inconsistencies in the database, a number of robust conclusions have emerged from the review. Firstly, it is very clear that most of the biotic attributes identified in the review have a beneficial impact on ecosystem service provision. The main exceptions are mortality rate; the impact of non-native species; the impact of plantations on freshwater supply; and a limited number of examples where monocultures provide a better service than polycultures (e.g. in eight of the studies on timber plantations).
Many studies look only at the entire area of a habitat such as forest or wetland, or the abundance of a particular species or functional group. Other biotic attributes are rarely studied, and these often have to be inferred by the reviewer rather than being explicitly mentioned. Nevertheless, it is possible to identify several distinct clusters of attributes that provide ecosystem services in different ways.
The most commonly identified cluster of attributes relates to the physical amount of vegetation within an ecosystem: habitat area, primary productivity, above- and below-ground biomass, stem density, species size/weight, growth rate, and successional stage. These tend to have beneficial impacts on a particular group of services: atmospheric regulation, water flow regulation, mass flow regulation, water quality regulation and air quality regulation. The reasons for this are fairly obvious. For atmospheric regulation, the amount of carbon stored in an ecosystem is directly proportional to biomass, so larger and older trees provide a better service. For water flow regulation, trees intercept rainwater and absorb groundwater through evapo-transpiration, and coastal vegetation provides a physical barrier to storm waves and tsunamis. For mass flow regulation, vegetation roots stabilize soil and thus prevent erosion, especially on steep slopes, and for water quality regulation the area of vegetated land, especially forests or wetlands, is important for filtering water, absorbing excess nitrogen and preventing runoff of polluted water into streams. For air quality regulation the leaves of trees and bushes trap and absorb pollution. The major exception was for freshwater provision, where this cluster of attributes can have a negative effect, with 20 studies finding that forests can reduce stream flows downstream. However, this depends on the type of forest and the local hydrology: the negative effects are mainly related to plantations of fast-growing species such as pine and eucalyptus in dry climates, but there are also five positive examples where cloud forests capture moisture from the air, and old growth forests improve groundwater infiltration.
The second cluster focuses on the presence or abundance of particular species or functional groups. This is particularly important for the provision of freshwater fishing, timber, species-based recreation, pollination and pest regulation. Again, the reasons are fairly obvious as these services rely on the existence of commercially important species of fish and timber, species of interest to eco-tourists, and species or functional groups of pollinating insects such as bees, or pest predators such as wasps or hoverflies. A number of species-specific traits are significant in determining which species are most effective for the provision of certain services, especially species size (for species-based recreation and fishing), and behaviour for pest predators and pollinators.
Although articles discussing other biotic attributes are less common, there is a growing body of research into the effect of a cluster of diversity-related indicators. These include species richness, species population
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 84 diversity, functional richness and functional diversity, but also structural complexity and landscape diversity. These studies illustrate the two classic mechanisms through which diversity can contribute to ecosystem services. The first is niche complementarity, where efficiency is maximised because different organisms occupy different ecological niches. For example, carbon storage can be greater in a forest where a mixture of different trees and shrubs with a range of heights and root depths maximises use of the available light, water and nutrients. The second mechanism is the selection effect, where the presence of a wide range of different species improves the chance that one of them will be a high performer, such as a large species of tree (good for carbon storage) or a particularly effective pest predator. Interestingly, both of these mechanisms are shown to be important in different circumstances, sometimes for the same ecosystem service (see the reports on atmospheric regulation and water quality regulation in Annex 5). Surprisingly few articles mention the role of diversity in underpinning the resilience of the ecosystem to environmental change, but this was not a major focus when selecting the search terms for the literature review. It is possible that more literature could be identified using a specific search including the term “resilience”.
Because diversity can lead to more biologically productive ecosystems, it tends to benefit services that depend on the physical amount of biomass, especially timber production, freshwater fishing, atmospheric regulation, water flow regulation and water quality regulation. However, there is a separate group of services that benefit from diversity itself, regardless of its impact on productivity. Pollination and pest regulation services are greatly enhanced by the existence of a community consisting of a variety of species that can pollinate different flower shapes and sizes, prey on different species, or are active at different seasons or different times of day. Diversity also benefits species-based recreation where, for example, eco- tourists like to see a wide variety of wildlife.
At the landscape level, habitat diversity is also of great importance for maintaining effective populations of pollinators and pest predators close to cultivated crops, as these organisms depend on natural or semi- natural habitats to provide alternative sources of food and shelter when crops are harvested. Structural diversity and complexity is also important to provide a variety of feeding, breeding and over-wintering sites for beneficial insects, and also improves landscape aesthetics, atmospheric regulation, water quality regulation and water flow regulation.
Abiotic factors also affect the delivery of ecosystem services, but the impact varies depending on the context, with examples of both positive and negative effects as well as a large number of studies where the direction of the impact is unclear. The comment fields in the review database were important for this analysis, as it is not possible to assign a consistent direction of impact to a factor such as “soil”, “geology” or “slope”. These also reveal differences of approach between reviewers, with some reviewers recording every mention of an abiotic factor (contributing to the large number of “unclear” studies) whereas others recorded only cases where the factor was shown to affect the ability of the ecosystem to deliver the service.
The review showed that ecosystems are vulnerable to climatic factors such as precipitation and temperature, and exposure to conditions outside a given range can diminish their ability to provide ecosystem services. The analysis of abiotic factors is therefore useful to help to predict how the delivery of ecosystem services might be affected by future environmental change. In particular, a number of studies cite examples of how climate change could impair the delivery of ecosystem services in the future. Food
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 85 production seems particularly vulnerable, as crop yields could be reduced due to higher temperatures, reduced water availability, desertification and salination of groundwater supplies. Higher temperatures can also increase fish mortality, increase emissions of volatile organic compounds from vegetation, degrade coral reefs (which provide flood protection), cause more forest fires (thus releasing carbon into the atmosphere) and increase algal blooms (bad for water quality and fishing). Rising sea levels, linked to global warming, can inundate wetlands, leading to loss of water flow and water quality regulation services. Droughts (recorded as decreased water availability or precipitation) can dry out wetlands, leading to loss of services including atmospheric regulation, and they can also damage the ability of trees to provide carbon storage and air quality regulation. Higher temperatures are also linked to more evaporation of ground and surface water, leading to reduced freshwater provision especially when this increases the salinity of groundwater supplies, and to reduced snow cover, decreasing the supply of freshwater from meltwater. However there were isolated examples of benefits from climate change, including increased crop and timber production and carbon storage due to enhanced plant growth at higher altitudes or more northern latitudes (i.e. in what were formerly colder climates), and a CO2 fertilisation effect that could boost crop yields in some cases.
In view of the likely increase in extreme weather events as a result of climate change, it is interesting to note the discussion in the water flow regulation report (Annex 5). A number of studies appear to show that the ability of forests to provide flood protection decreases, eventually to zero, for more extreme rainfall events. However, recent work queries this interpretation of the data and claims that forests reduce both the frequency of floods of a certain size, and the size of downstream peak flows, for all rainfall events.
In some studies, it appears that ecosystems deliver greater benefits in more testing circumstances. For example, the benefits of vegetation in stabilising soil appear to be greater on steeper slopes (Mandal et al., 2013), and wetland vegetation absorbs more excess nitrogen when concentrations in the water are higher (Christen and Dalgaard, 2013).
5.3 The role of human intervention in protecting and enhancing ecosystem services
Human activities are shown to have a range of positive and negative impacts, and many studies cite a mix of both. Although ecosystems are suffering widespread degradation from activities such as intensive agriculture, deforestation and urban development, there are also many examples of ways in which ecosystem services can be enhanced through protection, restoration and sustainable management of ecosystems. For example, organic agriculture can be beneficial for pollination and pest regulation; forests can be planted to provide mass flow regulation; and wetland restoration can improve water quality regulation. There are also examples of trade-offs and negative feedbacks within services, such as the potential damage and disturbance to wildlife caused by excessive eco-tourism activities, though there are opportunities to minimise this through careful regulation and sustainable management.
The role of human intervention is particularly important because of the significant level of interactions between ecosystem services. Interactions are identified in 40% of studies, though these are not always investigated explicitly. The majority of interactions (56%) are positive, and this highlights studies where ecosystems have multiple benefits. For example, forests can provide atmospheric regulation, water flow regulation, mass flow regulation and aesthetic benefits. However, there are also some important trade-offs
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 86 between services, especially between provisioning services and regulating or cultural services. Management that benefits one service often has negative impacts on another. Commonly cited examples include fertiliser application, which benefits food and timber production, but has negative impacts on water quality and freshwater fishing. Similarly, harvesting forests for timber has a negative impact on atmospheric regulation, water flow regulation and (for clear-cutting) aesthetic landscapes. Sometimes management has short-term benefits, but may have adverse consequences in the long-term. For example, intensive farming may produce high crop yields but it could also cause a long-term decline in populations of pollinators and pest predators, threatening future food security. Improved analysis of these interactions could help decision-makers to develop management strategies that exploit synergies and balance trade-offs more effectively.
6 Conclusions
6.1 Outputs for OpenNESS
This review has synthesised a large and varied mass of literature into a single database with a simple and logical structure, enabling a detailed analysis of the way in which natural capital stocks (both biotic and abiotic) contribute to ecosystem service flows.
This knowledge base can now be used to improve the design and selection of data for the ecosystem service mapping and modelling methods being developed and applied within the OpenNESS case studies (Task 3.2), and the accompanying guidance being developed to assist the case studies (Task 3.4). Ultimately, a refined version of these tools, datasets and guidance will form part of the Oppla web-platform (www.oppla.eu) to make them accessible for future application by the wider user community (Work Package 6).
Useful outputs of the database and analysis for the rest of the OpenNESS project include: A simple and effective typology for classifying ecosystem service providers, biotic and abiotic attributes and the links and interactions between them; A useful set of indicators of ecosystem service supply, which can now be refined from the original JRC indicator list as a result of this exercise; Information on the contribution of different ecosystems and land cover types to different ecosystem services; Identification of the multiple benefits and synergies provided by many ecosystems, especially forests, which contribute to all 13 ecosystem services investigated here; Identification of potential negative impacts, interactions and trade-offs between and within ecosystem services; Information on the mechanisms by which natural capital contributes to ecosystem services, including habitat area, species abundance, species traits and the role of functional diversity; Information on the way in which abiotic factors affect ecosystem condition and the delivery of ecosystem services, especially concerning the threat posed by future climate change; Preliminary indication of the importance of thresholds for ecosystem service supply, e.g. percentage of forest area in a catchment required to deliver flood protection or meet water quality standards;
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 87
Examples of ways in which human activities can improve or protect ecosystem service delivery; Awareness of areas where data is lacking, which can be used to guide future research (see below).
The review emphasises the importance of conserving biodiversity and natural capital in order to continue to deliver robust and resilient ecosystem services in a changing world. It also provides valuable information on the effectiveness of different species, habitats and management techniques – for example, the selection of wetland species for excess nutrient removal, the value of old-growth forests for carbon storage, or the establishment of semi-natural habitats around crops to boost populations of beneficial insects.
Using this information base, local decision-makers can identify opportunities to protect and enhance vital ecosystem services in their area, maximising synergies and minimising trade-offs between services, whilst national and regional policy-makers can design effective policies to facilitate these goals at a wider scale. The need for careful policy design is particularly urgent in order to balance trade-offs between provisioning and regulating or cultural services, and to protect ecosystems against future climate change and urban development. To preserve a balanced mix of ecosystem services as well as a healthy underlying level of biodiversity to sustain future services, it will be necessary to protect certain key habitats such as wetlands and old-growth forests from over-exploitation.
6.2 Data gaps and recommendations for future work
As discussed above, the review identified a number of gaps in the literature concerning the relationship between natural capital and ecosystem services. To address the needs of decision-makers, future research should aim to: Extend geographical coverage to include more studies relevant to Africa, Asia, South America and Oceania; Address the link between biodiversity and ecosystem services more explicitly, especially through studies that examine the role of functional diversity in delivering services and providing resilience to change; Examine the impact of ecosystem condition on ecosystem service provision; Investigate potential thresholds in the biophysical state of ecosystems, beyond which the future delivery of ecosystem services could be compromised; Study interactions between ecosystem services, including synergies and trade-offs and the implications for decision-makers, for example with respect to land use management, or options to minimise negative interactions such as pollution or over-exploitation of resources ; Evaluate ways in which human management and related policies can protect and enhance ecosystem services and build resilience to change.
Finally, the review yielded a number of useful recommendations for improvements to the OpenNESS literature review methodology. These have been summarised in an internal project document which is available on request. They include potential enhancements to the database and additional guidance for reviewers to improve the consistency of the literature review. One major area for improvement relates to the choice of indicators, as it is clear that the JRC indicators for mapping ecosystem services (Egoh et al., 2012) did not adequately cover all of the ecosystem services reviewed, especially for the services freshwater fishing, food provision, and mass flow regulation. Some of the services failed to cite any of the
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 88 indicators found in the JRC review, and a number of important additional indicators were identified which could be included in future work.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 89
Annex 1: Supplementary literature search on ecosystem services and biophysical thresholds
Erik Stange, Norwegian Institute for Nature Research - NINA
The literature review presented in this deliverable included examination of how selected studies address the possible existence of thresholds. Reviewers collected information on legal boundaries, safe minimum standards and biophysical thresholds. Only 63 studies (out of 709) explicitly mentioned the existence of thresholds for specific ecosystem services (some studies mentioned more than one threshold, so 68 thresholds were identified in total). Of these, 50 studies addressed biophysical thresholds (53 thresholds). Two-thirds of these address ecosystem services related to water: 17 investigated freshwater provision, nine investigated water quality regulation, and five investigated water flow regulation (flood protection). Biophysical thresholds were also mentioned in eight studies on food production, and between one and three examples were found for each of the other nine services reviewed.
Biophysical thresholds pertaining to ecosystem service delivery are a topic of great interest because their existence essentially defines the limits for sustainable use of natural capital. Thresholds are a subcategory of many non-linear relationships that are common within ecological systems. The simplest definition of a threshold or discontinuity is a rapid state change occurring as a consequence of a smooth and continuous change in an independent variable (Luck, 2005). In the context of resource use and management, thresholds are often associated with tipping points and their subsequent regime shifts: concepts that involve large abrupt and persistent changes in the structure and function of a system (Biggs, 2009). There is a vast literature on the presence and dynamics of thresholds in ecology and socio-ecological systems, and thus we reasoned that the 50 articles found in the main literature review were only a very small subset of the articles in the literature that addressed how thresholds in a system’s biophysical attributes influence ecosystem service delivery.
To determine the information available in the existing scientific literature regarding threshold relationships between an ecosystems’ biophysical attributes and the ecosystem services they provide, a supplementary search of articles included in the ISI web of science online database was conducted, using “ecosystem AND service AND threshold” as subject search terms. These search criteria resulted in 264 articles. The abstracts of these articles was first examined and then the main text to ascertain whether these articles revealed any relevant information; this resulted in 80 articles which treated the concept of thresholds in a way that was relevant to ecosystem services and natural capital. It was never intended that this simplistic search would produce an exhaustive list of all instances in the literature where researchers pursued questions pertaining to biophysical thresholds and ecosystem processes and functions. The initial result (264 hits) is a full order of magnitude lower than the results found using “ecological AND threshold” as subject search terms (2948 hits). Furthermore, a search using “ecosystem AND threshold” yielded even a greater number: 3254 hits.
Even though there is a vast literature on ecological thresholds, little has been connected to ecosystem services as yet. Nevertheless, the thresholds addressed in these thousands of articles have almost certain relevancy in the context of ecosystem services, even if the authors did not address ecosystem services explicitly. “Ecosystem service” is also a recently coined term (Braat and De Groot, 2012), and this is reflected in the publication dates of the articles identified in our supplementary review. With the exception
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 90 of Carpenter et al. (1999), all articles were published within the past 10 years (Figure 1). The most comprehensive effort to date connecting the broader literature on thresholds with ecosystem service delivery is a database maintained by The Resilience Alliance titled “Thresholds and Regime Shifts in Ecological and Social-Ecological Systems” (Resilience Alliance and Santa Fe Institute, 2004). Presently the database includes 278 references which address 103 examples of thresholds which all have relevance to ecosystem service delivery. The examples are also categorised according to whether a state change has occurred (demonstrated) or is merely proposed.
Figure 1: Publication dates for the 80 articles reviewed for addressing biophysical thresholds relevant to ecosystem service provision.
It is also noteworthy that only one article found through the “ecosystem AND service AND threshold” ISI search was among the 50 articles identified through the main literature review. One possible explanation for this discrepancy is that the search terms used in the supplementary review are more restrictive than the criteria reviewers used for identifying biophysical thresholds in the main reveiw. “Threshold” is often used interchangeably with “limit”, even though the two terms are not completely synonymous (a threshold is a type of limit, but not all limits can be considered thresholds according to the definition provided earlier). An ISI search that used “ecosystem AND service AND limit” yielded 1136 results. The 50 articles identified in the main review may have either addressed biophysical limits that could also be thresholds without the articles’ authors referring to them as thresholds, or the use of the word “threshold” could have been buried in the manuscript sufficiently that it was undetectable by the ISI search engine.
Only 28 of the articles reviewed in this supplementary search represented original research that attempted to quantify or otherwise assess thresholds that pertained to either biophysical attributes or the associated ecosystem services. Many articles addressed thresholds and ecosystem services either by alluding to the possible or probable existence of thresholds in specific systems, by identifying them as a component of dynamics in a modelled system outside of the context of specific services, or by calling for increased research that would be capable of identifying real-world thresholds. The remaining articles (26) were review or concepts articles that referenced data from previous studies that identified thresholds. Quantifying the number of articles referenced in these review articles was outside of the scope of this effort, but very few cited articles that had also turned up in our supplementary search were found,
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 91 indicating that the cited research regarding a given biophysical threshold did not explicitly utilise “ecosystem services” as a term.
The 28 articles that attempted to quantify thresholds focused on an array of ecosystem types: rangeland and grasslands (6 articles), agro-ecosystems (4), rivers/streams/wetlands (4), marine (3), urban (3), estuarine (2), alpine (2), and various additional terrestrial ecosystems. Studies were located on all continents, with focal areas ranging from small plots of unspecified size at a single field station to areas greater than 350,000 km2. The biophysical attributes were also quite varied, with landscape complexity or fragmentation being the only attribute found in more than one article. The drivers affecting the attributes were less varied. Land use and management, overharvesting, overgrazing, urbanisation and development were identified as the drivers. Only two articles identified climate change as moving biophysical attributes towards or past a threshold, and only one article cited invasive species.
Of the articles that are able to identify threshold values for the biophysical attributes, many provide evidence that the systems have either partially or fully passed a threshold. Two articles indicate that thresholds have already been passed, and 12 describe heterogeneous systems where thresholds have been passed in some places but not others. An additional seven articles provide evidence that their systems are approaching thresholds.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 92
Annex 2: Analysis of variation of biotic attributes by location and spatial scale
Analysis of whether biotic attributes vary depending on spatial scale and location is limited by the small number of database entries for the larger scales, which creates a bias. For example, 42 studies on mass flow regulation are at the local scale, with only one each at continental or sub-continental scale and three each at national or sub-national scale. Due to the larger number of studies at small scales, it is more likely that a large range of attributes and ESPs have been recorded by reviewers, whereas for larger scales the substantially lower number of studies entered into the database means it is less likely that such a large range of attributes and ESPs have been recorded.
Despite this limitation, one difference between spatial scales is apparent (though perhaps not surprising): studies at large scales (i.e. global, continental and national) tend to focus more on entire communities or habitats, and to study community and habitat level attributes. In contrast, local scale studies tend to investigate a broad range of ESPs including specific species and functional groups as well as entire communities/habitats (Figure 1). This is because smaller scale studies (often including field or laboratory experiments) are capable of detailed comparisons of particular species or functional groups, which is more difficult at larger scales.
For example, local and sub-national scale studies of atmospheric regulation investigate carbon storage by specific species populations (e.g. pine plantations), functional groups (e.g. grasses), and entire communities (e.g. comparing pasture and forest). However, at the national and global scales, only forest - a single community/habitat - is considered in the literature reviewed. This tendency for communities to be the only ESP at larger scales applies to most of the ecosystem services, so it is likely that this is not simply an artefact of the low number of studies at large scales. The main exceptions to this are for food provision, freshwater fishing and pollination, where the focus at all scales is on specific species populations, i.e. different food crops, fish species or pollinating insects, or, for pollination, functional groups (e.g. flower-visiting insects). Figure 2 shows the distribution of ESP by spatial scale with these three species-oriented services omitted: the tendency for larger scales to focus more on the community as ESP is more pronounced.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 93
Local Entire community or habitat
Two or more communities or Sub-national habitats Dominant community National Single functional group
Continental Two or more functional groups
Single specific species population Global
Two or more specific species 0% 20% 40% 60% 80% 100% populations % of studies
Figure 1: ESP by spatial scale (continental and sub-continental scales are combined due to the small number of entries at sub-continental scale)
Local Entire community or habitat
Two or more communities or Sub-national habitats Dominant community National Single functional group
Continental Two or more functional groups
Single specific species population Global
Two or more specific species 0% 20% 40% 60% 80% 100% populations % of studies
Figure 2: ESP by spatial scale with food production, freshwater fishing and pollination omitted (continental and sub-continental scales are combined due to the small number of entries at sub-continental scale)
To analyse whether biotic attributes are linked to spatial scale or location, a simplified dataset was compiled (Table 1) by combining the continental and sub-continental scales, both of which were sparsely populated; removing the “other” category; deleting sparsely populated categories (sapwood; leaf N content); and combining similar attributes into the following attribute clusters:
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 94
Biomass aboveground biomass belowground biomass productivity wood density Community/habitat area presence of a specific community/habitat type community/habitat area Community age / successional stage community/habitat age successional stage Diversity species richness functional richness functional diversity species population diversity Species or functional group abundance presence of a specific species abundance of a specific species abundance of a specific functional group presence of a specific functional group
Visual inspection of Table 1, in which the most commonly cited attributes are shaded in darker shades of green, does not reveal any obvious variation in the frequency distribution of biotic attributes by spatial scale. This is confirmed when the results are plotted graphically, as shown in Figure 3.
Table 1: Simplified table showing number of studies mentioning biotic attributes vs. spatial scale, for all ecosystem services combined.
Sub- Local national National Continental Global Total Community/habitat area 180 105 18 7 23 333 Community/habitat structure 82 73 11 5 13 184 Community age/successional stage 49 30 2 1 4 86 Litter/crop residue quality 16 4 1 21 Biomass 88 32 7 2 8 137 Stem density 17 6 2 1 26 Mortality rate 15 7 2 2 4 30 Species/functional group abundance 212 82 17 1 26 338 Species size/weight 49 15 2 3 7 76 Population growth rate 9 7 1 2 19 Flower-visiting behavioural traits 11 4 1 3 19 Predator behavioural traits 9 2 3 14 Life span/longevity 5 1 1 4 11 Diversity 140 70 20 1 20 251 Landscape diversity 9 6 3 18 Total 891 444 83 23 122 1563
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 95
250 Local Sub-national 200 National
150 Continental Global
100 Numberof studies studies Numberof
50
0
Figure 3: Line graph showing frequency of citation of biotic attributes for each spatial scale.
Similarly, there is no apparent variation in the pattern of biotic attributes between studies located in different continents (Table 2; Figure 4).
Table 2: Simplified table showing number of studies mentioning biotic attributes vs. continent, for all ecosystem services combined.
North South Row Labels Europe America America Africa Asia Oceania Global Total Community/habitat area 113 82 18 9 58 30 23 333 Community/habitat structure 72 33 10 10 29 17 13 184 Community age/successional stage 17 19 10 2 16 18 4 86 Litter/crop residue quality 8 2 1 2 3 4 1 21 Biomass 31 44 5 10 31 8 8 137 Stem density 6 9 2 2 6 1 26 Mortality rate 5 10 3 7 1 4 30 Species/functional group abundance 97 96 16 26 56 21 26 338 Species size/weight 10 23 7 7 18 4 7 76 Population growth rate 4 3 1 7 2 2 19 Flower-visiting behavioural traits 5 5 2 2 2 3 19 Predator behavioural traits 7 1 3 3 14 Life span/longevity 3 4 4 11 Diversity 77 80 8 21 33 12 20 251 Landscape diversity 5 4 2 3 1 3 18 Total 460 415 80 92 268 126 122 1563
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 96
120 Europe 100 North America 80 South America
60 Africa Asia 40
Number of studies studies of Number Oceania 20 Global 0
Figure 4: Line graph showing frequency of citation of biotic attributes for studies located on each continent.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 97
Annex 3: Indicators for ecosystem assessment
The majority of the ecosystem service literature reviewed (83%) used assessment methods which were based on primary data, such as direct field observations or public statistical databases. Mechanistic, correlative and conceptual models were used very rarely (Figure 1Error! Reference source not found.).
1% 6% 10% Conceptual model Correlative model Mechanistic model Primary data
83%
Figure 1: A breakdown of the assessment methods used in the studies reviewed. Only 81% of the studies recorded indicators, and 37% discussed an indicator which had not been identified previously by the JRC review on indicators for mapping ecosystem services (Egoh et al., 2012), as shown by the significant number of entries labelled “other” in Table 1Error! Reference source not found.. While the JRC indicators were useful for services such as atmospheric regulation, they were not relevant for research on freshwater fishing, where there was no mention of either of the two JRC indicators (water supply or river salinity) in the reviewed literature, but additional indicators (including nutrient concentrations, water quality or oxygen concentrations) were used in 15 of the articles. Interesting results were also found for timber production. For the four JRC indicators which had been identified for this service, “timber production” was mentioned in all articles citing indicators, whereas “raw material” was cited only once, and there was no mention of the other two indicators: “fuel wood” or “reeds”.
Table 1: JRC indicators for ecosystem assessment cited in the literature for the various ecosystem services.
Ecosystem service Indicator name No. of % of studies studies Freshwater fishing Other 15 31%
Timber production Timber production 40 95% Raw material 1 2%
Freshwater provision Water supply (precipitation) 29 53% Water supply 16 29% Water supply (surface water) 9 16% Water supply (ground water) 4 7% Land cover 11 20%
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 98
Ecosystem service Indicator name No. of % of studies studies Soil characteristics 9 16% Litter containment 1 2% Distance to water resources 1 2% Other 38 69%
Food production Crop yield 37 74% (cultivated crops) Agricultural production 9 18% Commodity production value 4 8% Livestock production 3 6% Net Primary Production 2 4% Land cover 2 4% Fish catch 1 2% Productivity index 1 2% Other 39 78%
Air quality regulation Pollutant concentration 39 78% Tree cover 37 74% Air purification 31 62% Deposition velocity 25 50% Other 22 44%
Atmospheric regulation Carbon storage 42 70% (carbon sequestration) Above ground biomass 42 70% Below ground biomass 28 47% Soil carbon 35 58% NPP 15 25% Soil characteristics 15 25% Forest biomass 14 23% Nutrient flux 7 12% Vegetation map 6 10% Land cover 5 8% Other 6 10%
Mass flow regulation Soil erosion 40 80% (erosion protection) Erosion control 7 14% Soil retention 4 8% Water retention 4 8% Soil characteristics 3 6%
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 99
Ecosystem service Indicator name No. of % of studies studies Land cover 2 4% Land Use 2 4% Geomorphology 2 4% Precipitation 1 2% Other 2 4%
Water flow regulation Flood attenuation 34 57% (flood protection) Hydrological flow 31 52% Land cover 15 25% Soil characteristics 8 13% Vegetation map 8 13% Water holding capacity 8 13% Flood prevention 7 12% Slope 6 10% Water filtration 5 8% Annual flood 5 8% Digital Elevation Model (DEM) 5 8% Water regulation 4 7% Flood control 3 5% Erosion 2 3% Nutrient retention 2 3% Water quality 2 3% Disturbance prevention 1 2% Other 5 8%
Water quality regulation Water quality 43 72% Nutrient retention 37 62% Water quality regulation 36 60% Water characteristics 30 50% Vegetation map 10 17% Slope 2 3% Other 5 8%
Pollination Bumblebee density 36 71% Land cover 33 65% Habitat 26 51% Land Use 22 43% Crop yield 21 41%
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 100
Ecosystem service Indicator name No. of % of studies studies Cost of bees 1 2% Other 38 75%
Pest regulation Pest density 18 30% Land cover 7 12% Pest control (expenses) 3 5% Other 17 28%
Recreation (species-based) Fish abundance 4 7% Land cover 4 7% Forest birds density 2 3% Forest birds richness 2 3% Protected Areas 1 2% Farmland birds richness 1 2% Other 40 66%
Aesthetic landscapes Aesthetic value 31 51% Scenic beauty 12 20% Landscape and nature diversity 5 8% Rare species 1 2% Other 22 36%
A list of the additional indicators cited multiple times in the literature is provided in Error! Reference source not found.. Many services, especially pollination and species-based recreation, cited a broad range of indicators (41 and 59 additional indicators, respectively); however the majority of these were cited only once. Those which appear to be particularly important, however, being cited multiple times, include catch/yield for freshwater fishing (10 counts); distance to natural/native vegetation (13 counts) and pollinator visitation rates (17 counts) for pollination; evapotranspiration (14 counts) and water yield (11 counts) for freshwater provision; and visitor numbers for species-based recreation (10 counts).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 101
Table 2: Additional indicators for ecosystem service assessments found in the literature which are cited more than once. Those cited ten times or more have been highlighted in grey.
Ecosystem service Additional indicators cited in the literature Frequency Freshwater fishing Catch/ yield 10 Fish size/weight 4 Catch per unit effort 2 Freshwater provision Evapotranspiration 14 Water yield 11 Stream flow 9 Catchment/watershed area 3 Leaf Area Index 3 Catchment water balance 2 Plant water use 2 Sap flux density 2 Food production (cultivated Soil characteristics 6 crops) Crop biomass 3 Water use efficiency 3 Ecosystem services value 2 Cost of production 1 Air quality regulation Greenness 3 Mixing/boundary layer height 3 VOC emission factors 3 BVOC emission factors 2 Hydrocarbon emissions 2 Leaf Area Index 2 Pollination Pollinator visitation rates 17 Distance to native/natural vegetation 13 Fruit/seed set 7 Abundance of bees 6 Pollinator abundance 6 Bee diversity 3 Floral/flower abundance 3 Floral/flower diversity 3 Flower density 3 Pollen grain deposition per visit 3 Crop type 2 Floral resource 2 Hive density 2 Plant reproductive success 2 Pollination efficiency 2 Recreation (species-based) Visitor numbers 10 Visitor satisfaction 9
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 102
Ecosystem service Additional indicators cited in the literature Frequency Visitor expenditure 6 Angling/fishing effort 4 Species abundance 4 Willingness to pay 4 Length of trip 3 Employment 2 Fledging success (birds) 2 Frequency of visit 2 Mortality rate 2 Recreation/ecotourism opportunities 2 Travel cost 2 Travel time (distance to site) 2 Aesthetic landscapes Willingness to pay 4 Naturalness 3 Perception 3 Property values 3 Biodiversity 2 Landscape coherence 2 Landscape preferences 2
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 103
Annex 4: Policies
Only 13% of the studies reviewed cite a particular policy. Some of these cite multiple policies, with 121 examples identified in total by the reviewers. However, policy is discussed more often than this implies, with many articles describing existing national/local level policies, or discussing policy recommendations in general terms. Of the 24 EU policies covered in the database, only the following nine are mentioned in the literature, and these are rarely cited: Ambient Air Quality Directive Biodiversity Strategy Birds Directive Common Agriculture Policy Environmental Impact Assessment Habitat Directive Nitrates Directive Soil Thematic Strategy Water Framework Directive (WFD)
Policies are cited most frequently for water flow regulation, followed by air quality regulation, atmospheric regulation, food production, pollination, species-based recreation and freshwater provision (Table 1Error! Reference source not found.).
Policies categorised as “other” are identified most frequently in the literature, occurring 83 times in the review. These policies are relatively diverse (a complete list is shown in Table 2Error! Reference source not found.). The most commonly mentioned are measures associated with the UNFCCC, such as REDD and the Kyoto Protocol (including Clean Development Mechanism (CDM) schemes) for the service of atmospheric regulation; the RAMSAR Convention on wetlands; and various Biodiversity Action Plans.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 104
Table 1: Counts of policies cited in the literature per ecosystem service. Ambient Air Quality directive Biodiversity Strategy Birds Directive Common Agriculture Policy Environmental Impact Assessment Habitat Directive Nitrates Directive Soil Thematic Strategy WaterFramework Directive (WFD) Other Total
Freshwater fishing 1 1 Timber production 0 Freshwater provision 1 1 1 11 14 Food production (crops) 1 1 1 1 1 1 9 15 Air quality regulation 4 9 13 Atmospheric regulation 14 14 Mass flow regulation 1 1 3 5 Water flow regulation 3 4 1 5 8 21 Water quality regulation 1 1 4 2 8 Pollination 1 11 12 Pest regulation 1 1 Recreation (species-based) 1 1 1 12 15 Aesthetic landscapes 2 2 Total 4 4 1 5 1 8 1 2 12 83 121
Table 9: Additional policies cited in the literature
Ecosystem Service Additional policies cited
Freshwater fishing World Commission on Dams (policy principles and guidelines) Timber production None Identified Freshwater provision CBD CDM (Kyoto, carbon sequestration projects including afforestation and reforestation) Integrated Water Resources Management IUCN
National ecological networks (China) National Water Acts
Payment for Watershed Services (discussed)
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 105
Ecosystem Service Additional policies cited
RAMSAR Convention REDD+ Soil and water conservation programmes (Otago, New Zealand) Spatial Biodiversity Assessments Sustainable forest management
UNFCCC (increased interest in afforestation, e.g. of páramo grassland in the Andes) Food production Agricultural policies leading to specialisation (Sweden, though subsidiaries, (cultivated crops) price regulations, legislation and extension) Brazil’s Proambiente Pilot program (Pecosystem services scheme, Amazonia) Food Harvest 2020 Strategy (Ireland; DAFF, 2010) Payments for carbon sequestration/ Carbon offsetting Subsidies schemes, e.g. for irrigated farming RAMSAR Convention Air quality regulation City programmes including Barcelona’s Energy, Climate Change and Air Quality Plan; NYC million tree planting initiative District air quality management plans (US) EU specific standards, such as: The Euro Standards on road vehicle emissions; Directive 2010/75/EU on industrial emissions; Directive 94/63/EC on VOCs Kyoto Protocol National/regional Air Standards (China, US national clear air act, US South Coast) Policy to use urban forests to improve air quality (Santiago Regional Metropolitan government, Chile) Atmospheric regulation Carpathian Convention (carbon sequestration) CDM Forest-based carbon offset initiatives Kyoto Protocol Logging ban policies (China) REDD
Sustainable forest management (SFM) (Ministerial Conferences on the Protection of Forests in Europe (MCPFE) guidelines) UN FAO Forest Resource Assessments Urban forestry planning and management (e.g. Hangzhou, China) Mass flow regulation Environmental Law controlling the width of riparian vegetation (erosion protection) (Brazil) Surface Mining Control and Reclamation Act (North America) Water flow regulation Biodiversity Action Plans (flood protection) CDM Declaration of Arles (to take measures to reduce future flood risks, which include land management and forestry) Forestry Stewardship Council timber standards Kyoto Protocol
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 106
Ecosystem Service Additional policies cited
Ministerial Conferences on the Protection of Forests in Europe (MCPFE) guidelines (for the sustainable management forests) Protected forests (e.g. headwaters water resource conservation, Taiwan) RAMSAR Convention UK Countryside Stewardship Scheme Water quality regulation National initiatives: Integrated Mangrove Aquaculture System (China); Great Lakes Restoration Initiative (N. America). Pest regulation Agri-environment schemes (UK) Pollination Agri-environment schemes Conservation areas/ nature reserves
Conservation of Agricultural Resource Act (California) EU organic farming regulation (2092/91/EEC )
Organic Food Act (California, US)
UKs 2020 Biodiesel Target (part of policy to reduce GHG and other emissions from transport) Recreation (species-based) Action Plan for Biodiversity and Nature Conservation (Denmark)
Buffer zone regulations (Nepal government, 1996 to establish buffer zones around existing protected areas) Endangered Species Act Policies (e.g. for lions) Implementation of regulations relating to listed threatened of protected species (TOPS, South Africa) Local laws (e.g. ban of catch and release fishing in Germany) Natura 2000 Network Preventative policy against crocodile attacks (Australia) Recreational hunting as a policy for population control of feral pigs (US, California) River Catchment Management Plans (e.g. for River Spey) Rottnest Island Authority Act (1987, Western Australia, establishes the Rottnest Island Authority as a statutory body to control and manage the Island, reporting to the Minister for Tourism) Tourism Development Strategies Tourist hunting regulations (e.g. Tanzania, 2010) Use of permits (e.g. to assess risk to bat species from tourism, to reduce impacts of swimming with dolphins, Australia) Aesthetic landscapes Act on Urban Parks and Green Spaces (Korea) Creation and Management of Forest Resources Act (Korea) Establishment and Management of Forest Resource Act (Korea) Framework act on forestry (Korea) Red grouse hunting policy
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 107
Annex 5: Short reports for each ecosystem service
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 108
Freshwater fishing
Tibor Erős and Ágnes Irma György, Centre for Ecological Research, Hungarian Academy of Sciences
The literature search
Altogether 49 articles on freshwater fish and fisheries were reviewed, including 31 BESAFE articles and 18 additional articles. The original BESAFE search used the following terms: biodiversity OR “biological diversity” OR species OR habitat* OR genetic* OR trait* OR functional* OR landscape OR richness OR abundance
AND the two sets of terms listed below.
Ecosystem service/disservice terms Additional service related terms
“Freshwater fishing” Trout “Quantity of caught fish” Salmon “Fish yield” Carp “Fisheries management” Freshwater eel “Aquaculture performance” Perch “Impact of fish polyculture on yield” “Impact of fish monoculture on yield” “Effect of stocking density on freshwater fishing” “Relationship between fish abundance and yield”
For the OpenNESS review, 18 articles that were strictly about intensive aquaculture (monoculture or polyculture) in the original BESAFE database were replaced by new, more relevant articles using the keywords “fish” and “ecosystem services”. Nonetheless, a lot of articles dealt with aquaculture and fish production (almost 30%); a good deal were about culture-based fisheries, recreational fishing (angling) and fish communities in natural waters (altogether 37%), while some articles (14%) were reviews or synopsises on fish communities and fisheries at greater spatial and temporal scales (e.g. ‘Freshwater fish in Scotland in the 18th, 19th and 20th centuries’, Maitland, 1977; ‘Potential impacts of global climate change on freshwater fisheries’, Ficke et al., 2007). More than half (59%) of the articles involved were published after the year 2000 (Figure 1).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 109
6
5 4 3 2 Number of Articles 1 1975 1980 1985 1990 1995 2000 2005 2010 2015 Years
Figure 1: Publication dates of the 49 journal articles reviewed on freshwater fishing.
Background/context
The articles varied greatly in terms of their spatial scale. Most of the studies (41%) were local or sub- national (29%), while the remaining articles focused on a larger spatial scale (18% global, 8% national and 4% continental, see Figure 2). Owing to the high number of articles on fish production, studies conducted in Asia were the most frequent. This was followed by a high amount of studies from North America, which focused mainly on recreational fishing. There were also a fair number of articles on experiments in Europe as well as articles reviewing fish populations and fisheries on a global scale. Only one study was from Australia and one from Africa. The type of evidence presented in the articles was independent of the location of the studies (Fisher’s exact test, p>0.05).
The majority of the studies, 78% (38 articles), were snapshots; 20% (10 articles) analysed long-term changes in fish stocks in lakes and reservoirs, and there was also one on scenario analysis (Figure 2).
Time scale Spatial scale Scenario analysis 2%
Global 18% Long- Continental term 4% 20% Local 41% National 8% Sub- Snapshot national 78% 29%
Figure 2: Percentage of studies at each temporal and spatial scale for freshwater fishing.
The reviewed studies mainly dealt with rivers and lakes (including fish ponds) (76%, 37 articles), but the condition was usually not mentioned (61%, 30 articles); in the rest the condition of the ecosystem was
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 110
unfavourable (Table 1). Note that regarding fish ponds and reservoirs, the interpretation of the condition is problematic since it is generally based on the demands of the fisheries and not of the ecosystem. In fact, we suggest that future meta-analyses should broaden the scope from freshwater fishing and use other keywords. A focus on fish rather than on freshwater fishing can help, because lots of articles deal with fish as an ecosystem provider and these are about natural systems, rather than artificial or human modified habitats (fish ponds, reservoirs).
In seven cases, more than one ecosystem type was mentioned within an article. Not surprisingly, the ecosystem type most often mentioned together with rivers and lakes was wetlands; it was present in five studies. In addition to rivers and lakes only one ecosystem type, sparsely vegetated land, was mentioned by itself in one article. In 32% of the articles (16 articles) no natural ecosystems were mentioned, only artificial ponds (mono- or polyculture). Eleven articles also only focused on experimental ponds and laboratory experiments, and even these were absent in one article which was about fish fillet traits and body measurements (Rutten et al. 2005). In only one article was the condition of the ecosystem favourable (Mollot et al. 2003), and in another one the ecosystem’s original condition was destroyed (impounded river sections with fish stocking, Maceina and Reeves 1996).
Table 1: Ecosystem types present in the review of the ecosystem service freshwater fishing.
