Introduction

The mapping and assessment of ecosystem services (ES) is one of the core actions (Action 5) of the European Union’s (EU) Biodiversity Strategy, which aims at “halting the loss of biodiversity and the degradation of ecosystem services in the EU by 2020, and restoring them in so far as feasible, while stepping up the EU contribution to averting global biodiversity loss”1. To support the implementation of Action 5 a working group on Mapping and Assessment of Ecosystems and their Services (MAES) was established2,3. These efforts place the European Union on the course to achieve its global commitments under the Convention on Biological Diversity, particularly the Aichi Target 11 on conservation of biodiversity and ES of terrestrial, inland water, coastal and marine areas through the expansion of protected areas; and the Target 14 on restoring and safeguarding essential ES.

Approaches to ES mapping are abundant and vary on aim and rational, type of ES analysed, spatial scale, and source of information2,4,5. Several mapping methodologies’ reviews are available in literature4,5,6,7,8,9. According to Martínez-Harms and Balvanera5 review, which focus on social-ecological assessments of ES, the most commonly mapped ES are carbon storage, carbon sequestration, food production and recreation. The most frequently used method is the causal relationships based on the understanding of ES and readily available information. Other methods for mapping ES are extrapolation of primary data (e.g., field data, surveys, and census data), expert knowledge, regression models and look-up tables. Regional (103–105 km2) and national (105–106 km2) spatial scales are the most common analysed; and land cover variables, topographical information and spectral vegetation indices are frequently used as source of information5.

The process of mapping and assessing ES requires a first stage of identification and classification of ES, which according to Haines‐Young and Potschin10 is foreseen as conceptually and technically challenging. The lack of understanding on a common definition of the ES concept, together with the variety of purposes, applications (e.g. environmental accounting, ES mapping, ES valuing) and disciplines involved (e.g. ecology, sociology, economy, geography), contribute to the debate around ES definition and classification. In this context, in 2009, the European Environment Agency (EEA) promoted the development of a consistent classification of ES – the Common International Classification of Ecosystem Services (CICES) – compatible with the design of Integrated Environmental and Economic Accounting methods. Coordinated by the University of Nottingham, CICES provides a “common base” for comparison across Europe11.

This classification system has been adopted by the MAES working group and will be used through Europe, by Member States, in the mapping and assessment of the state of ecosystems and their services3.

When applied to complex coastal regions, these mapping and classification approaches must be adapted to their biophysical and sociocultural characteristics, governance framework and to the scale of analysis.

Ria de Aveiro coastal region is a complex socio-ecological system given the multiplicity of institutions, organizations and stakeholders involved in its governance, together with the variety of ecosystems included (marine, freshwater, forest, agro-ecosystems). Therefore, and due to its complexity, at a regional scale, the case study of Ria de Aveiro coastal region presents an opportunity to explore the application of such approaches. In addition, this case study is relevant in the European and national context, not only for the coastal lagoon itself and its rich and diverse natural capital, but also for its central national geographic location in the Atlantic coastal zone, between two metropolitan areas (Porto and Lisbon). Moreover, the presence of the Aveiro harbour (considered as one of the most important maritime gates of the Iberian Peninsula and the south of Europe), which allows and strengthens both the maritime and terrestrial transport connections with a number of national and international cities12, turns this coastal area pertinent.

Having Ria de Aveiro coastal region as a show case, the main objectives of this paper are: i) to define the exact geographic boundaries of a complex coastal region; ii) to identify and classify the ES provided; and iii) to map the ES provided by a complex coastal region. In this context, the criteria used to define the case study boundaries, as it involves different ecosystem typologies and a complex governance framework are discussed; the provided ES are identified; and the most suitable indicators that will serve as basis to map the ES provided are defined (Fig. 1). This approach follows the principles of integrated and ecosystem-based management approaches, acknowledging the complexity and the interspecies relationship within ecological systems, but also accounting for social and governance objectives13. It aims to contribute to the discussion on ES classification and mapping at regional scale.

Figure 1: Overview of the conceptual approach:
figure 1

From the boundaries definition to the ES classification and mapping. Drawings were generated with Microsoft Visio Professional 2016.

