Climate change assessment impacts on the coastal area of Maliakos Gulf, Greece

Coastal ecosystems such as estuaries, salt marshes, coastal lakes, lagoons, dunes and intertidal flats, are usually linked to economic and social development, prosperity and flourishing civilization, due to the fertile soil of the riparian zone, the abundance of freshwater and the existence of the sea (Turner et al. 1996). They provide habitat, flood protection, erosion control, water purification, carbon sequestration and opportunities for recreation and environmental and cultural education and research (Barbier et al. 2011), while they contribute to human welfare, both directly and indirectly, and therefore represent part of the total economic value of the planet (Costanza et al. 1997). Simultaneously, coastal wetlands and the economic and social systems that depend upon them, are particularly vulnerable to human-induced changes, such as intensive cultivation, industrial sprawl and physical alteration of the environment (Vitousek et al. 1997; Millennium Ecosystem Assessment 2005). Anthropogenic climate change is expected to exacerbate impacts of other drivers of degradation of wetlands and lead to a reduction in the ecosystem services provided by them (Harley et al. 2006). Especially, sea level rise (SLR) due to global climate change that results in warming of oceans (thermal expansion), melting of ice sheets and glaciers on land, and reduction of liquid water storage on land, is expected to have a significant impact on coastal wetlands and near-coastal systems, since it affects the total wetland area, inundation frequency and duration, and wetland productivity (Morris et al. 2002). The quantitative assessment of the potential physical and economic degradation of wetlands and wetlands' services provided due to human-induced climate change and SLR can be accomplished using a coherent scenario-based approach (Mehvar et al. 2018).

Many studies have focused on the quantification of the wetlands at risk due to SLR and attempted an estimation of the economic loss associated with adversely impacted wetlands. These efforts usually employ a geographic information system (GIS) to assess the inundated areas (e.g., Brander et al. 2012; Blankespoor et al. 2014), while the valuation of vulnerable wetlands (or other land uses) is usually accomplished based on a simple direct costing approach. This approach evaluates the cost by multiplying a quantity loss (land) with the value of the item lost (Bosello et al. 2012). Other more sophisticated methodologies try to include in the analysis more complicated aspects such as the changes in relative prices responding to an initial land loss and the investment in coastal protection and adaptation measures. These analyses are usually performed with models, e.g., FUND (Climate Framework for Uncertainty, Negotiation and Distribution) that is an integrated assessment model of climate change (Tol 1997), DART, a dynamic Computable General Equilibrium (CGE) model (Deke et al. 2001), DIVA (Dynamic Interactive Vulnerability Assessment) model (Hinkel & Klein 2009), and CIAM (Coastal Impact and Adaptation Model; Diaz 2016). Other experts use meta-analysis as a tool to estimate the economic value of wetlands, based on the statistical synthesis of existing studies (Brouwer et al. 1999; Woodward & Wui 2001; Brander et al. 2006; Ghermandi et al. 2008).

The assessment of SLR impact on coastal wetlands on regional scale varies depending on the methodological approach and the special localized particularities. For example, Roebeling et al. (2013) investigated the consequences of SLR and coastal erosion to the ecosystem services value in Europe and reported that for future coastal erosion projection (2006–2050) the total area lost ranges between 3,700 and 5,800 km², coastal wetland areas being affected most severely, leading to a coastal ecosystem service values decrease of 20.10–19.40 × 10⁹ € per year by 2050. Yoskowitz et al. (2017) calculated that fresh and salt marsh area lost in Galveston Bay, Texas, USA wetlands due to 0.69 m SLR by the year 2100 (compared to 2009), will be 21% and 12% of its extent, respectively, which will lead to a value loss of 40 × 10⁶ and 11 × 10⁶ US$ annually, respectively. Mehvar et al. (2019) estimated that the inundation area of the Western coastal area of Bangladesh for climate change scenarios RCP2.6, RCP6.0, and RCP8.5 in 2100 ranges between 0.25 m, through 1.18 m, to 1.77 m SLR respectively, leading to an US$ 0–1 million to US$ 16.5–20 million loss of the total value of ecosystem services.