Condition Unfavourable Favourable not Total Recovering Declining Destroyed mentioned Woodland and forest 1 1 Sparsely vegetated land 1 1 Urban 1 1 Wetlands 1 2 2 5 Rivers and lakes 1 1 4 1 30 37 Coastal 1 1 2 Marine inlets and 1 1 2 transitional waters Total 3 1 9 1 35 49
Linkages between natural capital and the service
In most of the articles (63%) the Ecosystem Service Providers (ESPs) belonged to two or more species populations and in 20% to a single specific species population. Most frequently, the ESPs were commercially important fish species (e.g. Nile tilapia - Oreochromis niloticus, carp - Cyprinus carpio), and in a few studies the entire community of fish, or fish as the dominant community.
Concerning the main biotic attributes, the species-related attributes were the most frequently mentioned (Figure 3). Species abundance was the most important (55%, 27 articles) together with species size/weight (45%, 22 articles), which are the primary parameters describing fish stocks; of these the direction was positive in 48% and 86% of studies, respectively. Stocking density was the most important factor
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 111
influencing the yield of fish species: catch rates usually increased with higher fish densities, though some studies mentioned that a threshold could be reached beyond which over-stocking led to reduced fish size and increased mortality (e.g. de Silva, 2003; Lorenzen, 1995; Smith et al., 2012). Bigger fish were preferred both by fishermen and anglers. Species richness (24%, 12 articles) was mentioned mainly in articles on aquaculture and the direction was mostly positive: the polyculture systems allowed an increase in total production with the same amount of supplied feed, thus improving the systems’ sustainability.
Mortality rate was cited in 22% (11) of the articles and had a negative direction in 73% (8) of these 11 articles. High fish densities usually increased mortality (e.g. Lorenzen, 1995). The presence of certain species in some studies also increased the mortality of commercially/recreationally important fish species which were the main providers of the service, for example, the fish community suffered huge losses from introduced non-indigenous sea lamprey (Petromyzon marinus) predation when the lamprey was abundant in Lake Superior (Lawrie, 1978).
Functional group related biotic attributes were not relevant in the reviewed articles, while community/habitat related attributes appeared in 27% (13) of the reviewed articles. Community/habitat area and community/habitat structure had a primary effect on the ecosystem service in 62% and 31% of the articles, respectively, and the direction was diverse. Specifically, a correlation between the surface area of water bodies and fish yield was frequently cited.
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Primary productivity Aboveground biomass Landscape diversity Species richness Presence of a specific species type Species abundance Species size/weight Population growth rate Life span/longevity Mortality rate 0 5 10 15 20 25 30 Number of studies
Negative Both (Positive & Negative) Positive Unclear Figure 3: Biotic attributes contributing to freshwater fishing.
Abiotic factors were mentioned in 43% (21) of the articles. Nutrient availability and water quality were most frequently identified (31% of the articles mentioned one or the other; 20% of the articles mentioned each): these are the main factors influencing primary production, and thus the carrying capacity of water bodies. Other factors such as temperature and O2 concentration were only occasionally mentioned (Figure 4).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 112
Temperature Precipitation Snow Water availability Evaporation Nutrient availability Water quality Slope Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 4: Abiotic factors influencing freshwater fishing.
Human input and management
Human input was mainly identified as being direct, and intensity was nearly equally split between extensive and intensive. In the freshwater fishing category a lot of articles dealt with aquaculture and culture-based fisheries, and thus the most frequent direct input was fish stocking and other fisheries management activities. Increased fish stock and yield owing to human impact (mainly stocking) was considered as a positive result, but in most of the studies it was not due to, or related with, the improvement of the ecosystem. Also habitat modification had a significant role: reservoirs and artificial fish ponds were created to produce higher amounts of fish. In 33% of the studies (16 articles) the impact was positive, in 24% (12 articles) no impact was mentioned, in 18% (9 articles) the impact was negative and in 16% (8 articles) both positive and negative impact was observed. Positive impact was typical particularly in the case of stocking density, resulting in higher yields, whereas overfishing and habitat degradation had a strong negative impact on yields.
Human input and management was direct in most of the articles (65%, 32), the most frequent management method being stocking. 24% of the articles did not mention any human input and in 8% (4 articles) both direct and indirect input was applied. Indirect input in itself was mentioned in only one article (4%), where the abundance of submerged aquatic plants had a positive correlation with catch rates of largemouth bass and a negative correlation with average weight of fish caught, thus fish stock structure may be influenced by the management of macrophytes (Maceina and Reeves, 1996).
Human management was intensive in 31% of articles (15 articles), mainly examining aquacultures; typical management types covered were stocking, rearing and feeding of certain commercially important species and regulating water quality parameters (e.g. O2 concentrations, nutrients) artificially. Extensive management was characteristic for culture-based fisheries (20%, 10 articles), where fish are artificially
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 113
stocked in natural waters but fish growth and propagation depends primarily on naturally available food sources and environmental parameters. 16% of the articles contained both intensive and extensive management methods, while in 27% (13 articles) the intensity was not mentioned.
Thresholds
A threshold was found only in one article, belonging to the ‘other biophysical’ category, regarding fish weight in a journal article about fillet traits and body measurements of Nile tilapia for the European market (Rutten et al. 2005).
Indicators
No indicators were cited from the JRC report (Egoh et al., 2012) in the reviewed articles, but in 31% (15 articles) other indicators were cited. The main indicator (used in 10 articles) was fish catch (mainly weight/area). Furthermore, indicators describing physiological conditions of fish were cited in three articles, physical and chemical parameters of the water body in one, and cost (willingness to pay to increase ecosystem services) in another one.
Policies
Only one article mentioned a policy, the World Commission on Dams, 2000 in Dugan et al. (2010).
Interactions with other ecosystem services
Recreation was mentioned the most (in 20% of the articles). In most of the studies (60%) the relationship was positive: fish stocks rich in fish, especially large game fish, are favoured by anglers, but as a “side effect” incidental introductions of non-indigenous, non-game fish species is also taking place all over the world and causing unexpected problems, such as changes in population structure and food webs or the disappearance of native species in natural environments.
Water flow regulation schemes such as irrigation, drainage and dam construction have a negative relationship with freshwater fishing in most of the studies, because water quality, connectivity, fish habitats and spawning grounds usually deteriorate or diminish, resulting in transformed, often impoverished, i.e. less diverse, ecosystems (e.g. Craig et al., 2004).
Since food production was specified as cultivated crops, the relationship with food production was not as pronounced as it may have been without this limitation, because there are many articles on aquaculture and culture-based fisheries that are aiming to produce higher quantities of fish for food.
The interlinkages between the above mentioned ecosystem services are based primarily on empirical evidence.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 114
Overall observations on the review
Most of the articles (63%, 31 articles) were based on empirical evidence, and 27% (13 articles) on both empirical and modelling. Only 8% (4 articles) focused solely on modelling and in 2% (1 article) the basis was not mentioned. Direct evidence was given in 67% of the articles (33), and indirect in 8% (4), while in 24% (12 articles) the direction of the evidence was not mentioned.
Regarding the applied methods, most articles used both qualitative and quantitative (51%, 25 articles) methods, one third of the articles used only quantitative methods (33%, 16 articles), 14% (7 articles) were based only on qualitative evidence and in one article (2%) the method was not mentioned.
In relation to observation types, almost all the articles applied multiple observations (82%, 40 articles), the observation type was not mentioned in one article, while the rest (16%, 8 articles) were based on single observations. The four questions covered the evidence appropriately.
Articles on aquaculture could have been classified within the food production category instead of freshwater fishing, since they are about strictly controlled environments which are designed for the commercial production of fish. More articles about culture-based, artisanal fisheries or even industrial fisheries in natural freshwaters might have represented a better basis for ecosystem services analysis in the freshwater fishing category. The inclusion of fish catch and fish abundance as indicators for freshwater fishing are recommended in future reviews, and the temporal scale question should include a category for review articles (e.g. ‘review’) too.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 115
Timber production
Linda Meiresonne, Institute for Nature and Forest Research (INB0), Belgium
The literature search
The journal articles to be reviewed were selected in the following steps (Table 1).
Table 1: Search steps for journal articles dealing with ecosystem services ‘timber production’ and biotic attributes.
Steps Description of literature search
1 The primary search started during 2012, as part of the BESAFE project. When the Web of Science was searched with the terms specific for the ecosystem service: ‘wood, wood production, timber, timber production’, there were 1618 hits.
2 Then biodiversity terms were added: ‘biodiversity OR biological diversity OR species OR habitat* OR genetic* OR trait* OR functional* OR landscape OR richness OR abundance’. This search resulted in 903 hits.
3 As the sample was still too big, the term ‘tree diversity’ was added, resulting in 20 hits.
4 To be relevant for this analysis, articles were selected on specific cases, comparing wood or timber production of different tree species in specific forest conditions. This reduced the sample further to 11 relevant hits. Two meta-analyses on the topic were found: Piotto et al. (2008) and Zhang et al. (2012a). These articles were considered not relevant as they did not contain specific cases, but contained usable secondary references.
5 Snowballing resulted in 16 additional references, bringing the total to 27 references.
6 Where the article is describing cases with more than one tree species combination, resulting in contradicting results (i.e. positive or negative or unclear effect of species abundance or richness), these relationships were entered as separate studies in the database. This resulted in 8 additional records.
7 For the OpenNESS review, the same search was updated in 2014. This resulted in 7 additional articles, bringing the total to 42 studies in the database.
Note: the review article of Cardinale et al. (2012) in Nature (‘Biodiversity loss and its impact on humanity’) reports on the links between biodiversity and ecosystem services. For wood production 53 data points were reported. However, these data originate from only three source articles, which are all included in our review.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 116
9 8 7 6 5 4 3
Number of Articles 2 1 1985 1990 1995 2000 2005 2010 2015 2020 Years
Figure 1: Publication dates of the 42 journal articles reviewed for the ecosystem service of timber production.
Background/context
Spatial scale and location Most studies were at a local (26) or sub-national scale (16) (Figure 2). Only one study was at a national level. The locations of the studies were well spread over the climate types and continents of the world (Map 1, Table 2). However, we see that studies dealing with plantations were mainly conducted in the tropics.
Temporal scale Regarding the temporal scale, most studies (28) were snapshot research on the local to subnational scale (Figure 2). Nine studies were on an annual basis, while five studies looked at the long-term.
Time scale Spatial scale National 2%
Long- term 12% Sub- national Annual 36% 21% Local 62% Snapshot 67%
Figure 2: Percentage of journal articles at each temporal and spatial scale for timber production.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 117
Map 1: Location of studies on timber provision.
Table 2: Location of studies for timber production, split into forest or plantation
Climate type N Continent Country Forest Plantation Boreal 3 Europe Finland 1 North America Canada 2 Temperate 17 North America USA 3 2 Canada 3 South America Chile 3 Oceania Australia 3 Europe Germany 1 Sweden 2 Mediterranean 5 Europe Spain 3 North America USA 1 Asia Iran 1 Tropical 17 North America Costa Rica 6 Puerto Rico 1 Panama 5 Oceania Australia 3 Asia Philippines 1 Africa Kenya 1 Total 42 19 23
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 118
Ecosystem types All ecosystem types for ‘timber production’ are – not surprisingly - woodland and forest (Table 3). In most of the articles, the condition of the ecosystem was not mentioned.
Table 3: Ecosystem types present in the review of timber production
Ecosystem Type Condition Count Woodland and forest Condition not mentioned 32 Favourable 5 Unfavourable - recovering 4
Linkages between natural capital and the ecosystem service
The articles in the database assessed timber production of monocultures and/or timber production of mixed forests. All kind of comparisons are made: mono versus mono, mono versus mix, and mix versus mix.
Ecosystem Service Provider (ESP) classes The ESP consisted of two or more specific species populations (30 studies), two or more functional groups (5 studies), or a single specific species population (7 studies).
Most of the studies (30, i.e. 71 %) concerned two or more specific species populations. More than half of these were plantations for timber production. They consisted of two or more tree species, planted in pure and mixed proportions. In the case of two tree species (A and B), these trials often use a ‘replacement series design’ of: 100% A, 75% A + 25% B, 50% A + 50% B, 25% A + 75% B and 100% B: e.g. the Australian studies of Bauhus et al. (2004) with Eucalyptus globulus ssp. Pseudoglobulus and Acacia mearnsii and Bristow et al. (2006) with Eucalyptus pellita and Acacia peregrine. Pure versus 50% A/50% B mixtures: e.g. in Amoroso and Turnblom (2006) with Douglas-fir (Pseudotsuga menziesii) and Western hemlock (Tsuga heterophylla) in the Pacific Northwest of the USA.
The remaining studies with two or more species populations were forests: e.g. the study of the boreal forest of Southern Finland (Scots pine (Pinus sylvestris) and silver birch (Betula pendula) - Hynynen et al. (2011), or the spruce-dominated forests of Canada (with the most frequent tree species being: black spruce (Picea mariana), white spruce (Picea glauca), red spruce (Picea rubens) and balsam fir (Abies balsamea) - Lei et al. (2009)).
Five studies focused on two or more functional groups. Most of them were forests, with two or more dominant tree species and a large number of accompanying tree species. Examples are pine-dominated Mediterranean forests in Catalonia, Spain (Vila et al., 2003), and a tropical plantation in the Philippines (77 species belonging to 32 families and 57 genera - Nguyen et al., 2012).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 119
Biotic attributes All studies focused on species type in relation to forests and plantations with economically important tree species, such as (among others) conifers and eucalyptus.
Species richness was the most important attribute (38 studies, 90%), with a positive relationship in 28 studies (67%), evenly divided over forests and plantations (Figure 3). An example of a strong positive relationship between species richness and timber production was found in the Douglas-fir/western hemlock (Pseudotsuga menziesii/Tsuga heterophylla) forests of Oregon, USA (coexisting with many other tree species in natural stands, in particular Alnus rubra, Thuja plicata and Acer macrophyllum) and in the mixed-conifer forests of California with Pinus ponderosa, Pinus jeffreyi, Pinus lambertiana, P. menziesii, Abies concolor and Libocedrus decurrens as the dominant tree species (Liang et al., 2007). In Costa Rica it was found that in 15 - 16 year-old tropical forest plantations with native tree species on abandoned pasture land, mixed plantations performed better than pure stands for all variables considered (such as basal area, total volume and aboveground biomass), regardless of species composition (Piotto et al., 2010).
A negative relationship was found in 8 studies (5 in forests, 3 in plantations). In forests, the negative relationship between species richness and timber production was mainly recorded in rather old-growth forests. This was the case in the boreal forest in Southern Finland (Hynynen et al., 2011) with a negative impact of birch competition on the growth of pine trees. A similar conclusion was found in the temperate deciduous forests in Germany, where forest stands differing in tree taxa diversity and grouped into three diversity levels were studied: (i) four forest stands of European beech Fagus sylvatica contributing with 85% to 100 % to the total tree basal area; (ii) four stands mainly consisting of beech, lime (Tilia cordata and T. platyphyllos) and ash Fraxinus excelsior; (iii) four stands with five dominant tree taxa (beech, lime, ash, hornbeam (Carpinus betulus) and maple (Acer pseudoplatanus and A. platanoides) (Jacob et al., 2010). Finally, in north-west North America, mixed plantations of conifer species Douglas-fir (Pseudotsuga menziesii) and Western hemlock (Tsuga heterophylla) had a negative impact on timber productivity, in comparison with pure stands (Amoroso and Turnblom, 2006).
In 4 cases the relationship was unclear.
Species abundance was the second most important biotic attribute, with a positive relationship in 8 studies, a negative relationship in 1 study (forest) and an unclear relationship in 1 study.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 120
Community/habitat structure Successional stage Primary productivity Aboveground biomass Stem density Species richness Functional richness Functional diversity Presence of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Mortality rate Sapwood amount Leaf N content Other 0 5 10 15 20 25 30 35 40 45 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 3: Biotic attributes contributing to timber production.
The impact of the soil is the most frequently cited abiotic attribute in some studies, with seven studies showing a positive contribution, while six studies showed unclear contributions (Figure 4).
Temperature Precipitation Evaporation Soil Slope Other
0 10 20 30 40 50 Number of studies Negative Positive Unclear
Figure 4: Abiotic factors influencing timber production.
Human input and management
The only type of human input considered in the studies was the fact that some of the studied forests are planted and managed by humans.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 121
Thresholds
There was no information about thresholds.
Indicators
The following indicators were proposed for timber production from ‘Indicators for mapping ecosystem services: a review’, JRC, Egoh et al, (2012): Raw material Timber production Fuel wood Reeds
The indicators in the reviewed articles referred to timber production, such as growth of the trees (height, diameter, circumference) or the production of wood (m³ or tonnes).
Policies
Policies were not explicitly mentioned in the reviewed articles.
Interactions with other ecosystem services
No information about the linkages between timber production and other ecosystem services was found.
Overall observations on the review
It was very obvious that nearly all studies followed the same type of research: based on empirical data, with direct and quantitative measurements, often with multiple observations.
The study of the influence of ESPs, biotic and abiotic attributes on timber production was never explicitly linked to an ‘ecosystem service story’. All the studies concerned ‘timber production’ only. There was never a link with other ecosystem services.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 122
Freshwater provision
Laura Mononen, Finnish Environment Institute
The literature search
In total, 55 journal articles were entered in the database. Of these, 49 were from the BESAFE review. One article was not available through any channel. An additional search with the same search terms as BESAFE was undertaken. The search resulted in thousands of hits even when results were refined by adding other environmental and biodiversity-related search terms. In the end, six additional articles matched with the contents in the BESAFE articles between the years 2012 and 2013 (Figure 1). The search was conducted in June and July 2014.
Used search terms and results:
Water provision OR Fresh water OR Freshwater OR Water suppl* OR Drinking water OR Drinkable water OR freshwater Time span was limited to results from years 2012-2013. This search gave 50,298 results which was a large number of hits. As in BESAFE, additional search terms related to biodiversity were added to refine the results.
Biodiversity OR “biological diversity” OR species OR habitat* OR genetic* OR trait* OR functional* OR landscape OR richness OR abundance This search gave also a large number of results (16,793). The same additional search terms were used to refine the results as follows: Forest* OR vegetation OR soil OR sapwood OR leaf* The search narrowed the results to 2,038 hits.
As in BESAFE, at this point search terms were changed and a new search was conducted: Water provision OR Fresh water provision OR Freshwater provision OR Water suppl* OR Drinking water provision OR Drinkable water provision OR freshwater provision This search resulted in 17,121 hits. This search was also refined with biodiversity related terms. This time there were 3,861 hits.
For both searches, results were ordered according to their relevance.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 123
9
7
5
3 Number of Articles 1 1980 1985 1990 1995 2000 2005 2010 2015 Years
Figure 1: Publication dates of the 55 journal articles reviewed on freshwater provision.
Background/context
The term ‘Freshwater provision’ was indirectly interpreted from the selected articles and, in the majority of the studies, water levels, water yield, runoff or recharge rate was considered to correspond with freshwater provision. Inland waters were not directly discussed in the articles. The focus was on discovering how changes in land use or vegetation cover composition affect water yields or runoff, and indirectly the availability of freshwater.
The 55 articles reviewed were concentrated in areas where water scarcity is an issue; for example Australia (12), New Zealand (5), South Africa (4) and Hawaii (4). Elsewhere studies were conducted in South America and a few in Europe. Many studies were review articles and some assessed the issue at larger scales. Some studies indicated concern over growing demand for drinking water (e.g. Blumenfeld et al. 2009). Nevertheless, most studies focused on examining water levels, runoff, or recharge, which were often used indirectly as indicators to assess the availability of freshwater. Decreases in water yield due to reforestation were most accentuated in dry regions with low water availability and precipitation (Rey Benayas et al., 2007).
The majority of the studies were snapshots (Figure 2). However, long-term studies are appropriate when effects of forest structure (age structure of the trees) on catchment water levels are studied before and after land or forest management practices. The effects vary greatly at different phases of forest management. Results of the review article of Bosch and Hewlett (1982) summarised that water yields increase after clear-felling, and after reforestation water yields start decreasing in proportion with the growth rate of new stands. Later studies also support these findings (e.g. Cornish and Vertessey, 2001; Webb and Kathuria, 2012). The reasons for these clear-cutting effects relate to reduced canopy interception and forest floor interception (e.g. Bosch and Hewlett, 1982; Putuhena and Cordery, 2000). The same phenomenon is in question with total change in land use; conversion of grasslands to forests increased the evapotranspiration rate and may result in reduced water yield (e.g. Duncan 1995; Bruijnzeel, 2004; Fahey and Jackson, 1997) and changing tree species in afforestation (e.g. Nosetto et al., 2005). In the study of Ruprecht and Schofield (1989), conversion of native jarrah forests to agricultural area in Australia resulted in an increase of streamflows by 30% rainfall/year because of reduced interception loss and reduced transpiration. Cavaleri and Sack (2010) discovered that difference between water use of native and invasive species is greatest in hot and wetter climates.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 124
Scenario Time scale Spatial scale analysis 7% Sub- continental Global 16% Long- 2% term National Local 38% Seasonal 16% 7% 2%
Snapshot Sub- 75% national 37%
Figure 2: Percentage of journal articles at each temporal and spatial scale for freshwater provision.
Most of the studies were conducted as catchment comparisons (sub-national or local studies were 75% of all those reviewed, see Figure 2). Of all ecosystems, forests were represented in the majority of the review articles (44/55), focusing mainly on the effects of afforestation (with native or invasive tree species) on water availability. Studies concerning grasslands (mentioned in 23 articles) and croplands (11) were related to land use change, as in many cases grasslands or croplands were converted or planned to be converted to forests. Especially for grasslands, 8 out of 11 articles indicated the situation in grasslands as unfavourable and declining (Table 1). For forests, the situation is two-fold since in some of the studies forests were cut to replant with new, often alien, species (e.g. Bruijnzeel 2004; Kagawa et al., 2009; Cavaleri and Sack,. 2010; Nie et al., 2011).
Concern over tropical montane cloud forests was raised in four studies (Hamilton, 1995; Gomez-Peralta et al., 2008; Brauman et al., 2010; Saenz and Mulligan, 2013). Cloud-affected forests are delicate ecosystems and human-induced threats in the forms of conversion to agricultural land, dams and deforestation are the biggest concern in these areas. Also, changes in climate causing the cloud belt to rise were identified in a study that was conducted in Peru (Gomez-Peralta et al., 2008). In general, cloud forests have a positive effect on water availability (Gomez-Peralta et al., 2008; Brauman et al., 2010). Concern over the delicate páramo ecosystem in the Andean region was also addressed (Buytaert et al., 2007).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 125
Table 1: Ecosystem types present in the review of freshwater provision.
Condition Ecosystem Favourable Unfavourable not Total No evidence Recovering Declining mentioned Woodland and forest 8 1 3 7 25 44 Heathland and shrub 1 3 4 Grassland 3 1 8 11 23 Cropland 3 1 2 5 11 Urban 1 3 4 Wetlands 2 2 2 6 Rivers and lakes 1 1 5 7 Coastal 2 2 Total 17 3 4 21 56 101
Linkages between natural capital and the ecosystem service
Changes in land cover produced a clear response in water levels, so community/habitat area was the main biotic attribute identified (Figure 3). Forests use more water than shorter vegetation because of the greater capacity of trees, with their longer roots, to access soil water (Petheram et al., 2002; Nosetto et al., 2005; Blumenfeld et al., 2009) and because trees exchange vapour with the atmosphere more effectively (Nosetto et al., 2005). At the species level there are significant differences in water use (Müller, 2009; Singh, 2012). For example, eucalyptus forest species seem to reduce runoff more than pine species (Farley et al., 2005). This was shown in multiple catchment comparisons. The meta-analysis of 94 catchment experiments by Bosch and Hewlett (1982) found a 40 mm change in water yield per 10% change in vegetation cover for pine and eucalypt forests, a 25 mm change in deciduous forest and a 10 mm change in scrub. Increases in forest cover caused decreases in water yield, and vice versa. However, an updated meta- analysis of 145 experiments by Sahin and Hall (1996) found somewhat lower changes in water yield per 10% change in vegetation cover: 20-25mm in coniferous forest, 17-19 mm in deciduous-hardwood, 6 mm in eucalypt forest, and 5-9 mm in scrub. Age was also cited as having an effect on the water use of plants, which seems to be higher in older forests (Cornish and Vertessey, 2001; Sun et al., 2006). Biodiversity was found to have a positive effect on water provision (Chan et al., 2006).
In many studies, the results showed a rapid increase in water levels when forest management practices start and old forest is cleared away through timber production (Waterloo et al., 2007; Komatsu et al., 2008). Generally, water levels start declining once introduced vegetation starts using more water, especially in the fast-growing phase of the trees (Webb and Kathuria, 2012). A linkage to stem density was also found, as higher stem density indicates higher water use of the trees (Kagawa et al., 2009; Webb and Kathuria, 2012). Higher sapwood area was found to indicate higher water use of the trees as well (Kagawa et al., 2009). Thinning had a positive effect on water yield in the study by Webb and Kathuria (2012) where tree density was decreased after 14 years of plantation growth, resulting in higher streamflow for six years. Similar results were obtained in mountain ash forests in the study of Jayasurya et al. (1993).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 126
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Successional stage Primary productivity Aboveground biomass Belowground biomass Stem density Species richness Presence of a specific functional group Presence of a specific species type Species population diversity Species size/weight Population growth rate Sapwood amount Other 0 5 10 15 20 25 30 35 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 3: Biotic attributes contributing to freshwater provision.
In general, abiotic factors affecting water levels were identified comprehensively (Figure 4). In most of the studies, precipitation (52), soil (42), evaporation (or evapotranspiration) (39), temperature (27), and slope (26) were mentioned, and in many studies these factors were assessed. Important, but minor, attention to geology (9), snow (9) and wind (7) was given. Rainfall was cited as having a positive effect on water levels and drinking water availability. In contrast, evaporation, or evapotranspiration, has a negative affect by releasing more moisture to the atmosphere. Elevation was also found to affect water yields as in higher altitudes the evaporation conditions are different (Bruijnzeel, 2004). Total annual rainfall and net precipitation increases at higher elevations where water inputs also come from fog interception (Gomez- Peralta et al., 2008). Soil characteristics were often mentioned, but not often concretely assessed. As indicators, soil characteristics were used in nine studies. Buytaert et al. (2007) found that soil destruction during the afforestation process may hinder infiltration and stimulate surface runoff and, later, decrease low flows. Soils were also considered as indirect indicators with soil characteristics being measured in some studies where soil moisture was measured (e.g. Holdsworth and Mark 1990; Brujnizeel 2004; Dierick and Höscher 2009), or soil characteristics assessed (Nosetto et al. 2005; Petheram et al. 2002).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 127
Temperature Precipitation Snow Water availability Evaporation Wind Nutrient availability Water quality Soil Geology Slope Other 0 10 20 30 40 50 Number of studies Negative Both (Positive & Negative) Positive Unclear
Figure 4: Abiotic factors influencing freshwater provision.
Human input and management
Direct human inputs were mentioned many times. In particular, conversion of grasslands to forests or replacement of native forest with invasive species or logging showed direct human input to water levels, however, the effect on water availability may also be indirect. Climate change was mentioned as an indirect human input that directly affects water availability (Bruijnzeel, 2004; Cavaleri and Sack, 2010; Garmendia et al., 2012; Mitchell et al,. 2012; Bangash et al., 2013).
By choosing the right forest management practices, water levels can be controlled, for example, by choosing suitable tree species for the site (Dierick and Hölscher, 2009). Conversion of land, land preparation, plantation and thinning are the main human induced factors affecting the water levels or the availability of freshwater (Fahey and Watson, 1991).
Thresholds
Thresholds found in the reviewed articles were categorised as ‘other biophysical thresholds’, but they could also indicate indirectly ‘safe minimum standards’. The intensity of afforestation seemed to affect the change in water yield, as in the study of Fahey and Watson. (1991) where two thirds of the watershed was planted, resulting in a 20% reduction in water yield. As discussed in the section on links between natural capital and ecosystem services, the meta-analyses by Bosch and Hewlett (1982) and Sahin and Hall (1996) found varying changes in water yield per 10% change in vegetation cover, depending on the vegetation type. However, Bosch and Hewlett (1982) found that for changes in vegetation cover affecting less than 20% of the catchment area, changes in stream flow cannot be detected using standard measurement techniques. Sahin and Hall (1996) report that the relationship between changes in water yield and vegetation cover that they derived was valid only for vegetation changes affecting at least 20-25% of the catchment area, and preferably above 40%. Similar results were reported by other authors. Hornbeck et al. (1993) found that the reduction in basal area must exceed 25% before responses to water yield could be
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 128
seen in measurements. Williams et al. (2012) suggest that the changes seem not to be detectable if harvesting affects less than 20% of the forested catchment area. Observations in the Amazon also suggest that deforestation should exceed 20% of the area before changes in water yield are significant (Nosetto et al., 2005). However, for all these studies, it is not clear whether there is a biophysical threshold or whether it is just that standard measurements cannot detect smaller changes in stream flow.
Farley et al. (2005) found differences between areas with different shares of runoff of precipitation. In regions where natural runoff is 30% of the mean annual precipitation, they estimated that runoff would change to half of precipitation when afforestation takes place. In regions where runoff is less than 10% of mean annual precipitation, afforestation is expected to result in complete loss of runoff. In the study of Petheram et al. (2002), it was discovered that precipitation over a threshold of 250 mm would result in 30% of water above the threshold becoming recharge. Bren and Hopmans (2007) found a limit of 900 mm of annual precipitation has to be reached before any runoff could be generated.
Indicators
The JRC indicators that were used in studies included water supply (precipitation) (n=29), water supply (14), land cover (11), surface waters (9), soil characteristics (9), water supply groundwater (4) and distance to water resources (1). Precipitation was used many times as measured data or as given meteorological data, as primary data or in a correlative model with other measured data. Precipitation in general had a positive effect on water levels, whilst in contrast, evaporation had a somewhat negative effect.
Water supplies were assessed as being a surrogate with runoff (e.g. Egoh et al., 2009), streamflow (Putuhena and Cordery, 2000; Webb and Kathuria, 2012), and annual water yield (Ruprecht and Stoneman, 1996). Also ground water levels have been assessed (e.g. Nie et al., 2012; Egoh et al., 2009; Ruprecht and Schofield, 1989). Water provision was also assessed through the precipitation-evaporation balance (e.g. Chan et al., 2006; Muller, 2009). Water storage capacity has been studied by Vertessy et al. (2001) with sapwood area index, sap flow, leaf area index, rainfall interception and soil and litter evaporation measurement. In the study of Takahashi et al. (2011) water storage capacity was also assessed through canopy parameters. In a recent study of Baral et al. (2013), the provisioning of water was measured in terms of the filtering, retention and storage of freshwaters available for human use. This study discussed water provisioning as an ecosystem service.
Policies
The policies that were mentioned in the literature review were mainly local or national policies to improve water availability and quality, for example, Integrated Water Resources Management (IWRM) (Blumenfeld et al., 2010), subsidies for water conservation (Holdsworth and Mark, 1990) or The National Water Act (Le Maitre et al,. 2002). Other policies mentioned related to climate change and forests: United Nations Framework Convention on Climate Change (Buytaert et al., 2007), Kyoto protocol (Farley et al., 2005) and Sustainable Forest Management (Blumenfeld et al., 2010). The remaining policies that were mentioned were biodiversity-related (Egoh et al., 2008; 2009).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 129
Interactions with other ecosystem services
From the ecosystem service perspective, the studies were closely linked to the trade-off between timber production and water provisioning because in many studies the effects of faster growing invasive species on water levels were studied and compared with catchments with native cover. Water quality was mentioned when drinking water was mentioned because clean water indicates freshwater (e.g. Webb, 2009; Blumenfeld et al,. 2010). Forests were noted to improve water quality (Bruijnzeel, 2004; Sun et al,. 2006; Egoh et al., 2008). However, changes in land use, e.g. forest management practices, may alter the water quality conditions (Ruprecht and Stoneman, 1996), also in the form of nutrients (Holdsworth and Mark, 1990). Forest soils and vegetation capture has positive effects in the form of flood control and the impact is positive for water resources (Bruijnzeel, 2004; Sun et al., 2006; Blumenfeld et al,. 2010). This also means that less sediment and other impurities transfer to water flows, and water quality improves (Blumenfeld et al., 2010).
Afforestation was found in many studies to have a positive impact on atmospheric regulation. However, the impact of atmospheric regulation on water yields was difficult to interpret (unclear in 10 articles). The results vary between tree species and other site conditions. A negative relationship was clearly indicated in three studies (Buytaert et al., 2007; Cavaleri and Sack, 2010; Simonit and Perrings, 2013), whilst a positive effect was clear in five studies (Hamilton, 1995; Chan et al. 2006; Kagawa et al,. 2009; Webb, 2009; Saenz and Mulligan, 2013). It is, however, notable that the actual trade-off between ecosystem services was not studied in these articles.
Overall observations on the review
For the purpose of this review, the interpretation of freshwater was challenging to find from the articles. It is important to acknowledge that there were differences in what was considered as freshwater in these articles, so they are not all comparable with each other.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 130
Food production (cultivated crops)
Laura Miguel Ayala, ALTERRA, The Netherlands
The literature search
NOTE: The food production review did not follow the same protocol as the other reviews as no biodiversity search terms were used. The review focuses more on abiotic factors.
The reviewed articles were compiled in the EndNote programme. The EndNote database contains 70 articles, among those only 50 were suitable and entered into the OpenNess database. In the EndNote programme articles were grouped per keyword (/group of keywords) search. The most used keywords and keyword combinations that expressed the link of natural capital with food production were:
-Food production AND: soil quality, soil nutrients, nutrient availability. -Food production AND: climate change, precipitation, temperature. -Crop Yield AND: organic matter, soil quality, food security. -Crop Yield AND: wetlands, water quality, water stress, soil salinity, water supply. -Crop Yield AND: soil type, shading trees, soil texture, soil biota. -Agriculture AND: delta
The most frequently used database was Web of Science with the option of narrowing down the first searches with more specific keywords. The search procedure consisted of ‘food production’ AND ‘other criteria’ (above listed). Scopus and Google Scholar were also used as support databases in some occasions. Some articles were found after reviewing the literature of other articles.
During the search a large amount of general and theoretical (not empirical) articles on ecosystem services and agriculture/food production were found. There were also several studies based on literature reviews and meta-analyses and not as many on individual case studies in the hits obtained in a preliminary search.
The articles reviewed were published within a large range of time. However, the search was focused on the most recent publications available and more than 70% of the articles reviewed (36) were published after the year 2000 (Figure 1).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 131
9
7
5
3 Number of Articles 1 1975 1985 1995 2005 2015 Years
Figure 1: Publication dates of the 50 journal articles reviewed on food production
Background/context
The studies had a wide geographical spread, but more than half were in North America or Europe (Map 1).
Map 1: Location of studies on food production.
The studies presented in the reviewed articles focused on diverse temporal scales with a predominance of ‘seasonal’ (Figure 2), where the ecosystem service of food production was studied in the context of the crop growing period and harvesting. Some studies presented long-term historical data, and to a lesser extent future projections (scenario analysis), a momentary condition in a certain location (snapshot), and annual studies.
As Figure 2 shows, more than half of the locations mentioned in the articles (124 in total) referred to a sub- national scale and a considerable amount of the studies were carried out at a local scale. Only in a few cases did the locations of the studies cover national or sub-continental scales.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 132
Time scale Spatial scale Continental Global 2% 1% National Scenario Snapshot 6% analysis 16% 16% Local 39%
Long- Sub- term national Seasonal 24% 52% 36% Annual 8%
Figure 2: Percentage of articles at each temporal and spatial scale for food production.
Table 1 summarises the range of ecosystem types found in the review of food production. ‘Cropland’ was present in almost all of the articles reviewed due to the fact that it is a key ecosystem type in food production/crop yield. Its condition was rarely mentioned except in a few cases where the ecosystem status was declining. Other ecosystem types often mentioned and importantly linked to the production of food were ‘wetlands’, ‘grassland’, and ‘woodland and forest’.
Table 1: Ecosystem types present in the review of food production.
Ecosystem Favourable Unfavourable Condition Total - declining not mentioned Woodland and forest 6 6 Heathland and shrub 1 1 Grassland 1 6 7 Cropland 1 7 39 47 Urban 1 3 4 Wetlands 1 6 3 10 Rivers and lakes 1 1 Coastal 1 1 Marine inlets and transitional waters 1 1 Total 2 17 59 78
Linkages between natural capital and the ecosystem service
The ecosystem service providers were mainly classified as either ‘single specific species population’ or ‘two or more specific species populations’. One study focuses on functional groups (perennials and annuals). The most frequently mentioned were: fruits, potato, legumes, rice, soybean, wheat, maize, sunflower, nuts, sugar beet and barley.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 133
The review found little information about biotic attributes due to the different search protocol used. However, abiotic factors were found to be very important attributes for food production (Figure 3). In some of the articles they had a direct link with crop yield (some of these interactions are further explained below). The most important ones were: Temperature and precipitation: reduction of the wet season and increasing temperatures can affect crop yield by e.g. altering planting dates, shortening the growing season, generating soil water stress, etc. Nutrient availability, which contributes to the fertility status of the soil. Soil: on some occasions yield declines were related to soil properties, which were affected by prolonged soil wetness or soil nutrient depletion. Some of the articles mentioned the role of microorganisms in food production, through soil biota experiments in the field and its effect on crop yield (human induced). Water availability and water quality: especially in the case of rainfed crops, the yield can be limited by the availability of soil water during the summer growing season. Salinity in irrigated agriculture can also have negative effects, causing a reduction in the weight and number of fruits yielded.
Temperature Precipitation Snow Water availability Evaporation Nutrient availability Water quality Soil Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 3: Abiotic factors influencing food production.
Human input and management
Food production is an ecosystem service which usually involves direct human input and management (44 out of 50 articles). In the studies reviewed, human interventions were oriented towards increasing food production to meet food security, coping with climate change impacts in agriculture, or keeping up with the increasing market demand and trade due to the increase in world population growth, among other factors.
Human input was mentioned on several occasions with the purpose of enhancing crop yield through, for example, fertilizer application. Other interventions referred to the use of machinery and its impacts (e.g.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 134
soil compaction and its effects on crop yield), or upstream dam construction affecting the downstream recharge of wetlands and groundwater, which negatively affects the water supply and quality of agricultural fields in the wetland area. Intensification of agriculture (19 articles) or agricultural expansion were also mentioned in some of the studies.
The impact of human activities was not always clearly specified in the articles reviewed. It was considered to be a positive impact on 6 occasions and a negative impact in 11 (out of 50) articles.
Thresholds
Thresholds were not usually defined in the articles. The few thresholds that were specifically mentioned belonged to the categories of ‘safe minimum standards’ and ‘other biophysical’ and referred to: the decreasing value of ecosystem services with the increasing value of agricultural services; crop exposure to certain temperatures; organic matter content in the soil; soil thickness; groundwater levels; and soil salinity (in terms of electrical conductivity). No other categories of thresholds were found.
Indicators
The most frequently cited indicator was ‘crop yield’ (usually in kg/ha or tons/ha, primary data), which was used on 33 occasions. Some of the articles mentioned ‘agricultural production’ or ‘commodity production’ (e.g. in $/ha).
Other indicators cited were: water productivity (kg food/m3 water evapotranspired), food security (average daily production of calories/household member), livestock production (e.g. cow ownership), agricultural land (ha), and parameters considered to measure soil productivity (e.g. organic matter, soil moisture content).
Policies
The studies that were set up in Europe mentioned some of the following policies: Biodiversity Strategy, Habitat Directive, Common Agriculture Policy, Nitrates Directive, Water Framework Directive and Environmental Impact Assessment. In the studies located outside Europe, the policies listed in the database were not applicable.
Interactions with other ecosystem services
The ecosystem services most mentioned in relation to food production were freshwater provision, water flow regulation (flood protection) and water quality regulation. The availability of water seems to be of high significance in terms of quantity and quality for both rainfed and irrigated agriculture. The costs of obtaining water of sufficient quality to ensure food production are increasing. In cases of increased demand of water and aquifer overexploitation, e.g. due to lack of rain or changes in rainfall patterns, the risk of higher salinity of groundwater may increase in coastal areas due to groundwater intrusion. Shortage of freshwater for agriculture can lead to lower yields.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 135
Water quality is seriously threatened by the agricultural use of fertilizers and pesticides, which can also leach to adjacent ecosystems (e.g. causing nitrogen fertilizer pollution or heavy metal accumulation) and thus reduce production. Wetlands play an important role in water quality regulation due to their cleansing and recycling capacity. Wetlands are also buffer ecosystems against extreme events such as droughts and flooding.
Mass flow regulation (erosion protection) was mentioned in some of the articles. According to these, soil erosion decreases soil productivity through the decrease in soil organic matter content. Severely eroded soils showed significantly lower yields than slightly eroded or moderately eroded soils. Crop maturity (e.g. corn) can be delayed by increased erosion.
Atmospheric regulation (carbon sequestration) was mentioned in some studies in relation to food production. Some of the articles registered large carbon sequestrations from agricultural land uses, e.g. from olive and almond farming, while vineyards are net CO2 emitters.