Full size image

Methods and Materials

Definition of the geographic area

Geographic boundaries are frequently defined by administrative boarders, (e.g. county, municipality, parish), statistical units (e.g. NUTS - Nomenclature of Territorial Units for Statistics) or by jurisdictional boundaries of planning and management tools through plans and programmes (e.g. strategic planning, land/maritime development, environmental planning). Nevertheless, these boundaries do not always correspond to the geography of human uses, ecosystem processes or boundaries, such as those based on biogeography, oceanography and/or bathymetry14. In addition, in the context of an integrated and ecosystem-based management approaches, such boundaries should be taken into account, and a balance between ecological, social and jurisdictional factors should be achieved15,16.

Therefore, the criteria for defining the study area include:

  1. i

    Analysis of the territory elements (e.g. land cover, topography, bathymetry) and identification of the physical boundaries of the main studied ecosystems and its interfaces, in order to include significant connective structures of the landscape (i.e., physical relationships that facilitate the link between different elements in the landscape or spatial settings across multiple scales, such as blue and green infrastructures) and thus adopt a system-wide approach17;

  2. ii

    Spatial planning and management tools that focus on the study area, particularly at regional or municipal level for land/maritime development, spatial planning, and/or environmental planning;

  3. iii

    Designated areas for nature conservation under international, supranational or national protected areas network (e.g. Ramsar Convention, Natura 2000 network, Nature Parks);

  4. iv

    Administrative and statistical boundaries (e.g. parish, municipality, NUTS III), depending on the detail of the assessment as well as the available information;

  5. v

    Existence, availability and scale of spatial data, which may condition the enforcement of previous criteria.

Identification and classification of ecosystem services

Both the ES concept and the ES classification have evolved over time with varying attention for the ecosystem basis or the economic use, as summarized in Table 1.

Table 1 ES definitions (adapted from Vandewalle et al.49 Braat and de Groot19 Häyhä and Franzese20).
Full size table

Despite the extensive debate about its adequacy18,19,20, the Millennium Ecosystem Assessment (MA) definition (Table 1) and classification of ES have been widely adopted. Nevertheless, this definition has also been considered broad and ambiguous21,22, and alternative definitions have been proposed (Table 1).

In 2009, the EEA fostered and supported the development of CICES, which was adopted by the MAES working group, and will be used by Member States during the implementation of the EU Biodiversity Strategy3. For this reason, and given that CICES is consistent with existing and accepted typologies of ES (such as MA23 and TEEB24), and establishes a basis for comparison between ES assessments in different ecosystems and countries25, the final version of CICES (V4.3) is the classification system used in this research.

CICES follows a hierarchical structure as a way to allow its users to select the most appropriate level of detail required to their application. At the highest hierarchical level (called ‘Sections’) there are three broad categories11. Below these major ‘Sections’ are nested a series of ‘Divisions’, ‘Groups’ and ‘Classes’, as shown in Table 2. CICES considers the outputs of ecosystems dependent on living processes. Abiotic outputs are classified separately and are hierarchically divided in ‘Sections’, Divisions’ and ‘Groups’. CICES V4.3 does not include abiotic materials and renewable abiotic energy in its main working matrix, but as an accompanying matrix. In the present study both matrixes will be considered.

Table 2 Summary of the ES provided by the case study area (CW denotes coastal waters; TW denotes transitional waters; FW denotes freshwaters; TE denotes terrestrial ecosystems (including agro-ecosystems); • denotes presence of the ES; *stands for adaptations of CICES V4.3 during the mapping process).
Full size table

Mapping of ecosystem services

From the variety of ES mapping methodologies consulted, the methodology adopted relates to Burkhard et al.4 and Medcalf et al.26 methodologies.

Similarly to Burkhard et al.4, the adopted mapping approach relies on the principle that ecosystems differ in their capacity to provide ES. Besides, as Medcalf et al.26, it assumes that there is a significant set of available information that can be used to either map a given ES or provide a proxy for the service so that it can be mapped. Like Burkhard et al.4, this approach uses existing spatial data on habitats and land use/land cover (LU/LC) to demonstrate ecosystems’ capacity to provide ES in a spatial manner. This is complemented with alternative data on management plans, administrative procedures and legal instruments (e.g., designated areas for, or areas restricted to, certain activities), biophysical aspects (e.g., soil typology, evapotranspiration rates, position in the landscape, like next to a watercourse), and human activities (e.g., recreational areas, shellfish collecting areas), similarly to Medcalf et al.26, enabling to reduce the uncertainty associated and provide maps of actual ES. This approach uses a set of qualitative indicators that are assigned to each ES and abiotic outputs to indicate its presence. A Geographic Information System (GIS) based approach is used to map these ecological, biophysical and socioeconomic features that illustrate provisioning, regulating and maintenance, and cultural services. These maps can be organized by ‘Division’ or ‘Group’ (under CICES classification) depending on their complexity in order to obtain clear and visually attractive maps, easily understandable by technicians, planners and other stakeholder groups.