Previous attempts concerning the assessment of the impact of SLR to coastal zones due to human-induced climate change in Greece did not usually manage to include wetlands in the analysis. Dalianis et al. (1997) estimated the economic impact of 1.0 m SLR by the year 2100 in Greece to be $3.7 × 10⁶, without including the damage from losses to wetlands. Under the research program PESETA, the land lost and the direct and indirect costs of SLR for Europe and Greece for the 2020s and 2080s was estimated, but without examining the effect on wetlands land lost and changes (Bosello et al. 2012). Similarly, Klaoudatos et al. (2015) estimated the area lost and the economic impact of SLR in the Cyclades Islands, Greece for SLR 0.3, 0.6 and 1.0 m without including wetlands. Only Kontogianni et al. (2012) included wetlands in the estimation of financial loss due to SLR for Greece (138 × 10⁶ € and 247 × 10⁶ € for SLR 0.5 and 1.0 m in 2100, respectively).

The scope of this study was to investigate the impact of the climate change-driven SLR specifically on coastal wetlands and in relation to other socio-economic sectors. In order to accomplish that, three SLR scenarios based on Intergovernmental Panel on Climate Change (IPCC) future projections were examined. For each scenario the land area lost was calculated and then depending on the land use (agricultural, settlement, wetland) a value was attributed, so as to estimate the final economic consequences, while also estimating the consequences of damage to the infrastructure (road network) of the area. The main objective was to estimate the inundated areas and the financial impacts of each scenario examined emphasizing wetlands in particular, so as to quantify their sensitivity in relation to other sectors, and to propose mitigation measures to minimize the possible adverse effects. The area chosen for this application was the regional coastal wetlands of Spercheios River and Maliakos Gulf system in Greece.

The Spercheios River basin and Maliakos Gulf are located in the central part of Greece. The wider area is characterized by intense tectonic activity, the presence of geothermal resources, high rates of sediment deposition and small tidal sea level fluctuations (Kraft et al. 1987; Apostolopoulos 2005). Nowadays, human activities such as agriculture and, to a lesser extent, industrial activities (small manufacturing units of agricultural products, olive oil refineries, etc.) impose significant pressures on the inland water resources, while water abstractions mainly for irrigation, water flow regulations and modifications (small weirs, distributor, canalization and partial diversion of the original route of the river especially close to its estuary) comprise the most important hydromorphological alterations of the area (HCMR 2015). These engineering interventions have led to the destabilization of the coastal zone and the alteration of the coastline during the last decades (creation of a new deltaic system in the northern part of the estuary and decrease of the rate of advance or even a retreat of the coastline in the southern part (Pechlivanidou 2012)). Maliakos Gulf is a semi-enclosed, shallow, marine embayment with limited wind-generated waves, subjected to pollution pressures derived mainly from aquaculture and mussel farms, and poor land use management practices in Spercheios River basin (HCMR 2015).

The Spercheios River and Maliakos Gulf wider area have been included in many environmental protection networks, which in many cases overlap (Natura 2000, CORINE biotopes, Wildlife Refuges, etc.). Additionally, in the study area, five coastal wetlands have been recorded in the inventory conducted by Greek Biotope/Wetland Centre (EKBY; Zalidis & Mantzavelas 1995), some of which are not protected under national legislation (Skarfia marsh, Almyropotamos estuary and Vromolimni lagoon). Skarfia comprises typical and extensive salt marshes with a remarkable variety of halophyte communities that host large populations of migratory waterbirds that often use this habitat for wintering and nesting. These marshes provide protection against seawater intrusion and coastal erosion. Vromolimni lagoon is also used by migratory waterbirds for wintering, while Almyropotamos estuary is characterized by typical wetland vegetation (Ministry of Environment Energy and Climate Change – MEECC 2013; Figure 1; Table 1). It should be noted that Livari, located close to the mouth of the river and where the land meets the sea, is actually a small embayment with depths up to 5.0 m and consists of a natural sea park for the reproduction of fish and the development of their juveniles (Georghiou 1996). Livari wetland is not expected to be affected by sea level rise.

The IPCC is an international body established by the World Meteorological Organization (WMO) and United Nations Environment Programme (UNEP) to provide policymakers with regular assessments of the scientific basis of climate change, its impacts and future risks, and options for adaptation and mitigation. Based on the most recent comprehensive assessment report reviewing the latest climate science (Fifth Assessment Report – AR5), four different 21st century pathways of greenhouse gas (GHG) emissions and atmospheric concentrations, air pollutant emissions and land use changes can be projected. These pathways can be described by four Representative Concentration Pathways (RCPs), which include a stringent mitigation scenario (RCP2.6), two intermediate scenarios (RCP4.5 and RCP6.0) and one scenario with very high GHG emissions (RCP8.5). These emission scenarios are directly related to assumptions concerning the economic and population growth, lifestyle, technological development, future land use and land cover and the political decisions and actions regarding the upcoming climate changes (IPCC 2014).