Overall observations on the review
The evidence presented in the articles was mainly empirically based (26 articles), or a combination of both empirical and modelling (14 articles). Only a few were purely modelling studies, or did not explicitly mention the method used. In 37 of the articles the evidence was direct, and in 11 it was indirect (using a surrogate). Regarding the observations taken, almost all the articles registered multiple observations (44 articles). The evidence was mostly quantitative (35 articles) or a combination of both quantitative and qualitative (13 articles).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 136
Air quality regulation
Erik Stange, Norwegian Institute for Nature Research - NINA
The literature search
Air quality regulation was a service that was new for the OpenNESS WP3 review. The title of this service has many synonyms, in addition to a number of alternative meanings that are largely unrelated to ecosystem services. The following process was used to identify the 50 articles included in our review:
1. ISI search term “Air Quality Regulation”: 1025 hits 2. Refine search with term with Web of Science category “Environmental Studies”: 88 hits, and “Environmental Sciences”: 350 hits. In both cases, most articles had very low ecosystem service (ES)-natural capital (NC) relevance (articles pertained to policy issues, technology, etc.) 3. Search result with “air quality regulation” and “ecosystem” as added search term: 48 hits. Low degree of ES-NS relevance: 9 articles, based on contents of titles/ abstracts. Contained a number of reviews that facilitated both snowball and reverse snowball search techniques. 4. Search result with “air quality regulation” and “service” as added search term: 54 hits. Again low degree of ES-NS relevance: 3 new articles that allow for snowball and reverse snowball searches. 5. Search result with “air quality regulation” and “biological diversity” as added search term: 3 hits. No new relevant articles. 6. Search result with “air quality regulation” and “biodiversity” as added search term: 18 hits. No new relevant articles. 7. New search, with “air quality” and “ecosystem service”: 94 hits. Moderate degree of relevance to an assessment of the “contributions of natural capital stocks to ecosystems services”: approximately 45 (containing many reviews), with the bulk of the remainder being policy and valuation articles. However the results did provide ample material to conduct an intelligent search. 8. Intelligent search approach: Through a review of the articles found above, we found the following useful synonyms: air pollution removal/absorption/mitigation, urban forest/ forestry/ ecosystem/ vegetation, particulates. Continued searching of ancestors and descendants produced a final count of 50 articles from the primary literature (Figure 1).
8
7 6 5 4 3
Number of Articles 2 1 1985 1990 1995 2000 2005 2010 2015 2020 Years
Figure 1: Publication dates of the 50 journal articles reviewed on air quality regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 137
Background/context
Studies included in this review investigated the extent to which vegetation (primarily trees) is capable of reducing components of atmospheric pollution such as ozone (O3), particulate matter (PM10), sulphur dioxide (SO2), carbon monoxide (CO) and nitrogen oxides (NOX). Ground level ozone, particulates and nitrogen oxides are recognised as the most significant air pollutants in terms of health impacts, and these pollutants tend to be particularly problematic in urban areas where a growing proportion of the world’s population now lives. In Europe, it is estimated that 75% of Europe’s population lives in urban areas, and up to 60% of the urban population has been exposed to dangerously high levels of ambient ozone since 1996 (EEA 2009). Ecosystems can improve air quality in a number of ways. Urban trees can influence the microclimate of cities through shading and evapotranspiration, changing wind patterns, modifying boundary layer heights, and reducing energy consumption related to heating and cooling buildings. Tree foliage can also intercept airborne pollutants, increasing dry deposition and thus removing pollutants from the atmosphere. This can occur either through the plant stomata, where gaseous pollutants are absorbed by the plant (stomatal deposition), or through interception of airborne particles on the vegetation surfaces (non-stomatal deposition). Alternatively, some groups of vegetation generate higher levels of biogenic volatile organic compounds (BVOC), such as isoprene and mono-terpene, which are emissions that can act as precursors of secondary air pollutants such as ground level ozone and particulate matter. Some trees periodically producer large volumes of pollen, which can have important negative effects on air quality for the residents who live nearby.
The articles generally all either identified or acknowledged most or all of these mechanisms by which ecosystems can influence air quality. Yet the primary aim of most articles was an attempt to quantify either the effects vegetation had on dry deposition (with regards to either non-stomatal or stomatal) or the potential deleterious effects plant BVOC emissions had on air quality. The vast majority of the literature reviewed (all but 4 articles) detailed studies that investigated air quality impacts of vegetation either in a particular urban setting (a single city), or comparisons of effects between different cities. Consequently, the spatial scales tended to be predominantly local or occasionally regional when investigations addressed the impacts of peri-urban forests (Figure 2). Only one article investigated air quality regulation on a national scale, which happened to be in Great Britain. Europe and North America were the most common areas studied (Map 1). Only seven articles focused on urban areas located on other continents.
Most studies (26) included in the review used a snapshot temporal scale, which we defined as a study duration ranging from 1 month up to a single complete year. Eight studies used an annual temporal scale, which we defined as at least a complete year but not more than 2. Six studies were classified as long-term, ranging from seven to 30 years (Figure 2).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 138
Time scale National Spatial scale 2% Global 1% Sub- national 4% Scenario analysis 18% Long- term 8%
Snapshot Annual 58% 12% Local 93%
Seasonal 4%
Figure 2: Percentage of articles at each temporal and spatial scale for air quality regulation.
Map 1: Location of studies on air quality regulation.
The ecosystem types addressed by the studies in the review were again dominated by those found in and around urban environments (Table 1). Only six articles did not explicitly mention the urban environment. Trees are the dominant vegetation category capable of influencing deposition rates, and therefore woodlands and forests were also a common focus of the studies (30 studies). Other ecosystem types were occasionally included when the research approach involved quantifying all types of ground cover (vegetation) and then determining the corresponding contributions to air quality improvements. Studies only rarely addressed the condition of the ecosystems they mentioned. In only six cases did authors describe the conditions of urban ecosystems as “unfavourable and declining”.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 139
Table 1: Ecosystem types present in the review of air quality regulation.
Condition Unfavourable Ecosystem Favourable not Total Recovering No change Declining mentioned Woodland and forest 2 3 25 30 Heathland and shrub 1 5 6 Grassland 1 7 8 Cropland 8 8 Sparsely vegetated 5 5 land Urban 2 1 6 35 44 Wetlands 4 4 Rivers and lakes 6 6 Coastal 1 1 Marine inlets and 1 1 transitional waters Total 3 5 2 6 97 113
Linkages between natural capital and the ecosystem service
Ecosystem service providers (ESP) Urban trees are the primary vegetation capable of having an impact on air quality, and hence urban trees were the main ESP addressed in nearly all studies included in the review. Because trees are the dominant component of woodland habitat, it was often difficult to ascertain whether the authors treated urban trees as a dominant community, an entire community or habitat (i.e., “urban woodlands”), or a single functional group. Plausible definitions for these ESP categories overlapped, and one could easily combine them into a single category in the case of this service. Studies which we chose to place in either the two or more communities or habitats (5) or two or more functional groups (4) could also potentially be combined into a single category. These studies often compared differences in either particulate absorption between dominant tree types (i.e., conifers vs. broadleaves) or differences between groups of trees that varied in the BVOC emission rates. Depending on the context, these groups could be considered either a functional group or a habitat type, and the differences in terms of how many studies are assigned to each category might have little actual distinction. The studies which were classified as looking at either a single species or population (5) or two or more specific populations (5) all identified the air quality regulation attributes of particular species, either because the samples used for measurements were all of the same species (e.g., Freer-Smith et al., 1997), or because species specific-attributes were identified and compared across two or more species (e.g., Owen et al., 2003). However even in these cases, the effective ESP, urban trees, was the same.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 140
Biotic attributes Leaf area index (LAI) and canopy cover were frequently cited as two biological attributes of urban woodlands that have the greatest positive impacts on provision of air quality improvement (shown as “Other” in Figure 3). When leaves intercept airborne particulates, deposition of particulates for a given area increases asymptotically with increasing leaf abundance. Canopy cover could be treated by the articles’ authors as either an attribute of an individual tree (DBH and corresponding crown area e.g., Donovan et al., 2005; Millward and Sabir, 2011; Paoletti et al., 2011; Baro et al., 2014), a function of species abundance within a study area (e.g., McPherson et al., 1998; Yin et al., 2011), or the community or habitat area of urban woodlands (e.g. Powe and Willis, 2004; McDonald et al., 2007; Jim and Chen, 2008). Additionally, community and habitat area was treated by authors in either relative (i.e., the per cent of land area covered by trees e.g., Lovasi et al., 2013) or absolute terms. Authors generally regard LAI as a close correlate of canopy cover, with some articles treating LAI explicitly as a distinct attribute (Owen et al., 2003; Escobedo and Nowak, 2009; Pugh et al., 2012), or expressing leaf density in terms of above ground biomass (Benjamin et al., 1998; Donovan et al., 2005).
Many studies also addressed some of the potentially deleterious effects of urban tree abundance on air quality. Tree species vary in their BVOC emission rates (Taha, 1996; Donovan et al., 2005), and trees with higher BVOC emissions can contribute to higher concentrations of ground level ozone. Some studies treated groups with differing BVOC emission rates as different functional groups (Alonso et al., 2011) and addressed the functional diversity of urban woodlands (Nowak et al., 1998). However, studies included in the review that addressed the ozone forming potential of urban trees all concluded that the net effect of trees on air quality was positive (e.g., Morani et al., 2011).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 141
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Primary productivity Aboveground biomass Stem density Landscape diversity Species richness Functional richness Functional diversity Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Life span/longevity Mortality rate Other 0 5 10 15 20 25 30 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 3: Biotic attributes contributing to air quality regulation.
More recently published articles investigated the effects the physical structure of urban trees have on air quality with regards to air flow patterns. Five articles discussed evidence that trees planted along corridors with high levels of automobile traffic are capable of creating a canyon that prevents atmospheric mixing (Jim et al., 2008; Pugh et al., 2012; Wania et al., 2012; Amorim et al,. 2013 and Salmond et al., 2013). It is unclear whether this should be treated by our review as a biological effect (species abundance) or a purely abiotic effect with regards to the physical structure the trees create.
Abiotic attributes Temperature was the most commonly cited abiotic attribute, with most studies concluding that it had a negative effect on air quality (Figure 4). For many of these studies, this was because the formation of ground level ozone and other secondary pollutants is greater at higher ambient temperatures. Warmer temperatures were often associated with increased drought stress, and decreased gaseous uptake (stomatal deposition). Yet the temperature effects on air quality improvement are decidedly non-linear, and the relationship for a given study depends on the temperature ranges experienced by the location. For lower temperatures, increasing temperatures can increase plant metabolic activity and positively impact air pollution mitigation. This relationship can then become negative at higher temperatures for the reasons listed above. Consequently, temperature was listed as both positive and negative by eight studies. Only three studies viewed temperature as an abiotic attribute with a positive effect on air quality.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 142
Wind was the next most frequently cited abiotic attribute that influenced how ecosystems influence air quality (Figure 4). For the most part, wind was treated as having a positive effect because the authors contended that wind speeds increased dry deposition rates. Yet the relationship between wind speeds and deposition rates is exceedingly complex and is heavily influenced by the spatial arrangement of vegetative structures and wind directions. Wind can also re-suspend previously deposited particles. An additional six studies therefore listed the role of wind speed on air quality as either both positive and negative, and one article (Nowak et al. 2013) treated it as only negative.
Primary abiotic effects addressed in the reviewed literature also included evaporation, or more specifically evapotranspiration—which is a combined biotic and abiotic effect. Eight studies listed this as having a positive effect on pollution mitigation, and in four studies it was concluded that its effect was unclear. Other abiotic attributes that were recognised as having an impact on biotic regulation of air quality were often related to solar radiation (both with regards to potential albedo effects and as a stressor that would diminish vegetation’s gaseous uptake), air humidity (which could also be categorised as water availability), spatial dynamics of ozone concentrations, boundary layer height for atmospheric mixing and the physical structure of the built environment. Effects of these attributes were either positive (e.g., higher albedo from vegetation lowers surface temperatures and production of secondary pollutants like O3 and NOx), negative (e.g., solar radiation stresses plants), both or unclear.
Temperature
Precipitation
Water availability
Evaporation
Wind
Soil
Slope
Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 4: Abiotic factors influencing air quality regulation.
Human input and management
The ecosystem contributions to removal of air pollution required an interpretation of human management that might differ in important ways from the interpretation of management involved in provision of other ecosystem services. As stated above, the vast majority of articles reviewed addressed air quality regulation in urban environments as provided by urban and peri-urban trees. Urban trees are predominantly managed
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 143
in the sense that most trees are planted in specific locations and also often receive continued care (e.g. watering, pruning and replacement). Peri-urban woodlands also benefit from management in the sense that their continued existence is the result of conservation measures that have prevented these areas from being developed. Accordingly, human management serves to establish or maintain the biophysical structure that ultimately provides the service. This category of management will obviously differ from ecosystem services like provision of foodstuffs, where management is required in conjunction with the ecosystem processes and functions of the ESP. In both cases, management and human inputs facilitate delivery of the ecosystem service, but they occur at different steps along the cascade model.
Thresholds
Information from the articles in this review regarding thresholds generally involved addressing whether municipalities were in compliance with standards for pollutant concentrations (O3, PM10, NOX, CO). Eight studies mentioned such thresholds for compliance in a way that could be interpreted as “safe minimum standards”, and an additional four studies were categorised as mentioning “legal boundaries”. In some cases, this is an arbitrary distinction, because standards set by the World Health Organization often also become a threshold for legislated policy. Specific standards often varied by the governing body responsible for the geographic region where the study took place (U.S. state standard, U.S. national standard, EU member state standard, etc.). These types of thresholds may have minimal relevance in terms of comparisons across ecosystem services types, especially because the discussions do not concern thresholds of the biophysical structure itself (i.e., how much urban land area needs to be forested or “green” to ensure provision of the air regulation service?) Moreover, these thresholds and the variables that they concern are most heavily influenced by industrial activity that is the predominant source of air pollution.
Indicators
Air purification was a frequently cited indicator, with 24 studies using mechanistic models and seven studies providing primary data. Likewise, many studies used mechanistic models to report deposition velocity (17), or provide data for this indicator (8). Twenty-nine studies employed primary data on pollution concentrations, because the models used in the majority of investigations required this type of data. An additional nine studies addressed pollutant concentrations with mechanistic models. Tree cover data was also an important component of the modelling for many studies, with 31 studies using primary data as a model input and five studies using mechanistic models to arrive at estimates of tree cover.
Other variables that could be interpreted as indicators of ecosystem service delivery were grouped into the following categories: production of BVOC, market values of air purification, traffic volume, pollutant particle trapping efficiency (%), canopy resistance, mixing height and vegetation biomass.
Policies
Policies were listed in thirteen of the articles reviewed. Four articles explicitly mentioned the Ambient Air Quality directive. Nine articles mentioned other policy initiatives. These policies included the 2010 EU directive on industrial emissions, the Kyoto emissions agreement, the US National Ambient Air Quality standards (2 studies), China’s national air standards, and the NYC million tree planting initiative.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 144
Interactions with other ecosystem services
Authors of the studies included in the review often listed many other ecosystem services that were complementary to air quality regulation, and in only one isolated occasion did they mention an associated ecosystem disservice: pest regulation, in the sense that urban trees and woodland provide habitat for various pest species. Among the specific ecosystem services included in the OpenNESS review, the studies mention the positive contributions urban trees and other vegetation made to aesthetic attributes (18 times), atmospheric regulation (14 times), mass flow regulation or erosion protection (6 times), water flow regulation or flood protection (12 times), recreation (7 times) and timber production (3 times). The most frequently mentioned ecosystem service of all, however, was microclimate regulation (21 times), which was classified as “other.” Trees contribute to microclimate regulation by providing shading and wind protection, which—among other benefits—also feeds back into air quality improvements through decreasing temperatures and energy consumption and limiting ozone formation. None of these linkages were addressed in a quantitative sense or supported by empirical evidence.
Overall observations on the review
The lasting impression from this review is that the evidence supporting the benefits of an urban area’s green infrastructure is not particularly strong, nor is there reason to believe that the magnitude of the positive impact is very large. Studies of this service are heavily dependent on modelling: 30 studies were categorised as “modelling” and eight were categorised as “empirical and modelling.” And while almost all models were parameterised and run with actual data of pollution levels and ground cover data, the results are difficult to verify and in only a few cases connected to human health statistics that could test whether reductions have any true human benefit. (Some of the eight “empirical and modelling” studies could actually be better categorised as “modelling.”) To paraphrase from Setälä et al. (2013), the majority of studies suggesting that urban vegetation is an efficient remover of airborne pollutants are based on large- scale modelling in which the removal of the pollutant is simply a function of deposition velocity, pollution concentration, and some parameters describing forest structure (such as biomass and leaf area index). Such models are based on data from plant chamber studies performed in the field or in greenhouses where factors influencing plant physiology and their physical/chemical processes have been studied, as well as on studies measuring fluxes of air pollutants often above the tree canopy (e.g. Duyzer et al., 1995; Hargreaves et al., 1992). Only a few studies exist in which pollutant fluxes have been quantified within forest or at the tree canopy where the uptake of pollutants happens.
It was difficult to establish clear criteria which would determine whether the evidence should be categorised as direct or indirect. All but one study were quantitative (three were both quantitative and qualitative), and most studies used multiple observations in the sense that studies were replicated either across time or space. But even when modelling results demonstrated air quality improvements, the absolute values were quite small with respect to total pollutant concentrations and most authors were quick to assert that this particular ecosystem services was complementary to many others (see above) that are likely to be of greater importance. A review of urban ecosystem services (Pataki et al., 2011,) stated this reviewer’s overall impression best: “the removal of atmospheric pollutants by vegetation is one of the most commonly cited ecosystem services, yet it is one of the least supported empirically.”
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 145
Atmospheric regulation
Alison Smith and Sen Li, Environmental Change Institute, University of Oxford, UK
The literature search
50 articles were entered into the database from BESAFE and 10 additional articles were entered into the database to cover the period from 2012 to 30 June 2014 (Figure 1). Keywords were: (“Carbon storage” OR “Carbon sequestration” OR “Carbon loss” OR “Carbon emissions”) AND (Biodiversity OR “Biological diversity” OR Species OR Habitat* OR Genetic OR Trait* OR Function* OR Landscape OR Richness OR Abundance) AND (Tree* OR Soil* OR biomass)
12 10 8 6 4 2 Number of articles 0 1995 2000 2005 2010 2015 Publication year
Figure 1: Publication dates of the 60 journal articles reviewed on atmospheric regulation.
Background/context
Most of the studies were experimental measurements of vegetation biomass (which was then used to estimate carbon storage), carried out at a particular site or a set of sites within a region. Therefore the majority of the studies, 38 out of 60, were at the local level, with several of the studies covering multiple locations so that there were 45 local locations altogether. Fifteen studies were at the sub-national level, and only four were national, with just two being global (Figure 1). One modelling study had no location.
The geographical spread was very wide, with all continents being well represented (Map 1). There were four studies in Africa, 15 in Asia (of which nine were in China), 2 in Australasia, 23 in the Americas (spread across North, Central and South America) and 13 in Europe. The evidence did not appear to vary with the location of the study.
Most of the studies (58%) were snapshots of the amount of carbon stored at a particular time. However, 28% were long-term, having repeated the measurements over a period of several years, and 7% were seasonal or annual, with 8% being scenario analysis (Figure 1).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 146
Map 1: Location of studies on atmospheric regulation.
Time scale Spatial scale Annual National Global 3% Long- 3% 6% term 20% 0%
Scenario Sub- analysis national 6% 22%
Local 69%
Grand Total 71% Figure 2: Percentage of studies at each temporal and spatial scale for atmospheric regulation.
The main focus was on forests, which formed 62% of the 68 ecosystems examined (a number of studies looked at more than one ecosystem). The condition of the ecosystems was very rarely explicitly mentioned, but in 12 out of the 42 cases where forests were studied, we inferred that the condition was favourable, either because it was in a protected area, or because the article stated that it was an unlogged old-growth forest (Table 1). In one case the forest was destroyed (by fire), and in two cases we inferred that it was unfavourable but recovering (suffering from the effects of past timber harvesting, but now being protected or restored).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 147
The next most commonly studied ecosystem was grassland, forming 19% of the ecosystems examined, followed by cropland (10%). Heathland was only studied twice, and urban, coastal, wetland and sparsely vegetated ecosystems just once each. Again, condition was rarely mentioned for these ecosystems, although we inferred a favourable condition for one experimental grassland plot and one restored heathland site, and an unfavourable but recovering condition for a badly degraded grassland site that was being restored.
Table 1: Ecosystem types present in the review of atmospheric regulation.
Condition Unfavourable Ecosystem Favourable Destroyed not Total - recovering mentioned Woodland and forest 12 2 1 27 42 Heathland and shrub 1 1 2 Grassland 1 1 11 13 Cropland 7 7 Sparsely vegetated land 1 1 Urban 1 1 Wetlands 1 1 Coastal 1 1 Total 14 3 1 50 68
Linkages between natural capital and the ecosystem service
Ecosystem Service Provider (ESP) In 87% of the studies, the ecosystem service provider (ESP) was the entire community / habitat area (39 studies), or two or more communities/habitats (13 studies). This reflects the typical study format of measuring the carbon storage for an entire site, often comparing two or more sites against each other, e.g. intact forest vs. logged forest, or forest vs. grassland.
In 10% of studies, the ESP was one or more specific species populations. This was for studies of carbon storage in individual tree species, e.g. Pondorosa pines (Laclau, 2003) or Miscanthus grass (Zimmerman et al., 2012), or comparisons of different tree species (Kaul et al., 2010; Hector et al., 2011). Just two studies examined different functional groups: Steinbeiss et al. (2008) compared plots sown with different mixes of grasses, small herbs, tall herbs and legumes, and Cahill et al. (2009) compared C3 grasses and C4 grasses.
Biotic attributes The biotic attributes are shown in Figure3. The most commonly mentioned attributes are above- and below-ground biomass, which are positively correlated with carbon storage. This is not surprising, as most of the studies estimated carbon storage directly from measurements of biomass, typically by multiplying by a factor of around 0.5. It is notable that above-ground biomass was mentioned more often: 38 times vs. 26 times. This reflects the greater difficulty of measuring below-ground biomass, which usually requires
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 148
destructive sampling (i.e. digging up roots), whereas above-ground biomass can be estimated more easily by measuring parameters such as tree diameter and height. However, as a significant portion of biomass is contained below-ground, this also points to a weakness in the evidence base. For the two instances of a negative relationship between carbon storage and above-ground biomass, one refers to a study where carbon in a particular soil horizon was negatively correlated with above-ground biomass (Hollingsworth et al., 2008) and the other to carbon emissions from decaying biomass following a forest fire (Haugaasen et al., 2003).
Similarly, it is self-evident that primary productivity and population growth rate are positively correlated with carbon storage, and mortality rate (e.g. from forest fires or pests such as the bark beetle) is negatively correlated.
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Successional stage Primary productivity Aboveground biomass Belowground biomass Stem density Litter/crop residue quality Species richness Functional richness Functional diversity Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Population growth rate Life span/longevity Natality rate Mortality rate Wood density Leaf N content Predator behavioural traits (biocontrol) Other
0 5 10 15 20 25 30 35 40 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 3: Biotic attributes contributing to atmospheric regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 149
The next most common attributes were related to community/habitat. Presence of a particular community/habitat type or community/habitat area were linked to carbon storage in 30% of the studies. Again, this reflects the predominant study approach of measuring carbon storage at a specific site. Unsurprisingly, the studies typically found that carbon storage was higher in forests than in grassland or cropland, and higher in undisturbed old-growth forests than in logged forests. There was also a strong positive correlation with community age and structure, with 19 studies finding that older forests or plantations stored more carbon, not just because the trees were larger, but also because there was a greater variety of vegetation exploiting different heights and root depths in more mature natural forest. Similarly, 13 studies found that more structurally complex forests and those at a later successional stage stored more carbon – again because of the richer variety of trees, shrubs and understorey vegetation exploiting the available resources, and thus producing more biomass per unit area. However, the relation with stem density was not always clear cut: at higher stocking densities in plantation forests, tree growth could suffer as a result of competition, so that overall carbon storage could be reduced.
Ten studies found a positive correlation with species type, either for studies of individual tree species in plantations or experimental plots, or for natural forests where carbon storage is dominated by a few particular species of large or fast growing trees. Related to this, 11 studies found a positive relationship with species size and/or weight, and five with wood density. Three studies found a positive correlation of litter quality with soil carbon storage, though in three other studies the relationship was unclear.
The influence of species richness is clear from this review, with 15 studies showing a positive correlation with carbon storage, vs. just one showing a negative link, and one unclear. Fewer studies addressed functional diversity, but there were still eight showing a positive relationship, with four unclear and one negative; two showing a positive link with functional richness; and seven with a link to a particular functional group.
The ten newly added articles provide interesting new findings that contribute to the on-going debate over the link between biodiversity and carbon storage, with some studies showing a positive link and others showing that low-diversity plantations can offer very high carbon storage provided that the species planted are large. Gonzalez et al. (2014), for example, find a link between species richness and carbon storage for old growth and secondary forest in Peru. This is probably also linked to the later successional stage and hence larger size of the trees in the old growth forest (though the relationship holds true for secondary forest sites as well). Yet Wardle et al. (2012) find that although both carbon storage and biodiversity increase in old-growth boreal forests with the time since the last fire, there is no causal relationship, as both are driven by other factors (mainly the shift to more conservative plant traits as the ecosystem matures, plant composition shifts to polyphenol producers which immobilise nutrients, and declining nutrient availability).
Potter and Woodall (2014) shed further light on the apparent conflict with their analysis of data from across North America that shows a positive correlation between live above-ground biomass and species richness or functional diversity on low productivity sites with low stocking density, but a weaker or negative relationship as the productivity and stocking density of the sites increases. They conclude that niche complementarity is important to maximise carbon storage on low productivity sites, and on high productivity sites at early successional stages. However, as tree density increases towards the self-thinning
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 150
stage on higher productivity sites, the sites that support a high abundance of high carbon species (typically larger trees) are capable of storing more carbon, even if biodiversity is low. They highlight the importance of maintaining a high phylogenetic diversity in order to maintain the carbon storage service by providing community resilience to climate change, especially on low quality sites that are subject to higher levels of stress (e.g. from droughts).
Another potential explanation is suggested by Tran Van Con et al (2013) who show that species richness, species diversity and structural diversity are all correlated to above-ground biomass (AGB) (and therefore carbon storage). They hypothesise that species richness is correlated to AGB where forests are structurally complex, as the niche complementarity theory may apply when there are several canopy layers to exploit. This could help to explain why some studies have found that species richness does not appear to be correlated to biomass in structurally simpler forests such as those in Central Europe, and in the study by Wardle et al. of boreal forests (above). The finding is supported by the negative relationship they found between AGB and species evenness: as evenness tends to 1, indicating structurally simpler forests, the AGB decreases. They also find that the relationship between biodiversity and AGB is strong within sites, but weak when all sites are combined, indicating that the relationship may be obscured when combining results from sites with different environmental characteristics. This may also help to explain why some studies have found no correlation.
Cadotte (2013) provides further evidence for the theory of niche complementarity with an experiment showing that mixtures of prairie grassland species that are widely separated in evolutionary terms (high phylogenetic diversity) produce more biomass than mixtures of close relatives.
Lang’at et al. (2013) show how niche complementarity can work in young mangrove forests. They found that plots containing a mixture of fast growing, shallow rooted Avicennia and two slower growing, deeper rooted species produced higher root biomass than would be expected from the monoculture yields (this is called overyielding). They proposed that this is because plots containing Avicennia alone would suffer competition between individuals for root space, but that this competition is reduced when some of the individuals are slower growing, deeper rooted specimens from other species. After four years, the yield of the other two species was neither enhanced nor suppressed, but it is possible that suppression by the more vigorous and tolerant Avicennia could occur as the stand matures.
Cavanaugh et al (2014) also find evidence that both niche complementarity and the selection effect are important. Analysis of data from 11 natural tropical forests showed that aboveground C storage increased with both taxonomic diversity and functional dominance, specifically the dominance of trees with large maximum diameters. They propose three reasons to explain why the study found a relationship between diversity and carbon storage that was not apparent in other studies: larger plot sizes in their study (>1 ha); the global nature of the study (which contradicts the suggestion of Tran Van Con et al. that the relationship may be obscured when comparing sites with very different characteristics); and the focus on mature forests, because complementarity may increase over time.
Several articles address the issue for agroforestry systems – especially coffee plantations. Hager (2012) finds that species richness is correlated with increased carbon storage for agroforestry coffee farms in Costa Rica. However, Richards and Mendez (2013) find that initial species richness is correlated with the
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 151
carbon sequestration rate but not with overall carbon stock, in coffee agroforestry farms in El Salvador. It is also important to note, as Hager (2012) points out, that coffee plantations contain species that have been deliberately selected by the farmer, and therefore these studies cannot be used to infer the effects of biodiversity on carbon storage for unmanaged forests. Nevertheless, there could be important implications for future management of farms in order to increase carbon storage, possibly driven by Payment for Ecosystem Services (PES) schemes. Hager also discusses the potential for N-fixing leguminous shade trees to increase soil carbon storage by increasing productivity and thus litter fall.
Hulvey et al. (2013) employ a meta-analysis to investigate whether forests planted for carbon storage should include a mix of species. They find that tree species mixtures (of up to 5 species) produce equal or greater above-ground biomass compared to monocultures of the most productive species. Possible mechanisms include: reduced intraspecific competition, resource-use complementarity, reduced disease/pest damage, facilitation (for example, through the inclusion of N fixers or induced greater light- use efficiency) and altered nutrient cycling. They also highlight that mixtures that include nitrogen fixers are more productive than monocultures without N fixers.
Abiotic attributes Precipitation was mentioned 14 times, and water availability three times. Three studies found that precipitation and/or water availability increased carbon storage as trees grew better when more moisture was available (Hantanaka et al., 2011), root mass increased (Meier and Leuschner, 2010) and soils stored more carbon in cool, damp conditions (Beier et al., 2009; Persiani et al 2008). Three more studies implied that droughts or generally drier conditions associated with climate change might reduce carbon storage, by increasing fire risk (Conard and Ivanova, 1997; Haugaasen et al., 2003) and affecting the establishment of young trees (Law et al., 2003). In the remaining eight studies, however, the nature and direction of the effect was unclear.
The influence of temperature was more mixed, depending on whether heat or cold was the limiting factor for plant growth for the ecosystem and habitat in question. Lavorel and Grigulis (2012) found that alpine plant biomass decreased at higher altitudes due to temperature (cold) stress, whereas the study of European heathlands by Beier et al. (2009) found greater soil carbon storage in the cold, wet conditions of the UK compared to hotter conditions, and Conard and Ivanova (1997) cite the potential increase in fire risk with hotter, drier conditions as a result of climate change. Three studies cite both positive and negative effects. For example, Zhao et al. (2009) find that higher temperatures could increase the growth rate of trees in timber plantations while shortening the rotation length, but if temperatures increase above 2oC, this trend could reverse. Similarly, Meier and Leuchsner (2010) found that higher temperatures could increase the growth rate of vegetation, but droughts could damage forest productivity. Seidl et al. (2008) found that the bark beetle that damages trees develops more rapidly up to its optimum temperature of 30oC, but then decreases above that, and cannot develop at all at very high temperatures (above 38.9oC).
Soil type was mentioned 17 times, with more fertile soils contributing to higher productivity (Balvanera et al., 2005), but in most cases the relationship was unclear. Hager (2012) mentions soil carbon loss due to erosion on steep slopes. Other abiotic factors received little mention. Five articles mention nutrient availability, but the effect is mixed: Lavorel and Grigulis (2012) and Asuming-Brenpong et al. (2008) find that higher nitrogen levels stimulate plant growth and therefore increase carbon storage, but Beier et al.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 152
(2009) find that nitrogen deposition can also stimulate conversion of shrubland to grassland, with a negative impact on carbon storage.
Temperature Precipitation Water availability Evaporation Wind Nutrient availability Soil Slope Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 4: Abiotic factors influencing atmospheric regulation.
Human input and management
The predominant type of human input was logging and over-harvesting of forest products, which had a negative impact on carbon storage (e.g. Kirby and Potvin, 2007; Law et al., 2003; Qi et al., 2010), even for selective logging, because this led to the extraction of the largest trees and selective depletion of those species (Balvanera et al., 2005). Other negative human inputs arose from grazing (Beier et al., 2009; Klummp et al., 2009; Persiani et al., 2008), forest fires (Conard and Ivanova, 1997) and urban development (Yang Bei et al., 2011).
However, examples of positive impacts were also found where there was management to promote biodiversity and ecosystem services, such as on organic coffee farms (Hager, 2012), under no-till farming to enhance soil carbon (Mishra et al., 2010) or when fencing off grassland areas to protect from over-grazing (Sun et al., 2011).
In nine cases, both positive and negative effects were found: either for experimental systems that compared different treatments (e.g. the study of different planting regimes by Asuming-Brenpong et al., 2008), or for cases where forest or grassland that had been damaged by logging or overgrazing was now being protected or restored (e.g. Cahill et al., 2009; Glenday, 2008; Hector et al., 2011).
Thresholds
Only one article found a threshold: Chen (2006) reported that carbon storage increases with species richness but that it may saturate at a low number of species, after which it increases more slowly.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 153
Indicators
Table 2 shows the number of times that each indicator type was used. It can be seen that the most frequently used indicators were primary data on above-ground biomass (42 times), below-ground biomass (28 times) and soil carbon (35 times). These measurements were then used to derive carbon storage. Typically, the researchers would estimate above-ground biomass by measuring the diameter at breast height (DBH) of trees, and sometimes the height as well. They would then input these measurements into various “allometric equations”, which are experimentally-derived equations (from previous studies) that relate DBH and height to volume or mass for each species. For below-ground biomass, either similar equations could be used, or, for more accurate measurement, the trees could be dug up and the roots actually weighed – clearly a difficult task for large trees, as well as being destructive. For shrubs, herbaceous plants, dead wood and leaf litter, the usual approach was to destructively sample a sub-plot by harvesting and weighing. For soil carbon, cores were used to sample the soil at various depths, and the samples were dried and analysed to separate out the organic carbon from the inorganic. Some studies split this carbon pool into root biomass, microbes and fungi, soil humus, etc. “Forest biomass” was inferred in some studies simply by adding the above-ground and below-ground biomass (and sometimes soil carbon) for a forest area.
Carbon storage was usually estimated by multiplying the above- and below-ground biomass by carbon factors, usually around 0.48-0.50 (sometimes varying by species or forest type). Soil carbon was measured directly.
Table 2: Indicators used in atmospheric regulation studies.
Indicator Conceptual Correlative Mechanistic Primary data Total model model model Above ground biomass 1 3 3 35 42 Below ground biomass 1 2 4 21 28 Carbon storage* 2 6 13 21 42 Forest biomass 3 3 8 14 Land cover 5 5 NPP 4 11 15 Nutrient flux 1 6 7 Soil carbon 1 4 30 35 Soil Characteristics 2 13 15 Vegetation map 1 1 4 6 Other 7 7 Total 5 17 33 161 216 *Note: there is a lack of consistency in the categorisation of the method used to estimate carbon storage: most studies either used primary data (soil carbon) or estimated it as around 0.5 x biomass (classified as either conceptual, correlative or mechanistic). Studies varied considerably in the completeness of the sampling. Some studies looked only at above- ground biomass for the larger trees (typically DBH >10cm), whereas others exhaustively sampled every possible carbon pool.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 154
Several of the studies also used secondary indicators: land cover, vegetation maps and soil characteristics were often used, either in the process of obtaining detailed empirical measurements, or as a proxy for estimating carbon storage over a wider area. Some used NPP to estimate annual carbon sequestration. Often fast-growing young forests might have higher rates of annual carbon sequestration, whereas mature forests might absorb less carbon each year, but store far higher amounts per hectare. Some studies (mainly those focusing on soil carbon) also looked at the impact of nutrient fluxes such as nitrogen. The results of these studies were often very complex and hard to interpret, with no clear relationships between nutrient flux and carbon storage or sequestration emerging.
A variety of “Other” indicators were identified, not covered by the JRC classes. However, these were not used to estimate carbon storage: they were mainly indicators of functional traits such as leaf digestability (Lavorel and Grigulis, 2012), structural complexity (Keeton et al., 2010) or % canopy cover (Mills and Cowling, 2006), or of other important processes such as disturbance by fire (Jonsson and Wardle, 2010).
Policies
None of the pre-selected policies were explicitly mentioned in the articles. However, we recorded 14 studies in the “Other” category, five of which were mentions of REDD and/or the Clean Development Mechanism, one mention of the UNFCCC and one of the Kyoto protocol. There were some mentions of local/regional policies: Keeton et al (2010) mentioned the Carpathian Convention 2003, and Chen et al. (2010) says that there are local logging bans in place to protect forests. There were also some mentions of theoretical policy approaches that could be applied, e.g. Zhao et al (2010) say that urban forestry planning and management in China could be a method to mitigate climate change, and Hulvey et al. (2013) suggest a two-pronged strategy for designing carbon plantings including: (1) increased tree species richness; and (2) the addition of species that contribute to carbon storage and other target functions.
Interactions with other ecosystem services
A variety of other ecosystem services were shown to interact with carbon storage, as shown in Figure 5. The most obvious interaction was with timber provision, mentioned in 19 studies. This is an interesting case, as sustainably managed forests can obviously provide wood as well as storing carbon, yet when trees are harvested this clearly reduces the amount of carbon stored. Seven studies showed the negative impact of timber harvesting on carbon storage in forests. Some of these were based on empirical measurements (e.g. Glenday, 2008; Hector et al., 2011) or model estimates (Balvanera et al., 2005) of carbon storage in logged forests compared to undisturbed forests, whereas some were qualitative statements (e.g. Langat et al., 2013). Seven more studies mentioned the role of timber plantations in providing wood as well as storing carbon, though only two of these refer to examples of sustainable production: sustainable plantations on former pasture (Ruiz-Jaen and Potvin, 2011), and timber trees grown on coffee plantations (Hager, 2012). However, Laclau (2003) states that long rotation times for pine plantations can help to maximise carbon storage as well as timber provision, and Kaul et al. (2010) claim that plantations can relieve pressure on natural forests.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 155
Water quality regulation
Water flow regulation (flood protection)
Timber production
Recreation (species-based)
Potable water (quantity) Positive Pest regulation Both (Positive & Negative) Other Negative Mass flow regulation (erosion protection) Unclear Food production (cultivated crops)
Air quality regulation
Aesthetic landscapes
0 2 4 6 8 10 12 14 16 18 20 Number of studies
Figure 5: Interactions between atmospheric regulation and other ecosystem services.
Kirby and Potvin (2007) identify the potential for both positive and negative impacts: managed forest stores 86% more carbon than pasture (335 vs 46 Mg C/ha) but the preferred species for harvest are also those with highest carbon storage, so carbon depletion of the forest could occur through selective extraction. However, this could be mitigated by species management. In four cases, the net impact is unclear, e.g. Hulvey et al. (2013) note that planting species mixtures to maximise carbon storage could result in a trade- off if lower numbers of timber species are planted as a result; and Zhang et al. (2012b) describe pulpwood plantations with short rotation times (10 years) where carbon storage would be short-lived. However, Tolbert et al. (1999) claim that even for short-rotation coppice or miscanthus plantations for biofuel, carbon storage underground could be enhanced compared to pasture.
Food production is also frequently mentioned, with four articles mentioning the negative impact of forest clearance for agriculture on carbon storage, and one (Bunker et al., 2005) mentioning the potential loss of fruit production when natural forests are replaced with timber plantations. There are three positive interactions: Lavorel and Grigulis (2012) and Mishra et al. (2010) mention the benefits of increased soil carbon / soil fertility for food production in grasslands/croplands, and Yen and Lee (2011) mention the role of bamboo plantations in providing bamboo shoots to eat. Furthermore, two studies identify trade-offs between potential positive and negative impacts: Mills and Cowling (2006) describe how unsustainable browsing by goats has depleted soil carbon in shrublands, but how restoration can help to support a sustainable level of grazing, whereas Hager (2012) finds that organic coffee farms store more soil carbon, but that yield is lower – but this can be addressed by planting more leguminous trees to boost both nutrient cycling and carbon storage.
A wide range of other services are mentioned, mainly relating to the positive benefits of forests both for carbon storage and a range of other ecosystem services including aesthetic landscapes, air quality regulation, mass flow regulation (erosion protection), water flow regulation (flood protection) and
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 156
freshwater provision. “Other” services include medicinal products, rubber production, reduction of fire risk, the role of urban forests in cooling buildings, and maintenance of biodiversity/wildlife habitat as a supporting service. Most of these are based on qualitative statements citing other studies, typically in the introductions or conclusions of the studies.
Overall observations on the review
Whereas comments from the original BESAFE reviewers reveal some difficulty in identifying 50 relevant articles, the database update for 2012-2014 revealed a wealth of new material. It was difficult to restrict the update to just 10 new articles, and as a result a number of potentially very useful articles had to be omitted. The reviewer would recommend that the update be extended if sufficient time and resources become available.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 157
Mass flow regulation (erosion protection)
Frédéric Archaux, IRSTEA, France
The literature search
This literature search was carried out independently from the previous BESAFE review. It is interesting to note that the two independent 50-hits reviews did not share any common references! When listing first authors, only four first authors were in common. Among the 365 authors listed in the 100 studies, only 18 were in common. Part of the lack of correspondence is due to studies published since the BESAFE review: 13 of the 50 OpenNESS articles were published during 2013 - 2014 (Figure 1). The following report is based only on the second (OpenNESS) review.
Only seven hits were found on Scopus with the keyword “mass flow regulation” and none proved to be suitable. As a result, a complementary search was done with keywords “soil erosion AND quantification” providing 354 hits but still more references were needed to reach the target of 50. A third search was thus performed, with “soil erosion AND ecosystem service” keywords delivering 380 hits, most of which differed from the previous ones. A few (<5) references were finally identified from the reference lists of some articles.
This three round search was necessary because many studies about mass flow regulation used available models (sometimes calibrated with local data but not always), and empirical studies quantifying soil erosion were favoured over modelling studies (although modelling studies were not systematically rejected).
Two meta-analyses were included in the review and it was tried as much as possible to avoid including in this review references used in the meta-analyses. Finally, some journals for which a pdf of the studies could not be obtained immediately were systematically removed, which may have partly biased the review.