Data acquisition

An important phase in ES mapping is data acquisition, database development, and quality assessment26,27. In order to implement the proposed methodology, several data needs to be collected and analysed regarding physical, ecological, socioeconomic, territorial and structural aspects of the study area and its surroundings. Because ecosystems are one of the primary landscape units that provide ES, it is important to analyse their spatial distribution28. Other features that contribute to the study area characterization and the accuracy of ES maps are: morphology of the territory (elevation and bathymetry); river network and water bodies; geology; soil typology; LU/LC; and social and economic data (e.g., population density, buildings and infrastructures), which provide information regarding the major demands alongside the main drivers of change and pressures29. Therefore, the resulting geodatabase is composed by multi-source geospatial data in GIS format, which needs to be assessed regarding its spatial coverage, data projections, suitability, date and frequency of updates27.

Study area

Coastal territories are considerably diverse and complex, particularly those integrating estuaries or coastal lagoons such as Ria de Aveiro, as they are transition areas between freshwater and marine systems, and between aquatic and terrestrial systems30. Moreover, coastal areas are characterized by an intense human presence and activity, being subject to powerful and growing pressures and impacts.

The study area is located in the northwest coast of Portugal (40°38′N, 08°45′W) and is integrated in the Vouga river catchment area (368,521 ha) (Fig. 2). Ria de Aveiro coastal region integrates urban areas and a wide range of natural and semi-natural habitats, namely coastal and halophytic habitats (e.g., coastal dunes, coastal lagoon, salt marshes, salt pans, intertidal flats), agro-ecosystems, woodlands and freshwater rivers and lakes31. The marine part of the study area is mostly composed by infralittoral fine sand and circalittoral fine sand32. Due to the great diversity in habitats and bird species, the study area has been integrated in the Nature 2000 network as a Special Protection Area (SPA) and a Site of Community Importance (SCI). It also incorporates the São Jacinto Dunes Nature Reserve, the Ramsar Site Pateira de Fermentelos Lake, and Águeda and Cértima Valleys. This region generates significant economic benefits originated from land and maritime activities such as agriculture and livestock, fishing, maritime port activity, industry, tourism and recreation30.

Figure 2: Ria de Aveiro study area and its main territorial elements.
figure 2

Map tiles by Stamen Design (http://stamen.com) under CC BY 3.0 (http://creativecommons.org/licenses/by/3.0/). Data by OpenStreetMap (http://openstreetmap.org), licensed under CC BY-SA. The license terms can be found on the following link: http://creativecommons.org/licenses/by-sa/3.0/. Sources of ESRI World Topographic Map: Esri, HERE, DeLorme, Intermap, increment P Corp., GEBCO, USGS, FAO, NPS, NRCAN, GeoBase, IGN, Kadaster NL, Ordnance Survey, Esri Japan, METI, Esri China (Hong Kong), swisstopo, MapmyIndia, {copyright, serif} OpenStreetMap contributors, GIS User Community. Pictures copyright: 1 ©Lisa Sousa, 2 ©Nuno Rodrigues, 3 ©Ana Lillebø, 4 ©Célia Laranjeira.

Full size image

The governance framework of Ria de Aveiro coastal region is characterized by the involvement of a variety of government departments (e.g., Portuguese Environmental Agency – APA I.P; Regional Development Coordination Commission of the Centre – CCDRC; Aveiro Region Intermunicipal Community – CIRA), non-governmental agencies and other stakeholders (e.g. land-owners, fishermen associations, sports associations). In addition, 11 municipalities have jurisdiction over different parts of the case study30,33. The spatial planning and management of Ria de Aveiro coastal region is performed by programs and plans of national, regional, inter-municipal and municipal levels. Figure 3 presents a schematic representation of the spatial incidence (marine and terrestrial) and governance framework of the relevant plans and programmes for the study area.

Figure 3: Territorial incidence of case study’s relevant planning instruments for both terrestrial and marine space (after Sousa et al.30).
figure 3

Drawings were generated with Microsoft Visio Professional 2016.