Based on AR5 the SLR is likely to range between 0.40 m and 0.63 m globally in 2100 in relation to 1986–2005, while the worst scenario (RCP8.5) predicts an increase between 0.45 m and 0.82 m until 2100 (IPCC 2014). Local relative SLR can deviate substantially from the global mean due to a number of processes. Especially in the Mediterranean, which is a semi-enclosed basin, connected to the Atlantic Ocean through the Strait of Gibraltar, global climate modelling attempts to assess future sea level change have not delivered a consistent signal (Slangen et al. 2017). In addition to the thermosteric component, the projected contribution from an increase in salinity can be taken into account for the estimation of sea level change in the Mediterranean. The halosteric contribution can lead to −0.10 mm/yr sea level fall due to water contraction, while the addition of salt to the basin in terms of mass can lead to +0.12 mm/yr sea level rise (Jordà & Gomis 2013). As a simplification, these two contradicting effects can be neglected and Mediterranean mean sea level change can be restricted to its thermosteric component (Slangen et al. 2017).

Based on regional models, the SLR in the Mediterranean basin is expected to be lower than the global mean. Based on recent studies, thermosteric SLR ranges between 0.02 and 0.07 m in 2050 compared to 1951–2000 (Carillo et al. 2012), between 0.03 and 0.61 m in 2100 compared to 1950–2000 (Marcos & Tsimplis 2008), between 0.45 m and 0.60 m in 2099 compared to 1961–1990 (Adloff et al. 2015), or between 0.21 and 0.53 m (RCP4.5) and 0.25 and 0.80 m (RCP8.5) compared to 1985–2005 (Vousdoukas et al. 2017). The total SLR in the Aegean is expected to range between 0.14 and 0.274 m by 2040–2050 relative to 1990–2000 (Galassi & Spada 2014). Finally, based on Marcos et al. (2016), the average sea level of the Mediterranean basin is expected to reach 1.02 m (RCP8.5) by the end of the 21st century, taking into account the changes in the Northeast Atlantic sea level caused by circulation and thermal expansion, the difference between the sea level in the Mediterranean and in the northeast Atlantic, and the terrestrial ice melt (Table 2).

For the purposes of this study, it was assumed that in the Mediterranean SLR is likely to range by the year 2100 between 0.3 m (stringent mitigation scenario) and 1.0 m (extremely pessimistic scenario that that does not include any specific climate mitigation target). Therefore, the consequences of SLR of 0.3 m (mild scenario – Sc1), 0.6 m (moderate scenario – Sc2) and 1.0 m (extreme scenario – Sc3) were examined, in order to cover all the projections reviewed. The study area covers the inner, western part of Maliakos Gulf and also the outer, eastern part (between Cape Tapia north and Cape Knimis south).

It should be noted that in the SLR scenarios' implementation, only the climate change-driven SLR was examined, while sea level changes due to natural causes, such as tectonic activity and eustatism, increased sediment transport rate from a river to the coastal area, and coastal morphology and geological structure were not taken into consideration. These sea level drivers, although in some cases cannot be considered negligible, could not be included in the analysis because of the uncertainty that arises due to lack of required information for the specific region. Additionally, the impacts from tectonic activities in the area may cause changes in the sediment transport behaviour but in a much longer-term way than potential changes caused by SLR. Moreover, the anthropogenic factors tend to impose adverse effects on the amount of sediment supplied to the coastal area since water overexploitation and riverbed material extraction are common practices in this particular area (Dimitriou & Stavroulaki 2018).

