3 Classification systems and their different typologies

3.1 Presentation of five selected classification systems

Although in the ecological literature, the key terms “ecosystem process”, “ecosystem function”, “ecosystem service” and “benefit” have been subject to various and sometimes contradictory interpretations, a wide range of authors have attempted to provide a systematic typology and comprehensive framework for integrated assessment and valuation of ecosystem goods and services (see Daily, 1997*; de Groot et al., 2002*; MEA, 2005*; de Groot, 2006*; Boyd and Banzhaf, 2007*; Fisher and Turner, 2008*). However, because of the dynamic and complexity of ecosystems a single, consistent classification typology is difficult to develop (Costanza, 2008*). There are many useful ways to classify ecosystem goods and services dependent on the different purposes of use.

Since a pluralism of typologies exists, we only illustrate some selected examples, which demonstrate different approaches and developments to classify ecosystem functions and services. Five different classification systems are presented, which are applied in many assessments and are used often as basis for further classification developments (Table 1). We have selected studies, which have shaped and differentiated the ecosystem service research community from the beginning (Costanza et al., 1997*; Daily, 1999*; MEA, 2003*) as well as typologies aiming at integrating the concept of ecosystem services into landscape planning and management within a European context (Bastian and Schreiber, 1999*; de Groot et al., 2010*). In addition, two further classification approaches are presented, which show examples for further developments and adaption of the current typologies in the literature for regional as well as international integrated landscape planning projects.

Costanza et al. (1997*) tried to estimate the current economic value of renewable ecosystem services for 16 biomes, based on published studies and a few original calculations. For the purposes of this analysis the selected ecosystem services were categorised into 17 major groups. According to Costanza et al. (1997*) ecosystem services represent the benefits humans derive, directly or indirectly, from ecosystem functions. Some ecosystem services are the product of more than one function, and one single function can contribute to two or more services. The classified ecosystem services represent the basis for further studies (e.g. de Groot et al., 2002*; MEA, 2005*; de Groot, 2006*).

According to Daily (1999*) natural ecosystems and their related biodiversity are seen as capital assets that will yield a wide range of life-supporting goods and services over time. Benefits, which derive from ecosystems, will therefore enhance human welfare. In order to support sustainable ecosystem service management, Daily (1997) developed a conceptual framework for assessing ecosystem services and their trade-offs and revised it two years later (Daily, 1999*). The “new” classification system encompasses the production of goods, regeneration processes, stabilizing processes, life-fulfilling functions, and conservation of options.

Using the definition of (Costanza et al., 1997*) [see Section 2 on definitions], the Millennium Ecosystem Assessment (MEA, 2003*) provides a simple typology of services that has been widely taken-up in the international research and policy literature. Four broad types of service are suggested: “Provisioning services”, “Regulating services”, “Cultural services” and “Supporting services”. This classification is understandably not meant to fit all purposes, which has been pointed out for contexts regarding environmental accounting, landscape management and valuation, for which alternative classifications have been proposed (e.g. Boyd and Banzhaf, 2007*; Wallace, 2007*; Fisher and Turner, 2008).

Following de Groot et al. (2010*) ecosystem functions are intermediate between processes and services and can therefore be defined as the “capacity of ecosystems to provide goods and services that satisfy human needs, directly and indirectly (de Groot, 1992*). The provided typology is mainly based on the MEA (2003*) and de Groot (2006*). Four broad types of services are distinguished: “provisioning services”, “regulating services”, “habitat or supporting services” and “cultural and amenity services”. This classification concept was established aiming at integrating the concept of ecosystem services and values into landscape planning, management and decision making (de Groot et al., 2010*).

Bastian and Schreiber (1999*), that are well known in the German speaking community base their classification approach on a long lasted research history in landscape functioning and management. The so-called landscape functions are divided into three groups: “production functions” (economic functions), “regulation functions” (ecological functions) and “habitat function” (social function). Each group is again classified into main-functions and sub-functions so that the cause and effect chains and interactions between land-use demand on the one hand and landscape structure on the other hand are observable (Bastian, 1991, 1997; Bastian and Schreiber, 1999*).

3.2 Comparison of different typologies

Whereas Costanza et al. (1997*), the MEA (2005*), and de Groot et al. (2010*) focus on ecosystem services, Bastian and Schreiber (1999) refer to landscape functions (Table 1). Daily (1999), in comparison to them, includes in her classification both goods, processes and functions.