Full size image

Results

The case study boundaries definition

Ecosystem-based management is an integrated approach to management that recognises that human uses and ecosystem health are interdependent. Although it considers ecological, social and cultural objectives, ecological sustainability is the primary goal of management34,35. To act in accordance with the objectives of this approach, the delineation of the case study boundaries took into account the concepts of structural connectivity (based entirely on landscape structure36, i.e. not considering, for now, the functional responses) and complementarity between the different natural and semi-natural systems – marine, transitional, riverine, and terrestrial, including human-shaped ecosystems (agro-ecosystems). However, authors acknowledge the importance of functional connectivity, which is of paramount importance when considering habitats sensitivity and vulnerability, on risk assessment. To comply with the considered concepts, the study area includes, besides the Ria de Aveiro coastal lagoon (the focus of this research): i) the ecological structures considered complementary to the lagoon – e.g., the coastal strip between Furadouro (in the North) and Praia de Mira (in the South) and the lagoon’s margins, responsible for its shape; and the Pateira de Fermentelos freshwater lake, an important wetland from the conservation point of view; and ii) the connective structures – e.g., rivers that flow into the lagoon, riparian corridors, and agricultural fields (such as ‘bocage’) between the lagoon and the Pateira de Fermentelos, which are components of the landscape that can facilitate the biological flows, i.e., which provide important functional connectivity.

Nevertheless, adjustments to the marine and terrestrial boundaries were made in order to match the existing management and spatial planning framework30,37 (Fig. 3), namely the Vouga Estuary Program (under development) and the Coastal Zone Program (CZP) for the stretch Ovar – Marinha Grande:

  1. i

    inclusion of the estuary margins (50 m measured from the limit of the water body);

  2. ii

    adjustments in the marine boundary up to a depth of 30 m in order to include the marine area covered by the CZP.

Additionally, adjustments were made to include the limits of the SPA Ria de Aveiro, the SCI Ria de Aveiro, the Ramsar Site Pateira de Fermentelos Lake, and Águeda and Cértima Valleys.

As result of the adjustments, the study area comprises a total area of 62,535ha of which 30,779ha are marine and 31,756ha are terrestrial. Its elevation ranges between 0–20 m in most of the study area, reaching 80 m in the proximity of Pateira de Fermentelos.

Ria de Aveiro coastal region - ecosystem services identification

A total of 59 ES and abiotic outputs were initially considered in this analysis, following the CICES system for both ES and abiotic outputs. The identification of the ES provided by the study area was based on Barbier et al.38, Maltby et al.39, Salomidi et al.40, ADAPT-MED41, Liquete et al.8, Sousa et al.42, and Lillebø et al.31.

For the purpose of presenting the ES identified both in a concise way and without losing detailed information, the study area was divided in four major ecosystem typologies, similar to those defined by Maes et al.3: coastal waters, transitional waters, freshwaters, terrestrial ecosystems (including agro-ecosystems). ES related to green infrastructures within urban areas were not considered in this study since the current spatial scale does not allow the degree of detail required for that type of analysis.

The ES and abiotic outputs delivered by the Ria de Aveiro coastal region were identified and briefly described in Table 2 (a detailed description is given in Supplementary Tables S1 and S2). Regarding abiotic outputs, it was identified the provision of mineral nutritional substances (e.g. marine salt), associated to transitional waters; non-metallic materials (e.g. sand and gravel), associated to coastal waters; and weather regulation, both associated to transitional and freshwaters.

Spatial distribution of ecosystem services

A set of qualitative indicators, summarized in Table 3, were identified and assigned to each ES and abiotic outputs. This selection took into account the existence and accessibility of spatial data, which serves as a proxy of the service so it could be mapped. Only the ES and abiotic outputs identified in Table 2 were considered in Table 3; however, not all have an assigned indicator due to the lack of spatial data (e.g. hunting, under the ES class ‘wild animals and their outputs’) or due to the nature of the ES (e.g. inspiration and sense of place under ‘aesthetic’ ES class, and traditional boats under ‘heritage, cultural’). Additionally, during the ES and abiotic outputs mapping exercise, some issues were identified regarding the application of the CICES (see Discussion section), resulting in small adaptations of CICES V4.3 table, highlighted in Table 2 with an asterisk. The full list of indicators used to map the ES and abiotic outputs delivered by Ria de Aveiro coastal region, along with the typology and source of data used are given in Supplementary Tables S1 and S2.

