Ecosystems and the services they provide are essential in ensuring human well-being and providing the basis for sustainable socio-economic development (South African Department of Environmental Affairs (DEA) 2011). The global population is predicted to grow from 7.3 billion in 2015 to 11.2 billion in 2100 (United Nations, Department of Economic and Social Affairs (UN DESA) 2015). Approximately 60% of the world’s ecosystem services (17 of the 21 assessed), including fisheries and fresh water, is already degraded or used unsustainably (Millennium Ecosystem Assessment (MEA) 2005). Globally, humans are at risk of pushing earth systems out of the fairly stable, 11 700-year long Holocene epoch, to a state likely to be much less hospitable for anthropogenic development (Intergovernmental Panel on Climate Change (IPCC) 2012, 2013; Steffen et al. 2004, 2015). This unsustainable development path is expected to have implications for ecosystem services and human well-being (McGranahan et al. 2005). In Africa, where the population is expected to increase from 1.2 billion to 4.3 billion between 2015 and 2100 (UN DESA 2015), the capacity of natural resources to produce ecosystem services has already declined as a result of inappropriate management, threatening the social and economic value of these services for the human societies they support (Munang et al. 2011).

It was clear from the Millennium Ecosystem Assessment (MEA 2005) that the management and protection of ecosystem services is vital to support human development. Subsequent to the findings of the MEA has been the rapid growth in science and policy attention on the subject (Convention on Biological Diversity (CBD) 2010; Daily et al. 2009; Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) 2012; TEEB Foundations 2010). Several policies and strategies focused on the conservation of biodiversity – defined as the variability, diversity and ecological complexes amongst living organisms, species and ecosystems (Republic of South Africa 2004) – now explicitly also include objectives to monitor, assess and protect ecosystem services (e.g. CBD 2010, EC 2010). For example, IPBES was established with the key objective to ensure long-term human well-being and sustainable development, through the conservation and sustainable use of biodiversity and ecosystem services (CBD 2010, IPBES 2012). The Economics of Ecosystems and Biodiversity (TEEB) initiative undertaken by the UN Environment Programme and initiated in 2007 (TEEB Foundations 2010) was aimed at mainstreaming the values of biodiversity and ecosystem services into decision-making at all levels (Costanza et al. 2014). Many countries have welcomed the findings from TEEB and plan to take some of the recommendations forward (Kumar & Martinez-Alier 2011).

However, many challenges remain in translating scientific knowledge generated through biodiversity and ecosystem services assessments to real-world decision making (Ruckelshaus et al. 2015). These include the identification of priority ecosystem service areas and the appropriate planning, protection and management of these through policy, or public and private sector practise (Goldman et al. 2008; Laurans et al. 2013).

Aligning ecosystem services and biodiversity objectives became a common conservation strategy to address these challenges, subsequent to the findings of studies on local to regional scale that identified the coincidence of ecosystem service value with biodiversity. The establishment of protected areas was identified as a tool to ensure biodiversity conservation and safeguard ecosystem services (Chan et al. 2006; Naidoo & Adamowicz 2005; Naidoo & Ricketts 2006; Turner et al. 2007). It is known that effectively managed protected areas augment ecosystem services linked to natural areas, such as water regulation, soil and climate (Willemen et al. 2013) and that high levels of biodiversity can play a significant role in either regulating processes and functions of ecosystem services or directly providing them (Balvanera et al. 2006; Diaz et al. 2005). Evidence shows that increased biodiversity has positive effects on ecosystem services (Balvanera et al. 2006; Diaz et al. 2005); however, it is still unclear whether the level of co-occurrence of ecosystems services and biodiversity could allow for synergies towards the achievement of objectives for both (Balvanera et al. 2006; Turner et al. 2007). In cities, there are limited options for managing biodiversity given competing demands for land. For example, in Durban, South Africa, environmental transformation and degradation continue to increase, as substantial human need and poverty still exist (Roberts & O’Donoghue 2013). This adds to the complexities in managing biodiversity and ecosystem services in an urban context.