The number of publications steadily increased since the early 1990s, a typical pattern when searching references from electronic data bases (Figure 1).
10 8 6 4 2 Number of articles 0 1990 1995 2000 2005 2010 2015 Publication year
Figure 1: Publication dates of the 50 journal articles reviewed on mass flow regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 158
Background/context
There was a clear scale bias among the 50 selected references. Indeed, most of the studies were local (n=41), and studies were rarely at a sub-national (3) or national (3) scale, and very rarely at larger scales (sub-continental, n=1; continental, n=1; both comprised the Mediterranean area) (Figure 2).
Regarding the temporal scale of the studies (Figure 2), most used a snapshot (n=35); fewer based their study on a long-term perspective (7), or on annual (3) or seasonal (1) time scales. Finally four studies considered a scenario rather than a particular temporal scale.
Sub- Continental continental 2% Global 2% Scenario National 2% analysis 6% 8% Long- Sub- term national 6% 14% Annual 6% Seasonal 2% Snapshot 70% Local 82%
Time scale Spatial scale
Figure 2: Percentage of studies at each temporal and spatial scale for mass flow regulation.
All continents were not represented equally in the final set of studies (Table 1, Map 1). Actually, 42% of the studies were from Europe, often from mountain or Mediterranean areas where soil erosion is a frequent issue. The database contains similar numbers of studies for Asia, North and South America (ranging from 6 to 9). Many Chinese publications were not considered because they were based on models rather than quantification and it was preferred to focus more on empirical findings. As found in previous review work by AlterNet, the protocol based on Scopus probably missed many case studies from the South published in open sources and in the grey literature (e.g. national technical reports), so that the figures should be interpreted with caution. Except for these general feelings, there was no relationship between the evidence and the location of the study.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 159
Table 1: Location of studies on mass flow regulation.
Continent Sample size Europe 21 Asia 9 North America 8 South America 6 Africa & Middle-East 3 Australia 1 Global 2
Map 1: Location of studies on mass flow regulation.
The factors affecting the service depended on the ecosystem considered. In forests, soil erosion following fires or logging was compared to undisturbed sites. Soil erosion in grassland was frequently compared to that in crop fields (e.g. Schuller et al., 2003); the impact of management (e.g. grazing) was also considered. Wind erosion was mostly studied on sparsely vegetated land (e.g. Field et al., 2011).
Ecosystem condition was explicitly mentioned in only 21% of the studies, and difficult to define from the study site description. For the studies for which ecosystem condition was provided, there was no strong patterns, no category being more frequently reported than another (Table 2).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 160
Table 2: Ecosystem types present in the review of mass flow regulation.
Condition Unfavourable Ecosystem Favourable Destroyed not Total No evidence Recovering Declining mentioned Woodland and 3 3 1 13 20 forest Heathland and shrub 2 1 12 15 Grassland 1 1 1 2 20 25 Cropland 1 20 21 Sparsely vegetated 1 1 1 3 6 land Urban 2 2 Wetlands 2 2 Total 7 1 6 3 2 72 91
Linkages between natural capital and the ecosystem service
Most studies considered a land use or a management type as a whole, although interpretation of differences in soil erosion among land uses or among management types was usually done in terms of vegetation differences. For instance, Neris et al. (2013) compared soil erosion in burnt forest areas compared to unburnt areas. As a result, the respective roles of vegetation and soil in mass flow regulation were rarely disentangled.
Most of the studies considered the role of the entire (plant and animal) community (n=36) as the Ecosystem Service Provider (ESP) rather than specific components. Nevertheless, a few restricted the study to: the dominant community (n=2), a single functional group (n=4, e.g. trees in Borrelli et al., 2014) or multiple functional groups (n=2, e.g. earthworms and termites in Brown et al., 2010), a single species (n=4) or multiple species (n=2, e.g. Stipa tenacissima, Rosmarinus officinalis and Anthyllis cytisoides grass/shrubs in Bochet et al., 1998). For instance, Acharya et al. (2007) looked at the soil retention effect of the planted Fodder grass (Setaria anceps).
Three studies dealt with animals as “soil engineers”: earthworms and termites (Jouquet et al., 2012), prairie dogs (Martínez-Estévez et al., 2013) and dung beetles (Brown et al., 2010).
Resulting from the generally holistic view of the studies, the most frequently attributed biotic trait of the ESP was “presence of a community/habitat” (e.g. ground vegetation), followed by community structure (Figure 3). Yet several studies also considered some abundance measure, such as habitat area, biomass and species abundance.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 161
Aboveground biomass Belowground biomass Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Successional stage Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Other
0 5 10 15 20 25 30 35 40 45 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 3: Biotic attributes contributing to mass flow regulation.
Three main abiotic features were regularly cited as driving the intensity of the service (Figure 4). The amount of precipitation was by far the major erosive force (n=15). Slope and wind were also frequently shown to affect soil erodibility (in 9 and 4 studies, respectively). Soil type also influenced the service, with soils varying in their propensity to erode (5 studies). Geology and nutrient availability were two features of soils potentially affecting the service of mass flow regulation.
Precipitation Snow Wind Nutrient availability Soil Geology Slope Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 4: Abiotic factors influencing mass flow regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 162
Human input and management
It was not easy to assess human input and management in these studies. These were interpreted as any action performed by humans on ecosystems. These mostly included land use changes and management practices, such as: forestry (forest clearing, logging), agriculture (cropping, grazing, tillage, burning). Additional areas of human influence included the introduction of invasive species, forest fires and habitat restoration (native forests, grass).
Thresholds
None of the studies interpreted their results in terms of thresholds, though one study recorded the use of a user-defined critical, or threshold, value for the CTI (compound topography index), which combines slope with overland flow magnitude and concentration to represent the erosive power of overland flow. Three reasons for the general lack of thresholds may be invoked. One is the general lack of legislation about tolerable soil erosion that studies could refer to (see ‘Policies’ below). A second is that soil retention, which can be relatively easily measured, is not directly the main service of use to humans (such as food production or water quality), and few studies related soil erosion to these services. Finally, it may be that there is no threshold, or that threshold levels may be strongly context-dependent.
Indicators
Not surprisingly, although partly due to the deliberate emphasis on observational studies during the literature search, by far the main indicator used to assess the mass flow regulation service was soil erosion, followed to a much lesser extent by soil retention or erosion control. A number of studies used equations that integrate other (indirect) indicators to estimate the potential soil erosion, such as land use/land cover, soil properties (geomorphology, soil characteristics, water retention) and (rarely) the amount of precipitation (Table 3).
Table 3: Indicators used in articles on mass flow regulation.
Secondary indicator Sample size Land cover 6 Vegetation map 0 Erodibility 3 Land use 3 Soil characteristics (+geomorphology) 5 Soil retention 2 Slope 13 Water flow (precipitation) 1
Policies
Interestingly, while soil erosion may have significant impacts at local and landscape scales, policies were very rarely mentioned in the studies. Two European studies cited the Common Agricultural Policy and the
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 163
Soil Thematic Strategy (the latter at a time when it was not approved yet). Three additional publications mentioned a German regional land use planning law and two national ones, namely a Brazilian environmental law and a North American surface mining law. These findings suggest that the soil control service is not well covered by existing legislation, despite the fact that a number of studies have shown the significant socio-economic impact of soil erosion.
Interactions with other ecosystem services
A number of studies quantified different variables that could be interpreted as services. For instance, Jouquet et al. (2012) showed that earthworm casts significantly decreased water conductivity and increased water pH; a result that could be interpreted as a positive effect on water quality. Eight of the pre- selected list of ecosystem services were related to the mass flow regulation service. The interlinkages were mostly identified based on empirical evidence, but in some cases they were part of the model applied to the case study (e.g. Jackson et al., 2013), so that these studies do not provide evidence for a real link. Generally, the ecosystem services acted in synergy (n=32); they were rarely antagonistic (3). In a limited number of studies, the relationship was unclear (7) (Table 4). The two main reasons are that soils support a number of supporting ecosystem services such as food and raw material production, and that soil erosion is often caused by water flow and is thus also connected to water quality (as erosion pollutes water).
As mentioned before, water flow regulation was logically by far the most frequently cited ecosystem service (n=18) acting in synergy with mass flow regulation, as runoff is frequently the main cause of soil erosion. Factors limiting soil erosion by favouring water infiltration into the soil (e.g. dung beetles, Brown et al., 2010) were considered to be positively related to water flow regulation.
Table 4: Interactions between mass flow regulation and other ecosystem services.
Ecosystem services linked with mass flow regulation Positive Negative Unclear Total Water flow regulation (flood protection) 16 1 1 18 Food production (cultivated crops) 4 1 2 7 Other 4 0 2 6 Atmospheric regulation (carbon sequestration) 4 0 1 5 Timber production 1 1 0 2 Freshwater provision (quantity) 1 0 0 1 Water quality regulation 1 0 0 1 Pest regulation 0 0 1 1 Aesthetic landscapes 1 0 0 1 Total 32 3 7 42
Higher soil protection was also associated with higher food production or with higher potential for food production in four studies (e.g. higher grain yield, Acharya et al., 2007; higher organic matter and better nutrient cycling, Mandal et al., 2013). There was little conflicting evidence (slightly antagonistic in one study and unclear in two studies), suggesting limited variability in the relationships between ecosystem services related to mass flow regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 164
In forests, fires naturally reduced timber production while increasing soil erosion by rain. In restored forests where soil retention was higher, CO2 efflux increased with restoration age leading to a positive interaction with the service of atmospheric regulation (Larkin et al., 2013). In a study combining various previously published models (Lorencová et al., 2013), grassland appeared to limit soil erosion compared to crop fields, as well as to sequester more CO2 in the soil.
Four services were less frequently cited or could be implied from the presented results, such as water quality regulation, freshwater provision, pest regulation and aesthetic landscapes (which were mostly positively related to mass flow regulation).
Interactions were also cited with other aspects of regulation and maintenance ecosystem services including plant diversity, habitat, habitat connectivity, drought risk regulation, soil organic matter and nutrient content. Some of these topics deserve further study to guarantee the generality of the findings.
Overall observations on the review
Erosion was sometimes assessed experimentally for general purposes. It might have been useful to separate observational from experimental studies.
All studies quantified soil erosion; one combined a qualitative and a quantitative assessment. Most of them (n=39) were based on empirical data, seven used models (four the Universal Soil Equation, two its revised version; GISCAME and InVest were also cited); four studies combined empirical data and models. As explained above, these figures should not be seen as representative of the real ratio between empirical and modelling studies as a deliberate choice was made to favour empirical studies.
Except for the seven modelling studies (i.e. that did not quantify soil erosion), the majority of the references relied on multiple observations (n=38; only seven used single observations). Evidence was direct for 35 studies; the 15 remaining ones included the seven modelling studies. The eight others used either soil depth measurements or 137Cs as indirect proxies for soil erosion.
It was not always possible to define natural capital when dealing with mass flow regulation. Many comparisons were made between land uses (e.g. grassland vs crop fields), or between land management practices for the same land use (e.g. grassland with varying grazing intensity). For instance, are fertilised and grazed pastures more “natural” than crop fields? Similarly, several studies comparing the impact of the tillage system on soil loss in crop fields were not included in the review, because it was considered that crop fields with no tillage (often compensated for by more frequent use of herbicides) are not more natural than crop fields with tillage.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 165
Water flow regulation
Alison Smith and Sen Li, Environmental Change Institute, University of Oxford, UK
The literature search
48 articles were entered into the database from BESAFE and 12 additional articles were entered into the database to cover the period from 2012 to 30 June 2014 (Figure 1). As the original search methodology had been relatively unsuccessful (identifying only 24 articles), slight modifications were applied for the updated search: 1. The term “wetlands” was omitted from the first search field, to reduce the number of articles mentioning wetlands but not flooding. 2. For simplification, the multiple strings involving the word “flood” were deleted from the first search field and replaced simply with “flood*”. The term “water flow regulation” was added. 3. The final search term was altered to include more words specific to the water flow regulation service.
Table 1: Keywords and terms used in the updated Web of Science search.
Ecosystem service/disservice Biodiversity terms Additional service related terms Flood* Biodiversity “Peak flow” “water flow regulation” “Biological diversity” Attenuation Species Storage Habitat* Protection Genetic Defence Trait* Prevention Function* Runoff Landscape Evapotranspiration Richness Infiltration Abundance “Vegetation cover”
8
6
4
2
Number of articles 0 1985 1990 1995 2000 2005 2010 2015 2020 Publication year
Figure 1: Publication dates of the 60 journal articles reviewed on water flow regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 166
Background/context
Ecosystems can provide a flood protection service in several ways: (i) Forests and other vegetation intercept rainfall, increasing the rate at which it evaporates before reaching the ground; (ii) Plants suck up water from the ground and release it into the air through transpiration; (iii) Plant roots can increase the rate at which water infiltrates into the ground; (iv) Wetland areas, when not completely saturated, can act as a sink for surface runoff; (v) Coastal vegetation can reduce the speed and height of waves, providing protection against floods during storms and tsunamis.
The articles studied reflect this range of situations, though there was a strong bias towards investigating the role of forests in reducing surface runoff, typically through “paired catchment” studies that compared empirical measurements of stream flow in similar-sized forested and unforested catchments. Figure 2 shows the ecosystem categorisations: note that there are 88 ecosystem records from 60 studies, partly because some studies covered more than one ecosystem type, and partly because some received a multiple classification, e.g. “forest” and “wetland” for mangrove forest, or “wetland” and “coastal” for saltmarsh. Of the 88 ecosystem records, 37 were forests, 18 were wetlands and just three were coastal (including one on coral reefs), with the rest including rivers and lakes, cropland, grassland, heathland, sparsely vegetated land and urban (a green roof).
There was little information on ecosystem condition, though in some cases we inferred that the condition was favourable because the ecosystem was undisturbed, protected or restored. In other cases we inferred it was unfavourable because the article stated that it had been degraded, typically due to logging, overgrazing or other human interference.
Woodland and forest Wetlands Rivers and lakes Grassland Favourable Cropland Unfavourable - no evidence Heathland and shrub Unfavourable - recovering Coastal Unfavourable - declining Sparsely vegetated land Condition not mentioned Urban
0 10 20 30 40 Number of studies
Figure 2: Classification of ecosystem types and conditions for the service of water flow regulation.
The studies spanned all continents, but there was a distinct bias towards Europe (26 studies, of which 13 were in the UK) and North America (9 studies). Of the other studies, 10 were in Asia (of which four were in China), five were in South or Central America, five were in Australasia and just two were in South Africa (Map 1). There was no indication that the evidence varied with the location of the study.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 167
Map 1: Location of studies on water flow regulation.
Most of the studies were at the local scale (32), with several spanning multiple local sites, typically to enable comparison of sites with different land cover (e.g. forested vs. unforested). In some cases, the studies examined catchments that were large enough or widely dispersed enough to categorise the study as sub-national or national, and three studies covering catchments in different countries were categorised as continental. Five studies were global, spanning data from a wide range of countries (Figure 3).
To capture a range of rainfall conditions, most studies used long-term datasets on stream flow, or modelled different rainfall or wave height scenarios. However, some were snapshots, e.g. studies of coastal vegetation height or roughness (Figure 3).
Time scale Spatial scale Continent al 5% Global 9% Scenario National Snapshot 3% analysis 22% Seasonal 22% 3%
Sub- Annual national 5% Local 58% Long- 25% term 48%
Figure 3: Percentage of studies at each temporal and spatial scale for water flow regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 168
Linkages between natural capital and the ecosystem service
Ecosystem service providers (ESPs) In most of the studies, the flood protection service was provided by the entire community or habitat, or more than one community or habitat (such as a catchment containing both forest and grassland). In two studies, a large catchment was dominated by one community type (e.g. boreal spruce forest). For a few studies, the service was provided by specific species populations, such as in forests or plantations dominated by one or two tree species (Figure 4).
3% 12%
3% Dominant community
Entire community or habitat 23% Two or more communities or habitats 59% Single specific species population Two or more specific species populations
Figure 4: Ecosystem service providers (ESPs) in the studies reviewed on water flow regulation.
Biotic attributes No studies were found that specifically set out to investigate the impact of biodiversity on flood protection. However, many studies addressed the impact of land cover on flood risk, from which information could be extracted on the relation between water flow regulation and community/habitat area, age and structure (Figure 5).
Studies of river catchments typically found a positive relationship between the percentage of the catchment covered by forest and reduced surface runoff and/or a reduction in peak stream flow. For example, a review of data from 56 developing countries by Bradshaw et al. (2007) found that flood frequency and duration decreased as natural forest cover increased, though forest cover explained only 14% of the variation in flood frequency. They predicted that a 10% natural forest loss would increase flood frequency by 4-28% and duration by 4-8%. Similar effects were observed by Clark (1987), Fahey and Jackson (1997), Iroume et al. (1995), Ogden et al. (2013) and many others. Vegetation cover also reduced runoff in shrublands (Schmittner and Giresse, 1996; Lana-Renault et al., 2014; Kabanda and Palamuleni, 2013).
Many studies found that wetlands and floodplains could reduce flooding by providing an area to absorb runoff and overspill from rivers (e.g. Ming et al., 2007; Posthumus et al., 2010). However, the evidence was mixed, with a review by Bullock and Acreman (2003) finding that floodplain wetlands reduced flooding in
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 169
23 out of 28 studies, but headwater wetlands only reduced flood peaks in 30 out of 66 studies, with 27 studies finding the opposite. This was thought to be because headwater wetlands are often saturated.
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Successional stage Aboveground biomass Belowground biomass Stem density Litter/crop residue quality Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Species abundance Species size/weight Population growth rate Other
0 10 20 30 40 50 60 Number studies Negative Both (Positive & Negative) Positive Unclear
Figure 5: Biotic attributes contributing to water flow regulation.
Vegetation structure was not always mentioned explicitly, but many articles described the role of the tree canopy in physically intercepting rainfall and thus reducing surface runoff in rain-fed catchments. Carlyle- Moses and Price (2007), for example, estimated that pine-oak forest intercepted 15% of rainfall. In snow- dominated catchments, tree canopy reduced or delayed snowmelt through shading (Green and Alila, 2012; Nedkov and Burkhard, 2012) or reduced the accumulation of snow through interception (Lin and Wei, 2008; Moore and Wondzell, 2005).
For coastal vegetation such as salt marshes, the height and roughness was important in providing protection from storm surges (Möller et al., 2007). Mangrove forests provide a physical barrier against storm surges due to increased flow resistance (Mazda et al., 1997), and so do coral reefs, with tall, branching corals providing greater wave attenuation (Ferrario et al., 2014). Woodlands on floodplains also slow the passage of water, with the physical structure of the vegetation having an important effect (Thomas and Nisbet, 2006). Martin (1999) also notes that intercropping with trees reduces surface runoff by providing preferential infiltration at the stem base.
Some negative impacts were also cited. Lee and Shih (2004) found that mangroves blocked coastal channels, causing raised water levels and increasing flood risk, and Zavaleta (2000) found that Tamarisk
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 170
(Tamarix sp.), an invasive woody shrub, increases sedimentation in river channels, leading to increased frequency and severity of flood damage. Erskine and Webb (2003) found that invasive willow plants could reduce stream channel capacity, but they also found that large woody debris in streams could slow the passage of flood peaks.
Community / stand age was also significant. Studies that compared stream flow for forests of different ages typically found a greater reduction in stream flow as forests matured and trees grew larger (Hümann et al., 2011; Putuhena and Cordery, 2000). This was linked to an increase in rainfall interception and evapo- transpiration in rain-fed catchments, or a decrease in snow melt in snow-dominated catchments. These factors are linked to some other attributes that were cited occasionally: successional stage and above- ground biomass.
The role of forests in increasing infiltration and thus reducing surface runoff is linked to below-ground biomass (greater root density; Lange et al., 2013) and litter quality (more humus improves infiltration; Gholzom and Gholani, 2012; Schmittner and Giresse, 1996).
A handful of studies found that a specific functional group had an effect. For example, Mazda et al. (1997) found that mangrove species with aerial roots provided greater storm protection, and Hümann et al. (2011) found that conifer forests reduced runoff more than deciduous forests, possibly due to their deeper root systems.
Several studies also identified species-specific impacts: Lange et al. (2013) found that beech forests had higher infiltration than spruce forests due to greater root density. A number of studies find that species size is important, e.g. Anderson et al. (2006) find that taller vegetation delays peak discharge in rivers.
Abiotic attributes The most significant abiotic factor, unsurprisingly, is precipitation (Figure 6). A number of studies conclude that the impact of forest cover decreases with the size of the rainfall event, with forest cover having limited effect for extreme events (e.g. Bathurst et al., 2007; Bruijnzeel, 2004; Cheng et al., 2002; Clark, 1987; Moore and Wondzell, 2005). However, Green and Alila (2012) dispute this view. They argue that decades of previous studies using paired catchment data for forested and unforested catchments are faulty. Previous work found that the difference in flood abatement between harvested and unharvested catchments diminished for larger floods. However, this new approach instead compares the whole frequency distribution of floods for harvested and unharvested catchments, finding that both the frequency and magnitude of floods increases after harvesting. In other words, “while the largest floods may or may not be increasing in magnitude relative to the chronologically paired control catchment floods, the small and moderate floods are being elevated in magnitude as a result of harvesting to become some of the largest floods.” They find that the frequency of the largest floods tends to double or triple after harvesting. For the snow melt dominated catchments in North America that are the focus of their study, they ascribe the main reason for this finding to the shading effect of the forest canopy, which delays snow melt. In a follow-up article from the same team, Schnorbus and Alila (2013) provide further evidence for this with a modelling study of a snow-dominated area.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 171
It is not clear from the articles above whether the same principle applies to rain-driven floods as to snow melt floods. However, Lana-Renault et al. (2014) found that vegetation cover exerts a significant rain-driven flood prevention role even for exceptional events, though this may have been influenced by exceptionally dry conditions that reduced the level of the water table before the storm. They found that forests delay and reduce the peak flow, but may not affect total runoff over time. Similarly, a study in Panama by Ogden et al. (2013) found that even during the largest storm on record the forest catchment produced about 35% less total runoff than the mosaic catchment.
Wet conditions also reduce the ability of wetlands (Smakhtin and Batchelor, 2005) and green roofs (Stovin et al., 2012) to absorb further rainfall.
Temperature Precipitation Water availability Evaporation Wind Water quality Soil Geology Slope Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 6: Abiotic factors influencing water flow regulation.
Other abiotic factors include the negative effect of future temperature rise on coral reefs (Ferrario et al., 2014) and, via sea level rise, on wetlands (Temmerman et al., 2012; Wamsley et al., 2010). Soil and geology was also found to be important, influencing the infiltration capacity. For example, Bruijnzeel (2004) found that soils with a shallow impeding layer had a high runoff coefficient even when forested, whereas Ming et al. (2007) found that wetland soils had very high porosity (up to 89%), providing high water storage capability. Slope was also important in some cases. Slope increases runoff speed in Alpine areas with little vegetation (Lana-Renault et al., 2014), and Green and Alila (2012) also found that steep slopes maximise absorption of incoming shortwave radiation (insolation) so the shading effect of tree canopy is more important.
Wind can increase wave size and velocity during storms, causing loss of the coastal marsh (Möller et al., 2007) and can also reduce rainfall interception by shaking water from trees (Carlyle-Moses and Price, 2007. Other factors include slope aspect (the protective effect of forests from snow melt is lower on shaded, north facing slopes according to Green and Alila, 2012) and drainage ditch density (forests reduce runoff
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 172
more where there is a dense network of drainage ditches that would otherwise promote fast runoff, say Schnorbus and Alila, 2013).
Human input and management
Human impacts were identified in 41 of the 60 studies, with the impacts being predominantly negative (17 cases), though they were positive in six studies, both positive and negative in eight studies and unclear in 10 studies. The main type of impact is deforestation, for timber extraction, agriculture or urban development, which increases flood risk (e.g. Bathurst et al., 2007; Lin and Wei, 2008; Yang et al., 2013). Apart from loss of tree cover, forestry activity can also increase flood risk though compaction of the soil with heavy machinery (Moore and Wondzell, 2005) or construction of drainage ditches (Robinson et al, 2003). Wetlands can also be damaged, e.g. through over-grazing of coastal marshes (Temmerman et al., 2012), and grassland can suffer compaction from over-grazing (Ford et al., 2012). But positive impacts can result from restoring forests (Bruijnzeel, 2004), mangroves (Mazda et al., 1997), wetlands (Acreman et al., 2007), coral reefs (Ferrario et al., 2014) or riparian ecosystems (Zavaleta, 2000).
Sometimes human impacts can be partially reversed when farmland is abandoned, but long-term effects on soil structure may persist. Lana-Renault et al. (2014) find that intensive farming in the Pyrenees has degraded the soil, increasing the risk of flash floods, but abandonment of farmland in the 1950s has decreased the frequency and magnitude of flood events in this region as trees and shrubs have colonised fields. However, in areas that were previously farmed, the soil is still degraded and the flood prevention capacity is reduced.
Thresholds
Only three examples of thresholds were found. In the study of the adverse impact of mangroves on raising the water level in coastal channels, Lee and Shih (2004) found that beyond a 20% mangrove removal rate there is no further improvement (reduction) of the flood risk. In their study of wave attenuation by coastal vegetation, Möller et al. (2007) found that attenuation decreases with inundation depth; extrapolation suggests that there would be no difference between salt marsh and sand flat at inundation depths >3.7m and little difference at depths >1.3m. And Schnorbus and Alila (2013) found that annual maximum peak flow in forest catchments starts to increase only after a threshold of 20-30% of the catchment area has been harvested.
Indicators
The indicators from the JRC (2012) report were not very helpful, as there was considerable overlap between some of the indicators (e.g. flood attenuation, flood control, flood prevention, or water regulation, water holding, water recharge) whereas other indicators were unclear (e.g. annual flood). In reality, the main indicator cited in the articles was peak stream flow, which we equated to hydrologic flow and/or flood attenuation, cited 32 and 34 times respectively. Estimates of attenuation of wave height, for the coastal protection studies, were also classified as flood attenuation. In addition, there were eight studies that estimated water holding capacity (for wetlands and flood plains). Other common indicators included data on land cover (% forest area) or soil characteristics (regarding porosity / infiltration rate). The
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 173
vast majority of studies provided primary data, although there were a few modelling studies (but even these were calibrated with empirical data).
Only a few other indicators (not included in the JRC report) were cited: these were all one-off occurrences, including number of overtopping incidents; sediment loading downstream of a wetland; stemflow; evapotranspiration rate and root density.
Policies
The Common Agriculture Policy, the Water Framework Directive and the Habitat Directive were each mentioned four times, and the Soil Thematic Strategy once. Other policies included the Ramsar Convention on Wetlands (mentioned twice), UK Biodiversity Action Plans (twice), and single mentions of the Clean Development Mechanism (for promoting afforestation) and the “Declaration of Arles” to reduce flood risks in Europe.
Interactions with other ecosystem services
A number of other ecosystem services were linked to water flow regulation (Figure 7). In particular, there is often a negative correlation with the service of freshwater supply, because the absorption of water by forests tends to reduce stream low-flows as well as peak flows, which can exacerbate water shortages in regions where water supply is limited (e.g. Farley et al., 2005). Several of the studies provided empirical evidence for this, as the catchment studies often measured both peak flows and low flows. However, some authors found that forests can both reduce peak flows and enhance low flows, perhaps through improving infiltration of surface runoff, thus providing a long-term store of groundwater to be released over time (e.g. Bruijnzeel, 2004; Ogden et al., 2013).
The other main negative interactions / trade-offs are with the services of timber and food production, as forest clearance for timber or agriculture tends to increase flood risk. However, some articles cited the establishment of timber plantations on former grassland as a positive benefit in terms of reduction of flood risk. These, and the other interactions listed below, were generally discussed in qualitative terms (or at the most basic level of % land cover).
However, there are many positive synergies with other ecosystem services. Thirteen articles cite the positive benefits of forests and wetlands for water quality regulation, by filtering out impurities such as excess nutrients and other pollutants (e.g. Yang et al., 2010, found that wetlands reduced N and P loads by 23%). Ten articles mentioned the benefits of wetlands, forests and mangroves for landscape aesthetics (e.g. Foote et al., 1996; Acreman et al., 2011), and a number mentioned the importance of these ecosystems as wildlife habitats, and for species-based recreation such as bird watching or eco-tourism (e.g. Ewel et al., 1998; Foote et al., 1996) and for pollination (Acreman et al., 2011). However, there can also be trade-offs. Acreman et al. (2011) found that raising the water level in wetlands would have a beneficial effect on wildlife and biodiversity but would reduce the flood storage capacity, and Ford et al. (2012) point out that grazing is bad for flooding but good for forbs (flowering plants) that contribute to pollination and landscape aesthetics.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 174
There are also strong synergies with mass flow regulation, as forests, mangroves and coastal vegetation protect against erosion. The role of forests in stabilising steep slopes is particularly important (e.g. Cheng et al., 2002; Lana-Renault et al., 2014). There are also good synergies with atmospheric regulation, as forests and wetlands store carbon. The article on green roofs also mentioned the benefits for urban air quality regulation (Stovin et al, 2012). Other interactions include the role of urban green roofs in providing evaporative cooling (Stovin et al., 2012).
Freshwater fishing Timber production Potable water (quantity) Food production (crops) Air quality regulation Atmospheric regulation Negative Mass flow regulation Both (Positive & Negative) Water quality regulation Positive Pollination Pest regulation Unclear Recreation (species-based) Aesthetic landscapes Other
0 5 10 15 Number of studies
Figure 7: Interactions between water flow regulation and other ecosystem services.
Overall observations on the review
It should be noted that a reduction in runoff or peak stream flow can only be equated to reduced flood risk in areas where flooding is a problem. For some of the articles transferred from the BESAFE database, this did not appear to necessarily be the case. For example, Farley et al. (2005) found that stream flow decreased substantially with forest cover, but the focus of the study was on the negative impacts for reduced water availability rather than flood risk.
Three articles identified during the new search for 2012-2014 found no significant correlation between forest area and flood risk. These could not be included in the database as there is no opportunity to select “none” or “no significant impact” in the dropdown box that indicates the relationship between attribute traits and ecosystem service provision.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 175
Water quality regulation
Roy Haines-Young1 and Alison Smith2 1CEM, School of geography, University of Nottingham, UK 2Environmental Change Institute, University of Oxford, UK
The literature search
Forty four records were transferred from BESAFE to OpenNESS, and a further 16 records were added to cover the period 2012-2014 (Figure 1), using the same search terms as for the BESAFE review (Table 1).
Table 1: Search terms used for water quality regulation.
Ecosystem service/disservice Biodiversity terms Additional service related terms Water quality Biodiversity Tree* Water regulation “Biological diversity” Soil* Water purification Species Forest* Nutrient* retention Habitat* Vegetation Nutrient* translocation Genetic Plant* Trait* Pollutant* Function* Wetland* Landscape Microorganism* Richness Accumulation Abundance Sediment*
10 8 6 4 2 Number of articles 0 1990 1995 2000 2005 2010 2015 2020 Publication year
Figure 1: Publication dates of the 60 journal articles reviewed on water quality regulation.
Background/context
Studies were mainly at local and sub-national scales, including experimental studies of the impact of vegetation type on water quality in wetlands and studies of the impact of riparian buffer zones (Figure 2). A
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 176
number of large-scale land use studies were also included which looked at the impact of deforestation or agricultural practices on water quality in rivers and lakes.
Over half of the studies were snapshots, with the remainder being spread evenly between seasonal, annual and long-term (Figure 2). Almost half of the studies (27/60) were located in North America, with 13 in Asia and only eight in Europe. Evidence does not seem to vary with location.
Time scale Scenario Spatial scale analysis Global 5% National 2% Long- 3% term 17%
Local 49% Annual Sub- 13% Snapshot national 53% 43% Seasonal 15%
Figure 2: Percentage of studies at each temporal and spatial scale for water quality regulation.
A fairly even spread of ecosystem types is covered (Table 2), including wetlands (26), woodlands (23), grassland (18) and cropland (15). Information on condition is sparse: most articles do not use the formal definitions and the conditions were therefore interpreted by the reviewer from information in the articles, but this was only possible in 10 studies.
Table 2: Ecosystem types and conditions for water quality regulation.
Condition Unfavourable Ecosystem Favourable Destroyed not Total - recovering mentioned Woodland and forest 1 2 20 23 Heathland and shrub 2 2 Grassland 1 1 16 18 Cropland 15 15 Sparsely vegetated 2 2 land Urban 7 7 Wetlands 2 1 1 22 26 Rivers and lakes 11 11 Coastal 1 1 Total 4 3 3 95 105
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 177
Linkages between natural capital and the ecosystem service
In over three quarters of the studies reviewed, the ecosystem service provider was one or more entire communities or habitats (typically forests or wetlands). However, 13 articles studied individual species or groups of species, which were often assemblages of nitrogen-removing plants (e.g. common reed, Phragmites Australis) in experimental wetlands. One article studies the role of two functional groups (submerged macrophytes and zooplankton) (Figure 3).
Entire community or habitat 15%
Two or more communities or habitats 7% 2% 43% Dominant community
10% Two or more functional groups
Single specific species population
23% Two or more specific species populations
Figure 3: Ecosystem service providers for water quality regulation (% of studies).
The review identifies a number of ways in which ecosystems such as forests, wetlands and grassland can improve water quality: 1. Permanent vegetation reduces soil erosion compared to bare ground or farmland (e.g. Caroll and Tucker, 2000); 2. Vegetation and marshes can trap sediment before it reaches water courses (e.g. Rashin et al., 2006, de Souza et al 2013); 3. Vegetation can absorb and adsorb excess nutrients and other impurities (e.g. Ahmad et al., 2014; Christen and Dalgaard, 2013); 4. Soils can host de-nitrifying bacteria that break down nitrates from fertiliser runoff into harmless nitrogen gas (e.g. Martin et al., 1999); 5. Vegetation roots can improve infiltration, allowing more impurities to be filtered out by the soil and preventing pollution of adjacent streams and lakes (e.g. Christen and Dalgaard, 2013).
Because of the role of vegetation in preventing erosion, physically trapping sediment and absorbing pollution, biotic attributes related to the amount of vegetation are found to have a positive impact. By far the most commonly cited attributes are community/habitat area (41 counts) and presence of a specific community/habitat (27 counts), but other attributes include above- and below-ground biomass, primary productivity, stand age, stem density and species size or weight (Figure 4). One interesting exception was reported by de Souza et al. (2013) who found a negative correlation between stream water quality and forest age, tree size and successional stage. In their study, more mature riparian forests with fewer but large trees resulted in worse stream water quality than younger forests with many, smaller trees. They
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 178
suggested that this was partly due to the increased surface roughness and higher stem density of the immature forest, which results in greater interception of overland flow, and partly to higher phosphorous absorption by the younger, faster-growing trees; they state that forests can eventually become a net source of remobilised phosphorous as they mature.
Several articles refer to the abundance of highly effective species, such as California bulrush (Hogan et al., 2013), poplar (populus sp.) and willow (salix sp.) (Christen and Dalgaard, 2013) or seagrass (Zostera marina) (Moore, 2004), or functional groups such as mangroves (Peng et al., 2013). However, Doherty et al. (2014) reported a negative correlation between species size and water quality in the case of invasion of poorly drained constructed wetlands by Typha (bulrush), where the tall vegetation and thick litter layer prevented the formation of moss and algal mats, so that prolonged inundation led to highly erodible soil and the wetland became a net source of solids, nitrogen and phosphorous. However, this was not the case in well drained wetland, which provided better flow regulation, erosion prevention, biodiversity and water quality.
Ten studies find an impact from various types of diversity, including species richness, species population diversity, functional richness and functional diversity. The impacts are predominantly positive and seem to be related to the ability of more diverse mixtures to be more productive, and therefore take up more nutrients, due to niche complementarity (i.e. exploitation of a wider range of resources) (Fisher et al, 2009; Cardinale, 2011). However in two cases the impact is unclear, with Cardinale et al. (2011) stating that there is no evidence that polycultures out-perform the most efficient monocultures. Similarly, Weisner and Thiere (2010) find that wetlands dominated by a less diverse mix of tall, emergent vegetation are more efficient at nitrogen removal. These two studies therefore support the selection effect rather than the niche complementarity effect.
A more complex picture is reported by Tomimatsu et al. (2014) who study reed beds dominated by a single species (Phragmites, the common reed), where genetic diversity could be important. They found that productivity and nitrogen removal rate was significantly greater (by about 15%) in genotypic polycultures compared to that expected based on monocultures, though the advantage was less in the second year. This was due to enhanced uptake of nitrogen, driven by both complementary and competitive interactions among genotypes. Complementarity was more important in the first year of the experiment, but competition and selection were more important in the second year, with the most effective genotype dominating the mix. However, they hypothesise that complementarity may be more important in disturbed systems, e.g. if subject to stress from prolonged drought.
Abiotic factors were not covered extensively, though initial water quality, temperature, soil and slope were all mentioned, with the direction of influence varying between studies (Figure 5). For example, Tomimatsu et al. (2014) find that higher temperatures in summer speed up nitrogen removal in wetlands due to higher plant growth rates, but Rodrigo et al. (2013) find that warmer weather stimulates algal blooms.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 179
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Successional stage Primary productivity Aboveground biomass Belowground biomass Stem density Litter/crop residue quality Landscape diversity Species richness Functional richness Functional diversity Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Population growth rate Leaf N content Other
0 5 10 15 20 25 30 35 40 45 Number of studies Negative Both (Positive & Negative) Positive Unclear
Figure 4: Biotic attributes contributing to water quality regulation.
Temperature Precipitation Water availability Water quality Soil Geology Slope Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 5: Abiotic factors influencing water quality regulation.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 180
Human input and management
A range of human input and management practices were noted, with a mix of positive and negative impacts. Typically, negative impacts arose from deforestation, degradation of wetlands, increase in urbanisation and impervious surfaces, and influx of pollution from agricultural fertilisers, pesticides and sewage. However, positive impacts resulted from restoration of wetlands or construction of artificial wetlands, afforestation, protection of forests and wetlands, and planting of riparian buffer strips.
Thresholds
A significant proportion of the studies reviewed referred to some kind of threshold, most of which were related to biophysical properties, though some articles mentioned minimum safe standards for water quality.
One type of threshold is related to the percentage land use in catchment areas. Clapcott et al. (2012) find that nitrogen and phosphorous levels in streams in New Zealand tend to increase, and macroinvertebrate and fish population diversity and health decrease, above a threshold of 40-60% removal of indigenous vegetation. They also find a threshold related to the percentage of impervious surfaces in a catchment, with a negative effect on water quality observed at very low levels (any level >0%), and an exponential response between 0-20%, due to storm water input. They suggest that 10% is likely to be the upper threshold for maintaining stream integrity, and this conclusion is also supported by Haidary et al. (2013). They also cite other articles (not reviewed here) that suggest that the ecological status of rivers based on a biological index is impaired when over 69% of land use in catchments is for agriculture (Donohue, McGarrigle and Mills, 2006), and that 20% urbanisation and 50% agricultural development are thresholds for changes in habitat, biological and water quality indicators (Gergel et al., 2002).
A second type of threshold refers to species or functional diversity. Martin et al. (1999) find that minimum levels of richness, evenness and diversity were required for a microbial community to perform denitrification. However, Cardinale et al. (2011) found a saturation effect whereby ecosystem function ceased to improve after a certain level of species richness.
Indicators
The main indicators were direct measurements of water quality, typically concentrations of various forms of nitrogen and phosphorous and/or suspended sediments, and measurements of nutrient removal rates. Other indicators included land cover in the form of a vegetation map.
Policies
Coverage of the policy issue in the articles reviewed was limited. Most came from outside Europe and so did not mention EU policy directly. Only one article mentioned the Water Framework Directive, and only one mentioned the Habitats Directive and Biodiversity Strategy.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 181
Interactions with other ecosystem services
Links to other ecosystem services were mentioned in 19 studies (Figure 6), comprising a varied mix of both positive and negative interactions. For example, fertilisers from timber and crop production are a cause of water pollution (e.g. Clapcott et al., 2012), but cleaner water supplies are also beneficial for irrigation of food crops such as rice (Park et al., 2014), and riparian buffer strips can be a source of sustainably harvested timber (Christen and Dalgaard, 2014). Similarly, carbon is sequestered in forests and in wetland soils, but wetlands can also be a source of methane and nitrous oxide, two powerful greenhouse gases (e.g. Hefting et al., 2013; Mitsch et al., 2012). However, forests and wetlands also have mainly positive benefits for freshwater fishing, mass flow regulation, water flow regulation and aesthetic landscapes.
Freshwater fishing Timber production Potable water (quantity) Food production (crops) Atmospheric regulation Mass flow regulation Negative Water flow regulation Both (Positive & Negative) Pollination Positive Pest regulation Unclear Recreation (species-based) Aesthetic landscapes Other
0 2 4 6 8 Number of studies
Figure 6: Interactions of water quality regulation with other ecosystem services.
Overall observations on the review
Some of the BESAFE articles were rather old, and drew mainly on material from outside Europe with deforestation seeming to be over-represented. The datedness of the studies did not necessarily undermine the findings, but the way the material was reported did not necessarily use the notion of ecosystem services, so interpretation was needed. Evidence was mostly empirical and primary, but the strength of evidence on the issue of the contribution of biodiversity to the ecosystem service was weak or partial, and further analysis is required. The studies do not really enable the treatment of natural capital as opposed to ecosystem services to be distinguished.