Table 3 Summary of the indicators used to map the ES and abiotic outputs provided by the case study (NA denotes not applicable).
Full size table

Whenever possible and appropriate, data on administrative processes and legal instruments were used in combination with spatial data as a mean to achieve more accurate maps and consistent with the case study reality. This was especially the case of provisioning and cultural services (e.g. Salicornia sp. harvesting, fishing and shellfish collecting, archaeological sites, protected areas). The class ‘wild plants, algae and their outputs’ within division ‘nutrition’, for instance, refers to wild glasswort Salicornia sp., which is present in almost all salt pans; however, only one of them is certified and authorized to commercialize it. Thus, only this salt pan was mapped (Fig. 4). Another example is the class ‘wild animals and their outputs’ under the same division. The entire coastal lagoon has potential for fishing; however, this activity is forbidden in certain areas within Aveiro harbour jurisdiction to ensure the safety of navigation, people and goods (Legal Notice no. 01/2012). This results in the exclusion of these areas from the map (Fig. 4).

Figure 4: Spatial distribution of ES classes and abiotic outputs present in Ria de Aveiro coastal region under the division ‘nutrition’.
figure 4

(a) Detail of Ria de Aveiro central area. (b) Fishing restricted area (Public Notice no. 01/2012). Map generated with ArcGIS 10.

Full size image

The proposed indicators for mapping regulating and maintenance services are dominantly based on the presence and distribution of habitats units; however, in the case of the ES classes ‘hydrological cycle and water flow maintenance’ (within the division ‘mediation of flows’) and ‘weathering processes’ (within the division ‘maintenance of physical, chemical, biological conditions’) complementary data was considered to achieve a more accurate representation of the ES. Therefore, for the ‘hydrological cycle and water flow maintenance’ case, areas with higher evapotranspiration rates were considered beyond the riparian and alluvial habitats. For the ‘weathering processes’ case, the type of soil was combined with the land cover and floodplain areas (Fig. 5), i.e. the ES is represented by the areas which integrates both fluvisols and forests, or both fluvisols and floodplain areas. Although the climate and topography have influence in the weathering processes, they were not considered since these factors are quite regular within the study area.

Figure 5: Spatial distribution of ES classes present in Ria de Aveiro coastal region under the group ‘soil formation and composition’.
figure 5

(a) Detail of the spatial distribution of forest and floodplain areas. (b) Detail of the spatial distribution of fluvisols. Map generated in ArcGIS 10.

Full size image

A significant number of cultural services are geometrically represented by points, not only because of the reduced area (even at this scale) occupied by the ES (e.g. subaquatic archaeological sites), but also due to the nature of the ES. The ES class ‘entertainment’, for instance, refers to ex-situ experiences of the Ria de Aveiro coastal region through festivals and fairs. The location of such festivals and fairs is mapped despite the fact that the service is provided by the existence of salt pans in the coastal lagoon, for example, or the existence of a long tradition of fishing and shellfish collecting in the lagoon. Another example of this technical detail is the use of points to map birdwatching and landscape enjoyment under the ‘experiential use of plants, animals and land-/seascapes in different environmental settings’ class. The locations that allow a better enjoyment of the ES are mapped, despite the fact that is the landscape characteristics, or the birds’ diversity that provide the service (Fig. 6).

Figure 6: Spatial distribution of ES classes present in Ria de Aveiro coastal region under the division ‘physical and intellectual interactions with biota, ecosystems, and landscapes’.
figure 6

Map generated with ArcGIS 10.

Full size image

Discussion

The process of defining the study area boundaries, identifying and mapping the ES raised some challenges. The fact that coastal lagoons are interface ecosystems hampers the definition of strict boundaries, essential for the scope of this research. Therefore, there was the need to ensure the link between marine, transitional and riverine systems without losing focus on the main feature of this study, the Ria de Aveiro coastal lagoon. Boundaries definition is a crucial step for the identification and classification of ES, since it enables the understanding on the type of ecosystems present. This is particularly important when considering the existence of land-based recreation activities, for instance, driven by the coastal lagoon (e.g. bird watching, angling, walking), or for example the role of coastal dunes in contributing to the lagoon’s integrity. Moreover, the compliance of the study area with the legal framework is considered an advantage since it enables the integration of these results in the spatial planning and management tools, and facilitates funding acquisition for its implementation or development.