Another key challenge for environmental management, as emphasised by Escobedo et al. (2011), is the willingness of planners and decision-makers to assess ecosystem services against ecosystem disservices (Lyytimäki & Sipilä 2009). Ecosystem disservices are classified as the nuisances or negative effects to humans from natural or human-impacted ecosystems, such as biological hazards, invasive species, floods, storms, heat waves and pests (Lyytimäki & Sipilä 2009; von Döhren & Haase 2015). The consideration of ecosystem disservices may add a valuable dimension to the ecosystem services management debate; however, in this study we focussed only on beneficial ecosystem services.

In an attempt to identify synergies for the management and protection of selected ecosystem services with biodiversity, this paper uses the greater Durban area (also known as the eThekwini Municipal Area (EMA)) as a case study to identify important areas for ecosystem services provision (called ‘ecosystem service hotspots’) of 13 ecosystem services. We assesses the extent to which these areas are identified as conservation priorities in relation to the eThekwini Municipality Draft Systematic Conservation Plan (2012b); namely, (1) as critical biodiversity areas, (2) as conservation areas and (3) as part of the Durban Metropolitan Open Space System (D’MOSS). We then assess their distribution relative to various categories of ownership and propose opportunities for the protection and management of ecosystem service hotspots in an urban context through policy and practise.

Recent regional and international studies have only considered a limited number of ecosystem services (usually between four and six) (Chan 2006; Egoh et al. 2009; Jantz & Manuel 2013; Pan, Xu & Wu 2013). The availability in Durban of 13 ecosystem service layers (Glenday 2012, Table 1) therefore offered a good opportunity to analyse spatially a larger number of individual ecosystem services on the local scale. These ecosystem service layers were identified to inform the conservation planning process within the municipality and to facilitate the selection of priority areas where management efforts would yield maximum benefits. The current study aimed to further inform the conservation planning process by answering the following questions:

On average, 30.2% of ecosystem service hotspots are located within terrestrial critical biodiversity areas, 51.1% within D’MOSS, with nominal overlaps of 0.3%, 3.9% and 5.1% within estuaries and freshwater critical biodiversity areas and conservation areas, respectively (Table 2). All ecosystem services except phosphorus retention for dams were found to overlap with conservation areas.

Terrestrial critical biodiversity areas contain significant proportions of carbon storage (41.5%), all three flood attenuation services (47.5%–50.4%) and sediment retention service hotspots (28.4%–39.2%) (Table 2). Terrestrial critical biodiversity areas showed minor overlaps with nitrogen and phosphorus retention services (12.1%–19.8%), with the smallest overlap being with the water yield service hotspots (7%).

Despite that all ecosystem services except phosphorus retention for dams were found to overlap to some extent with conservation areas, given the limited areas of overlap, most ecosystem service hotspots are unprotected (Table 2). The small percentage overlap between 1% and 14% of hotspots with conservation areas highlights that substantial proportions of ecosystem services lie outside these areas, leaving them vulnerable to human impact and degradation.

All categories of ecosystem service hotspots were represented within D’MOSS, of which carbon storage (76%) and all flood attenuation hotspots (68%–77%) have the highest representation. About half of all three sediment retention ecosystem service hotspots were contained within D’MOSS (49%–55%), whilst water yield service was identified to have the smallest overlap with D’MOSS (11%).

With respect to the proportions of critical biodiversity and conservation areas found in ecosystem service hotspots, on average, ecosystem service hotspot areas in this study contained small percentages of critical biodiversity areas, with 8.8% of terrestrial critical biodiversity areas, 4.5% of estuary critical biodiversity areas and 6.5% of freshwater critical biodiversity areas (Table 3).

Most ecosystem service hotspots were unprotected (Table 2). Generally, conservation areas performed poorly with regards to the ecosystem services included in this study, with small percentages of conservation areas being contained within hotspots, except for carbon storage (80% of conservation areas overlap with carbon storage hotspot) (Table 3).