The update with 16 more recent articles partly addresses the concern about datedness, though few studies for Europe were found. There were still few studies that address the contribution of biodiversity to water quality regulation, other than through the simple metric of the entire community / habitat area.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 182
Pollination
Clara Veerkamp1 and Nina Fabrega2 1Netherlands Environmental Assessment Agency (PBL)/ Wageningen University, The Netherlands 2 European Commission, Joint Research Centre (JRC), Ispra, Italy
The literature search
The search terms used were:
biodiversity OR “biological diversity” OR species OR habitat* OR genetic* OR trait* OR functional* OR landscape OR richness OR abundance AND “Pollination services” OR Pollinators OR “Pollinator efficiency” OR “Flower visitors” OR Zoophilous AND “Wild plants” OR “Crop plants” OR Fruit OR “Seed set” OR Horticulture
In total, 51 articles were entered into the database consisting of 27 articles from BESAFE and 24 additional articles for OpenNESS. All OpenNESS articles except one were published before 2013 (Figure 1). Please note that some articles were entered as more than one database record (study), where there were different study sites or/and multiple ecosystems.
12 10 8 6 4 2 Number of articles 0 2000 2002 2004 2006 2008 2010 2012 2014 Publication year
Figure 1: Publication dates of the 51 journal articles reviewed on pollination.
Background/context
The studies spanned a wide range of locations from all over the world (Map 1), but most were conducted in Europe (36 studies, of which 15 were from Greece, eight from Germany and four from the Netherlands). 19 studies originate from North America (mainly from the USA: 15), 12 studies from Africa (mainly from South Africa with seven studies) and seven studies from Asia (Indonesia). Australia (three studies) and South America (one study from Chile) were less represented. Two studies were conducted at the global scale including data from a wide range of countries.
Most studies were conducted at the local scale (54), often spanning multiple local study sites to enable comparison of sites with different land cover or/and land use. Twenty studies were classified as ‘sub-
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 183
national’, and four studies as ‘national’. To estimate the pollination services most studies used data from a ‘snapshot’ (24), often referring to measurements over several months during the flowering season, followed by ‘seasonal’ data sets (10) (e.g. flowering season), ‘long-term’ data sets (5) and one scenario analysis (Figure 2).
Map 1: Location of studies on pollination.
Time scale Spatial scale Scenario Global 4% analysis National 2% 5% Long- term 10% Sub- Seasonal national 19% 25%
Local 66% Snapshot 69%
Figure 2: Percentage of studies at each temporal and spatial scale for pollination.
Agricultural areas benefit from pollination, so most studies were conducted in agricultural areas. These were mainly in croplands (37) ranging from coffee plantations and eggplant fields to apple orchards and oilseed rape fields, though 12 studies were in grasslands grazed by livestock or (semi-natural) grassland patches in an agricultural landscapes. Eleven studies investigated the pollination services in woodland and
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 184
forests (adjoining agricultural areas), four in heathlands (fynbos areas in South Africa) and one in sparsely vegetated land (alpine habitat). Four studies estimated pollination in urban areas (e.g. urban parks) (Table 1).
Not much information was given on ecosystem condition. When ecosystems are under management, we classified ecosystems as ‘unfavourable’ in the case of intensive conventional farming and ‘favourable’ in the case of extensive, organic farming, or areas under protection for conservation. In cases where no specific information was given about the management of state of the ecosystem the ecosystem was classified as “condition not mentioned”.
Table 1: Ecosystem types present in the review of pollination.
Unfavourable Condition not Ecosystem Favourable Total - no evidence mentioned Woodland and forest 2 9 11 Heathland and shrub 2 2 4 Grassland 4 1 7 12 Cropland 3 11 23 37 Sparsely vegetated land 1 1 Urban 1 3 4 Total 11 13 45 69
Linkages between natural capital and the ecosystem service
Ecosystem service provider (ESPs) The ecosystem service provider (ESPs) is the pollinator, in most cases (wild)bees and hoverflies. Twenty- four studies deal with a single functional group referring mainly to pollinators in general (mainly (wild)bees) and 16 studies refer to two or more specific species populations including different bees species populations. However, there are also six studies where the ESPs are two or more functional groups which could also be classified as two or more specific species populations (e.g. one reviewer classified a study with two different pollinators as two functional groups while another reviewer classified this as two different specific species populations).
Biotic attributes Biotic attributes are shown in Figure 3. The abundance and richness of a specific functional group (i.e. pollinators) or specific species of pollinators has a positive influence on the pollination services. “The reduction in abundance of pollinators, in combination with reduction in pollinator species richness, … resulted in 60-80% reduction in pollination function” (Balvanera et al. 2005).
To provide the pollination service, appropriate habitat for pollinators is needed. Loss of native vegetation decreases pollination services from pollinators. In this way, natural habitat and less intensively managed areas (e.g. organic farmland) support higher pollinator abundance and diversity than intensive agriculture, which results in a higher pollination function for croplands (e.g. Balvanera et al., 2005).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 185
Species abundance and richness in agricultural areas is greatly influenced by the proportion of (semi)natural land (e.g. grasslands, forest areas) in the surrounding landscape (e.g. Watson et al. 2011) and the distance to a (semi)natural habitat (e.g. Carvaleheiro et al., 2010; Albrecht et al., 2007). Albrecht et al. (2007) summarised that species richness of pollinators in intensively managed meadows decreases significantly with distance to (semi)natural areas, and Carvaleheiro et al. (2010) showed a decline in flower- visiting insects with distance to natural habitat (abundance and species richness declined by more than 80% of their maximum values over a distance of 500m).
Higher pollination services can be achieved by increasing landscape heterogeneity. “Residual bee diversity increased with increasing landscape heterogeneity” (Holzschuh et al., 2007) because this provides alternative resources among pollinators.
Articles also cited negative relationships. Species richness and diversity of native pollinators showed a negative relationship with increasing number of managed bees (A. mellifera) in highland coffee plantations in Mexico. Other studies, however, showed a positive impact of managed beehives on pollination services and thus crop yield (e.g. Geerts and Pauw, 2011).
Flower visitation behavioural traits such as foraging distance, flight range, pollinator size, and bee tongue length were also notable attributes influencing pollination success, as they condition the pollinator’s ability to access flowers. Such traits can either be positive or negative depending on other attributes such as location of the ecosystem, landscape diversity, isolation from natural habitat, and pollinator species present. For example, Bommarco et al. (2011) and Biesmeijer et al. (2006) point out the importance of tongue length and flower specificity. Short-tongued bee species are shown to be better adapted to exploit flowers while long-tongued species are more specialised (Bommarco et al., 2011).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 186
Community/habitat area Community/habitat structure Primary productivity Landscape diversity Species richness Functional richness Functional diversity Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Natality rate Flower-visiting behavioural traits (pollination) Other
0 5 10 15 20 25 30 35 Number of studies Negative Both (Positive & Negative) Positive Unclear
Figure 3: Biotic attributes contributing to pollination.
Abiotic attributes Several articles described climate conditions during the study time, but no clear relationship was given between climate and pollination services (Figure 4). Often counting of pollinators happened during sunny, dry days, as pollinators are less active during rainy days (e.g. Blanche and Cunningham, 2005).
Temperature Precipitation Wind Soil Slope Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive Unclear
Figure 4: Abiotic factors influencing pollination.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 187
Human input and management
Most reports described a negative impact by human input (16 studies). Fifteen studies are not clear about the impact of management on pollination services, and two studies show a positive impact by management on pollination.
Human input such as land management is a direct driver of changes in ecosystems, which influence pollination services by reducing natural habitat. The replacement of natural vegetation by crops and the use of pesticide and fertilizer causes loss of natural habitat and fragmentation, which decreases pollinator richness and diversity (e.g. Farwig et al., 2009, Balvanera et al., 2005, Vergara et al., 2009). Some studies also analyse the effect of managed beehives on pollination services in agricultural areas, which were found to have a positive impact on pollination (Geerts and Pauw, 2011). The conservation and protection of (semi)natural habitats such as in National Parks or urban gardens supports pollination services. Land management thus plays an important role in pollination services. Thereby, the intensity of management seems to be crucial. Less intensive farming (e.g. extensive, organic farming) supports pollinator’s abundance and diversity and thus benefits pollination services.
Thresholds
Thresholds were not explicitly mentioned or abundant throughout the literature reviewed. Only one threshold was found and classified as “other biophysical”. The identified threshold highlighted that there comes a point in ecosystems where additional bee visits no longer increase production/productivity (Greenleaf and Kremen, 2006). No other relevant thresholds were found.
Indicators
The most frequently cited indicators were ‘bumblebee density’ (35 studies), land cover (33), habitat (26), land use (22) and crop yield (21). When looking more closely at these indicators it is clear that the indicators from the JRC (2012) report were not always very useful, and indicators were sometimes interpreted differently to be more specific. For example, the indicator ‘bumblebee density’ was in most cases not given; rather the number of flower visiting insects (pollinators in general), or more specifically the number of specific pollinator species such as bees or hoverflies was given. The indicator ‘crop yield’ was often interpreted as the measured fruit set or/and seed set.
An important indicator cited, but not included in the JRC report, was the distance of agricultural fields to (semi)natural habitats, which is classified under the indicator category ‘others’.
Policies
The Biodiversity Strategy was mentioned once; other policy measurements were classified as ‘others’ (11 studies) such as agri-environment schemes, organic farming policy and the conservation of agricultural resources act.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 188
Interactions with other ecosystem services
18 studies described interactions with other ecosystem services: 17 with food production (cultivated crops), and one with pest regulation. While food production was identified as having a positive linkage with pollination services, the relationship between pest regulation and pollination was not clear.
Overall observations on the review
Overall evidence was mostly quantitative, based on multiple observations, direct and empirical. Several articles also specified that statistical analysis methods were used to obtain the evidence; in such cases the statistical analysis performed was commented in the “empirical data/modelled information” comment box. Ten articles were found to present both empirical and modelling data. For example, Winfree et al. (2007) used a model to separate and predict pollen provided by honey bees vs native bees. A total of seven articles provided evidence through literature review, such as Ricketts et al. (2006) and Garibaldi et al. (2011) which based their evidence on 23 and 29 previous pollination-related studies respectively.
Even though often not mentioned explicitly, pollination evidence was frequently related to agricultural production and crop yields. Therefore, in several cases the indicators and evidence presented were not strictly related to the pollinators but reflected the results of pollination through agriculture-related indicators. As a result, the use of the indicators was not always useful; partly because the given indicators were not used in the literature (e.g. bumblebee density was not given, but amount of pollinators collected during a specific time; or crop yield was not given in most cases, but the fruit set or seed set); and partly because important indicators were missing (e.g. distance between agricultural area and natural habitat; landscape diversity).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 189
Pest regulation
Victoria Wyllie de Echeverria and Alison Smith, Environmental Change Institute, University of Oxford, UK
The literature search
The 50 original BESAFE articles were entered into the database and the additional information extracted. A further 10 articles covering the period 2012-June 2014 were then added using the same keywords as for the BESAFE project:
“Natural pest control" OR "Pest control" OR "Biological control" OR "Biological pest control” AND Biodiversity OR “Biological diversity” OR Species OR Habitat* OR Genetic OR Trait* OR Function* OR Landscape OR Richness OR Abundance
Overall, 60 articles were analysed. In the following report, some findings are reported using the percentage of articles supporting the discussion points, and some using the absolute number of articles, depending on the situation.
10 8 6 4 2 Number of articles 0 1990 1995 2000 2005 2010 2015 Publication year
Figure 1: Publication dates of the 60 journal articles reviewed on pest regulation.
Background/context
Studies were classified as being in all spatial scales (Figure 2). Most of the studies focused on a local scale (46), followed by sub-national (7) and global (7). Temporally, most studies (45) were snapshots (Figure 2), with only one long-term study.
Most of the studies were located in North America (10) and Europe (28), with two studies using multiple locations in both North America and Europe (classified as global) (Map 1). Other studies were located in Brazil, Cameroon, Australia (3), Japan (3), Malaysia, the Philippines, Asia (a modelled ecosystem) and Vietnam. In addition, there was one study that bridged sites in Brazil, Indonesia and Malaysia, and five reviews that were classified as global.
The evidence did not appear to vary based on spatial scale. Almost all of the studies (95%) reported data based on empirical evidence, or on models based on a study site, while only three used a conceptual model that was not informed by a specific location. However, since most of the studies (67%) were based in
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 190
Europe and North America, it could be concluded that either most of the researchers work out of those countries, or that is where the majority of funding is located.
Map 1: Location of studies on pest regulation.
Time scale Scenario Continental Spatial scale Long- analysis 2% Sub- term 2% 3% continental 1% Global National 11% Annual 3% Seasonal 17% Sub- 3% national 11%
Local 72% Snapshot 75%
Figure 2: Percentage of studies at each temporal and spatial scale for pest regulation.
A range of ecosystem types were mentioned in the literature reviewed: cropland, grassland, heathland and shrub, rivers and lakes, urban, wetlands, woodland and forest (Table 1). Many articles mentioned more than one of these ecosystem types. As might be expected, cropland was mentioned most frequently (in 83% of studies), followed by woodland and forest (30%), and grassland (28%). All other ecosystem types
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 191
were mentioned in less than 10% of studies. Only five studies mentioned the condition of the ecosystem types.
Table 1: Ecosystem types present in the review of pest regulation.
Ecosystem Count Cropland 50 Condition not mentioned 47 Favourable 2 Unfavourable - declining 1 Grassland 17 Condition not mentioned 16 Favourable 1 Heathland and shrub 2 Condition not mentioned 2 Rivers and lakes 5 Condition not mentioned 5 Urban 5 Condition not mentioned 5 Wetlands 1 Condition not mentioned 1 Woodland and forest 18 Condition not mentioned 17 Favourable 1 Grand Total 98
Linkages between natural capital and the ecosystem service
Ecosystem Service Providers (ESP) The studies reviewed covered ecosystem service providers at all levels. The most frequently studied level was the functional group (single: 16 studies, two or more: 21 studies), because the majority of the studies (62%) researched the influence of an entire functional group or groups (such as wasps, spiders, beetles or birds) instead of narrowing it down to specific species. The individual species level was studied in 28% of studies (single species: 9, two or more: 8) and the community or habitat level was only examined in six studies (Figure 3).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 192
Entire community or habitat 13% 10%
Single functional group 15% 27% Two or more functional groups
Single specific species population
35% Two or more specific species populations
Figure 3: Ecosystem service providers for pest regulation (percentage of studies).
Biotic attributes The biotic attributes mentioned in the studies reviewed are shown in Figure 4. It can be seen that the community-level attributes dominate, with community presence, area and structure each being cited at least 25 times. This reflects the common approach of investigating the dependence of the pest regulation service on the features of a particular habitat, such as cropland or orchards. Functional group-level attributes are also discussed frequently, especially presence and abundance of particular functional groups, functional diversity and functional richness. A large number of species-level attributes are also recorded, but these are mentioned in fewer studies. Species abundance was mentioned by 17 studies, and species richness by 14 studies, but other species attributes received less attention. Predator behavioural traits were discussed in 15 studies. Most biodiversity attributes had a positive influence on the pest control service. The most commonly cited attributes are discussed in more detail below.
As noted above, the top three attributes are related to the entire community or habitat, typically referring to the landscape surrounding crop fields. Results described instances when pest predation on crops was decreased, and populations of natural enemies increased (positive directions). Some examples of landscape features correlated to pest reduction include: complex landscapes (Bianchi et al., 2006; Chaplin-Kramer and Kremen, 2012; Öberg, 2007); crop lands interspersed with and/or surrounded by semi-natural habitat (Letourneau et al., 2012; Rusch et al., 2012; Steingröver et al., 2010), including forest/woodland (Bianchi et al., 2006; Bianchi et al., 2008; Boccaccio and Petacchi, 2009; Mody et al., 2011; Vinatier et al., 2012), dense ground cover (Colloff et al., 2013), grassland/grassy strips (Legrand et al., 2011; Vinatier et al., 2012) and non-crop habitat (Drapela et al., 2008); good connectivity between patches (Bianchi et al., 2010; Boccaccio and Petacchi, 2009); increased vegetation height (Anderson et al., 2013), and diverse plant communities (Drapela et al., 2008; Kellerman et al., 2007; Krauss et al., 2011).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 193
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Community/habitat/stand age Successional stage Primary productivity Aboveground biomass Stem density Litter/crop residue quality Landscape diversity Species richness Functional richness Functional diversity Presence of a specific functional group Abundance of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Population growth rate Life span/longevity Natality rate Mortality rate Leaf N content Predator behavioural traits (biocontrol) Other 0 5 10 15 20 25 30 35 Number of studies Negative Both (Positive & Negative) Positive Unclear
Figure 4: Biotic attributes contributing to the service of pest regulation
Habitat management can therefore influence predator density (Xu et al., 2011; Xu et al., 2012) through modifications such as thicker ground cover (Colloff et al., 2013), creation of field edges (Krauss et al., 2011), and increasing the length of roadside strips (Drapela et al., 2008). Xu et al. (2012) found that the densities of frogs, carabids and spiders (all predators) increased in treatments that had habitat management. Daghela Bisseleua et al. (2013) found that herbivory of cocoa leaves decreased with increasing shade. Improving habitat in fields, such as cutting grass in orchards, strip harvesting crops, and intercropping was also found to increase predation (Iperti, 1999; Xu et al., 2011). Organic farms were found to increase evenness of natural enemy populations (Crowder et al. 2010), have less damage from fungal parasites (Drinkwater et al., 1995) and have higher abundance of pollinators (Krauss et al., 2011).
All of these instances provided refuges for natural enemies, either for breeding, alternative food sources or protection, to increase their effectiveness as predators. The majority of the studies (69%) found positive effects, but there were also studies that discussed negative or both negative and positive results. These often listed similar examples (complexity, agricultural practices, etc.), but reached different conclusions. For example, Olsen and Wäckers (2007) found that field margins could act as a sink, rather than a source,
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 194
for parasitoid predators, with higher predation rates occurring in the centre of crop fields rather than at the margins. Similarly, Rand and Tscharntke (2007) found that predator-prey ratios were higher in simpler landscapes, concluding that there could be higher spill-over of predators from semi-natural habitats such as field margins into crop fields in these situations, as predators sought alternative prey in the crops.
For species level traits, two main attributes were identified: species abundance and predator behavioural traits (related to bio-control). With species abundance, unsurprisingly, it was found that increased densities of predators generally suppressed pest levels, with examples including spiders (Drapela et al., 2011; Oelbermann and Scheu, 2009), carabids (Legrand et al., 2011), and predators of soybean aphid (Blauuw and Isaacs, 2012). Kamo et al. (2010) found that the rate of development and abundance could be used as a pest control indicator. Zhang et al. (2010) reported that if pest pressure was higher, habitat management had a greater advantage over other forms of pest control. One area where studies were unclear about the direction of the results was when looking at relative predator abundance levels in fields and surrounding habitat, although Nash et al. (2008) found that 95% of all predators were collected in the edges, while only 83% were collected in fields.
Several examples of predator traits leading to pest suppression and positive results are listed. The ability to disperse well, over long distances (such as ballooning for spiders - Drapela et al., 2011; Öberg, 2007), and speedily (Bianchi et al., 2006) were all listed as positive traits. Species that have the ability to form aggregations during dormancy are advantageous (Iperti, 1999), as they can hatch en masse and attack prey. Other positive behavioural traits include species that are able to feed on other sources of food when prey density is low, as this allows predators to have an alternative food source (Castañé et al., 2011; Vollhardt et al., 2010); specific positive foraging behavioural traits associated with bats that allow them to be good predators (Lee and McCracken, 2005); and two species of mite predators that forage symbiotically, in different places on the leaf and in different seasons to effectively control mites (González-Fernández et al., 2009). Two studies found that when bats (Kunz et al., 2011) and birds (Koh, 2008) were excluded from some sampling locations there was increased herbivory by pests.
As many studies focused mainly on functional groups, it is not surprising that the abundance of a specific functional group was mentioned by many studies, and is thus in the top seven attributes in Figure 4. Similarly to species abundance, the density of a functional group/s resulted in increased predation on pests (Blauuw and Isaacs, 2012; Koh, 2008). Roubah et al. (2014) commented on predator size: larger beetles equalled more predation on pests.
A number of studies found that species richness, functional richness and functional diversity are important. However, while many of these studies find that land use management can enhance predator diversity, fewer demonstrate that predator diversity reduces pest activity. Those that do (e.g. Wilby et al. 2005; Casulo and Nannini, 2013; Kellermann et al., 2007; Rush et al., 2010) attribute this to niche complementarity, with different predators attacking different prey sizes, life stages, population densities and behaviour (e.g. flying vs. ground dwelling). Macfayden et al. (2012) state that complementarity is only important where there is resource partitioning. A number of studies do not find that diversity increases predation (e.g. Rezende et al., 2013).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 195
Although most studies displayed positive findings, some studies did have unclear results. For example, Kromp (1999) found that management can enhance functional diversity and predator abundance, but that the actual impact on pest populations was unclear. Olson and Wäckers (2007) found that predator abundance in field margins and crops varied widely between species and in different years, and concluded that the benefits of field margins for pest control were unclear.
Attributes that did not have their own designation, and thus were classified under ‘other’, were quite varied, though on re-examination it might be possible to file some of these under an assigned attribute. However, some of them do bridge between two attributes. Several of the attributes classed under this category concerned the positive effects of some form of organic agriculture and habitat/crop management (Drinkwater et al., 1995; Geiger et al., 2010; Legrand et al., 2011; Öberg, 2007; Steingröver et al., 2010; Xu et al., 2011). Some of the more interesting attributes in the ‘other’ category include the findings that linking the shape and the area of natural habitat is important, where the optimal solution is small patches of land in ‘archipelago formations’ (Zhang et al., 2010); that early pest reduction is much more effective than later in the season (Chaplin-Kramer and Kremen, 2012) and that habitat connectivity may be more important for enemies than pests (Bianchi et al., 2006).
Abiotic attributes The most common abiotic attribute cited was temperature (Figure 5), though the direction varied with each study and was often unclear. It was found that more moderate temperatures, in these cases provided by wooded climates (Bianchi et al., 2006) and mulch (Xu et al., 2012), tended to be more favourable to natural enemies, while extremes of temperature, both cold and hot, were less favourable for predation rates (Casula and Nannini, 2013), increased predator and prey populations (Xu et al., 2011) and changed development patterns and feeding rates (Iperti, 1999).
The next most commonly mentioned attribute was precipitation, though no articles gave specific examples of how precipitation affected pest regulation, but simply noted that it varied between study years and/or sites. Kromp (1999) noted that soil factors influenced predator density, but the direction was unclear, as this comment was not further clarified.
Temperature Precipitation Soil
0 10 20 30 40 50 Number of studies Negative Positive Unclear Figure 5: Abiotic factors influencing pest regulation.
Human input and management
Despite many studies being centred on agricultural areas and human-modified systems, human influence was only directly discussed in 22 of the studies, with 14 of these citing a direct influence. Of the 11 studies
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 196
where the intensity was mentioned, five concerned intensive input (typically referring to intensive agriculture), and five both intensive and extensive input (often comparing intensive and extensive agricultural techniques), with one being unclear. The direction of impact was mixed, being positive in eight studies, negative in five studies, both positive and negative in six and unclear in three.
Examples of negative human impact include the depletion of surrounding semi-natural habitat (Woodcock et al., 2014), removal of shade-providing trees in cocoa plantations (Daghela Bissuleua et al., 2013), or more intensive agriculture reducing the ability of ecosystems to provide habitat for natural enemies (Anderson et al., 2013). Some examples of positive impacts included planting wildflower patches (Blauuw and Isaacs, 2012), mulching as an agricultural technique (Schmidt et al., 2004), thick groundcover between orchard rows (Colloff et al., 2013), and less intensive rotation (Rusch et al., 2013), all of which provided habitat for natural enemies and thus reduced pest levels.
Thresholds
Two examples of thresholds were found, though one (Rusch et al., 2012) was a classification of three thresholds for beetle abundance per plant that was used to formulate a model, and did not necessarily correspond to a real life threshold in the performance of the ecosystem service. The other was a threshold for land cover identified by Letourneau et al. (2012), who found that rates of pest parasitism by Tachnid flies dropped precipitously as percentage annual crop cover exceeded species-specific thresholds (51% for Madremyia saundersii and 38% for Voria ruralis).
Indicators
The most commonly cited indicator was pest density (used based on primary data in 16 studies and with models in two studies). Pest control expenses were cited in three studies, based on primary data, and the proxy indicator land cover was classified using primary data in seven studies. However, there were 17 studies that used indicators not included in the JRC (2012) report, which were therefore classified as “other”: these were mainly to do with the population or richness of pest predators / natural enemies, though some studies also cited crop yield or pest damage.
Policies
Only one of the studies mentioned policy. Woodcock et al. (2014) discussed the possibility that general agri-environmental schemes can increase the robustness of ecosystem services provided by pest control and pollination by conserving and even creating new ‘semi-natural’ habitats.
Interactions with other ecosystem services
Interactions with several other ecosystem services were identified: pollination, food production, recreation and four classified as ‘other’ (Figure 6). The positive interactions with food production and pollination are obvious, as pest regulation reduces damage to food crops, and the creation of semi-natural habitats to boost pest predators will also have benefits for pollinating insects (e.g. bees, hoverflies). A wide range of other services can be provided: for example, Kunz et al. (2011) list the services provided by bats, including pest control, fertilizer (guano), eco-tourism opportunities (recreation), pollination and seed dispersion,
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 197
medicine and cultural symbols. Tscharntke et al. (2011) point out that shade trees not only provide habitat for predators, but also help with climate adaptation: they can lower air and soil temperatures, protect against droughts, increase resilience, increase humidity levels and reduce water stress. However, there can also be some trade-offs: for example, planting trees for shade is beneficial for pest regulation but can reduce crop yield (Daghela Bisseleua et al., 2013); planting ground cover between crop rows provides habitat for pest predators but can compete with crops for nutrients and water (Colloff et al., 2013); predators themselves can sometimes damage crops (Castañé et al., 2011), and non-crop plants can provide food and refuge for pests as well as pest predators (Bianchi et al., 2006).
Food production (cultivated crops)
Pollination Negative Recreation (species-based) Both (Positive & Negative) Other Positive Unclear 0 5 10 15 20 Number of studies
Figure 6: Interactions between pest regulation and other ecosystem services.
Overall observations on the review
In most studies (34), the evidence was empirically based, nine were model based, nine were both empirically and model based and eight did not mention what the evidence was based on. Forty-four studies were informed by direct evidence, 13 by indirect and three did not mention the directness of the evidence. Most (52) studies used quantitative methods, four studies used both quantitative and qualitative methods, and four studies did not mention which methods were used. Most (50) studies based their data on multiple observations, five on single observations, and five did not mention the number of observations.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 198
Recreation (species-based)
Nina Fabrega Domenech, European Commission, Joint Research Centre (JRC), Ispra, Italy
The literature search
The literature review on recreation (species-based) as an ecosystem service contained 32 updated studies which were previously included in the BESAFE database and 29 additional studies which consisted of new articles; therefore the OpenNESS recreation literature review included a total of 61 articles. Of the 29 articles added only five were from the years 2012 to present (four articles from 2012 were already in the BESAFE database) (Figure 1).
12 10 8 6 4 2 Number of articles 0 1990 1995 2000 2005 2010 2015 Publication year
Figure 1: Publication dates of the 61 journal articles reviewed on recreation.
As recreation (species-based) was not a new ecosystem service being reviewed, no new search terms were generated for the review process. The same search terms used to obtain the BESAFE database articles were explored. Different combinations of the following previously established search terms were used:
Recreation OR tourism AND species OR habitat OR genetic OR trait OR functional OR landscape OR abundance. Recreation OR tourism OR angling OR hiking OR birding OR hunting OR photography OR bat OR fishing AND biodiversity OR biological diversity AND species OR habitat or genetic OR trait OR functional OR landscape OR abundance.
The same combination of search terms was also carried out by substituting the search term tourism for aesthetic value in order to refine results. The database used to perform the searches was SCOPUS.
Background/context
All articles consisted of temporal scales classified as snapshots in time (Figure 2). The majority of studies, specifically 50, were set in local locations (Figure 2) since the provision of recreation (species-based) ecosystem services generally took place in parks and/or reserves where the service was abundant; i.e. national parks, natural reserves, or specific bodies of water among others. In many studies the provision of the ecosystem service (recreation) was examined simultaneously at several local parks/reserves within one country or region. Therefore several studies had more than one local location.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 199
Furthermore, other studies concentrated on the provision of a specific type of recreational ecosystem service rather than being location-specific. Such studies were classified as having global or continental scales as they reviewed the provision of recreational ecosystem services in a general manner or in more than two locations. For example, Cisneros-Montemayor and Sumailia (2010) discussed the benefits of ecosystem-based marine recreational activities across the globe, while Lindsey et al. (2007) compared recreational wildlife viewing preferences among tourists in four different protected areas in South Africa.
Time scale Spatial scale
Continental 1% Global 7%
National 10% Sub- national 11%
Local 71%
Snapshot 100%
Figure 2: Percentage of studies at each temporal and spatial scale for recreation.
Evidence did not vary according to the location of the study, but mostly depended on the type of recreational (species based) ecosystem service that was being studied. Recreation ecosystem services, such as wildlife viewing and hiking were based on observations and tourist questionnaires/surveys, while recreational ecosystem services such as hunting or fishing were based on measured evidence such as catch per unit effort or exploitation rates.
Ecosystem types ranged from woodland and forest (9 studies) to coastal and marine ecosystems (17) and river and lakes (8). Other less common ecosystem types found included cropland, grassland, heathland and shrubland, sparsely vegetated land and wetlands (Table 1).
The most common and relevant information found about ecosystem condition was whether the ecosystem area entailed any type of conservation or management, i.e. the ecosystem consisted of a national park/reserve, or some type of conservation management was being implemented. This was the case for the majority of ecosystems described in the reviewed studies. However, the condition of ecosystems overall was not explicitly described. Only two studies, Wee and Tsang (2008) and Zhang (2012c), specifically described ecosystems that had been to some extent degraded, which, as a result, led to the implementation of management measures to ensure ecosystem conservation/recovery. The remaining studies focused on locations that were already protected or managed and therefore did not discuss conservation being implemented as a result of degradation, or describe the health of the ecosystem.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 200
Table 1: Ecosystem types present in the review of recreation.
Unfavourable - Condition not Ecosystem Total recovering mentioned Woodland and forest 9 9 Heathland and shrub 4 4 Grassland 1 1 2 Cropland 4 4 Sparsely vegetated land 1 1 Wetlands 4 4 Rivers and lakes 8 8 Coastal 1 13 14 Shelf 1 1 Open ocean 2 2 Total 2 47 49
Linkages between natural capital and the ecosystem service
Recreational ecosystem service providers (ESPs) consisted exclusively of whole areas/habitats and/or specific wildlife species. The most frequently mentioned ESP classes were two or more specific species populations (23 studies) and single specific species populations (20 studies). The third most frequently mentioned ESP was entire communities or habitats (14 studies). Examples of single species populations as ESPs included dolphins for wildlife viewing, hunting game such as deer or lions, and specific fish populations for recreational angling. In addition, frequently mentioned entire communities/habitats ranged from coral reef communities to national parks, and designated hunting areas.
The top three biotic attributes identified were: presence of a specific species type, species abundance and species richness (Figure 3). Therefore, species-related biotic attributes were the most relevant attributes to consider when analysing the provision of recreational related ecosystem services. The presence of specific species was shown to be crucial for recreational ecosystem services such as wildlife viewing, hunting or fishing, as the provision of these services depends solely on the presence of relevant species. For example, the presence of megafauna, such as elephants and lions, in natural reserves in Africa was the key factor in attracting tourists and ensuring tourist “satisfaction” throughout their wildlife viewing experience. As for hunting and fishing, the presence of a specific species was evidently crucial since without the presence of game species (i.e. feral pigs, deer, fish) the corresponding recreational ecosystem service of hunting or fishing would have ceased to exist in that particular ecosystem.
As for functional group biotic attributes, functional group diversity and richness were identified in two different studies. Functional group diversity was considered an important attribute when examining wildlife viewing as a recreational ecosystem service. Functional group diversity, such as high mammal diversity in protected areas in South Africa (Lindsey, 2007), was found to be the most important feature considered among tourists throughout their wildlife viewing experience.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 201
Finally, community/habitat-related attributes such as presence of a specific community/habitat type or community area were not so frequently identified, since the presence of specific species was usually the principal biotic attribute recognised. Nevertheless, the presence of a specific community/habitat type was acknowledged as important when considering, for example, the provision of recreational ecosystem services by whole coral reef communities. Similarly, community area was noted as being significant when considering ecosystems providing hunting as a recreational ecosystem service, given that the size of the permitted hunting area and access to the habitat of game species was deemed essential to the provision of the service.
Presence of a specific community/habitat type Community/habitat area Species richness Functional richness Functional diversity Presence of a specific functional group Presence of a specific species type Species abundance Species population diversity Species size/weight Natality rate Mortality rate Other
0 5 10 15 20 25 Number of studies Negative Both (Positive & Negative) Positive Unclear
Figure 3: Biotic attributes contributing to the service of species-based recreation.
Abiotic attributes identified as important to the provision of recreational ecosystem services comprised of weather-related attributes such as precipitation, temperature and wind, in addition to water quality, and nutrient availability (Figure 4).
Several other abiotic attributes that were not included as choices in the database were identified as important, including distance to urban areas.
Extreme weather conditions including high temperatures and strong winds were classified as having a negative impact on service provision because they decreased the opportunities for humans to take advantage of the recreational ecosystem service in question. All other abiotic attributes were considered to be both positive and negative depending on their state. For example, depending on the reported water quality of a lake, water quality could be classified as having a positive or negative influence for the provision of recreational fishing as an ecosystem service.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 202
Temperature Precipitation Wind Nutrient availability Water quality Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative)
Figure 4: Abiotic factors influencing species-based recreation.
Human input and management
Human management was a fundamental factor in the provision of recreational ecosystem services, because the provision of such services predominantly takes place in protected/managed natural areas. Even though human input and management was significantly present throughout the reviewed literature, details about the extent and intensity of human input and management were often lacking. As a result, any mention of a national park/reserve or conservation management was considered a direct human input.
Management plans that were proven to be inadequate to provide proper conservation, or poor management that was not carried out properly, were classified as negative human input. Negative conservation efforts were often found to be due to legislative bureaucracy, uncontrolled implementation of protection plans, poor adherence to regulations, and lack of monitoring. However, most human input management was classified as both positive and negative as usually, to a certain extent, there was always some type of benefit. Examples where human input and management was classified as both positive and negative include studies where habitat was disrupted or altered but there was a trade-off as a result of the human intervention which led to increased environmental knowledge/awareness or further contribution towards conservation and management efforts.
Management and conservation plans that did not mention any setbacks or conflicts were classified as positive human input. Examples of positive human input, other than successful conservation and/or conservation improvements, included: effective enforcement of regulations, improvement in overall monitoring, risk assessment, interdisciplinary cooperation of stakeholders, promotion of environmental education, and local community involvement and development.
The intensity of human management, defined as intensive or extensive according to the use of capital, was always classified as unclear. None of the reviewed studies directly addressed the use or source of capital spent for either conservation or management plans.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 203
Thresholds
Thresholds were not prominent throughout the literature reviewed regarding the provision of recreational ecosystem services. One article was found for each of the three threshold categories; other biophysical, safe minimum standards, and legal boundaries. The identified thresholds included phosphorus as a threshold for an oligotrophic and eutrophic lake (Arlinghaus and Mehner, 2003) classified within the biophysical category; a safe minimum standard threshold regarding how noise from boats may affect fish (Lewin et al., 2006); and lastly a poorly defined property rights issue (Zivin et al., 2000) classified as a legal boundaries threshold.
Indicators
There was a considerable variety of indicators used to quantify recreational (species-based) ecosystem services throughout the reviewed literature. Several of the indicators included in the JRC’s “Indicators for Mapping Ecosystem Services: A Review” (Egoh et al., 2012) report were used in the reviewed studies. Such indicators included: distance to scenic site, protected area, land cover, land use, accessibility, fish abundance, number of visitors, bird density and bird richness. Indicators not included in the JRC report were also found in 40 studiesThese often entailed monetary measures such as cost per trip, average cost per person, contribution to GDP, demand for the recreational ecosystem service, willingness to pay, and hunting lease prices. Other more subjective indicators which were obtained through surveys and questionnaires consisted of documenting tourist motivations for their visit. These indicators involved quantifying visitor viewing preferences, knowledge on area/species visited, and conservation interest.
In contrast, quantitative indicators mentioned included measurements such as catch per unit effort, reproductive success, number of nests, and density of travellers. Finally, some other indicators used for quantifying recreational wildlife viewing experiences that were cited but not mentioned in the JRC report were: frequency of citing, viewing time, number of vehicles, duration of visit, distance from residence, tourism disturbance, and wildlife viewing distance.
Policies
Overall, the presence of policies concerning the provision of recreational ecosystem services was not a main feature throughout the literature reviewed. However, some articles did explicitly mention some policies that stood out but did not go into much detail about their relevance. Policies cited in the reviewed literature ranged from EU directives to action plans, regulations, and development strategies.
EU directive policies were mentioned in two studies: the EU water framework directive (2000/60/EC) and EU Habitat directive (92/43/EEC) were both mentioned in Butler et al. (2009) as an opportunity to perform better informed management decisions as a result of the implementation of such directives. Apollonio et al. (2000) also made reference to the same EU Habitat directive in addition to citing the Birds directive (79/409/EEC) as framework for creating a network of sites intended for biological diversity conservation within the European Union.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 204
Other policies mentioned consisted of implemented regulations such as the Tanzanian Tourist Hunting Regulations of 2010, the enforced Nepal government buffer zone regulations in 1996, and the Rottnest Island Authority Act in 1987 which managed the island for recreation purposes by controlling catch without restricting levels of participation. Other less specifically described policies included the catch and release fishing ban in Germany, legislation making trapping practices cheaper in California, and a 2004-2009 Biodiversity Action Plan in Denmark.
Interactions with other ecosystem services
The provision of recreational (species-based) ecosystem services was in some cases linked to food production, freshwater fishing, pest regulation, shoreline protection, and pollination services. All relationships were positive although it is important to note that some had the potential to become negative without proper regulation. These interlinkages were often implied or only briefly mentioned. Therefore, empirical evidence linking recreation to other ecosystem services was generally not present.
When considering fishing as a recreational ecosystem service, often the species in question also had a commercial fishing value for food provisioning. For example, this was seen in the case of migratory fish species in Indian rivers which had both a commercial (food provisioning) and recreational ecosystem service value; such a connection is mentioned but not examined or explained in detail in Everard and Kataria (2010).
Pest regulation was linked to the recreational ecosystem services of hunting and wildlife conservation/viewing. Wildlife conservation/viewing was connected to pest regulation in the cases of bats and dragonflies. Reviewed literature showed that the protection of both species has the potential to provide important insect regulation services and consequently reduce agricultural pests. In addition, the conservation of bats was also a positive link favouring pollination ecosystem services (Pennisi, 2004). As for the link between recreational hunting and pest regulation, the relationship was clear in cases where hunting game, such as feral pigs (Zivin, 2000) among others, were considered a potential threat for agricultural land. Hence, allowing controlled recreational hunting both provided a recreational ecosystem service as well as controlling the growth of hunting game populations which simultaneously served as a form of pest control.
Overall observations on the review
Overall, most studies contained quantitative, direct empirical data for multiple observations. Some studies were also simultaneously accompanied by qualitative evidence. Furthermore, 11 studies also contained modelled or projected evidence. Studies that were based on surrogate evidence were not dominant (4 studies). These studies were general literature reviews about a specific recreational ecosystem service. Many studies used different statistical methodologies in order to provide empirical evidence as well as to model data. If applicable, statistical techniques were mentioned in the “comment” boxes provided.
Studies varied greatly according to the specific recreational ecosystem service being studied. Overall it is difficult to generalise about the quantification and study of recreation (species-based) ecosystem services since methodologies, indicators and analyses vary greatly according to the specific recreational ecosystem service considered. It is very different to examine megafauna wildlife viewing in Africa and recreational
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 205
fishing in Germany, nevertheless they are both considered recreational (species-based) ecosystem services. There is great range of ecosystem services encompassed within this category and as a result, the reviewed studies varied from being extremely specific including models and statistical analysis to being quite general and based on observations or literature reviews. Studies which were most helpful were those that encompassed a variety of stakeholders and demonstrated the potential interlinkages (positive and negative) and collaborations that can stem from conservation efforts, tourism, community development, and even sustainability. Demonstrating the value of a multi-disciplinary integrated approach towards the management of recreational ecosystem services is valuable for future decision-making and management plans regarding all type of ecosystem services and conservation efforts.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 206
Aesthetic landscapes
Francis Turkelboom and Els Lommelen, Institute for Nature and Forest Research (INB0), Belgium
The literature search
50 studies (45 articles) were entered from BESAFE and 11 additional articles were added, totalling 61 studies (56 articles) on aesthetic landscapes in the database (Figure 1).
Keywords used in Web of Science to conduct the literature review were: biodiversity OR habitat* OR landscape* AND tourism OR recreation OR *esthetic* OR appreciation OR valuation OR preference* OR perception*
15
10
5
Number of articles 0 1980 1985 1990 1995 2000 2005 2010 2015 2020 Publication year
Figure 1: Publication dates of the 56 journal articles reviewed on aesthetic landscapes.
Background/context
Spatial scale and location The majority of the studies reviewed (52 out of 61 studies) investigated aesthetic landscapes on a sub- national scale. As each study often covers different regions to strengthen evidence or to make a comparison, a total of 105 locations are covered, of which 95 (90%) are at the sub-national scale, with only six at the national scale and one or two each for the local, continental and global scale (Figure 2). Nearly all the study sites were located in Europe or North America (Map 1). This partly reflects the existence of cultural and linguistic barriers to the standard technique of carrying out quantified surveys of user preferences (e.g. based on written questionnaires and photographs) in some countries, where observational techniques such as those used in Cocks et al. (2012) may be more suitable, though possibly less objective.