CICES reveals to be suitable to classify ES of complex coastal regions, with minor adjustments according to the scale of analysis, data availability, and its biophysical and sociocultural characteristics. While establishing a common typology of ES – underpinned on clear concepts and principles – CICES is flexible enough to be adapted to different realities, conditions, scales (e.g. Belgium25, Ria de Aveiro - this study), and purposes (e.g. mapping, economic valuation). This is possible because i) CICES is based on clear and well defined concepts; ii) it follows a hierarchical structure that allows the adoption of different levels of detail according to the user’s interest; iii) each level of the hierarchical structure has been designed in a way that there is no overlapping nor redundancy; and iv) it is focused on “final” services or outputs from ecosystems that people use or value, in order to avoid double counting11,25. However, the ‘regulation and maintenance’ section is relatively specific and requires scientific in-depth knowledge about biological and physic-chemical processes, which can be a constraint to its use by decision-makers, technicians, planners, or even by general public.

During the mapping exercise, some issues arose driven by different reasons, which led to minor adaptations to the CICES V4.3 in order to better integrate the case study conditions, scale of analysis and available data:

  1. i

    Lack of spatially detailed information to distinguish the biota (micro-organisms, algae, plants, and animals) from the ecosystem. This is the case of the ES classes ‘filtration/ sequestration/ storage/ accumulation by microorganisms, algae, plants and animals’ and ‘filtration/ sequestration/ storage/ accumulation by ecosystems’ within the division ‘mediation of waste, toxics and other nuisances’. Thus, the ES was spatially represented by the ecosystem components (salt marshes and reed marshes, intertidal flats, coastal waters, riparian and alluvial forests) and referred as ‘filtration/ sequestration/ storage/ accumulation by biota and ecosystems’.

  2. ii

    Lack of supporting information for the ES group ‘water conditions’ regarding the underlying service to achieve such state. In the mapping exercise, this group was excluded since it was considered that the proposed indicators3 reflect an ecosystem status based on chemical indicators, specifically the indicators for the trophic status, and not the provided service. Therefore, it reflects the status due to environmental pressures (eutrophication) instead of the provided service. On the other hand, ‘water conditions’ in the scope of the WFD includes also the biological indicators and the priority substances.

  3. iii

    Insufficient knowledge (particularly about the species and their distribution) regarding “natural” biological control has led to the use of the abundance and distribution of alien species or host-species as proxy indicators for the ‘pest control’ (e.g. Maes et al.3). Again, the proposed indicator reflects the resistance to the environmental pressure (alien species) and not the provided service. However, despite of acknowledging the existence of alien species in the case study (Table 2), this service was not mapped to avoid its misinterpretation or undermine the communication with the technicians, the general public and other stakeholder groups. Nevertheless, authors acknowledge that the spatial distribution of the alien species, and its monitoring, is of paramount importance for the management of ecosystems, and should be considered during a vulnerability assessment.

  4. iv

    Ambiguity and subjectivity associated to the group ‘other cultural outputs’ of the division ‘spiritual, symbolic and other interaction with biota, ecosystems, and land-/seascapes [environmental settings]’. In line with the CICES adaptation to Belgium25, this ES group was perceived as part of a valuation analysis and therefore it was excluded from the mapping.

The varying quality, scale and accuracy of the collected data created a barrier, requiring data refinement, reclassification, and projection. Data preparation involved projection to the same coordinate system (in this case ETRS 89 - European Terrestrial Reference System 1989), and data refinement, particularly regarding data on habitat distribution, LU/LC, and seabed benthic habitats, which were collected for the study area from regional, national, and European sources, covering the years 2011, 2007, and 2014, respectively. These pre-existing, but scattered, data was used as starting point to display ecosystems’ spatial distribution, since they are the primary landscape unit that provide ES. Whenever possible, preference was given to more detailed information.

For the terrestrial area overlapping the SPA Ria de Aveiro it was used the habitat map produced by AMBIECO43 in the scope of the Characterization Study on Ria de Aveiro Ecological Quality for the Polis Litoral Ria de Aveiro. The habitat map production was based on rectified orthophoto images (from 2005 with spatial resolution of 0.5–1 m) from the Portuguese Geographic Institute (IGP), and complemented with latest information from Bing (2009) and GoogleEarth (2009, 2010, 2011). For the remaining terrestrial area it was used the second level (15 classes) of LU/LC for Continental Portugal (COS2007), produced by IGP44. COS2007 was produced based on visual interpretation of rectified and high resolution (50 cm) orthophoto images. COS2007 is in vector format and has a minimum mapping unit of 1ha.