The typology of the ecosystem goods, services and functions is among these five broadly the same (except for the services of Costanza et al. (1997*), which are often used as the basis for further developments). The groups “provisioning services”, “production of goods” as well as “production function” represent the presence of a large variety of living biomass, which provides many goods for human consumption e.g. food, raw materials and genetic material. “Regulation” or “regeneration processes” relate to the capacity of ecosystems to regulate essential ecological processes and life support systems. Whereas Daily separates the group “stabilizing processes” from “regeneration processes”, the MEA introduces the group “supporting services”. In contrast to the others, de Groot et al. (2010*) include in their system the group “habitat or supporting” services, which are limited to two services (gene pool protection and nursery habitat). Thereby it is stressed that ecosystems provide refuge and reproduction-habitat that support ecological balance and evolutionary processes. Bastian and Schreiber also include “habitat function” but in the terms of social functions, that can be compared with the “cultural services” and “life-fulfilling functions” of the other authors. Although the typologies of these selected classification systems seem to be similar, the allocation of the services is varying, due to the different definitions of ecosystem goods, services, processes and functions and due to the different purposes of the assessments.

3.3 The problem of double counting

According to Wallace (2007), most of the proposed classification systems confuse ends with means. It should probably be distinguished between the benefit people enjoy and the mechanisms that give rise to that benefit. Assessed against these properties, any classification system containing both ecosystem processes and the outcomes of those processes within the same set will produce redundancy (Wallace, 2008*). The fact that different ecosystem functions can deliver similar or equal services may lead to double counting in the assessment of the total value of ecosystems. Particularly, the regulation services are often still included in other services (Hein et al., 2006*). For instance, “pollination”, which is among others important for the maintenance of fruit production, is already included in the service “production of food”. Therefore, Hein et al. (2006*) propose to include only regulation services if they provide a direct benefit to people living in the area or/and if they have an impact outside the ecosystem of consideration. Costanza et al. (1997*) suggest establishing a general equilibrium framework that could directly incorporate the interdependence between ecosystems functions and services. Another approach to avoid double counting is distinguishing between final and intermediate goods, when valuating the total value (see Boyd and Banzhaf, 2007). Mäler et al. (2008) e.g. reorganized the MEA classification so that provisioning and cultural services are merged into a new category, final services, and the supporting and regulating services are melded into the category intermediary services. The reason for this is that both the cultural and provisioning services are affecting human well-being directly, whereas the two others are doing that only indirectly.

The TEEB project, which is mainly based on the MEA classification, shifted “supporting services” such as nutrient cycling and food-chain dynamics to ecological processes. The “habitat services”, instead, has been identified as a separate category to stress the importance of ecosystems to provide habitat for migratory species and gene-pool “protectors” (TEEB, 2010*).

3.4 Further developments of classification systems

There exists a wide range of other useful ways to classify ecosystem functions, goods and services, like the suggestions from Costanza (2008) to classify by “spatial characteristics” or by the “excludability/rivalness” status of ecosystem services. The following presented classification systems demonstrate examples how the concept of ecosystem services can be applied to advanced international sustainability impact assessment projects as well as a comprehensive framework for analysing landscape functions in a coherent system.

The Integrated Project SENSOR (Helming et al., 2008*) aimed at developing ex ante Sustainability Impact Assessment Tools to support decision making on policies related to multifunctional land use in European regions and abroad. In the course of this project the concept of Land Use Functions (LUFs) (Pérez-Soba et al., 2008), which are defined by the different land uses as the private and public goods and services, was developed. These functions include the most relevant economic, environmental and societal aspects of a region. Each LUF is characterised by a set of key indicators that assess the “impact issues” defined in the EU Impact Assessment Guidelines (European Commission, 2005). Nine LUFs were defined: The societal LUFs include “provision of work”, “human health” as well as “recreation and cultural functions”. Whereas the economic LUFs encompass “residential and land independent production”, “land-based production” and “transport functions”, the environmental LUFs cover “provision of abiotic resources”, “support and provision of biotic resources” and “maintenance of ecosystem processes”.

In comparison to other current classification systems a wide range of functions has been aggregated to three main function groups each again divided into three LUFs. On the on hand such a slim framework demonstrates a comprehensible communication tool to stakeholders, however on the other hand some loss of information has to be accepted. Great emphasis had put on reaching a balance between the main function groups within the assessments. However, this emerged very difficult as the assessments of the functions groups societal, economic and environmental are based on different methods as well as within different spatial scales.

Recently a classification based on the Land Use Function concept has been provided including two main groups, namely the active and passive landscape functions (Konkoly-Gyuró, in press). Whereas the passive functions are divided into “regulating and life sustaining functions” of the natural systems (environmental regulation, habitat protection, biomass generation) and the “potentials” (biomass, row material production and provision of territory for the different land uses and provision of information and aesthetics), the active functions are the services provided by human activities and artificial territories (settlements, infrastructure networks, recreation- and agricultural surfaces etc.). Considering the core idea of this concept, namely focusing on natural as well as human introduced landscape functions, it can be concluded that the benefits derived from non-natural landscapes transformed by human activities have also be taken into account into decision making. This coincides with the recently emerged approach that well-being can be understood as socio-cultural constructions of modernity, which often comply with the economic system (Eckersley, 2005). However, it is questionable if human transformed landscape functions are equally important as functions derived from natural ecosystems.