The evidence seems to vary depending on the location of the study, though this cannot be verified because there are no studies where countries from different continents are compared (only immigrants and locals within one country). In European or Western countries, it seems that people generally prefer natural elements and the absence of manmade features. This is a general conclusion for the review, as the majority of the articles were European studies. For example: a study from the Netherlands shows that people tend
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 207
to have a strong appreciation for wild nature with wild animals, in this country where wild nature has completely disappeared. On the other hand, some articles in countries where nature dominates the landscape, for instance in African countries, show that people tend to prefer open landscapes or cultivated landscapes and/or the presence of human elements. However, the geographic location is not the only explanatory factor for the evidence. Other important explanatory factors are, for example, the country of origin of immigrant respondents, and user-related, professional or cultural backgrounds of the respondents. There can thus be a large variation of the evidence within one geographic location depending on the respondent group. It is clear that the same landscape (features) can be appreciated in very different ways depending on the socio-cultural background of the inhabitants and/or visitors.
Map 1: Location of studies on aesthetic landscapes.
Temporal scale Most of the studies (51), focused on snapshots in time, whilst a few other studies were based on scenario analyses (Figure 2).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 208
Time scale Spatial scale Global 1% Continental Local 2% 1% National Seasonal Scenario 6% 2% analysis 15%
Sub- Snapshot national 83% 90%
Figure 2: Percentage of studies at each temporal and spatial scale for aesthetic landscapes
Ecosystem types Eleven out of the 12 ecosystem types were mentioned in the reviewed articles (Table 1). Woodland and forest were mentioned most often (31 studies), followed by urban (19), rivers and lakes (12) and grassland (11). The condition of the ecosystem type was only mentioned twice: unfavourable–declining for grassland, and favourable for woodland and forest (Gómez-Limón and de Lucío Fernández, 1999).
Table 1: Ecosystem types present in the review of aesthetic landscapes.
Unfavourable Condition not Ecosystems Favourable Total - declining mentioned Woodland and forest 1 30 31 Heathland and shrub 4 4 Grassland 1 10 11 Cropland 9 9 Sparsely vegetated land 2 2 Urban 19 19 Wetlands 5 5 Rivers and lakes 12 12 Coastal 3 3 Marine inlets and transitional waters 2 2 Open ocean 2 2 Total 1 1 98 100
Linkages between natural capital and the ecosystem service
Ecosystem service providers (ESPs) The ecosystem service provider of aesthetic landscapes is most often a single habitat (48 studies) and otherwise usually two or more habitats (12 studies). As these habitats are considered to be the providers
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 209
of beauty, recreational value and other indicators (see below), it was preferred to use the term ESP ‘habitat’, rather than ESP ‘community’.
Biotic attributes The most commonly mentioned biotic attributes are habitat-related attributes (Figure 3), especially habitat structure (48 studies/42 articles), presence of a specific habitat type (15 studies) and habitat area (6 studies).
Presence of a specific community/habitat type Community/habitat area Community/habitat structure Successional stage Aboveground biomass Species richness Presence of a specific species type Species abundance Other
0 10 20 30 40 50 60 Number of studies Negative Both (Positive & Negative) Positive Unclear Figure 3: Biotic attributes contributing to aesthetic landscapes.
Habitat structure is positively correlated with the appreciation indicators in 35 studies (30 articles), negatively correlated in one study, both positively and negatively correlated in three studies, and the correlation is unclear in nine studies. Landscapes with a diverse structure of natural elements are indeed generally appreciated by a lot of people. Exceptions are some user groups that prefer less dense vegetation as a response to their connection with their environment, e.g. traditional farmers have a strong preference for open landscapes (Van den Berg et al., 1998; Swaffield and Foster, 2000; Pinto-Correia et al., 2011; Gómez-Limón and de Lucío Fernández, 1999; Van den Berg and Koole, 2006); black Americans usually show less appreciation for dense vegetation with a sense of enclosure (Kaplan and Talbot, 1988); while immigrants from Islamic countries show low preferences for wild and unmanaged landscapes (Buijs et al., 2009). In Arizona pine forests, the most structure-rich habitat of a choice experiment is one that is generally found to be less attractive, for instance a less dense stand with downed trees (Brown and Daniel, 1986). The correlation of habitat structure was classified as unclear in cases where half-open landscapes are preferred, above very open or very closed landscapes (e.g. students in Tveit, 2009). Also classified as such were cases where appreciation for rich/poor habitat structure differed between groups, and the link with habitat structure was less clear for some of the groups (Swaffield and Foster, 2000; Pinto-Correia et al., 2011; Gómez-Limón and de Lucío Fernández, 1999). One study on the link between presence of natural elements (which was the focus of the study) and community/habitat structure (Acar and Sakici, 2008) was also classified as ‘correlation unclear’.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 210
The presence of a specific habitat type is positively correlated with landscape aesthetics or appreciation in 12 studies, both positively and negatively correlated in one study and the correlation is unclear in two studies. So on the whole, people appreciate forests, mountains, agricultural landscapes, a number of other habitat types, and some trees at their workplace area (Heyman, 2012; Kaplan and Austin, 2004; Püschel- Hoeneisen and Simonetti, 2012; Kienast et al., 2012; de Aranzabal et al., 2009; Darmstad et al., 2006; Howley et al., 2012; Kaplan, 2007). Proximity of forests and parks has a significant positive effect on housing prices and households are willing to pay for changing agriculture into nature (Bowman et al., 2012; Liekens et al., 2013). In contrast , in Newmarket, afforestation was perceived to be negative as it was an exotic tree species and it was considered as 'not a part of the landscape of the area' (Ni Dhubháin et al,. 2009).
Habitat area is positively correlated with landscape aesthetics in five studies and the correlation is unclear in one study. So in general there is a preference for larger areas of urban green (Snep et al., 2009), extension of protected areas (Czajkowski et al., 2014), spacious rice paddy landscapes (Natori and Chenoweth, 2008) and households are willing to pay to enlarge a nature area (Liekens et al., 2013). Two other articles deal with preferences for more natural landscape features (Cocks et al., 2012; Schmitz et al., 2007).
One study deals with aboveground biomass, where more biomass is more appreciated, and one with successional stage, where climax vegetation was more appreciated.
Although the focus of this review was on landscapes, presence of species was also included in some studies if they influenced the appreciation of landscapes. When species are mentioned (13 studies), the attribute species richness (9 studies) is most often mentioned in the literature on aesthetic landscapes, followed by presence of a specific species type (4 studies), and species abundance (2 studies). These species-linked attributes only have a positive impact on the service in 10 studies, both a positive and negative impact in one study, and an unclear relationship in two studies. Thus, as could be expected, most people appreciate the presence of a variety of species, a high biodiversity and/or the presence of rare or remarkable species. Ja-Choon et al. (2013) found that one group of urban dwellers, a group consisting mainly of people from small and medium sized cities without children, showed a negative preference for a rich level of biodiversity and a positive preference for an average level of biodiversity. The majority of urban dwellers (mainly with children) in this study were however more interested in landscapes which had more species. Unclear correlations with species richness or presence of a specific species type were found when respondents considered other attributes to be more important in their choice. For instance, from the point of view of tourists, rare flora does not have a big influence on the attractiveness of a Taiwanese forest, because the uniqueness of forest landscapes and scenery, special climate phenomena, accessibility and accommodation were considered as more important contributors to attractiveness (Lee et al., 2010). Similarly, in a Swedish study, respondents preferred a species-poor ornamental park, despite their ability to correctly estimate species richness (Qiu et al., 2013).
Abiotic attributes Abiotic attributes mentioned in articles on aesthetic landscapes are: water availability, water quality, slope, relief, distance and climate/weather (Figure 4). Water availability is mentioned in nine studies (7 articles). The availability of water is mostly perceived as positive, but twice perceptions were both positive and
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 211
negative. People like a river with a forested buffer in their neighbourhood, but they do not like a river without a buffer. The same study shows that association-owned water features in the neighbourhood have a positive effect on housing prices, whereas the proximity to a public lake has a negative effect on home values (Bowman et al., 2012). Water quality has always a positive effect on visitors of aquatic ecosystems (Huppert et al., 2003; Hadwen et al., 2012; Kreitler et al., 2013).
Slope and relief are often appreciated (Swaffield and Foster, 2000; García-Llorente et al., 2012; Ja-Choon et al., 2013), which could be linked to the fact that mountains are appreciated for their beauty (Kienast et al., 2012). Yao et al. (2012b), on the other hand, found that visual preference increased as the variety of topography decreased. However, their study area in China was generally flat and topography was not a marked element in the visual quality. As could be expected, distance is always negatively correlated with preference or willingness to pay (Kaplan, 2007; Liekens et al., 2013; Kienast et al., 2012). Some specific weather conditions were appreciated by users, e.g. waves for surfing or micro-climate in bays (Lee et al., 2010; de Aranzabel et al., 2009; Ruiz-Frau et al., 2013).
Water availability Water quality Geology Slope Other
0 10 20 30 40 50 Number of studies
Negative Both (Positive & Negative) Positive
Figure 4: Abiotic factors influencing aesthetic landscapes.
Human input and management
With respect to aesthetic landscapes, human input and management was considered to be the strength of changes humans made to the natural landscape, such as agriculture and urbanisation. A positive impact means that an intensive agricultural landscape or an urban region is preferred more than a natural landscape. The human input is considered direct if the focal area has been managed, and indirect if a managed adjacent parcel potentially affects the appreciation. The intensity of human impact (intensive/extensive) is estimated from the description in the article: intensive are large fields of cultivation, often monocultures or this might be used for a city centre as well. Extensive management refers to traditional farming, forestry and small towns or visual elements of human activity/presence in a landscape.
Human impact is often disliked (16 studies/15 articles, e.g. Heyman, 2012). Examples of anthropic elements which are negatively appreciated are: buildings, parking areas, infrastructure, roads (Qiu et al., 2013; Kaplan et al., 2006; Acar and Sakici, 2008; Kienast et al., 2012; Kaplan, 2007; Hadwen et al., 2012), and clear-cuts in forestry, or a retention of less than 15% of the trees (Ribe, 2006 and 2009). Traditional,
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 212
more extensive farming landscapes were preferred over more intensive modern farming landscapes (Howley et al., 2012). Also crowding is disliked, for instance by white people in Chicago urban parks (Gobster, 2002), or by divers who preferred to visit places with less divers (Schuhmann et al., 2013).
Positive appreciation for human influences on the landscape was found in six studies. This appreciation was often linked to elements that were considered as typical for the region, such as traditional agriculture, cultural buildings and tree lines, and historic-artistic heritage (Marzetti et al., 2011; van Berkel et al., 2014; de Aranzabel et al., 2009). Furthermore, people are willing to pay to make nature more accessible via trails (Liekens et al., 2013), and divers appreciate the presence of wrecks as this is linked to a high biodiversity (Ruiz-Frau et al., 2013).
In the six articles where human impact was appreciated both positively and negatively, appreciation depended on which user group was considered. For instance, farmers and immigrants prefer managed landscapes over spontaneous nature (Buijs et al., 2009; van den Berg et al., 1998).
In nine articles, the correlation between landscape appreciation and human impact was unclear, often because only the consequences of the human impact were evaluated, not the visual quality of the human impact. External access, for instance, has a positive effect on the attractiveness of the forest (Lee et al., 2010), but it is not clear if the road infrastructure as such is appreciated. Kaplan and Talbot (1988) found that blacks favour built compounds, not because they particularly like buildings, but they like openness and less dense nature, which is often found around buildings.
Thresholds
No information was found on thresholds.
Indicators
Of the indicators from the JRC report (Egoh et al., 2012), aesthetic value (31 studies), scenic beauty (12 studies), landscape and nature diversity (5 studies) and rare species (1 study) are cited. The difference in meaning between aesthetic value and scenic beauty is however not always clear. In some articles, both words are used interchangeably with reference to the respondent’s preference (e.g. Thorn et al., 1997).
In the present report ‘scenic beauty’ is used whenever words like scenery, beauty or views were used; other wordings referring to preference with respect to visual qualities of landscapes were classified within ‘aesthetic value’. The term ‘preference’ is mentioned in at least 15 articles. As a preference with regard to visual issues (pictures, landscapes) has almost the same meaning as aesthetic value, it has been added to aesthetic value, instead of adding a third indicator with a similar meaning as aesthetic value and scenic beauty. In the same way, appreciation (2 studies) was considered the same as aesthetic value.
Twenty-two studies (21 articles) mentioned indicators that were not included in the JRC report. Also here the term preference appears, but here it is used with respect to recreational value. On the whole, recreational value seems an important indicator, as at least 11 articles use one or more indicators that are related to recreation. Other indicators are rather specific and linked to the main goal of the article, for
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 213
instance overall acceptability with respect to clearings (Ribe, 2006). An indicator to score naturalness or wilderness has been used three times (van den Berg and Koole, 2006; Huang, 2013; Natori and Chenoweth, 2008). Other indicators are very indeterminate terms such as perception (Ní Dhubháin et al., 2009) or very specific features that have been scored on a certain place such as the cleanness of water (Huppert et al., 2003) or an estimation of the species richness (Qiu et al., 2013). Finally, housing prices and psychological well-being are used as indicators (Bowman et al., 2012; Daniel et al., 2012).
Indicators for aesthetic landscapes are often scored by using a ‘Likert scale’ (33 studies), a score table with the lowest value (being 0 or a negative value) for dislike or no appreciation at all, and the highest value for totally like or total appreciation. Other score techniques that are often used are prioritising (11 studies), ranking (5 studies), willingness to pay (5 studies) and visitor counts (1 study).
Policies
Only three policies were mentioned: the Act on Urban Parks and Green Spaces, the Framework Act on Forestry (Ja-Choon et al., 2013) and the red grouse hunting policy (Black et al., 2010).
Interactions with other ecosystem services
The ecosystem services that were linked with aesthetic landscapes were food production (cultivated crops, 8 studies), timber production (8 studies), freshwater fishing (1 study), sea fishing (2 studies) and red grouse hunting (1 study).
The link between aesthetic landscapes and food production is positive in one study, negative in two studies, both positive and negative in three studies, and unclear in two studies. The three studies that are both positively and negatively correlated and one of the studies that is negatively correlated are directly referring to the conclusion that certain user groups appreciate or dislike agricultural landscapes (see above).
Timber production is positively correlated to aesthetic value only once; in another seven articles the correlation is unclear. In Shillelagh (Ireland), where a forestry based industry generated employment, the forest and its related industry was considered to be part of both the local history and the traditional landscape and therefore appreciated a lot (Ní Dhubháin et al., 2009). In most other articles linked to timber production, the appreciation for different alternative management types is compared, so timber production as such is not questioned.
Freshwater fishing is appreciated by some groups, but the general opinion about it is unclear (Larson et al., 2013). In the case of sea fishing, the importance has been studied compared to other sectors: sea fishing and recreation have similar economic impacts in Wales (Ruiz-Frau et al., 2013), whereas it is declining in importance relative to tourism, recreation and retirement industries in the Pacific Northwest (Huppert et al., 2003). Activities of red grouse shooting do not have any influence on visitor’s valuation of the landscape (Black et al., 2010).
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 214
Overall observations on the review
In the majority of the articles, quantitative methods were used (30 studies), followed by qualitative methods (16 studies) and both quantitative and qualitative methods (7 studies). Evidence was always based on multiple observations (54 studies) and mostly on direct evidence (48 studies compared to 3 for indirect evidence), though one of them was actually based on expert judgement. Most studies were based on empirical data (48 studies), and few on modelling (1 study) or both empirical data and modelling (2 studies).
Focussing on the soundness of the methodology, the most useful studies have a focus on aesthetics (not on recreation) and provide solid evidence by using methods that measure the focal features with no risk of misinterpretation. A commonly used method is a questionnaire with photographs. There are multiple good examples of this method (e.g. Van den Berg et al., 1998; Ribe, 2009), but a very informative one is Pinto-Correia et al. (2011) because this article clearly describes where to take care while selecting and manipulating photos by showing their own learning process. This article moreover distinguishes between different user groups, which is important as they often show different preferences. Liekens et al. (2013) is an inspiring example of a method based on willingness to pay (WTP), with also an elaborated description of related methods. Bowman et al. (2012) is an original approach in which people are asked to bid for houses with different environmental condition. All three articles exactly measure what they aim to know, as the options of the choice experiments only differ in the parameter they want to measure. In this respect, methods based on visitor’s rates, visitor-employed photography (VEP) and some questionnaires with photographs are less accurate, as it is often unclear why people choose a certain option.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 215
Annex 6: The OpenNESS database tool
A new database tool was developed to store and analyse the material from the OpenNESS literature review. The tool is a Windows application with a local MS Access database that was already filled with details of the earlier BESAFE study, containing 630 papers. Each reviewer filled in their own copy of the database with details from the 60 selected papers for the 13 chosen ecosystem services.
The interface allows users to fill in details of each paper using the following tabs: Reference, Interaction with other ecosystem services, Spatial scale and locations, Temporal scale, Ecosystem types & conditions, Ecosystem Service Provider Classes, Attribute trait classes, Abiotic factors, Human input and management, Indicators, Thresholds, Policies, Evidence, and Comments. Screenshots for each of these tabs are shown below. Most of the entry fields were enhanced with a comment field to gather more information when applicable. Spatial data can be entered using Bing Maps and are stored as a bounding box combined with a scale: Local, (Sub)National, (Sub)Continental, and Global.
During the reviewing phase all reviewers sent in their database several times for checks and creating summaries for progress purposes. For this, the tool was enhanced to combine all the completed reviews and extract data into Excel sheets, and to generate maps and network diagrams.
Technical details
The tool is a .NET 3.5 32-bit !WinForms application developed in C# with Visual Studio 2012. The components are NHibernate (OR/M) and !StructureMap (Dependency Injection). The database was generated using NHibernate and filled with lookup values from an Excel-sheet. This mechanism proved to be very flexible during the design phase, as fields and values could be easily changed. The user interface is generated at run-time and each entry field (textbox, checkbox or dropdown) is linked to a field in the database. Maps are generated using Google Maps and an overlay of spatial locations. Network diagrams showing the interactions between ecosystem services are created with a custom version of the Diagram.NET library.
Screenshots
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 216
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 217
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 218
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 219
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 220
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 221
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 222
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 223
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 224
Domain Model
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 225
References
Acar, C. and Sakıcı, C. (2008), ‘Assessing landscape perception of urban rocky habitats’, Building and Environment, 43 (6), pp. 1153–1170. Acharya, G. P., McDonald, M. A., Tripathi, B. P., Gardner, R. M. and Mawdesley, K. J. (2007), ‘Nutrient losses from rain‐fed bench terraced cultivation systems in high rainfall areas of the mid‐hills of Nepal’, Land Degradation and Development, 18(5), pp. 486-99. Acreman, M.C., Fisher, J., Stratford, C.J., Mould, D.J. and Mountford, J.O. (2007), ‘Hydrological science and wetland restoration: some case studies from Europe’, Hydrological and Earth System Science, 11(1), pp. 158-69. Acreman, M.C., Harding, R.J., Lloyd, C., McNamara, N.P., Mountford, J.O., Mould, D.J., Purse, B.V., Heard, M.S., Stratford, C.J. and Dury, J. (2011), ‘Trade-off in ecosystem services of the Somerset Levels and moors wetlands’, Hydrological Sciences Journal, 56 (8), 1543. Ahmad, S.S., Reshi, Z.A., Shah, M.A., Rashid, I., Ara, R. and Andrabi, S.M.A. (2014), ‘Phytoremediation Potential of Phragmites australis in Hokersar Wetland - A Ramsar Site of Kashmir Himalaya, International Journal of Phytoremediation, 16(12), pp. 1183-1191. Albrecht, M., Duelli, P., Mueller, C., Kleijn, D. and Schmid, B. (2007), ‘The Swiss agri-environment scheme enhances pollinator diversity and plant reproductive success in nearby intensively managed farmland’, Journal of Applied Ecology 44 (4), pp. 813–822. Alonso, R., Vivanco, M. G., González-Fernández, I., Bermejo, V., Palomino, I., Garrido, J. L., ... and Artíñano, B. (2011), ‘Modelling the influence of peri-urban trees in the air quality of Madrid region (Spain)’, Environmental pollution, 159(8), pp. 2138-47. Amorim, J. H., Rodrigues, V., Tavares, R., Valente, J. and Borrego, C. (2013), ‘CFD modelling of the aerodynamic effect of trees on urban air pollution dispersion’, Science of the Total Environment, 461, pp. 541-551. Amoroso M. and Turnblom E. (2006), ‘Comparing productivity of pure and mixed Douglas-fir and western hemlock plantations in the Pacific Northwest’, Can. J. For. Res. 36, pp. 1484–1496. Anderson, Annette, Tim Carnus, Alvin J. Helden, Helen Sheridan and Gordon Purvis (2013), 'The influence of conservation field margins in intensively managed grazing land on communities of five arthropod trophic groups', Insect Conservation and Diversity 6, pp. 201–11. Anderson B.G., I.D. Rutherfurd and A.W. Western (2006), ‘An analysis of the influence of riparian vegetation on the propagation of flood waves’, Environmental Modelling and Software, 21, pp. 1290-6. Apollonio, M., Baldaccini, N.E., Blasi, C., et al. (2000), ‘Natura 2000 Network in Italy: Guidelines and Management Trends’, In: QUADERNI DI CONSERVAZIONE DELLA NATURA, 18; 213-224. Italy: Ministero dell'Ambiente e della tutela del territorio. Arlinghaus, R., and T. Mehner. (2003), ‘Socio-economic characterization of specialized common carp (Cyprinus carpio L.) anglers in Germany, and implications for inland fisheries management and eutrophication control’, Fisheries Research 61 pp. 19–33.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 226
Ashworth, L., Quesada, M., Casas, A., Aguilar, R. and Oyama, K. (2009), ‘Pollinator-dependent food production in Mexico. Biological Conservation, 142(5), pp.1050-1057. Asuming-Brempong, S., Gantner, S., Adiku, S.G.K., Archer, G., Edusei, V. and Tiedje, J.M. (2008), 'Changes in the biodiversity of microbial populations in tropical soils under different fallow treatments', Soil Biology and Biochemistry, 40, 11, pp. 2811-2818. Badano E.I. and C. H. Vergara (2011), ‘Potential negative effects of exotic honey bees on the diversity of native pollinators and yield of highland coffee plantations’, Agricultural and Forest Entomology 13 (4), pp. 365-72. Bai, Y., Zhuang, C., Ouyang, Z., Zheng, H. and Jiang, B. (2011), ‘Spatial characteristics between biodiversity and ecosystem services in a human-dominated watershed’, Ecological Complexity, 8(2), pp. 177-83. Balvanera, P., Kremen, C. and Matrinez-Ramos, M. (2005), ‘Applying community structure analysis to ecosystem function: examples from pollination and carbon storage’, Ecological Applications, 15(1), pp. 360- 75. Balvanera, P., Pfisterer, A.B., Buchmann, N., He, J.S., Nakashizuka, T., Raffaelli, D. and Schmid, B. (2006), ‘Quantifying the evidence for biodiversity effects on ecosystem functioning and services’, Ecol. Lett., 9, pp. 1146-56. Balvanera, P., Siddique, I., Dee, L., Paquette, A., Isbell, F., González, A., Byrnes, J., O’Connor, M.I., Hungate, B.A. and Griffin, J.N. (2014), ‘Linking biodiversity and ecosystem services: current uncertainties and the necessary next steps’, BioSci., 64, pp. 49-57. Bangash, R. F., Passuello, A., Sanchez-Canales, M., Terrado, M., López, A., Elorza, F. J., ... & Schuhmacher, M. (2013), ‘Ecosystem services in Mediterranean river basin: Climate change impact on water provisioning and erosion control’, Science of The Total Environment, 458–460, pp. 246–255. Baral, H., Keenan, R. J., Fox, J. C., Stork, N. E., & Kasel, S. (2013), ‘Spatial assessment of ecosystem goods and services in complex production landscapes: A case study from south-eastern Australia’, Ecological Complexity, 13, pp. 35–45 Baró, F., Chaparro, L., Gómez-Baggethun, E., Langemeyer, J., Nowak, D. J. and Terradas, J. (2014), ‘Contribution of ecosystem services to air quality and climate change mitigation policies: The case of urban forests in Barcelona, Spain’, Ambio, 43(4), pp. 466-479. Bastian, O. (2013), ‘The role of biodiversity in supporting ecosystem services in Natura 2000 sites’, Ecol. Indic., 24, pp. 13-22. Batary, P., Baldi, A., Sarospataki, M., Kohler, F., Verhulst, J., Knop, E., ... and Kleijn, D. (2010), ‘Effect of conservation management on bees and insect-pollinated grassland plant communities in three European countries’, Agriculture, ecosystems and environment, 136(1), pp. 35-39. Bathurst, J. et al. (2011), ‘Forest impact on floods due to extreme rainfall and snowmelt in four Latin American environments 1: field data analysis', Journal of Hydrology, 400, pp. 281-91. Bauhus J., van Winden A. and Nicotra A. (2004), ‘Aboveground interactions and productivity in mixed- species plantations of Acacia mearnsii and Eucalyptus globulus’, Can. J. For. Res. 34, pp. 686–694. Beckett, K. P., Freer‐Smith, P. H. and Taylor, G. (2000), ‘Particulate pollution capture by urban trees: effect of species and windspeed’, Global change biology, 6(8), pp. 995-1003.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 227
Beier, C., Emmett, B. A., Tietema, A., Schmidt, I. K., Peñuelas, J., Láng, E. K., ... and Spano, D. (2009), ‘Carbon and nitrogen balances for six shrublands across Europe’, Global Biogeochemical Cycles, 23(4). Benjamin, Michael T. and Winer, Arthur M. (1998), ‘Estimating the ozone-forming potential of urban trees and shrubs’, Atmospheric Environment, 32 (1), pp. 53–68. Bianchi, F.J.J.A., Booij, C.J.H. and Tscharntke, T. (2006), ‘Sustainable pest regulation in agricultural landscapes: A review on landscape composition, biodiversity and natural pest control’, Proceedings of the Royal Society B, 273, pp. 1715-27. Bianchi, F.J.J.A., Goedhart, P.W. and Baveco, J.M. (2008), ‘Enhanced pest control in cabbage crops near forst in the Netherlands’, Landscape Ecology, 23, pp. 595-602. Bianchi, F.J.J.A., Schellhorn, N.A., Buckley, Y.M. and Possingham, H.P. (2010), ‘Spatial variability in ecosystem services: Simple rules for predator-mediated pest suppression’, Ecological applications, 20, pp. 2322-33. Biesmeijer, J.C., Roberts, S.P.M., Reemer, M., Ohlemueller, R., Edwards, M., Peeters, T., ... and Kunin, W.E. (2006), ‘Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands’, Science, 313(5785), pp. 351-4. Biggs, R., et al. (2009), ‘Turning back from the brink: Detecting an impending regime shift in time to avert it’, Proc. Natl. Acad. Sci. USA, 106, pp. 826–31. Bini, C., Gemignani, S. and Zilocchi, L. (2006), ‘Effect of different land use on soil erosion in the pre-alpine fringe (North-East Italy), Ion budget and sediment yield’, Science of the total environment, 369(1), pp. 433- 40. Bisseleua, H. B. D., Fotio, D., Missoup, A. D. and Vidal, S. (2013), ‘Shade tree diversity, cocoa pest damage, yield compensating inputs and farmers' net returns in West Africa’, PloS one, 8(3), e56115. Björklund, J., Limburg, K. E. and Rydberg, T. (1999), ‘Impact of production intensity on the ability of the agricultural landscape to generate ecosystem services: an example from Sweden’, Ecological economics, 29(2), pp. 269-291. Black, J., E.J. Milner-Gulland, N. Sotherton and S. Mourato (2010), ‘Valuing complex environmental goods: landscape and biodiversity in the North Pennines’, Environmental Conservation, 37 (02), pp. 136-146. Blanche, R. and Cunningham, S.A. (2005), ‘Rain forest provides pollinating beetles for Atemoya crops’, Journal of Economic Entomology 98(4), pp. 1193-201. Blauuw, B. and Isaacs, R. (2012), 'Larger wildflower plantings increase natural enemy density, diversity, and biological control of sentinel prey, without increasing herbivore density', Ecological Entomology, 37, pp. 386–94. Blumenfeld, S., Lu, C., Christophersen, T. and Coates, D. (2009), ‘Water, Wetlands and Forests. A Review of Ecological, Economic and Policy Linkages’, CBD Technical Series No. 47. Secretariat of the Convention on Biological Diversity and Secretariat of the Ramsar Convention on Wetlands, Montreal and Gland. Boccaccio, L. and Petacchi, R. (2009), ‘Landscape effects on the complex of Bactrocera olear parasitoids and implications for conservation biological control’, BioControl, 54, p. 616.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 228
Bochet, E., Rubio, J. L. and Poesen, J. (1998), ‘Relative efficiency of three representative matorral species in reducing water erosion at the microscale in a semi-arid climate (Valencia, Spain)’, Geomorphology, 23(2), pp. 139-50. Bodin, O., M. Tengo, A. Norman, J. Lundberg, and T. Elmqvist (2006), 'The value of small size: Loss of forest patches and ecological thresholds in southern Madagascar', Ecological Applications, 16, pp. 440-51. Bommarco, R., Lundin, O., Smith, H. G. and Rundlöf, M. (2011), ‘Drastic historic shifts in bumble-bee community composition in Sweden’, Proceedings of the Royal Society B: Biological Sciences, 279(1727), pp. 309-15. Börner, J., Mendoza, A. and Vosti, S. A. (2007), ‘Ecosystem services, agriculture, and rural poverty in the Eastern Brazilian Amazon: Interrelationships and policy prescriptions’, Ecological Economics, 64(2), pp. 356- 73. Borrelli, P. and Schütt, B. (2014), ‘Assessment of soil erosion sensitivity and post-timber-harvesting erosion response in a mountain environment of Central Italy’, Geomorphology, 204, pp. 412–424. Bosch, J. and Hewlett, J. (1982), ‘A review of catchment experiments to determine the effect of vegetation changes on water yield and evapotranspiration’, Journal of Hydrology, 55(1–4), pp. 3–23. Boukal, D. S., Jankovský, M., Kubečka, J. and Heino, M. (2012), ‘Stock–catch analysis of carp recreational fisheries in Czech reservoirs: Insights into fish survival, water body productivity and impact of extreme events’, Fisheries Research, 119, pp. 23-32. Bowman, T., J.C. Tyndall, J. Thompson, J. Kliebenstein and J.P. Colletti (2012), ‘Multiple approaches to valuation of conservation design and low-impact development features in residential subdivisions.’ Journal of Environmental Management, 104, pp. 101-113. Braat, Leon C. and Rudolf de Groot (2012), ‘The ecosystem services agenda: bridging the worlds of natural science and economics, conservation and development, and public and private policy’, Ecosystem Services, 1(1), pp. 4-15. Brack, C. L. (2002), ‘Pollution mitigation and carbon sequestration by an urban forest’, Environmental pollution, 116, S195-S200. Bradshaw, C.J.A, Sodhi, N.S., Peh, K.S.-H. and Brook, B.W. (2007), ‘Global evidence that deforestation amplifies flood risk and severity in the developing world’, Global Change Biology, 13, pp. 2379-2395. Brauman, K. A., Freyberg, D. L. and Daily, G.C. (2010), ‘Forest structure influences on rainfall partitioning and cloud interception: A comparison of native forest sites in Kona, Hawaii’, Agricultural and Forest Meteorology 150, pp. 265–275. Brauman, K.A., Freyberg, D.L. and Daily, G.C. (2012), ‘Potential evapotranspiration from forest and pasture in the tropics: A case study in Kona, Hawaii’, Journal of Hydrology, s440–441, pp. 52–61. Bren, L. and Hopmans, P. (2007), ‘Paired catchments observations on the water yield of mature eucalypt and immature radiata pine plantations in Victoria, Australia’, J. Hydrol., 336, pp. 416-429. Bristow M., Vanclay J., Brooks L. and Hunt M. (2006), ‘Growth and species interactions of Eucalyptus pellita in a mixed and monoculture plantation in the humid tropics of north Queensland’, Forest Ecology and Management, 233(2-3), pp. 285-294.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 229
Brown, T.C. and T.C. Daniel (1986), ‘Predicting scenic beauty of timber stands’, Forest Sci. 32, pp. 471-487. Brown, J., Scholtz, C. H., Janeau, J. L., Grellier, S. and Podwojewski, P. (2010),’ Dung beetles (Coleoptera: Scarabaeidae) can improve soil hydrological properties’, Applied Soil Ecology, 46 (1), pp. 9–16. Bruijnzeel, L.A. (2004), ‘Hydrological functions of tropical forests: not seeing the soil for the trees?’, Agriculture, Ecosystems and Environment, 104, pp. 185-228. Buijs, A.E., B.H.M. Elands and F. Langers (2009), ‘No wilderness for immigrants: Cultural differences in images of nature and landscape preferences’, Landscape and Urban Planning, 91 (3), pp. 113–123. Bullock, A. and Acreman, M. (2003), ‘The role of wetlands in the hydrological cycle’, Hydrology and Earth System Sciences, 7(3), pp. 358-389. Bullock, J., Aronson, J., Newton, A., Pywell, R. and Rey-Benayas, J. (2011), ‘Restoration of ecosystem services and biodiversity: conflicts and opportunities’, Trends in Ecol. and Evol., 26(10), pp. 541-549. Bunker, D.E., DeClerck, F., Bradford, J.C., Colwell, R.K., Perfecto, I., Phillips, O.L., Sankaran, M. and Naeem, S. (2005), ‘Species loss and aboveground carbon storage in a tropical forest’, Science, 310, pp. 1029-1031. Butler, C. D. and Oluoch-Kosura, W. (2006), ‘Linking future ecosystem services and future human well- being’, Ecol. and Society, 11(1), p. 30. Butler, J.R.A., A. Radford, G. Riddington and R. Laughton (2009), ‘Evaluating an ecosystem service provided by Atlantic salmon, sea trout and other fish species in the river Spey, Scotland: the economic impact of recreational rod fisheries’, Fisheries Research 96, pp. 259-66. Buytaert, W., Iniguez, V. and De Bievre, B. (2007), ‘The effects of afforestation and cultivation on water yield in the Andean paramo’, For. Ecol. and Manag., 251(1-2), pp. 22-30. Cabaraban, M. T. I., Kroll, C. N., Hirabayashi, S. and Nowak, D. J. (2013), ‘Modeling of air pollutant removal by dry deposition to urban trees using a WRF/CMAQ/i-Tree Eco coupled system’, Environmental Pollution, 176, pp. 123-133. Cadotte, M.W. (2013), 'Experimental evidence that evolutionarily diverse assemblages result in higher productivity', PNAS 110(22), pp. 8996–9000. Cahill, K.N., Kucharik, C.J. and Foley, J.A. (2009), ‘Prairie restoration and carbon sequestration: difficulties quantifying C sources and sinks using a biometric approach’, Ecological Applications, 19(8), pp. 2185-2201. Cardinale, B.J., Srivastava, D.S., Duffy, J.E., Wright, J.P., Dowing, A.L., Sankaran, M. and Jouseau, C. (2006), ‘Effects of biodiversity on the functioning of trophic groups and ecosystems’, Nature, 443, pp. 989-992. Cardinale, B. J., Matulich, K. L., Hooper, D. U., Byrnes, J. E., Duffy, E., Gamfeldt, L., ... and Gonzalez, A. (2011), ‘The functional role of producer diversity in ecosystems’, American journal of botany, 98(3), pp. 572-592. Cardinale, B.J., Duffy, J.E., Gonzalez, A., Hooper, D.U., Perrings, C., Venail, P., ... and Naeem, S. (2012), ‘Biodiversity loss and its impact on humanity’, Nature, 486, pp. 59-67. Carlyle-Moses, D.E. and Price, A.G. (2007), ‘Modelling canopy interception loss from a Madrean pine-oak stand, Northeastern Mexico’, Hydrol. Process., 21, pp. 2572-2580.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 230