In order to estimate the coastal area lost due to the SLR, detailed topographic data were necessary. The 0.0 m, 0.3 m, 0.6 m and 1.0 m contours of the study area were retrieved from the 1 × 1 m official digital terrain model (DTM) of the coastal zone, ‘navigable rivers’ and ‘large lakes’ of Greece (date of creation 2008–2009, National Cadastre & Mapping Agency – NCMA SA; after personal communication). The 0.0 m contour of the DTM was considered to be the sea water level of the current state. During this procedure the 1:5,000 scale analogue topographic maps of the Hellenic Military Geographical Service (HMGS) and the 2007–2009 large scale orthophotomaps – LSO with RMSExy <1.41 m (achieved <0.71 m) accuracy (National Cadastre & Mapping Agency SA – NCMA SA) of the area were also employed for higher accuracy during digitizing coastal industrial units, infrastructure such as roads, and vertical coastal structures such as docks and piers. In all cases the height of the docks and piers was considered to be higher than 1.0 m.

The settlement units' boundaries delineation was accomplished using the 1:50,000 scale analogue topographic maps of the HMGS, while the LSO (NCMA SA) were again employed for higher accuracy.

The topographic data processing was accomplished with ArcGIS 10.4 from ESRI software.

• VW is the wetland value (in dollar per square kilometre) at time t in region r;

• t denotes time;

• r denotes region;

• y is per capita income (in dollar per person per year) at time t in region r;

• d is population density (in person per square kilometre) at time t in region r;

• W^C is cumulative wetland loss (in square kilometre) at time t in region r;

• W₁₉₉₀ is the total area (in square kilometre) of wetlands in 1990 in region r;

• α is a parameter, the net present value of the future wetland services; note that we thus account present and future wetland values in the year that the wetland is lost;

• α= 21α’;

• α′ = 280,000 $/km², with a standard deviation of 187,000 $/km²; α is the average of the meta-analysis of Brander et al. (2006); the standard deviation is based on the coefficient of variation of the intercept in their analysis;

• β is a parameter, the income elasticity of wetland value; β = 1.16 (0.46, > 0; based on the results of meta-regression analysis of Brander et al. 2006);

• y₀ is a normalization constant; y₀ = 25,000 $/p/y;

• d₀ is a normalization constant; d₀ = 27.59;

• γ is a parameter, the population density elasticity of wetland value; γ = 0.47 (0.12, > 0, < 1); (based on the results of meta-regression analysis of Brander et al. 2006);

• δ is a parameter, the size elasticity of wetland value; δ = −0.11 (0.05,> −1, < 0); (based on the results of meta-regression analysis of Brander et al. 2006).

The estimation of wetlands' value was applied in the Fthiotida regional unit and 2013 was used as a reference year. As a simplification, it was assumed that population and per capita income does not change significantly compared to 2013 in the specific area. All necessary data concerning population and gross domestic product GDP (per capita income) were retrieved from Hellenic Statistical Authority (2013). The average Euro/US dollar exchange rate for the reference year 2013 was 1 € = 1.3281 $ (European Central Bank 2013).

In order to estimate the financial consequences of the land lost due to SLR, an objective unit value of the land outside urban designated areas was given, based on the adjusted property rates used for taxation purposes (‘the objective values’) given by the Greek Ministry of Finance. The value of the land varies, depending on the local community, the size, the public facilities in the area and the land use (annual crops, irrigated arable land, olive groves, vineyards, pastures, forest, mineral extraction sites or dump sites) (Greek Ministry of Finance 2008). In the specific case, for each land use a basic land value (BA) was attributed, based on the initial basic land value (ABA; depending on the local community) and the special basic land value (EBA; depending on the proximity to road network and seaside). It was assumed that the land has no building greater than 15 m² and that no further exploitation is possible due to legal prohibitions, which would attribute an additional value to the land if absent (AOik – land value and AD – value of possible further exploitation, respectively). Areas not cultivated were considered as pastures.

The cost assessment of SLR due to the land lost used for housing and tourism purposes in urban designated areas was also based on the adjusted property value rates used for taxation purposes in the specific area. The former study showed that the value of residential land in Maliakos Gulf's coastline ranges between 636 €/m² at Kourmousi and 758 €/m² at Kamena Vourla. The calculation was based on the initial basic land value and the value of buildings for each municipal community (Table 3; HCMR 2015).

In order to estimate the financial consequences of the SLR due to the income lost from agricultural activities, which are a major economic sector of the area, the economic return of each crop in Greece was estimated for the year 2013. The necessary data concerning the crop production and unit values at national scale were retrieved from Eurostat (2013). Information concerning the rental price of pastures was retrieved from the Official Journal of the Hellenic Republic (2012). Detailed information concerning the spatial distribution of the agricultural activities in the Spercheios River wider area was retrieved from the Greek Payment Authority of Common Agricultural Policy (CAP) Aid Schemes (OPEKEPE, after personal communication) for the year 2013.