Ecosystem services are not necessarily correlated with each other. For ecosystem service ranges, out of 78 possible combinations, only five showed a moderate correlation (between 0.2 and 0.5) and four showed a strong correlation (> 0.5) (Table 2-A1). The three flood attenuation services, i.e. for public infrastructure, private property and population protection, all had strong positive correlations with each other, with the highest correlations between services in this group. A strong positive correlation was also found between sediment retention services important to prevent clogging of storm drains and sedimentation to sewer pipes. Moderate correlations were found between carbon storage and water yield to dams. Water yield to dams also showed a moderate correlation with the sediment retention service to dams. Similarly, nitrogen retention service relative to dams also showed moderate positive correlations with phosphorus retention to dams and nitrogen retention to estuaries.

Correlations between ecosystem service hotspots were lower than those between ecosystem services ranges. Out of 78 possible combinations, only two combinations were moderately correlated (between 0.2 and 0.5) and three had strong correlations (> 0.5). Similar to the ranges, the sediment retention service hotspots showed moderate to strong correlations with each other (Table 3-A1). Nitrogen retention to dams also showed a moderate correlation with nitrogen retention to estuaries. This highlights that a combination of different areas are required to maintain the whole suite of ecosystem services.

The study identified positive correlations between the majority of the ecosystem services assessed and terrestrial critical biodiversity areas. Water yield to dams had the highest correlation with terrestrial critical biodiversity areas, followed by carbon storage and flood attenuation relevant to public infrastructure (Table 4-A1). Whilst overlap analyses indicated that terrestrial critical biodiversity areas generally contribute significantly to all ecosystem service hotspots, correlation analysis on average indicated weak correlations with all services, with weak negative correlations with sediment and phosphorus retention to dams (Table 4-A1). Whilst positive correlations generally imply that ecosystem service values increase with biodiversity values, given the weakness of the correlations, certain areas with high ecosystem service values, for the services assessed in this study, may not necessarily also have high biodiversity values and vice versa.

Whilst the percentage overlap of certain ecosystem service hotspots with critical biodiversity areas and conservation areas may seem negligible, when translated into area, large pieces of land significant for both critical biodiversity and ecosystem service hotspots are available for selection to be managed to satisfy the conservation objectives of both. For example, the lowest percentage overlap of a hotspot with terrestrial critical biodiversity areas was water yield service at 6%, amounting to approximately 7000 ha with high ecosystem service and biodiversity values.

On average, the majority of the ecosystem service hotspot areas either fall within the ITB land (44%) or on privately owned land (27%). This is followed by eThekwini Municipality administered land (7%) and state land (4%), with parastatal ownership constituting only 2% (Table 2, Figure 4, Table 5-A1). No ownership data were available for the remaining 17% of ecosystem service hotspot areas.

Significant proportions of nine ecosystem services hotspots occurred within ITB areas (between 41% and 86% of their area, Table 2, Table 5-A1). Amongst these, nitrogen retention to dams and estuaries (86% and 74% respectively), phosphorus retention to dams (72%) and water yield (64%) were significant. Moderate percentages (between 22% and 29%, Table 2, Table 5-A1) of four ecosystem service hotspots fall within areas for which ownership information was not available. As an example, Figure 5 shows the ownership distribution for the carbon storage hotspots.

The balance of the majority of ecosystem services was contained within private ownership (between 21% and 47%). Ecosystem service hotspots were generally poorly represented within municipal and state ownership categories, except for three sediment retention hotspots (sewer pipe, storm drains and harbour) of which between 15% and 16% are administered by the municipality.

The novel mapping approach used to produce the 13 original ecosystem service layers has certain limitations. For example, InVEST models do not explicitly consider surface-groundwater interactions or impacts of inter and intra annual variability, and some data are not locally calibrated, thereby limiting the exact quantification of services (Glenday 2012). However, the availability of these ecosystem service layers offered a unique opportunity to spatially analyse certain ecosystem services in Durban for the first time. This study is seen as an important step towards identifying potential management synergies between ecosystem services and biodiversity, and identifying strategies towards protecting ecosystem services in an urban context.

There are a number of factors that limit the effective management of ecosystem services in Durban. The vast majority of important ecosystem service areas are located outside of managed conservation areas and within communal lands that are currently under joint ITB and municipal administration (eThekwini Municipality 2012b), facing threats of habitat transformation, invasive alien plant invasions, rapid densification and pollution. Additionally, the generally weak correlations between ecosystem services included in this study and terrestrial critical biodiversity areas indicate that there is limited potential for the selection of areas supplying high levels of both ecosystem services and biodiversity. This finding highlights that more land may be required to conserve and manage both ecosystem services and biodiversity. In an urban context such as Durban, this is at the same time critically important and challenging.