Carpenter, S.R., Mooney, H.A., Agard, J., Capistrano, D., DeFries, R.S., Diaz, S., ... and Whyte, A. (2009), ‘Science for managing ecosystem services: beyond the Millennium Ecosystem Assessment’, PNAS, 106, pp. 1305-1312. Carroll, C. and A. Tucker (2000), ‘Effects of pasture cover on soil erosion and water quality on central Queensland coal mine rehabilitation’, Tropical Grasslands 34, pp. 254-62. Carvalheiro, L. G., Seymour, C. L., Veldtman, R. and Nicolson, S. W. (2010), ‘Pollination services decline with distance from natural habitat even in biodiversity-rich areas’, Journal of Applied Ecology, 47(4), pp. 810– 820. Castañé, C., Arnó, J., Gabarra, R. and Alomar, O. (2011), ‘Plant damage to vegetable crops by zoophytophagous mirid predators’, Biological Control, 59, pp. 22-29. Casula, Paolo and Mauro Nannini (2013), 'Evaluating the Structure of Enemy Biodiversity Effects on Prey Informs Pest Management', ISRN Ecology, Article ID 619393. Cavaleri, M.A. and Sack, L. (2010), ‘Comparative water use of native and invasive plants at multiple scales: a global meta-analysis’, Ecology 91(9), pp.2705-15. Cavanaugh, Kyle C., J. Stephen Gosnell, Samantha L. Davis, Jorge Ahumada, Patrick Boundja, David B. Clark, ... and Sandy Andelman (2014), 'Carbon storage in tropical forests correlates with taxonomic diversity and functional dominance on a global scale', Global Ecology and Biogeography, 23, pp. 563–573. Chan K.M.A., Shaw M.R., Cameron D.R., Underwood E.C. and Daily G.C. (2006), ‘Conservation planning for ecosystem services’, PLoS Biol., 4, pp. 2138-2152. Chaplin-Kramer, R. and Kremen, C. (2012), ‘Pest control experiments show benefits of complexity at landscape and local scales’, Ecological applications, 22, pp. 1936-1948. Chen, D.X., Li, Y.D., Liu, H.P., Xu, H., Xiao, W.F., Luo, T.S., Zhou, Z. and Lin, M.X. (2010), ‘Biomass and carbon dynamics of a tropical mountain rain forest in China’, Science China Life Sciences, 53(7), pp. 798-810. Chen, X. (2006), ‘Tree diversity, carbon storage and soil nutrient in an old-growth forest at Changbai Mountain, Northeast China’, Communications in Soil Science and Plant Analysis, 37, pp. 363-375. Chen, X. and Zong, Y. (1999), ‘Major impacts of sea-level rise on agriculture in the Yangtze delta area around Shanghai’, Applied Geography, 19(1), pp. 69-84. Cheng, J.D., Lin, L.L. and Lu, H.S. (2002), ‘Influences of forests on water flows from headwater watersheds in Taiwan’, Forest Ecology and Management, 165, pp. 11-28. Christen, Benjamin and Tommy Dalgaard (2013), ‘Buffers for biomass production in temperate European agriculture: A review and synthesis on function, ecosystem services and implementation’, Biomass and Bioenergy 55, pp. 53-67. Cisneros-Montemayor, A.M. and Sumaila, U.R. (2010), ‘A global estimate of benefits from ecosystem-based marine recreation: potential impacts and implications for management’, Journal of Bioeconomics 12 (3), pp. 245-268. Clapcott, J.E., Collier, K.J., Death, R.G., Goodwin, E.O., Harding, J.S., Kelly, D., Leathwick, J.R and Young, R.G. (2012), ‘Quantifying relationships between land-use gradients and structural and functional indicators of stream ecological integrity’, Freshwater Biology) 57, pp. 74–90.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 231
Clark, C. (1987), ‘Deforestation and floods’, Environmental Conservation, 14, pp. 67-69. Cocks, M.L., T. Dold and S. Vetter (2012), ‘‘God is my forest’ - Xhosa cultural values provide untapped opportunities for conservation’, South African Journal of Science 108 (5-6), pp. 52-59. Colloff, M.J., Lindsay, E.A. and Cook, D.C. (2013), ‘Natural pest control in citrus as an ecosystem service: Integrating ecology, economics and management at the farm scale’, Biological control, 67, pp. 170-177. Con, T. V., Thang, N. T., Ha, D. T. T., Khiem, C. C., Quy, T. H., Lam, V. T., ... and Sato, T. (2013), ‘Relationship between aboveground biomass and measures of structure and species diversity in tropical forests of Vietnam’, Forest Ecology and Management, 310, pp. 213-218. Conard, S.G. and Ivanova, G.A. (1997), ‘Wildfire in Russian boreal forests - potential impacts of fire regime characteristics on emissions and global carbon balance estimates’, Environmental Pollution, 98(3), pp. 305- 313. Cooke, S. J. and Suski, C. D. (2005), ‘Do we need species-specific guidelines for catch-and-release recreational angling to effectively conserve diverse fishery resources?’, Biodiversity and Conservation, 14(5), pp. 1195-1209. Cornish, P.M. and Vertessy, R.A. (2001), ‘Forest age-induced changes in evapotranspiration and water yield in a eucalypt forest’, Journal of Hydrology, 242 (1–2), pp. 43–63. Costanza, R., d’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., ... and van den Belt, M. (1997), ‘The value of the world’s ecosystem services and natural capital’, Nature, 387, pp. 253-260. Craig, J.F., Halls, A.S., Barr, J.J.F. and Beand, C.W. (2004), 'The Bangladesh floodplain fisheries', Fisheries Research, 66(2–3), pp. 271–286. Crowder, D.W., Northfield, T.D., Strand, M.R. and Snyder, W.E. (2010), ‘Organic agriculture promotes evenness of natural pest control’, Nature, 466, pp. 109-113. Czajkowski, M., A. Bartczak, M. Giergiczny, S. Navrud, T. Żylicz (2014), ‘Providing preference-based support for forest ecosystem service management’, Forest Policy and Economics 39, pp. 1–12. Daghela Bisseleua HB, Fotio D, Yede, Missoup AD, Vidal S (2013), 'Shade Tree Diversity, Cocoa Pest Damage, Yield Compensating Inputs and Farmers’ Net Returns in West Africa', PLoS ONE 8(3), e56115. Daily ,G.C. (ed) (1997), Nature’s services: societal dependence on natural ecosystems, Island Press, Washington DC. Daniel, T.C., A. Muhar, A. Arnberger, O. Aznar, J.W. Boyd, K.M.A. Chan, R. Costanza, T. Elmqvist, C.G. Flint, P.H. Gobster, A. Grêt-Regamey, R. Lave, et al. (2012), ‘Contributions of cultural services to the ecosystem services agenda’, PNAS 109(23), pp. 8812-19. De Aranzabal, I., M.F. Schmitz and F.D. Pineda (2009), ‘Integrating Landscape Analysis and Planning: A Multi-Scale Approach for Oriented Management of Tourist Recreation’, Environmental Management 44 (5), pp. 938-951. De Bello, F., Lavorel, S., Diaz, S., Harrington, R., Cornelissen, J., Bargett, R., ... and Harrison, P.A (2010), ‘Towards an assessment of multiple ecosystem processes and services via functional traits’, Biodivers. and Conserv., 19, pp. 2873-2893.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 232
De Groot, R.S., Alkemade, J.R.M. , Braat, L. , Hein, L.G. and Willemen, L.L.J.M. (2010), ‘Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision- making’, Ecol. Complex., 7, pp. 260-272. De Groot, R.S., Wilson, M.A. and Boumans, R.M.J (2002), ‘A typology for the classification, description and valuation of ecosystem functions, goods and services’, Ecol. Econ., 41, pp. 393-408. De Silva, S. S., Lin, Y. and Tang, G. (1992), ‘Possible yield-predictive models based on morphometric characteristics and stocking rates for three groups of Chinese reservoirs’, Fisheries research, 13(4), 369-380. De Silva, S.S. (2003), ‘Culture-based fisheries: an underutilised opportunity in aquaculture development’, Aquaculture 221 pp. 221–243. De Souza, A.L.T., Fonseca, D.G., Libório, R.A. and Tanaka, M.O. (2013), ‘Influence of riparian vegetation and forest structure on the water quality of rural low-order streams in SE Brazil’, Forest Ecology and Management 298, pp. 12–18. Detenbeck, N. E., Elonen, C. M., Taylor, D. L., Anderson, L. E., Jicha, T. M. and Batterman, S. L. (2004), ‘Region, landscape, and scale effects on Lake Superior tributary water quality’, JAWRA Journal of the American Water Resources Association, 40(3), pp. 705-720.. Díaz, S. and Cabido, M. (2001), ‘Vive la difference: plant functional diversity matters to ecosystem processes’, Trends Ecol. Evol., 16, pp. 646–655. Díaz, S., Fargione, J., Chapin, F.S. III and Tilman, D. (2006), ‘Biodiversity loss threatens human well-being’, PLoS Biol., 4(8), e277. doi:10.1371/journal.pbio.0040277. Díaz, S., Quétier, F; Cáceres, D., Trainor, S.F., Pérez-Harguindeguy, N., Bret-Harte, M.S., Finegan, B., Peña- Claros, M. and Poorter, L. (2011), ‘Linking functional diversity and social actor strategies in a framework for interdisciplinary analysis of nature's benefits to society’, PNAS, 108(3), pp. 895-902. Dierick, D. and Hölscher, D. (2009), ‘Species-specific tree water use characteristics in reforestation stands in the Philippines’, Agricultural and Forest Meteorology, 149 (8), pp. 1317–1326. Doherty, J.M., Miller, J.F., Prellwitz,S.G., Thompson,A.M., Loheide, S.P. and Zedler, J.B. (2014), ‘Hydrologic Regimes Revealed Bundles and Tradeoffs Among Six Wetland Services’, Ecosystems 17:, pp. 1026–1039. Donohue I., McGarrigle M.L. and Mills P. (2006), ‘Linking catchment characteristics and water chemistry with the ecological status of Irish rivers’, Water Research, 40, pp. 91–98. Donoso, Pablo J. and Lusk, Christopher H. (2007), ‘Differential effects of emergent Nothofagus dombeyi on growth andbasal area of canopy species in an old-growth temperate rainforest’, Journal of Vegetation Science, 18(5), pp. 675-684. Donovan, Rossa G., Stewart, Hope E., Owen, Susan M., MacKenzie, A. Robert and Hewitt, C. Nicholas (2005), ‘Development and Application of an Urban Tree Air Quality Score for Photochemical Pollution Episodes Using the Birmingham, United Kingdom, Area as a Case Study’, Environ. Sci. Technol., 39, pp. 6730- 6738. Dramstad, W.E., M. Sundli Tveit, W.J. Fjellstad and G.L.A. Fry (2006), ‘Relationships between visual landscape preferences and map-based indicators of landscape structure’, Landscape and Urban Planning 78 (4), pp. 465–474.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 233
Drapela, T., Frank, T., Heer, X., Moser, D. and Zaller, J.G. (2011), ‘Landscape structure affects activity density, body size and fecundity of Pardosa wolf spiders (Araneae: Lycosidae) in winter oilseed rape’, European Journal of Entomology, 106, pp. 609-614. Drapela, T., Moser, D. Zaller, J.G. and Frank, T. (2008), ‘Spider assemblages in winter oilseed rape affected by landscape and site factors’, Ecography, 31, pp. 254-262. Drinkwater, L.E., Letourneau, D.K., Worknes, F., van Bruggen, A.H.C. and Shennan, S. (1995), ‘Fundamental differences between conventional and organic tomato agro-ecosystems in California’, Ecological Applications, 5, pp. 1098-1112. Dugan, P. J., Barlow, C., Agostinho, A. A., Baran, E., Cada, G. F., Chen, D., …and Winemiller, K. O. (2010), 'Fish Migration, Dams, and Loss of Ecosystem Services in the Mekong Basin', AMBIO, 39, pp. 344–348. Duncan, M.J. (1995), ‘Hydrological impacts of converting pasture and gorse to pine plantation, and forest harvesting, Nelson, New Zealand’, Journal of Hydrology (NZ), 34(1), pp. 15-41. Durán Zuazo, V., Martínez, J. R. F. and Raya, A. M. (2004), ‘Impact of vegetative cover on runoff and soil erosion at hillslope scale in Lanjaron, Spain’, The Environmentalist 24 (1), pp. 39–48. Duyzer, J. H., Weststrate, J. H. and Walton, S. (1995). ‘Exchange of ozone and nitrogen oxides between the atmosphere and coniferous forest’, Water, Air and Soil Pollution, 85, pp. 2065–2070. Egoh, B., Reyers, B., Rouget, M., Richardson, D.M., Le Maitre, D.C. and van Jaarsveld, A.S. (2008), ‘Mapping ecosystem services for planning and management’, Agriculture, Ecosystems & Environment, 127 (1–2), pp. 135–140. Egoh, B., Reyers, B., Rouget, M., Bode, M. and Richardson, D.M. (2009), ‘Spatial congruence between biodiversity and ecosystem services in South Africa’, Biological Conservation, 142 (3), pp. 553–562. Egoh, B., Drakou, E.G., Dunbar, M.B., Maes, J., Willemen, L. (2012), ‘Indicators for mapping ecosystem services: a review’, JRC Scientific and Policy Reports, Publications Office of the European Union, 107pp, ISBN: 978-92-79-25821-3. Erskine, P. D., Lamb, D. and Bristow, M. (2006), ‘Tree species diversity and ecosystem function: can tropical multi-species plantations generate greater productivity?’, Forest Ecology and Management, 233(2), pp. 205-210. Erskine, W.D. and Webb, A.A. (2003), ‘Desnagging to resnagging: new directions in river rehabilitation in southeastern Australia’, River Research and Applications, 19, pp. 233-249. Escobedo, Francisco J. and Nowak, David J. (2009), ‘Spatial heterogeneity and air pollution removal by an urban forest’, Landscape and Urban Planning, 90, pp. 102–110. Everard M and Kataria G. (2011), ‘Recreational angling markets to advance the conservation of a reach of the Western Ramganga River’, Aquatic Conservation: Marine and Freshwater Ecosystems 21 (1), pp. 101– 108. Ewel, K.C., Twilley, R. and Ong, J.E. (1998), ‘Different kinds of mangrove forests provide different goods and services’, Global Ecology and Biogeography Letters, 7, pp. 83-94.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 234
Fagan, K.C., Pywell, R.F., Bullock, J.M. and Marrs, R.H. (2008), ‘Do restored calcareous grasslands on former arable fields resemble ancient targets? The effect of time, methods and environment on outcomes’, J. Appl. Ecol., 45, pp. 1293-1303. Fahey, B. and Jackson, R. (1997), ‘Hydrological impacts of converting native forests and grasslands to pine plantations, South Island, New Zealand’, Agricultural and Forest Meteorology, 84, pp. 69-82. Fahey, B.D. and Watson, A.J. (1991), ‘Hydrological impacts of converting tussock grassland to pine plantation, Otago, New Zealand’, Journal of Hydrology (NZ), 30(1), pp.1-15. Farley, K.A., Jobbágy, E.G. and Jackson, R.B. (2005), ‘Effects of afforestation on water yield: a global synthesis with implications for policy’, Global Change Biology, 11, pp. 1565-1576. Farwig, N., Bailey, D., Bochud, E., Herrmann, J. D., Kindler, E., Reusser, N., Schueepp, C. and Schmidt- Entling, M. H. (2009), ‘Isolation from forest reduces pollination, seed predation and insect scavenging in Swiss farmland’, Landscape Ecology 24 (7), pp 919-927. Ferrario, Filippo, Michael W. Beck, Curt D. Storlazzi, Fiorenza Micheli, Christine C. Shepard and Laura Airoldi (2014), ‘The effectiveness of coral reefs for coastal hazard risk reduction and adaptation’, Nature Communications, 5, pp. 3794. Ferris, R., Ellis, R. H., Wheeler, T. R. and Hadley, P. (1998), ‘Effect of high temperature stress at anthesis on grain yield and biomass of field-grown crops of wheat’, Annals of Botany, 82(5), pp. 631-639. Ficke A. D., Myrick, C. A. and Hansen, L. J. (2007), 'Potential impacts of global climate change on freshwater fisheries’, Reviews in Fish Biology and Fisheries, 17(4), pp. 581–613. Field, J. P., Breshears, D. D., Whicker, J. J., Zou, C. B. (2011), ‘Interactive effects of grazing and burning on wind- and water-driven sediment fluxes: Rangeland management implications’, Ecological Applications 21, pp. 22–32. Fischer, J., A. Zerger, P. Gibbons, J. Stott, and B. S. Law (2010), 'Tree decline and the future of Australian farmland biodiversity’, Proceedings of the National Academy of Sciences of the United States of America 107, pp. 19597-19602. Fisher, B., Turner, R.K. and Morling, P. (2009), ‘Defining and classifying ecosystem services for decision- making’, Ecol. Econ., 68, pp. 643-653. Fontaine, C., Dajoz, I., Meriguet, J. and Loreau, M. (2006), ‘Functional diversity of plant-pollinator interaction webs enhances the persistence of plant communities’, Plos Biol., 4(1), pp. 129-135. Foote, A.L., Pandey, S. and Krogman, N.T. (1996), ‘Processes of wetland loss in India’, Environmental Conservation, 23(1), pp. 45-54. Ford, Hilary, Angus Garbutt, Davey L. Jones, Laurence Jones (2012), ‘Impacts of grazing abandonment on ecosystem service provision: Coastal grassland as a model system', Agriculture, Ecosystems and Environment, 162, pp. 108– 115. Freer-Smith, P. H., Holloway, Sophy and Goodman, A. (1997), ‘The uptake of particulates by an urban woodland: Site description and particulate composition’, Environ Pollut. 95(1), pp. 27-35. Freiman, M. T., Hirshel, N. and Broday, D. M. (2006), ‘Urban-scale variability of ambient particulate matter attributes’, Atmospheric Environment, 40(29), pp. 5670-5684.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 235
Frye, W. W., Ebelhar, S. A., Murdock, L. W. and Blevins, R. L. (1982), ‘Soil erosion effects on properties and productivity of two Kentucky soils’, Soil Science Society of America Journal, 46(5), pp. 1051-1055. Gao, Y., B. L. Zhong, H. Yue, B. Wu, and S. X. Cao (2011), 'A degradation threshold for irreversible loss of soil productivity: a long-term case study in China', Journal of Applied Ecology 48, pp. 1145-1154. García-Llorente, M., B. Martín-López, I. Iniesta-Arandia et al. (2012), ‘The role of multi-functionality in social preferences toward semi-arid rural landscapes: An ecosystem service approach’, Environmental Science & Policy 19–20, pp. 136–146. Garibaldi, L. A., Steffan‐Dewenter, I., Kremen, C., Morales, J. M., Bommarco, R., Cunningham, S. A., ... and Klein, A. M. (2011), ‘Stability of pollination services decreases with isolation from natural areas despite honey bee visits’, Ecology Letters, 14(10), pp. 1062-1072. Garmendia, E., Mariel, P., Tamayo, I., Aizpuru, I. and Zabaleta, A. (2012), ‘Assessing the effect of alternative land uses in the provision of water resources: evidence and policy implications from southern Europe’, Land Use Policy, 29(4), pp. 761-770. Gaston, K.J. (2000), ‘Global patterns in biodiversity’, Nature, 405, pp. 220-227. Geerts, S. and Pauw, A. (2011), ‘Farming with native bees (Apis mellifera subsp. capensis Esch.) has varied effects on nectar-feeding bird communities in South African fynbos vegetation’, Population ecology, 53(2), pp. 333-339. Geiger, F., Bengtsson, J. et al. (2010), ‘Persistent negative effects of pesticides on biodiversity and biological control potential on European farmland’, Basic and Applied Ecology, 11, pp. 97-105. Gergel S.E., Turner M.G., Miller J.R., Melack J.M. & Stanley E.H. (2002), ‘Landscape indicators of human impacts to riverine systems’, Aquatic Sciences, 64, pp. 118–128. Gholzom, E.H. and Gholami, V. (2012), ‘A Comparison between Natural Forests and Reforested Lands in Terms of Runoff Generation Potential and Hydrologic Response (Case Study: Kasilian Watershed)', Soil and Water Res., 7(4), pp. 166–173. Glenday, J. (2008), ‘Carbon storage and emissions offset potential in an African dry forest, the Arabuko- Sokoke Forest, Kenya’, Environment, Monitoring and Assessment, 142, pp. 85-95. Gobster, P.H. (2002), ‘Managing urban parks for a racially and ethnically diverse clientele’, Leisure Sciences, 24, pp. 143–159. Gómez-Limón, J. and J.V. de Lucío Fernández (1999), ‘Changes in use and landscape preferences on the agricultural-livestock landscapes of the central Iberian Peninsula (Madrid, Spain)’, Landscape and Urban Planning, 44 (4), pp. 165–175. Gomez-Peralta, D., Oberbauer, S.F., McClain, M.E. and Philippi, T.E. (2008), ‘Rainfall and cloud-water interception in tropical montane forests in the eastern Andes of Central Peru’, Forest Ecology and Management, 255 (3–4), pp. 1315–1325. Gonzalez, Patrick, Benjamín Kroll, Carlos R. Vargas (2014), 'Tropical rainforest biodiversity and aboveground carbon changes and uncertainties in the Selva Central, Peru', Forest Ecology and Management, 312, pp. 78– 91.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 236
González-Fernández, J.J., de la Peña, F. et al. (2009), ‘Alternative food improves the combined effect of an omnivore and a predator on biological pest control: A case study in avocado orchards’, Bulletin of Entomological Research, 99, pp. 433-444. Green, K. and Alila, Y. (2012), ‘A paradigm shift in understanding and quantifying the effects of forest harvesting on floods in snow environments’, Water Resour. Res., 48, W10503. Greenleaf, S. S. and Kremen, C. (2006), ‘Wild bee species increase tomato production and respond differently to surrounding land use in Northern California’, Biological Conservation, 133(1), pp. 81-87. Greiner, M. and Hershner, C. (1998), ‘Analysis of wetland total phosphorus retention and watershed structure’, Wetlands, 18(1), pp. 142-149. Groffman, P.M. et al. (2006), ‘Ecological thresholds: the key to successful environmental management or an important concept with no practical application?’, Ecosystems, 9, pp. 1-13. Gruber, R. K., Hinkle, D. C. and Kemp, W. M. (2011), ‘Spatial patterns in water quality associated with submersed plant beds’, Estuaries and coasts, 34(5), 961-972. Hadwen, W.L., Boon, P.I. and Arthington, A.H. (2012), ‘Aquatic ecosystems in inland Australia: tourism and recreational significance, ecological impacts and imperatives for management’, Marine and Freshwater Research 63(4), pp. 325-340. Hager, Achim (2012), ‘The effects of management and plant diversity on carbon storage in coffee agroforestry systems in Costa Rica’, Agroforest Syst. 86, pp. 159–174. Haidary, Azam, Bahman Jabbarian Amiri, Jan Adamowski, Nicola Fohrer and Kaneyuki Nakane (2013), ‘Assessing the Impacts of Four Land Use Types on the Water Quality of Wetlands in Japan’, Water Resour Manage 27, pp. 2217–2229. Haines-Young, R. and Potschin, M. (2013), ‘Common International Classification of Ecosystem Services (CICES). Consultation on Version 4, August-December 2012’, Report to the European Environmental Agency, EEA Framework Contract no: EEA/IEA/09/003. Hamilton, L. S. (1995), ‘Mountain cloud forest conservation and research: a synopsis’, Mountain Research and Development, pp. 259-266. Hantanaka, N., Wright, W., Loyn, R.H. and Mac Nally, R. (2011), ‘Ecologically complex carbon - linking biodiversity values, carbon storage and habitat structure in some austral temperate forests’, Global Ecology and Biogeography, 20, pp. 260-271. Hargreaves, K.J., Fowler, D., Storeton West, R.L. and Duyzer, J.H. (1992), ‘The exchange of nitric oxide, nitrogen dioxide and ozone between pasture and the atmosphere’, Environmental Pollution, 75, pp. 53-59. Harrison, P.A., Berry, P.M., Simpson, G., Haslett, J.R., Blicharska, M., Bucur, M., Dunford, R., ... and Turkelboom, F. (2014), ‘Linkages between biodiversity attributes and ecosystem services: A systematic review’, Ecosystem Services 9, pp. 191-203. Harrison, P.A., Vandewalle, M., Sykes, M.T., Berry, P.M., Bugter, R., de Bello, F., Feld, C.K., ... and Zobel, M. (2010), ‘Identifying and prioritising services in European terrestrial and freshwater ecosystems’, Biodivers. and Conserv., 19, pp. 2791-2821.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 237
Hatanaka, N., Wright, W., Loyn, R. H. and Mac Nally, R. (2011), ‘Ecologically complex carbon’–linking biodiversity values, carbon storage and habitat structure in some austral temperate forests’, Global Ecology and Biogeography, 20(2), pp. 260-271. Haugaasen T., Barlow J. and Peres C.A. (2003), 'Surface wildfires in central Amazonia: short-term impact on forest structure and carbon loss', Forest Ecology and Management, 179, pp. 321-331. He, Y., Wei, Y., DePauw, R., Qian, B., Lemke, R., Singh, A., ... and Wang, H. (2013), ‘Spring wheat yield in the semiarid Canadian prairies: Effects of precipitation timing and soil texture over recent 30 years’, Field Crops Research, 149, pp. 329-337. Hector, A., Philipson, C., Saner, P., Chamagne, J., Dzulkifli, D., O’Brien, M., Snaddon, J.L., Ulok, P., Weilenmann, M., Reynolds, G. and Godfray, H.C.J. (2011), ‘The Sabah Biodiversity Experiment: a long-term test of tree diversity in restoring tropical forest structure and functioning’, Philosphical Transactions of the Royal Society B, 366, pp. 3303-3315. Hefting, Mariet M., Ronald N. van den Heuvel and Jos T.A. Verhoeven (2013), ‘Wetlands in agricultural landscapes for nitrogen attenuation and biodiversity enhancement: Opportunities and limitations’, Ecological Engineering 56, pp. 5– 13. Heyman, E. (2012), ‘Analysing recreational values and management effects in an urban forest with the visitor-employed photography method’, Urban Forestry and Urban Greening, 11(3), pp. 267-277. Hickley, P., Arlinghaus, R., Tyner, R., Aprahamian, M., Parry, K. and Carter, M. (2004), ‘Rehabilitation of urban lake fisheries for angling by managing habitat: general overview and case studies from England and Wales’, Rehabilitation, 4(4), pp. 365-378. Hoehn, P., Tscharntke, T., Tylianakis, J.M. and Steffan-Dewenter, I. (2008), ‘Functional group diversity of bee pollinators increases crop yield’, Proc. R. Soc. B-Biol. Sci., 275(1648), pp. 2283-2291. Hoffmann, C. M., Huijbregts, T., van Swaaij, N. and Jansen, R. (2009), ‘Impact of different environments in Europe on yield and quality of sugar beet genotypes’, European Journal of Agronomy, 30(1), pp. 17-26. Hogan, J.N., Daniels, M.E., Watson, Fred G., Oates, Stori C., Miller, Melissa A., et al. (2013), ‘Hydrologic and Vegetative Removal of Cryptosporidium parvum, Giardia lamblia, and Toxoplasma gondii Surrogate Microspheres in Coastal Wetlands’, Applied and Environmental Microbiology 79 (6), pp. 1859–1865. Holdsworth, D.K. and Mark, A.F. (1990), ‘Water and nutrient input: output budgets: effects of plant cover at seven sites in upland snow tussock grasslands of eastern and central Otago, New Zealand’, Journal of the Royal Society of New Zealand, 20(1), pp. 1-24. Hollingsworth, T.N., Schuur, E.A.G., Chapin III, F.S. and Walker, M.D. (2008), ‘Plant community composition as a predictor of regional soil carbon storage in Alaskan boreal black spruce ecosystems’, Ecosystems 11(4), pp. 629-642. Holzschuh, A., Dormann, C. F., Tscharntke, T. and Steffan-Dewenter, I. (2011), ‘Expansion of mass-flowering crops leads to transient pollinator dilution and reduced wild plant pollination’, Proc. R. Soc. B 278(1723), pp. 3444-51. Holzschuh, A., Steffan‐Dewenter, I., Kleijn, D. and Tscharntke, T. (2007), ‘Diversity of flower‐visiting bees in cereal fields: effects of farming system, landscape composition and regional context’, Journal of Applied Ecology 44(1), pp. 41-49.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 238
Hooper, D. U., Chapin III, F. S., Ewel, J. J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J. H., ... and Wardle D.A. (2005), ‘Effects of biodiversity on ecosystem functioning: a consensus of current knowledge’, Ecol. Monogr. 75(1), pp. 3-35. Hornbeck, J.W., Adams, M.B., Corbett, E.S., Verry, E.S. and Lynch, J.A. (1993), ‘Long-term impacts of forest treatments on water yield: a summary for northeastern USA’, Journal of Hydrology 150(2–4), pp. 323–344. Horne, P., Boxall, P. C. and Adamowicz, W. L. (2005), ‘Multiple-use management of forest recreation sites: a spatially explicit choice experiment’, Forest Ecology and Management 207(1), pp. 189-199. Howley, P., C.O. Donoghue and S. Hynes (2012), ‘Exploring public preferences for traditional farming landscapes’, Landscape and Urban Planning 104 (1), pp. 66–74. Huang, S.-C. L. (2013), ‘Visitor responses to the changing character of the visual landscape as an agrarian area becomes a tourist destination: Yilan County, Taiwan’, Journal of Sustainable Tourism 21 (1), pp. 154- 171. Hulvey, Kristin B., Richard J. Hobbs, Rachel J. Standish, David B. Lindenmayer, Lori Lach and Michael P. Perring (2013), 'Benefits of tree mixes in carbon plantings', Nature Climate Change 3, pp. 869-74. Hümann, M., Schüler, G., Müller, C., Schneider, R., Johst, M. and Caspari, T. (2011), ‘Identification of runoff processes - the impact of different forest types and soil properties on runoff formation and floods’, Journal of Hydrology 409, pp. 637-649. Huppert, D.D., R.L. Johnson, J. Leahy and K. Bell (2003), ‘Interactions between human communities and estuaries in the Pacific Northwest: trends and implications for management’, The Pacific Northwest Coastal Ecosystems Regional Study 26 (4), Part B, pp. 994-1009. Hutt, C. P. and Bettoli, P. W. (2007), ‘Preferences, specialization, and management attitudes of trout anglers fishing in Tennessee tailwaters’, North American Journal of Fisheries Management 27(4), pp. 1257- 1267. Hynynen, J., Repola, J. and Mielikäinen, K. (2011), ‘The effects of species mixture on the growth and yield of mid-rotation mixed stands of Scots pine and silver birch’, Forest Ecology and Management 262(7), pp. 1174-1183. Iperti, G. (1999), ‘Biodiversity of predaceous coccinellidae in relation to bio-indication and economic importance’, Ecological Research 25, pp. 1141-1149. Iroumé, A., Mayen, O. and Huber, A. (2005), ‘Runoff and peak flow responses to timber harvest and forest age in southern Chile’, Hydrological Processes, 21, pp. 37-50. Ja-Choon, Koo, Park Mi Sun and Youn Yeo-Chang (2013), ‘Preferences of urban dwellers on urban forest recreational services in South Korea’, Urban Forestry & Urban Greening 12 (2), pp. 200–210. Jackson, B., Pagella, T., Sinclair, F., Orellana, B., Henshaw, A., Reynolds, B., ... and Eycott, A. (2013), ‘Polyscape: A GIS mapping framework providing efficient and spatially explicit landscape-scale valuation of multiple ecosystem services’, Landscape and Urban Planning, 112, pp. 74-88. Jacob M., Leuschner C., Thomas F. (2010), ‘Productivity of temperate broad-leaved forest stands differing in tree species diversity’, Annals of Forest Science, 67(5), p 503.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 239
Jayasuriya, M.D.A., Dunn, G., Benyon, R. and O'Shaughnessy, P.J. (1993), ‘Some factors affecting water yield from mountain ash (Eucalyptus regnans) dominated forests in south-east Australia’, Journal of Hydrology, 150 (2–4), pp. 345–367. Jim, C. Y. and Chen, W. Y. (2008), ‘Assessing the ecosystem service of air pollutant removal by urban trees in Guangzhou (China)’, Journal of Environmental Management, 88(4), pp. 665-676. Jonsson, M. and Wardle, D.A. (2010), ‘Structural equation modelling reveals plant-community drivers of carbon storage in boreal forest ecosystems’, Biological Letters, 6, pp. 116-119. Jouquet, P., Janeau, J. L., Pisano, A., Sy, H. T., Orange, D., Minh, L. T.N. and Valentin, C. (2012), ‘Influence of earthworms and termites on runoff and erosion in a tropical steep slope fallow in Vietnam: A rainfall simulation experiment’, Applied Soil Ecology, 61, pp. 161–168. Kabanda, T.H. and L.G. Palamuleni (2013), ‘Land use/cover changes and vulnerability to flooding in the Harts catchment, South Africa’, South African Geographical Journal, 95, pp. 1, 105-116. Kagawa, A., Sack, L., Duarte, K. E. and James, S. (2009), ‘Hawaiian native forest conserves water relative to timber plantation: species and stand traits influence water use’, Ecological Applications 19(6), pp. 1429-43. Kamo, T., Tokuoka, Y. and Masahisa, M. (2010), ‘Influence of aphid-host plant pairs on the survivorship and the development of the multicolored Asian ladybird beetle: Implications for the management of vegetation in rural landscapes’, Ecological Research, 25, pp. 1141-1149. Kaplan R. (2007), ‘Employees’ reactions to nearby nature at their workplace: The wild and the tame’, Landscape and Urban Planning 82 (1–2), pp. 17–24. Kaplan, R. and M.E. Austin (2004), ‘Out in the country: sprawl and the quest for nature nearby’, Landscape and Urban Planning, 69 (2–3), pp. 235–243. Kaplan, R. and J.F. Talbot (1988), ‘Ethnicity and preference for natural setting: A review and recent findings’, Landscape and Urban Planning 15 (1–2), pp. 107–117. Kaplan, A., T. Taskın and A. Önenc (2006), ‘Assessing the visual quality of rural and urban-fringed landscapes surrounding livestock farms’, Biosystems Engineering 95(3), pp. 437–448. Kaul, M., Mohren, G.M.J. and Deadhwal, V.K. (2010), ‘Carbon storage and sequestration potential of selected tree species in India’, Mitigation and Adaptation Strategies to Global Change, 15, pp. 489-510. Keeton, W.S., Chernyavskyy, M., Gratzer, G., Main-Knorn, M., Shpylchak, M. and Bihun, Y. (2010), ‘Structural characteristics and aboveground biomass of old‐growth spruce–fir stands in the eastern Carpathian mountains, Ukraine’, Plant Biosystems, 144(1), pp. 148-159. Keitt, T. H. (2009), 'Habitat conversion, extinction thresholds, and pollination services in agroecosystems', Ecological Applications 19, pp. 1561-1573. Kellermann, J.L., Johnson, M.D., Stercho, A.M. and Hackett, S.C. (2007), ‘Ecological and economic services provided by birds on Jamaican Blue Mountain coffee farms’, Conservation Biology, 22, pp. 1177-1185. Kienast, F., B. Degenhardt, B. Weilenmann, Y. Wäger and M. Buchecker (2012), ‘GIS-assisted mapping of landscape suitability for nearby recreation’, Landscape and Urban Planning 105 (4), pp. 385–399. King, S. E. and Lester, J. N. (1995), ‘The value of salt marsh as a sea defence’, Marine Pollution Bulletin, 30(3), pp. 180-189.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 240
Kirby, K.R. and Potvin, C. (2007), ‘Variation in carbon storage among tree species: Implications for the management of a small-scale carbon sink project’, Forest Ecology and Management, 246, pp. 208-221. Klein, A. M., Steffan–Dewenter, I. and Tscharntke, T. (2003), ‘Fruit set of highland coffee increases with the diversity of pollinating bees’, Proceedings of the Royal Society of London. Series B: Biological Sciences, 270(1518), pp. 955-961. Klumpp, K., Fontaine, S., Attard, E., Le Roux, X., Gleixner, G. and Soussana, J.-F. (2009), ‘Grazing triggers soil carbon loss by altering plant roots and their control on soil microbial activity’, Journal of Ecology, 97, pp. 876-885. Koh, L.P. (2008), ‘Birds defend oil palms from herbivorous insects’, Ecological Applications, 18, pp. 821-825. Kohler, F., Verhulst, J., Van Klink, R. and Kleijn, D. (2008), ‘At what spatial scale do high‐quality habitats enhance the diversity of forbs and pollinators in intensively farmed landscapes?’, Journal of Applied Ecology, 45(3), pp. 753-762. Komatsu, H., Kume, T. and Otsuki, K. (2008), ‘The effect of converting a native broad-leaved forest to a coniferous plantation forest on annual water yield: a paired-catchment study in northern Japan’, Forest Ecology and Management, 255(3), pp. 880-886. Kosmas, C., Danalatos, N., Cammeraat, L. H., Chabart, M., Diamantopoulos, J., Farand, R., ... and Vacca, A. (1997), ‘The effect of land use on runoff and soil erosion rates under Mediterranean conditions’, Catena, 29(1), pp. 45-59. Krauss, J., Gallenbergen, I. and Steffan-Dewenter, I. (2011), ‘Decreased functional diversity and biological pest control in conventional compared to organic crop fields’, PLoS ONE 6(5), e19502. Kreitler, Jason, Michael Papenfus, Kristin Byrd and William Labiosa (2013), ‘Interacting Coastal Based Ecosystem Services: Recreation and Water Quality in Puget Sound, WA’, PLoS One 8(2), e56670. Kremen, C. (2005), ‘Managing ecosystem services: What do we need to know about their ecology?’, Ecol. Lett., 8, pp. 468-479. Kromp, B. (1999), ‘Carabid beetles in sustainable agriculture: a review on pest control efficacy, cultivation impacts and enhancement’, Agriculture, Ecosystems and Environment, 74, pp. 187-228. Kunz, T.H., E. Braun de Torrez, D. Bauer, T. Lobova and T.H. Fleming, (2011), 'Ecosystem services provided by bats', Annals of the New York Academy of Sciences. Laclau, P. (2003), ‘Root biomass and carbon storage of ponderosa pine in a northwest Patagonia plantation’, Forest Ecology and Management, 173, pp. 353-360. Lana-Renault, N., E. Nadal-Romero, M. P. Serrano-Muela, B. Alvera, P. Sánchez-Navarrete, Y. Sanjuan and J. M. García-Ruiz (2014), ‘Comparative analysis of the response of various land covers to an exceptional rainfall event in the central Spanish Pyrenees, October 2012', Earth Surface Processes and Landforms 39, pp. 581–592. Lang’at, Joseph K. Sigi, Bernard K. Y. Kirui, Martin W. Skov, James G. Kairo, Maurizio Mencuccini, Mark Huxham (2013), 'Species mixing boosts root yield in mangrove trees', Oecologia 172, pp. 271–278.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 241
Lange, Benjamin, Peter F. Germann and Peter Luscher (2013), ‘Greater abundance of Fagus sylvatica in coniferous flood protection forests due to climate change: impact of modified root densities on infiltration’, Eur J Forest Res 132, pp. 151–163. Larkin, D. J., Steffen, J. F., Gentile, R. M. and Zirbel, C. R. (2013), ‘Ecosystem changes following restoration of a buckthorn-invaded woodland’, Restoration Ecology, 22 (1), pp. 89–97. Larson, Silva, Natalie Stoeckl, Barbara Neil and Riccardo Welters (2013), ‘Using resident perceptions of values associated with the Australian Tropical Rivers to identify policy and management priorities’, Ecological Economics 94, pp. 9–18. Laterra, P., Orue, M. E. and Booman, G. C. (2012), 'Spatial complexity and ecosystem services in rural landscapes’, Agriculture Ecosystems and Environment 154, pp. 56-67. Lavorel, S. (2013), ‘Plant functional effects on ecosystem services’, J. Ecol., 101(1), pp. 4-8. Lavorel, S. and Grigulis, K. (2012), ‘How fundamental plant functional trait relationships scale-up to trade- offs and synergies in ecosystem services’, Journal of Ecology, 100, pp. 128-140. Law, B.E., Sun, O.J., Campbell, J., Tuyl, S.V. and Thornton, P.E. (2003), ‘Changes in carbon storage and fluxes in a chronosequence of ponderosa pine’, Global Change Biology, 9, pp. 510-524. Lawrie, A. H. (1978), ‘The fish community of Lake Superior’, Journal of Great Lakes Research, 4(3), pp. 513- 549. Lee, H.-Y. and Shih, S.-S. (2004), ‘Impacts of vegetation changes on the hydraulic and sediment transport characteristics in Guandu mangrove wetland’, Ecological Engineering, 23, pp. 85-94. Lee, Y.-F. and McCracken,G.F. (2005), ‘Dietary variation of Brazilian free-tailed bats links to migratory populations of pest insects’, Journal of Mammalogy, 86, pp. 67-76. Lee, C.-F., Huang, H.-I and Yeh, H.-R. (2010), ‘Developing an evaluation model for destination attractiveness: sustainable forest recreation tourism in Taiwan’, Journal of Sustainable Tourism 18 (6), pp. 811-828. Legrand, A., Gaucherel, C., Baudry, J. and Meynard, J.-M. (2011), ‘Long-term effects of organic, conventional and integrated crop systems on Carabids’, Agronomy for Sustainable Development, 31(3), pp. 515-524. Lei X., Wang W. and Peng C. (2009), ‘Relationships between stand growth and structural diversity in spruce- dominated forests in New Brunswick, Canada’, Canadian Journal of Forest Research, 39(10), pp. 1835-1847. Le Maitre, D.C., van Wilgen, B.W., Gelderblom, C.M., Bailey, C., Chapman, R.A. and Nel, J.A. (2002), ‘Invasive alien trees and water resources in South Africa: case studies of the costs and benefits of management’, Forest Ecology and Management, 160 (1–3), pp. 143–159. Letourneau, D. K., Bothwell Allen, S. G. and Stireman, J. O. III (2012), 'Perennial habitat fragments, parasitoid diversity and parasitism in ephemeral crops', Journal of Applied Ecology, 49, pp. 1405–1416. Lewin, W. C., Arlinghaus, R. and Mehner, T. (2006), ‘Documented and potential biological impacts of recreational fishing: insights for management and conservation’, Reviews in Fisheries Science, 14(4), pp. 305-367. Li, J. (1999), ‘An appraisal of factors constraining the success of fish stock enhancement programmes’, Fisheries Management and Ecology, 6(2), pp. 161-169.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 242
Liang, J., Buongiorno, J., Monserud, R. A., Kruger, E. L. and Zhou, M. (2007), ‘Effects of diversity of tree species and size on forest basal area growth, recruitment, and mortality’, Forest Ecology and Management, 243(1), pp. 116-127. Liekens, I., M. Schaafsma, L. De Nocker, S. Broekx, J. Staes, J. Aertsens and R. Brouwer (2013), ‘Developing a value function for nature development and land use policy in Flanders, Belgium.’ Land Use Policy 30, pp. 549– 559. Lin, Y. and Wei, X. (2008), ‘The impact of large-scale forest harvesting on hydrology in the Willow watershed of Central British Columbia’, Journal of Hydrology, 359, pp. 141-149. Lindsey, P. A., Alexander, R., Mills, M. G. L., Romanach, S. and Woodroffe, R. (2007), ‘Wildlife viewing preferences of visitors to protected areas in South Africa: implications for the role of ecotourism in conservation’, Journal of Ecotourism, 6(1), pp. 19-33. Liquete ,C., Piroddi, C., Drakou, E.G., Gurney, L., Katsanevakis, S., et al. (2013), ‘Current Status and Future Prospects for the Assessment of Marine and Coastal Ecosystem Services: A Systematic Review’, PLoS ONE 8(7): e67737. Lorencová, E., Frélichová, J., Nelson, E. and Vačkář, D. (2013), ‘Past and future impacts of land use and climate change on agricultural ecosystem services in the Czech Republic’, Land Use Policy, 33, pp. 183-194. Lorenzen, K. (1995), ‘Population dynamics and management of culture‐based fisheries’, Fisheries Management and Ecology, 2(1), pp. 61-73. Lovasi, G. S., O'Neil-Dunne, J. P. M., Lu, J. W. T. et al (2013), ‘Urban Tree Canopy and Asthma, Wheeze, Rhinitis, and Allergic Sensitization to Tree Pollen in a New York City Birth Cohort’, Health Perspect., 121(4), pp. 494-500.