For the marine part of the study area (up to 30 m depth) it was used the benthic habitat map from MESHAtlantic project (last updated in February 2014), available in the European Marine Observation and Data Network (EMODnet) website (http://www.emodnet-hydrography.eu/). The MESHAtlantic project covers over 356,000 km2 of seabed habitats of the European North Atlantic Ocean and used a broad-scale mapping method, proposed within the INTERREG MESH project, which is based on available information or on data derived from mathematical models of the marine environment. The broad-scale map is a map of the physical characteristics of the habitats with a 250 m grid resolution (which is roughly equivalent to a scale of 1:1,000,000) and uses the level 4 of the European Nature Information System (EUNIS) classification habitat types45.

In order to combine these three sources of information, the habitat classification was refined to harmonize the LU/LC taxonomies with the habitat classification (Fig. 7 and Supplementary Table S3).

Figure 7
figure 7

Detail of sources/typologies of data-(a) Habitats from AMBIECO/ PLRA, 2011; (b) Benthic habitats from MESHAtlantic, 2014; (c) COS2007, from IGP, 2010- used to obtain the final habitat map (d) for the Ria de Aveiro coastal region. Map generated with ArcGIS 10.

Full size image

All these challenges, along with the significant complexity of social-ecological systems and the scientific attempts to cope with that complexity46,47,48, contribute to a certain degree of uncertainty that must be internalized when using and communicating the results. Bellow we identify some of the sources of uncertainty:

  1. i

    Generalization and categorization used to reduce complex landscapes into a limited number of LU/LC, or habitat classes47. In the case study of Ria de Aveiro coastal region, whenever possible, it was given preference to habitat data, which was the most accurate available information. In the absence of more detailed information, the second level of COS2007 and the MESHAtlantic benthic habitat were used.

  2. ii

    Another source of uncertainty is the spatial and temporal mismatches of different sources of data. For example, in Ria de Aveiro coastal region, spatial data from different years and with different spatial resolutions had to be combined in order to fill the spatial gaps in habitat map. Moreover, even the most recent available data has more than five years.

  3. iii

    The ES classification system itself can be a source of uncertainty because of the ambiguity present in some classes, as previously referred. Some ES are difficult to assign to specific spatial units. For example, ‘aesthetic’ class and ‘other cultural outputs’ group within cultural services are appreciated in a very subjective manner and related to the landscape compositions47 and other social factors (e.g., cultural).

  4. iv

    Assumptions made in the course of the mapping, for example regarding the ecological status of the biotope and its ability to provide a certain ES.

Conclusions

The proposed framework proved to be suitable for addressing ES in complex coastal regions. On the one hand, it uses clear and objective criteria for delineating the geographic area, respecting the connectivity of natural systems but also the complexity of the governance framework, usual in these systems. Therefore, the analysis at a regional scale and the integration of several ecosystem typologies is seen as crucial for such socio-ecological systems. The use of the internationally accepted CICES classification, although adapted to the case study reality and scale, is seen as an advantage, allowing comparisons with other studies. On the other hand, the fact that the mapping approach is based on existing and available data, considers a wide range of ES and abiotic outputs, and uses mainstream software, means that this approach should be somehow easily replicable by technicians, and planners without investing large amounts of human and economic resources.

Exploring the application of CICES and an ES mapping approach based on qualitative indicators to Ria de Aveiro coastal region also allowed the identification of a number of pertinent issues. For instance, because the classes within regulation and maintenance services are relatively specific and require scientific-specific knowledge about biological and physic-chemical processes, CICES application by decision-makers, technicians, and planners can be demanding. Regarding the mapping approach, inspiration, and sense of place services within the ES class ‘aesthetic’ provided by the Ria de Aveiro coastal region, which is deeply present in the region30, still remains a challenge. Additionally, the mapping approach assumes that every part of a given ecosystem is of equal value with regard to its capacity to provide ES, without taking into consideration the ecosystems’ health, or the fact that certain habitats might have comparatively higher or lower potential to provide a certain service. The integration and analysis of additional information (e.g., ecosystem quality status data, the design of rules that could help grade the importance of different habitats capacity in providing ES) in the mapping process opens further opportunities.

Additional Information

How to cite this article: Sousa, L. P. et al. Ecosystem services provided by a complex coastal region: challenges of classification and mapping. Sci. Rep. 6, 22782; doi: 10.1038/srep22782 (2016).