The fact that the study found weak correlations between ecosystem services and that less than 5% of the total hotspot provisioning areas supply three or more ecosystem services may indicate that these services require separate considerations in conservation assessments (Egoh 2009) and may also limit the selection of areas that supply multiple services for management. However, this may only be true in light of the ecosystem services assessed in this study. The inclusion of additional ecosystem services, e.g. cultural and recreational services, may provide for greater selection of areas that supply multiple services for management.

For the management of ecosystem services and biodiversity to be successful, a multidisciplinary approach that involves all stakeholders is needed (Cowling et al. 2008). Varying ecological and social contexts may require different institutional mechanisms to protect ecosystem services, depending on locally based information. Thus, in the context of Durban, strategies for the protection of hotspots may even differ between the various ownership categories given the differences in their institutional structures. Furthermore, those ecosystem service hotspots for which ownership information was not available would also need to be considered as a lack of ownership information will hinder the formation of management partnerships.

Despite these challenges, there are numerous opportunities to enhance ecosystem service protection and management in the urban context. Where biodiversity priority areas overlap with important ecosystem service provisioning areas, conservation strategies aimed at biodiversity will safeguard ecosystem services (Egoh 2009) and those ecosystem services can be used as additional justification for biodiversity conservation.

Given the limited availability of human and financial resources for conservation and management, the small percentages of areas providing numerous ecosystem service hotspots – e.g. approximately 550 ha supplying five service hotspots (Figure 2) – do, however, allow for easy selection of areas where management efforts could yield greater benefits. In addition, overlap analyses of these 550 ha with critical biodiversity areas and D’MOSS would allow for the further refinement of areas to be prioritised for management where greater returns on management investments could be achieved.

Based on the findings that about a third of ecosystem services provisioning areas overlap with terrestrial critical biodiversity areas in Durban, there are numerous opportunities for the co-management of these areas. This is in line with the findings of a number of studies that indicated a direct relationship between ecosystem services and biodiversity (Chan et al. 2006; Egoh 2009; Singh 2002; Turner et al. 2007). However, based on the findings of this study that shows that ecosystem services can exist even in areas with low biodiversity and that ecosystem services cannot be fully accommodated in existing conservation areas and critical biodiversity areas, this justification cannot be used in all cases.

These findings highlight the need for an independent conservation strategy for ecosystem services in Durban which is complementary to biodiversity conservation. This strategy could include measures to ensure that ecosystem services are given consideration either through protected area expansion, in assessing development applications, through appropriate zoning, acquisition by management authorities and by implementing stewardship programmes in communally or privately owned areas. For co-management of ecosystem services and biodiversity, a prioritisation process may be required depending on the aim of management, to either select areas with high ecosystem service values and some biodiversity values, or high biodiversity values with some ecosystem service values. In addition, to safeguard ecosystem service hotspots lying outside of conservation areas and existing DMOSS, an ecosystem service management and protection strategy should, at least, also include the prioritisation of the 550 ha that have a hotspot richness of five or more and also consider the threats services face, as is the case for biodiversity where critically endangered features are given priority (Balvanera et al. 2001; Sisk et al. 1994).

The study identified win-win opportunities to co-manage ecosystem service hotspots and critical biodiversity and conservation areas and opportunities to manage areas supplying numerous ecosystem services. However, whilst these areas allow for easy prioritisation for greater returns on investment, the challenge lies in managing and protecting the large proportion of critical ecosystem service provisioning areas that are vulnerable to transformation outside of these managed areas. This raises the importance of establishing a link between planning and action that would incorporate priority ecosystem service areas within the municipal decision-making framework. This includes the identification of an ecosystem service management and protection strategy that facilitates the selection of ecosystem service priority areas for management and as locations for engagement with communal and private landowners through biodiversity stewardship; using ecosystem service hotspots as triggers for development assessment and approval processes; and the co-management of areas supplying both critical biodiversity and ecosystem services.