Luck G (2005) ‘An introduction to ecological thresholds – editorial’, Biological Conservation 124, pp. 299- 300. Luck, G.W., Harrington, R., Harrison, P.A., Kremen, C., Berry, P.M., Bugter, R., … and Zobel, M. (2009), ‘Quantifying the contribution of organisms to the provision of ecosystem services’, Biosci., 59(3), pp. 223- 235. Luck, G.W., Lavorel, S., McIntyre, S. and Lumb, K. (2012), ‘Improving the application of vertebrate trait- based frameworks to the study of ecosystem services’, J. Animal Ecol., 81(5), pp. 1065-1076. MA (2005) ‘Ecosystems and human well-being: synthesis’, Millennium Ecosystem Assessment, Island Press, Washington, DC. Mace, G.M., Norris, K. and Fitter, A.H. (2012), ‘Biodiversity and ecosystem services: a multilayered relationship’, Trends in Ecol. and Evol., 27, pp. 19-26. Maceina, M.J. and Reeves, W. C. (1996), 'Relations between submersed macrophyte abundance and largemouth bass tournament success on two Tennessee River impoundments’, Journal of Aquatic Plant Management, 34, pp. 33–38. Maetens, W., Vanmaercke, M., Poesen, J., Jankauskas, B.,Jankauskiene, G. and Ionita, I. (2012), 'Effects of land use on annual runoff and soil loss in Europe and the Mediterranean: A meta-analysis of plot data’, Progress in Physical Geography 36(5), pp. 597–651.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 243
Maitland, P. (1977), 'Freshwater fish in Scotland in the 18th, 19th and 20th centuries’, Biological Conservation, 12( 4), pp. 265–278. Mandal, D., Singh, R., Dhyani, S. K. and Dhyani, B. L. (2013), ‘Landscape and land use effects on soil resources in a Himalayan watershed’ CATENA 81(3), pp. 203–208. Mandelik, Y. and Roll, U. (2009), ‘Diversity patterns of wild bees in almond orchards and their surrounding landscape’, Israel Journal of Plant Sciences, 57(3), pp. 185-191. Manes, F., Incerti, G., Salvatori, E., Vitale, M., Ricotta, C. and Costanza, R. (2012), ‘Urban ecosystem services: tree diversity and stability of tropospheric ozone removal’, Ecological Applications, 22(1), pp. 349- 360. Martínez-Estévez, L., Balvanera, P., Pacheco, J. and Ceballos, G. (2013), ‘Prairie Dog Decline Reduces the Supply of Ecosystem Services and Leads to Desertification of Semiarid Grasslands’, PLoS ONE 8(10): e75229. Martin, P. (1999), ‘Reducing flood risk from sediment-laden agricultural runoff using intercrop management techniques in northern France’, Soil and Tillage Research, 52, pp. 233-245. Martin, T. L., Trevors, J. T. and Kaushik, N. K. (1999), ‘Soil microbial diversity, community structure and denitrification in a temperate riparian zone’, Biodiversity and Conservation, 8(8), pp. 1057-1078. Marzetti, S., M. Disegna, G. Villani and M. Speranza (2011), ‘Conservation and recreational values from semi-natural grasslands for visitors to two Italian parks’, Journal of Environmental Planning and Management, 54 (2), pp. 169-191. Mazda, Y., Magi, M., Kogo, M. and Hong, P.N. (1997), ‘Mangroves as a coastal protection from waves in the Tong Kind delta, Vietnam’, Mangroves and Salt Marshes, 1, pp. 127-135. McDonald, A. G., Bealey, W. J., Fowler, D., Dragosits, U., Skiba, U., Smith, R. I.. Donovan, R. G., Brett, H. E., Hewitt, C. N. and Nemitz, E. (2007), ‘Quantifying the effect of urban tree planting on concentrations and depositions of PM10 in two UK conurbations’, Atmospheric Environment, 41 (38), pp. 8455–8467. McPherson, E. Gregory, Scott, Klaus I. and Simpson, James R. (1998), ‘Estimating cost effectiveness of residential yard trees for improving air quality in Sacramento, California, using existing models’, Atmospheric Environment, 32 (1), pp. 75–84. Meier, I.C. and Leuschner, C. (2010), ‘Variation of soil and biomass carbon pools in beech forests across a precipitation gradient’, Global Change Biology, 16, pp. 1035-1045. Mertz, O., Ravnborg, H.M., Lövei, G.L., Nielsen, I. and Konijnendijk, C.C. (2007), ‘Ecosystem services and biodiversity in developing countries’, Biodivers. and Conserv., 16(10), pp. 2729-2737. Michaletz, P. H. (2009), ‘Variable responses of channel catfish populations to stocking rate: density- dependent and lake productivity effects’, North American Journal of Fisheries Management, 29(1), pp. 177- 188. Mills, A.J. and Cowling, R.M. (2006), ‘Rate of carbon sequestration at two thicket restoration sites in the Eastern Cape, South Africa’, Restoration Ecology, 14(1), pp. 38-49. Millward, A. A. and Sabir, S. (2011), ‘Benefits of a forested urban park: What is the value of Allan Gardens to the city of Toronto, Canada?’, Landscape and urban planning, 100(3), pp. 177-188.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 244
Milstein, A., Hulata, G. and Wohlfarth, G. W. (1988), ‘Canonical correlation analysis of relationships between management inputs and fish growth and yields in polyculture’, Aquaculture Research, 19(1), pp. 13-24. Ming, J., Xian-guo, L., Lin-Shu, X., Li-Juan, C. and Shouzheng, T. (2007), ‘Flood mitigation benefit of wetland soil - A case study in Momoge National Nature Reserve in China’, Ecological Economics, 61, pp. 217-223. Mishra, U., Ussiri, D.A.N. and Lal, R. (2010), ‘Tillage effects on soil organic carbon storage and dynamics in Corn Belt of Ohio USA’, Soil and Tillage Research, 107, pp. 88-96. Misra, A. (2013), ‘Climate change impact, mitigation and adaptation strategies for agricultural and water resources, in Ganga Plain (India)’, Mitigation and Adaptation Strategies for Global Change, 18 (5), pp 673- 689. Mitchell, P.J., Benyon, R.G. and Lane, P.N.J. (2012), ‘Responses of evapotranspiration at different topographic positions and catchment water balance following a pronounced drought in a mixed species eucalypt forest, Australia’, Journal of Hydrology, 440–441, pp. 62–74. Mitsch, William J., Li Zhang, Kay C. Stefanik, Amanda M. Nahlik, Christopher J. Anderson, Blanca Bernal, Maria Hernandez and Keunyea Song (2012), ‘Creating Wetlands: Primary Succession, Water Quality Changes, and Self-Design over 15 Years’, BioScience 62(3), pp. 237-250. Mody, K., Spoerndli, C. and Dorn, S. (2011), ‘Within-orchard variability of the ecosystem service 'parasitism': Effects of cultivars, ants and tree location’, Basic and Applied Ecology, 12, pp. 456-465. Möller, I., Spencer, T., French, J.R., Leggett, J. and Dixon, M. (2007), ‘The sea-defence value of salt marshes: field evidence from north Norfolk’, Water and Environment Journal, 15(2), pp.109-116. Mollot, R., Phothitay, C., Kosy, S., (2003), 'Hydrology, habitat and livelihoods on the floodplains of southern Lao PDR’, Proceedings of the 6th Technical Symposium on Mekong Fisheries, Pakse, Lao PDR 26-28 November 2003, pp. 155–176. Moore, K. A. (2004), ‘Influence of seagrasses on water quality in shallow regions of the lower Chesapeake Bay.’ Journal of Coastal Research, pp. 162-178. Moore, R.D. and Wondzell, S.M. (2005), ‘Physical hydrology and the effects of forest harvesting in the Pacific Northwest: A review’, Journal of the American Water Resources Association, 41(4), pp. 763-784. Morani, A., Nowak, D. J., Hirabayashi, S. and Calfapietra, C. (2011), ‘How to select the best tree planting locations to enhance air pollution removal in the MillionTreesNYC initiative’, Environ Pollut. 159(5), pp. 1040-7. Morris, J., Hess, T. M., Gowing, D. J. G., Leeds‐Harrison, P. B., Bannister, N., Vivash, R. M. N. and Wade, M. (2005), ‘A framework for integrating flood defence and biodiversity in washlands in England’, International Journal of River Basin Management, 3(2), pp. 105-115. Muller, J. (2009), ‘Forestry and water budget of the lowlands in northeast Germany - consequences for the choice of tree species and for forest management’, J. Water Land Dev., 13a, pp. 133–148. Muñoz, A. A. and Cavieres, L. A. (2008), ‘The presence of a showy invasive plant disrupts pollinator service and reproductive output in native alpine species only at high densities’, Journal of Ecology, 96(3), 459-467.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 245
Munyuli, M. T. (2012), ‘Micro, local, landscape and regional drivers of bee biodiversity and pollination services delivery to coffee (Coffea canephora) in Uganda’, International Journal of Biodiversity Science, Ecosystem Services and Management, 8(3), pp. 190-203. Nahuelhual, L., Carmona, A., et al. (2013), ‘article ‘Mapping recreation and ecotourism as a cultural ecosystem service: An application at the local level in Souhtern Chile’, Applied Geography 40, pp. 71-82. Nainggolan, D. Termansen, M. Reed, M.S. Cebollero, E.D. Hubacek, K. (2013), ‘Farmer typology, future scenarios and the implications for ecosystem service provision: A case study from south-eastern Spain’, Regional Environmental Change, 13(3), pp. 601-14. Nash, M.A., Thomson, L.J. and Hoffmann, A.A. (2008), ‘Effect of remnant vegetation, pesticides, and farm management on abundance of the beneficial predator Notonomus gravis (Chaudoir) (Coleoptera: Carabidae)’, Biological Control, 46, pp. 83-93. Natori, Y. and R. Chenoweth (2008), ‘Differences in rural landscape perceptions and preferences between farmers and naturalists’, Journal of Environmental Psychology 28 (3), pp. 250–267. Nedkov, S. and Burkhard, B. (2012), ‘Flood regulating ecosystem services - mapping supply and demand, in the Etropole municipality, Bulgaria’, Ecological Indicators, 21, pp. 67-79. Neris, J., Tejedor, M., Fuentes and J., Jiménez, C. (2013), ‘Infiltration, runoff and soil loss in Andisols affected by forest fire (Canary Islands, Spain)’, Hydrological Processes, 27 (19), pp. 2814–2824. Nguyen, H., Herbohn, J., Firn, J. and Lamb, D. (2012), ‘Biodiversity–productivity relationships in small-scale mixed-species plantations using native species in Leyte province, Philippines’, Forest Ecology and Management, 274, pp. 81-90. Ní Dhubháin, Á., M.-C. Fléchard, R. Moloney and D. O’Connor (2009), ‘Stakeholders’ perceptions of forestry in rural areas - Two case studies in Ireland’, Land Use Policy 26 (3), pp. 695–703. Nie, W. Yuan, Y., Kepner, W., et al. (2011), ‘Assessing impacts of Landuse and Landcover changes on hydrology for the upper San Pedro watershed’, Journal of Hydrology, 407 (1–4), pp. 105–114 Nosetto, M. D., Jobbagy, E. G. and Paruelo, J. M. (2005), ‘Land‐use change and water losses: the case of grassland afforestation across a soil textural gradient in central Argentina’, Global Change Biology, 11(7), pp. 1101-1117. Nowak, David J., McHale, Patrick J., Ibarra, Myriam, Crane, Daniel, Stevens, Jack C. and Luley, Chris J. (1998), ‘Modeling the Effects of Urban Vegetation on Air Pollution’, In: Gryning, Sven-Erik; Chaumerliac, Nadine, eds. Air pollution modeling and its application XII. New York; Plenum Press, pp. 399-407. Nowak, D. J., Crane, D. E. and Stevens, J. C. (2006), ‘Air pollution removal by urban trees and shrubs in the United States’, Urban forestry and urban greening, 4(3), pp. 115-123. Nowak, D. J., Hirabayashi, S., Bodine, A. and Hoehn, R. (2013), ‘Modeled PM2.5 removal by trees in ten US cities and associated health effects’, Environ Pollut. 178, pp. 395-402. Öberg, S. (2007), ‘Diversity of spiders after spring sowing - influence of farming system and habitat type’, Journal of Applied Entomology, 131, pp. 524-531.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 246
Oelbermann, K. and Scheu, S. (2009), ‘Control of aphids on wheat by generalist predators: Effects of predator density and the presence of alternative prey’, Entomologia Experimentalis et Applicata, 132, pp. 225-231. Ogden, Fred L., Trey D. Crouch, Robert F. Stallard and Jefferson S. Hall (2013), ‘Effect of land cover and use on dry season river runoff, runoff efficiency, and peak storm runoff in the seasonal tropics of Central Panama’, Water Resources Research, 49, pp. 8443–8462. Olson, D.M. and Wäckers F.L. (2007), ‘Management of field margins to maximize multiple ecological services’, Journal of Applied Ecology, 44, pp. 13-21. Owen, S. M., MacKenzie, A. R., Stewart, H., Donovan, R. and Hewitt, C. N. (2003), ‘Biogenic volatile organic compound (VOC) emission estimates from an urban tree canopy’, Ecological Applications, 13 (4) pp. 927- 938. Palomo, I., Montes, C., Martín-López, B., González, J.A., García-Llorente, M., Alcorlo, P. and García, C. (2014), ‘Incorporating the social-ecological approach in protected areas in the Anthropocene’, BioScience. DOI: 10.1093/biosci/bit033. Paoletti, E. (2009), ‘Ozone and urban forests in Italy’, Environmental Pollution, 157(5), pp. 1506-1512. Paoletti, Elena, Bardelli, Tommaso, Giovannini, Gianluca and Pecchioli, Leonella (2011), ‘Air quality impact of an urban park over time’, Procedia Environmental Sciences, 4, pp. 10–16. Park, Young-Seuk, Yong-Su Kwon, Soon-Jin Hwang and Sangkyu Park (2014), ‘Characterizing effects of landscape and morphometric factors on water quality of reservoirs using a self-organizing map’, Environmental Modelling & Software 55, pp. 214-221. Pataki, Diane E., Margaret M Carreiro, Jennifer Cherrier, Nancy E Grulke, Viniece Jennings, Stephanie Pincetl, Richard V Pouyat, Thomas H Whitlow, and Wayne C Zipperer (2011), Frontiers in Ecology and the Environment 9, pp. 27-36. Peng, Y., Guizhu Chen, Shiyu Li, Yu Liu and John C. Pernetta (2013), ‘Use of degraded coastal wetland in an integrated mangrove–aquaculture system: a case study from the South China Sea’, Ocean & Coastal Management 85, pp. 209-213. Pennisi, L.A., Holland, S.M. and Stein, T.V. (2004), ‘Achieving Bat Conservation through Tourism’, Journal of Ecotourism 10, 3(3), pp. 195-207. Persiani, A.M., Maggi, O., Montalvo, J., Casado, M.A. and Pineda, F.D. (2008), ‘Mediterranean grassland soil fungi: Patterns of biodiversity, functional redundancy and soil carbon storage’, Plant Biosystems, 142(1), pp. 111-119. Petes, L. E., A. J. Brown, and C. R. Knight (2012), 'Impacts of upstream drought and water withdrawals on the health and survival of downstream estuarine oyster populations’, Ecology and Evolution 2, pp. 1712- 1724. Petheram, C., Walker, G., Grayson, R., Thierfelder, T. and Zhang, L. (2002), ‘Towards a framework for predicting impacts of land-use on recharge: 1. A review of recharge studies in Australia’, Soil Research, 40(3), pp. 397-417. Pinto-Correia, T., F. Barroso, D. Surová and H. Menezes (2011), ‘The fuzziness of Montado landscapes: progress in assessing user preferences through photo-based surveys’, Agroforest Syst. 82, pp. 209-224.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 247
Piotto, D. (2008), ‘A meta-analysis comparing tree growth in monocultures and mixed plantations’, Forest Ecology and Management, 255(3–4), pp. 781–786. Piotto D., Craven D., Montagnini F., Alice F. (2010), ‘Silvicultural and economic aspects of pure and mixed native tree species plantations on degraded pasturelands in humid Costa Rica’, New Forests, 39(3), pp 369- 385. Posthumus, H., Rouquette, J.R., Morris, J., Gowing, D.J.G. and Hess, T.M. (2010), ‘A framework for the assessment of ecosystem goods and services; a case study on lowland floodplains in England’, Ecological Economics, 69, pp. 1510-1523. Potschin, M. and Haines-Young, R. (2011), ‘Ecosystem Services: Exploring a geographical perspective’. Progress in Physical Geography, 35(5), pp. 575-594. Potter, Kevin M. and Christopher W. Woodall (2014), 'Does biodiversity make a difference? Relationships between species richness, evolutionary diversity, and aboveground live tree biomass across U.S. forests', Forest Ecology and Management 321 (2014) 117–129. Powe, N. A. and Willis, K. G. (2004), ‘Mortality and morbidity benefits of air pollution (SO2 and PM10) absorption attributable to woodland in Britain’, Journal of Environmental Management, 70 (2), pp.119–128. Power, J. F., Sandoval, F. M., Ries, R. E. and Merrill, S. D. (1981), ‘Effects of topsoil and subsoil thickness on soil water content and crop production on a disturbed soil’, Soil Science Society of America Journal, 45(1), 124-129. Pretty, J. N., Noble, A. D., Bossio, D., Dixon, J., Hine, R. E., Penning de Vries, F. W. and Morison, J. I. (2006), ‘Resource-conserving agriculture increases yields in developing countries’, Environmental science and technology, 40(4), 1114-1119. Pugh, T. A. M., MacKenzie, A. R., Whyatt, J. D. and Hewitt, C. N. (2012), ‘Effectiveness of Green Infrastructure for Improvement of Air Quality in Urban Street Canyons’, Environ. Sci. Technol., 46 (14), pp 7692–7699. Püschel-Hoeneisen, N. and J.A. Simonetti (2012), ‘Forested habitat preferences by Chilean citizens: Implications for biodiversity conservation in Pinus radiata plantations’, Revista Chilena de Historia Natural 8, pp. 161-169. Putuhena, W.M. and Cordery, I. (2000), ‘Some hydrological effects of changing forest cover from eucalypts to Pinus radiata’, Agricultural and Forest Meteorology, 100, pp. 59-72. Qi et al. (2010), ‘Carbon sequestration in old-growth forest and plantations on Changbai Mountain in Northeast China’, Bioinformatics and Biomedical Engineering, 4th International Conference. Qiu, L., Lindberg, S. and Nielsen, A. B. (2013), ‘Is biodiversity attractive?—On-site perception of recreational and biodiversity values in urban green space’, Landscape and Urban Planning, 119, 136-146. Quiros, R. and Mari, A. (1999), ‘Factors contributing to the outcome of stocking programmes in Cuban reservoirs’, Fisheries Management and Ecology, 6(3), 241-254. Rand, T. and Tscharntke, T. (2007), ‘Contrasting effects of natural habitat loss on generalist and specialist aphid natural enemies’, Oikos, 116, pp. 1353-1362.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 248
Rashin, E. B., Clishe, C. J., Loch, A. T. and Bell, J. M. (2006), ‘Effectiveness of timber harvest practices for controlling sediment related water quality impacts’, Journal of the American Water Resources Association, 42(5), pp. 1307-1327. Resilience Alliance and Santa Fe Institute (2004), ‘Thresholds and alternate states in ecological and social- ecological systems’, Resilience Alliance. (Online.) http://www.resalliance.org/index.php/thresholds_database. Rey Benayas, J.M., Martins, A., Nicolau, J.M. and Schulz, J.J. (2007), ‘Abandonment of agricultural land: an overview of drivers and consequences’, CAB Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources, 2(057), 14 pp. Rey Benayas, J.M., Newton, A.C., Diaz, A. and Bullock, J.M. (2009), ‘Enhancement of biodiversity and ecosystem services by ecological restoration: A meta-analysis’, Science, 325, pp. 1121-1124. Ribe, R.G. (2006), ‘Perceptions of forestry alternatives in the US Pacific Northwest: Information effects and acceptability distribution analysis’, Journal of Environmental Psychology 26 (2), pp. 100–115. Ribe, R.G. (2009), ‘In-stand scenic beauty of variable retention harvests and mature forests in the U.S. Pacific Northwest: The effects of basal area, density, retention pattern and down wood’, J Environ Manage. 91(1), pp. 245-60. Richards, Meryl Breton and V. Ernesto Mendez (2013), 'Interactions between Carbon Sequestration and Shade Tree Diversity in a Smallholder Coffee Cooperative in El Salvador', Conservation Biology, 28, (2), pp. 489–497. Ricketts, T.H., Regetz, J., Steffam-Dewemter, I., et al. (2008), ‘Landscape effects on crop pollination services: are there general patterns?’, Ecol Lett. 11(5), pp. 499-515. Riedel, J., Dorn, S., Plath, M., Potvin, C. and Mody, K. (2013), ‘Time matters: Temporally changing effects of planting schemes and insecticide treatment on native timber tree performance on former pasture’, Forest Ecology and Management, 297, 49-56. Robinson, M., Cognard-Plancq, A.-L., Cosandey, C., David, J., Durand, P., Führer, H.-W. ... and Zollner, A. (2003), ‘Studies of the impact of forests on peak flows and baseflows: a European perspective’, Forest Ecology and Management, 186, pp. 85-97. Rodrigo, María A, Miguel Martín, Carmen Rojo, Sara Gargallo, Matilde Segura and Núria Oliver (2013), ‘The role of eutrophication reduction of two small man-made Mediterranean lagoons in the context of a broader remediation system: Effects on water quality and plankton contribution’, Ecological Engineering 61, pp. 371–382. Roubah, A. Lassere-Joulin, F., Amiaud, B. and Plantureux, S. (2014), 'Emergent effects of ground beetles size diversity on the strength of prey suppression', Ecological Entomology (2014), 39, 47–57. Rowe, L.K. and Pearce, A.J. (1994), ‘Hydrology and related changes after harvesting native forest catchments and establishing Pinus radiata plantations. Part 2. The native forest water balance and changes in streamflow after harvesting’, Hydrological Processes, 8(4), pp. 299 - 310. Ruiz-Frau, A., H.Hinz, G.Edwards-Jones & M.J.Kaiser (2013), ‘Spatially explicit economic assessment of cultural ecosystem services: Non-extractive recreational uses of the coastal environment related to marine biodiversity’, Marine Policy 38, pp. 90–98.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 249
Ruiz-Jaen, M.C. and Potvin, C. (2011), ‘Can we predict carbon stocks in tropical ecosystems from tree diversity? Comparing species and functional diversity in a plantation and natural forest’, New Phytologist, 189(4), 978-987. Ruprecht, J.K. and Schofield, N.J. (1989), ‘Analysis of streamflow generation following deforestation in southwest Western Australia’, Journal of Hydrology, 105 (1–2), pp. 1–17. Ruprecht, J. K. and Stoneman, G. L. (1993), ‘Water yield issues in the jarrah forest of south-western Australia’, Journal of Hydrology, 150(2), 369-391. Rusch, A., Valantin-Morison, M., Roger-Estrade, J. and Sarthou, J.P. (2012), ‘Using landscape indicators to predict high pest infestations and successful natural pest control at the regional scale’, Landscape and Urban Planning, 105, pp. 62-73. Rusch, Adrien, Riccardo Bommarco, Mattias Jonsson, Henrik G. Smith and Barbara Ekbom (2013), 'Flow and stability of natural pest control services depend on complexity and crop rotation at the landscape scale', Journal of Applied Ecology 50, pp. 345–354. Rutten M. J. M., Bovenhuis, H. and Komen, H., (2005), 'Genetic parameters for fillet traits and body measurements in Nile tilapia (Oreochromis niloticus L.)', Aquaculture, 246(1–4), pp. 125–132. Saenz, L. and Mulligan, M. (2013), ‘The role of Cloud Affected Forests (CAFs) on water inputs to dams’, Ecosystem Services, 5, pp. 69–77. Sahin, V. and Hall, M. J. (1996), ‘The effects of afforestation and deforestation on water yields’, Journal of hydrology, 178(1), pp. 293-309. Salmond, J. A., Williams, D. E., Laing, G., Kingham, S., Dirks, K., Longley, I. and Henshaw, G. S. (2013), ‘The influence of vegetation on the horizontal and vertical distribution of pollutants in a street canyon’, Science of the Total Environment, 443, pp. 287-298. Sande, S. O., Crewe, R. M., Raina, S. K., Nicolson, S. W. and Gordon, I. (2009), ‘Proximity to a forest leads to higher honey yield: Another reason to conserve’, Biological conservation, 142(11), pp. 2703-2709. Scheffer, M. (2009), Critical Transitions in Nature and Society. Princeton University Press. Schmidt. M.H., Thewes, U., Thies, C. and Tscharntke, T. (2004), ‘Aphid suppression by natural enemies in mulched cereals’, Entomologia Experimentalis et Applicata, 113, pp. 87-93. Schmittner, K.-E. and Giresse, P. (1996), ‘Modelling and application of the geomorphic and environmental controls on flash flood flow’, Geomorphology, 16, pp. 337-347. Schmitz, M.F., I. De Aranzabal and R.D. Pineda (2007), ‘Spatial analysis of visitor preferences in the outdoor recreational niche of Mediterranean cultural landscapes’, Environmental Conservation 34 (04), pp. 300-312. Schnorbus, M. and Alila, Y. (2013), ‘Peak flow regime changes following forest harvesting in a snow- dominated basin: Effects of harvest area, elevation, and channel connectivity’, Water Resources Research, 49, 517–535. Schröter, M., van der Zanden, E.H., van Oudenhoven, A.P.E., Remme, R.P., Serna-Chavez, H.M., de Groot, R.S. and Opdam, P. (2014), ‘Ecosystem services as a contested concept: a synthesis of critique and counter- arguments’, Conserv. Lett., DOI: 10.1111/conl.12091.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 250
Schuhmann, P. W., Casey, J. F., Horrocks, J. A. and Oxenford, H. A. (2013), ‘Recreational SCUBA divers' willingness to pay for marine biodiversity in Barbados’, Journal of environmental management, 121, 29-36. Schuller, P., Ellies, A., Castillo, A., Salazar, I. (2003), 'Use of 137Cs to estimate tillage- and water-induced soil redistribution rates on agricultural land under different use and management in central-south Chile', Soil and Tillage Research, 69(1-2), pp. 69-83. Seidl, R., Rammer, W., Jäger, D. and Lexer, M.J. (2008), ‘Impact of bark beetle (Ips typographus L.) disturbance on timber production and carbon sequestration in different management strategies under climate change’, Forest Ecology and Management, 256, pp. 209-220. Seppelt, R., Dormann, C. F., Eppink, F. V., Lautenbach, S. and Schmidt, S. (2011), ‘A quantitative review of ecosystem service studies: approaches, shortcomings and the road ahead’, J. Appl. Ecol., 48, pp. 630–636. Setälä, H., Viippola, V., Rantalainen, A. L., Pennanen, A. and Yli-Pelkonen, V. (2013), ‘Does urban vegetation mitigate air pollution in northern conditions?’, Environmental Pollution, 183, pp. 104–112. Shavit, O., Dafni, A. and Ne'eman, G. (2009), ‘Competition between honeybees (Apis mellifera) and native solitary bees in the Mediterranean region of Israel—Implications for conservation’, Israel Journal of Plant Sciences, 57(3), pp. 171-183. Simonit, Silvio and Charles Perrings (2013), ‘Bundling ecosystem services in the Panama Canal watershed’, PNAS 110(23), pp. 9326–9331. Singh, S. and Mishra, A. (2012), ‘Spatiotemporal analysis of the effects of forest covers on water yield in the Western Ghats of peninsular India’, Journal of Hydrology, 446–447, pp. 24–34. Smakhtin, V.U. and Batchelor, A.L. (2005), ‘Evaluating wetland flow regulating functions using discharge time-series’, Hydrol. Procsess., 19, pp. 1293-1305. Smallwood, C. B., Beckley, L. E. and Sumner, N. R. (2006), ‘Shore-based recreational angling in the Rottnest Island Reserve, Western Australia: spatial and temporal distribution of catch and fishing effort’, Pacific Conservation Biology, 12(3), pp. 238. Smith, J.A., Baumgartner, L.J., Suthers, I.M., Ives, M.C. and Taylor, M.D. (2012), ‘Estimating the stocking potential of fish in impoundment by modelling supply and steady-state demand’, Freshwater Biology 57, pp. 1482-1499. Snep, R.P.H., E.C. van Ierland and P. Opdam (2009), ‘Enhancing biodiversity at business sites: What are the options, and which of these do stakeholders prefer?’, Landscape and Urban Planning 91 (1), pp. 26–35. Sooknah, R. D. and Wilkie, A. C. (2004), ‘Nutrient removal by floating aquatic macrophytes cultured in anaerobically digested flushed dairy manure wastewater’, Ecological Engineering, 22(1), pp. 27-42. Steinbeiss, S., Beβler, H., Engels, C., Temperton, V.M., Buchmann, N., Roscher, C., Kreutziger, Y., Baade, J., Habekost, M. and Gleixner, G. (2008), ‘Plant diversity positively affects short-term soil carbon storage in experimental grasslands’, Global Change Biology, 14, pp. 2937-2949. Steingröver, E.G., Geertsema, W. and van Wingerden, W.K.R.E. (2010), ‘Designing agricultural landscapes for natural pest control: A transdisciplinary approach in the Hoeksche Waard (the Netherlands)’, Landscape Ecology, 25, pp. 825-838.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 251
Stovin, V., Vesuviano, G. and Kasmin, H. (2012), ‘The hydrological performance of a green roof test bed under UK climatic conditions’, Journal of Hydrology 414–415, pp. 148–161. Sun, D.S., Wesche, K., Chen, D.D., Zhang, S.H., Wu, G.L., Du, G.Z. and Comerford, N.B. (2011), ‘Grazing depresses soil carbon storage through changing plant biomass and composition in a Tibetan alpine meadow’, Plant, Soil and Environment, 57(6), pp. 271-278. Sun, G., Zhou, G., Zhang, Z., Wei, X., McNulty, S.G. and Vose, J.M. (2006), ‘Potential water yield reduction due to reforestation across China’, J. Hydrol., 328, pp. 548–558. Swaffield, S.R. and R.J. Foster (2000), ‘Community perceptions of landscape values in the South Island high country. A literature review of current knowledge and evaluation of survey methods’, Science for Conservation, 159, pp. 1173-2946. Szulczewska, B., R. Giedych, J. Borowski, M. Kuchcik, P. Sikorski, A. Mazurkiewicz, and T. Stanczyk (2014), 'How much green is needed for a vital neighbourhood? In search for empirical evidence’, Land Use Policy 38, pp. 330-345. Taha, H. 91996), ‘Modeling impacts of increased urban vegetation on ozone air quality in the South Coast Air Basin’, Atmospheric Environment, 30 (20), pp. 3423–3430. Taha, H., Douglas, S. and Haney, J. (1997), ‘Mesoscale meteorological and air quality impacts of increased urban albedo and vegetation’, Energy and Buildings, 25(2), 169-177. Takahashi,M., Giambelluca,T.W., Mudd, R.G., DeLay, J.K., Nullet, M.A. and Asner, G.P. (2011), ‘Rainfall partitioning and cloud water interception in native forest and invaded forest in Hawai‘i Volcanoes National Park’, Hydrological Processes, Special Issue: Hydrometeorology of tropical montane cloud forests, 25 (3), pp. 448–464. Temmerman, S., Mindert, B.D-.V. and Bouma, T.J. (2012), ‘Coastal marsh die-off and reduced attenuation of coastal floods: A model analysis’, Global and Planetary Change, 92, pp. 267-274. Thomas, H. and Nisbet, T.R. (2006), ‘An assessment of the impact of floodplain woodland on flood flows’, Water and Environ. J., 21, pp. 114-126. Thorn, A.J., R.C. Daniel, B. Orland and N. Brabyn (1997), ‘Managing forest aesthetics in production forests’, New Zealand Forestry 42 (2), pp. 21-23, 26-29. Tolbert, V.R., Joslin, J.D., Thornton, F.C., Bock, B.R., Pettry, D.E. and Bandaranayake, W. (1999), ‘Biomass crop production: benefits for soil quality and carbon sequestration’, New Zealand Journal of Forestry Science, 30, pp. 138-149. Tomimatsu, H., Nakano, K., Yamamoto, N. and Suyama, Y. (2014), ‘Effects of genotypic diversity of Phragmites australis on primary productivity and water quality in an experimental wetland’, Oecologia, 175, pp. 163–172. Tran Van Con, Nguyen Toan Thang, Do Thi Thanh Ha, Cao Chi Khiem, Tran Hoang Quy, Vu Tien Lam, Tran Van Do, Tamotsu Sato (2013), 'Relationship between aboveground biomass and measures of structure and species diversity in tropical forests of Vietnam', Forest Ecology and Management 310, pp. 213–218. Tscharntke, T. et al. (2011), ‘Multifunctional shade-tree management in tropical agroforestry landscapes: A review’, Journal of Applied Ecology, 48, pp. 619-629.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 252
Tveit, M.S. (2009), ‘Indicators of visual scale as predictors of landscape preference; a comparison between groups’, Journal of Environmental Management, 90 (9), pp. 2882–2888. UK NEA (2011), The UK National Ecosystem Assessment Technical Report. Cambridge: UNEP-WCMC. Van Berkel, Derek B. and Peter H. Verburg (2014), ‘Spatial quantification and valuation of cultural ecosystem services in an agricultural landscape’, Ecological Indicators 37, Part A, pp. 163–174. Van den Berg, A.E. and Koole, S.L. (2006), ‘New wilderness in the Netherlands: An investigation of visual preferences for nature development landscapes’, Landscape and Urban Planning, 78, pp. 362-372. Van den Berg, A.E., Vlek, C.A.J. and Coeterier, J.F. (1998), ‘Group differences in the aesthetic evaluation of nature development plans: a multilevel approach’, J. Environ.l Psychol., 18, pp. 141-157. Vergara, C.H. and Badano, E.I. (2009), ‘Pollinator diversity increases fruit production in Mexican coffee plantations: the importance of rustic management systems’, Agriculture, Ecosystems & Environment, 129 (1–3), pp. 117–123. Vertessy, R.A., Watson, F.G.R. and O’Sullivan, S.K. (2001), ‘Factors determining relations between stand age and catchment water balance in mountain ash forests’, Forest Ecology and Management, 143 (1–3), pp. 13–26. Vilà, M., Vayreda, J., Gracia, C. and Ibáñez, J. J. (2003), ‘Does tree diversity increase wood production in pine forests?’, Oecologia, 135(2), pp. 299-303. Vinatier, F., Gosme, M. and Valantin-Morison, M. (2012), ‘A tool for testing integrating pest management strategies on a tritophic system involving pollen beetle, its parasitoid and oilseed rape at the landscape scale’, Landscape Ecology, 27, pp. 1421-1433. Vollhardt, I.M.G., Bianchi, F.J.J.A., Wackers, F.L., Thies, C. and Tscharntke, T. (2010), ‘Spatial distribution of flowers vs. Honeydew resources in cereal fields may affect aphid parasitism’, Biological Control, 53, pp. 204-213. Wagner, E., B. R. Dumbauld, S. D. Hacker, A. C. Trimble, L. M. Wisehart, and J. L. Ruesink (2012), 'Density- dependent effects of an introduced oyster, Crassostrea gigas, on a native intertidal seagrass, Zostera marina’, Marine Ecology Progress Series 468, pp. 149-160. Wamsley, T.V., Cialone, M.A., Smith, J.M., Atkinson, J.H. and Rosati, J.D. (2010), ‘The potential of wetlands in reducing storm surge’, Ocean Engineering, 37, pp. 59-68. Wania, Annett, Bruse, Michael, Blond, Nadège and Weber, Christiane (2012), ‘Analysing the influence of different street vegetation on traffic-induced particle dispersion using microscale simulations’, Journal of Environmental Management, 94, (1), pp. 91–101. Wardle, David A., Micael Jonsson, Sheel Bansal, Richard D. Bardgett, Michael J. Gundale and Daniel B. Metcalfe (2012), 'Linking vegetation change, carbon sequestration and biodiversity: insights from island ecosystems in a long-term natural experiment', Journal of Ecology, 100, pp. 16–30. Wardynski, B. J., R. J. Winston, and W. F. Hunt (2013), 'Internal Water Storage Enhances Exfiltration and Thermal Load Reduction from Permeable Pavement in the North Carolina Mountains’, Journal of Environmental Engineering-Asce 139, pp. 187-195.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 253
Waterloo, M.J., Schellekens, J., Bruijnzeel, L.A. and Rawaqa, T.T. (2007), ‘Changes in catchment runoff after harvesting and burning of a Pinus caribaea plantation in Viti Levu, Fiji’, Forest Ecology and Management, 251 (1–2), pp. 31–44. Watson, J. C., Wolf, A. T. and Ascher, J. S. (2011), ‘Forested Landscapes Promote Richness and Abundance of Native Bees (Hymenoptera: Apoidea: Anthophila) in Wisconsin Apple Orchards’, Environmental Entomology 40(3), pp. 621-32. Webb, A.A. (2009), ‘Streamflow response to Pinus plantation harvesting: Canobolas State forest, southeastern Australia’, Hydrological Processes, 23 (12), pp. 1679–1689. Webb, A.A. and Kathuria, A. (2012), ‘Response of streamflow to afforestation and thinning at Red Hill, Murray Darling Basin, Australia’, Journal of Hydrology, 412–413, pp. 133–140. Wee, Y.C. and Tsang, K.C. (2008), ‘The changing face of birding in Singapore’, Nature in Singapore, 1, pp. 97–102. Wei, W., Chen, L., Fu, B. and Chen, J. (2010), ‘Water erosion response to rainfall and land use in different drought-level years in a loess hilly area of China’, Catena, 81(1), pp. 24-31. Weisner, S. E. B. and G. Thiere (2010), ‘Effects of vegetation state on biodiversity and nitrogen retention in created wetlands: a test of the biodiversity-ecosystem functioning hypothesis’, Freshwater Biology, 55(2), pp. 387–396. Williams, C. A., et al. (2012), ‘Climate and vegetation controls on the surface water balance: Synthesis of evapotranspiration measured across a global network of flux towers’, Water Resources Research, 48 (6). Winfree, R., Williams, N.M., Dushoff, J. and Kremen, C. (2007), ‘Native bees provide insurance against ongoing honey bee losses’, Ecol Lett. 10(11), pp. 1105-13. Winfree, R., Williams, N. M., Gaines, H., Ascher, J. S. and Kremen, C. (2008), ‘Wild bee pollinators provide the majority of crop visitation across land-use gradients in New Jersey and Pennsylvania, USA’, Journal of Applied Ecology 45(3), pp. 793–802. Woodcock, Ben A., Collin Harrower, John Redhead, Mike Edwards, Adam J. Vanbergen, Matthew S. Heard, David B. Roy and Richard F. Pywell (2014), 'National patterns of functional diversity and redundancy in predatory ground beetles and bees associated with key UK arable crops', Journal of Applied Ecology, 51, pp. 142–151. Xu, Q., Fujiyama, S. and Xu, H. (2011), ‘Biological pest control by enhancing populations of natural enemies in organic farming systems’, Journal of Food, Agriculture and Environment, 9(2), pp. 455-463. Xu, Q., Fujiyama, S. and Xu, H. (2012), ‘Pest control by enriching natural enemies under artificial habitat management along sidewalls of greenhouse in organic farming systems’, Journal of Food, Agriculture and Environment, 10(2), pp. 449-458. Yang Bai, Changwei Zhuang, Zhiyun Ouyang, Hua Zheng, Bo Jiang (2011), 'Spatial characteristics between biodiversity and ecosystem services in a human-dominated watershed', Ecological Complexity, 8, 177-183. Yang, J., McBride, J., Zhou, J. and Sun, Z. (2005), ‘The urban forest in Beijing and its role in air pollution reduction’, Urban Forestry and Urban Greening, 3(2), pp. 65-78.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 254
Yang, J., Yu, Q. and Gong, P. (2008), ‘Quantifying air pollution removal by green roofs in Chicago’, Atmospheric Environment, 42(31), pp. 7266-7273. Yang, W., Wang, X., Liu, Y., Gabor, S., Boychuk, L. and Badiou, P. (2010), ‘Simulated environmental effects of wetland restoration scenarios in a typical Canadian prairie watershed’, Wetlands Ecological Management, 18, pp. 269-279. Yao, Y., Cai, T., Wei, X., Mingfang Zhang, M. and Cunyong Ju, C. (2012a), ‘Effect of forest recovery on summer streamflow in small forested watersheds, Northeastern China’, Hydrol. Process., 26, pp. 1208- 1214. Yao, Y., X. Zhu, Y. Xu, H. Yang, X. Wu, Y. Li and Y. Zhang (2012b), ‘Assessing the visual quality of green landscaping in rural residential areas: the case of Changzhou’, Environ Monit. Assess. 184(2), pp. 951-67. Yen, T.-M. and Lee, J.-S. (2011), ‘Comparing aboveground carbon sequestration between moso bamboo (Phyllostachys heterocycla) and China fir (Cinninghamia lanceolata) forests based on the allometric model’, Forest Ecology and Management, 261(6), pp. 995-1002. Yin, Shan, Shen, Zhemin, Zhou, Pisheng, Zou, Xiaodong, Che, Shengquan and Wang, Wenhua (2011), ‘Quantifying air pollution attenuation within urban parks: An experimental approach in Shanghai, China’, Environmental Pollution, 159 (8–9), pp. 2155–2163 Yurova, A. Y. and Lankreijer, H. (2007), ‘Carbon storage in the organic layers of boreal forest soils under various moisture conditions: a model study for Northern Sweden sites’, Ecological modelling, 204(3), pp. 475-484. Zavaleta, E. (2000), ‘The economic value of controlling an invasive shrub’, AMBIO: A Journal of the Human Environment, 29(8), pp. 462-467. Zhang, W., van der Werf, W. and Swinton, S.M. (2010), ‘Spatially optimal habitat management for enhancing natural control of an invasive agricultural pest: Soybean aphid’, Resource and Energy Economics, 32, pp. 551-565. Zhang, Yu, Han Y. H. Chen and Peter B. Reich (2012a), ‘Forest productivity increases with evenness, species richness and trait variation: a global meta-analysis’, Journal of Ecology 100(3), pp. 742–749. Zhang, J., Guan, D.S. and Song, M.W. (2012b), ‘Biomass and carbon storage of Eucalyptus and Acacia plantations in the Pearl River Delta, South China’, Forest Ecology and Management, 227, pp. 90-97. Zhang, J.T., Xiang C. and Min, L. (2012c), ‘Effects of Tourism and Topography on Vegetation Diversity in the Subalpine Meadows of the Dongling Mountains of Beijing, China’, Environ Manage. 49(2), pp. 403-11. Zhao, M., Kong, Z.-H., Escobedo, F.J. and Gao, J. (2010), ‘Impacts of urban forests on offsetting carbon emissions from industrial energy use in Hangzhou, China’, Journal of Environmental Management, 91, pp. 807-813. Zhao, M., Xiang, W., Peng, C. and Tian, D. (2009), ‘Simulating age-related changes in carbon storage and allocation in a Chinese fir plantation growing in southern China using the 3-PG model’, Forest Ecology and Management, 257, pp. 1520-31. Zimmermann, J., Dauber, J. and Jones, M.B. (2012), ‘Soil carbon sequestration during the establishment phase of Miscanthus x giganteus: a regional-scale study on commercial farms using 13C natural abundance’, GCB Bioenergy, 4, pp. 453-61.
D3.1 – Database and operational classification system of ecosystem service-natural capital relationships 255
Zivin, J., Hueth, B.M., et al. (2000), ‘Managing a multiple-use resource: The case of feral pig management in California rangeland’, Journal of Environmental Economics and Management 39, pp. 189-204. Zuazo, V. D., Martínez, J. F. and Raya, A. M. (2004), ‘Impact of vegetative cover on runoff and soil erosion at hillslope scale in Lanjaron, Spain’, Environmentalist, 24(1), pp. 39-48.