The estimation of the economic consequences of SLR on the infrastructure and the road network of the study area was based on the average total construction cost per kilometre. Based on the audit performed by the European Court of Auditors (2013) that involved co-financed road projects in Germany, Greece, Poland and Spain in the period 2000–2013, the average total cost per kilometre varies from 10,941,402 €/km for motorways, through 6,225,187 €/km for express roads, to 4,159,281 €/km for two-lane roads. In this specific effort, the latter value was used in order to estimate the total cost of the damage on the local road network for each SLR scenario.

Other important public infrastructure (airports, ports) and industrial zones were not affected during the SLR scenarios examined and were not taken into consideration during this study.

In the case of Sc1 (0.3 m SLR), the terrestrial zone lost in the study area was estimated to be 6.2 km², 68% (4.2 km²) of which comprises wetlands (Spercheios Delta, Skarfia wetland and Spercheios River). Most of these areas belong to the Natura 2000 protection network (both in Bird Directive: GR2440005 and Habitat Directive: GR2440002) and have significant ecological value since they host, among others, 170 different bird species and a priority habitat. In this case, the cultivated area lost was estimated to be 2.4 km², since some agricultural activities (rice cultivation, pastures) are located in wetlands. The residential zone to be lost is 76,264 m², located mostly in Kammena Vourla area which is also a tourist resort in the particular area and any changes in the coastal area will have additional impacts on the specific industry. In the case of Sc2 (0.6 m SLR), the terrestrial zone to be lost will rise to 18.9 km², 41% (7.7 km²) of which comprises wetlands. The cultivated area lost was estimated to be 10.0 km² and the residential zone 141,996 m². Finally, in the case of Sc3 (1.0 m SLR) the total terrestrial zone lost will increase to 31.1 km² and the wetlands lost to 12.0 km² (39%). The estimated agricultural area lost was estimated to be 15.6 km² mainly in the lower part of Spercheios river plain and the residential zone that will be inundated reaches 513,396 m², with the most significant impacts in the Panorama area (Figures 2 and 3).

Spercheios Delta is expected to lose the greatest amount of area in all scenarios examined (Sc1: 3.67 km², Sc2: 6.73 km² and Sc3: 0.07 km²). This consists of 11%, 21% and 31% of the total wetland, respectively. Skarfeia wetland, on the other hand, is expected to suffer the greatest percentage lost in relation to the total wetland area, which in the case of Sc3 reaches 74%. Almyropotamos was not affected in Sc1, while in Sc2 and Sc3 it will lose 11% and 22% of the total area, respectively. Finally, Vromolimni lagoon was not affected in any SLR scenario examined (Figures 2 and 3(a)).

The main agricultural crop type that will be affected in the case of SLR is rice, since the area lost in rice fields will range from 0.77 km², through 3.57 km², to 4.69 km² (Sc1, Sc2 and Sc3, respectively). Pasture area lost will range from 0.64 km², through 1.12 km², to 1.58 km², while wheat and spelt area lost will range from 0.13 km², through 0.69 km², to 1.56 km² (Sc1, Sc2 and Sc3, respectively; Figures 3(b) and 4).

Concerning the consequences in urban areas, the greatest loss is expected at Panorama, Kamena Vourla, Kourmousi and Agia Marina settlements in all scenarios examined (Table 3; Figure 5).

Based on the results, the total economic impact of SLR in the Spercheios River and Maliakos Gulf system is expected to be 75.4 × 10⁶ €, 161.7 × 10⁶ € and 510.7 × 10⁶ € in scenarios Sc1, Sc2 and Sc3, respectively. This consists of 3.5%, 7.5% and 23.7% of the GDP of Fthiotida regional unit, respectively (Figure 6(a)).

The value of the wetland lost due to SLR for the reference year 2013 ranges from 3.48 €/m² (4.62 $/m²) for 0.3 m SLR, through 3.52 €/m² (4.68 $/m²) for 0.6 m SLR, to 3.58 €/m² (4.76 $/m²) for 1.0 m SLR. These values lead to an economic loss of 14.7 × 10⁶ €, 27.0 × 10⁶ € and 43.0 × 10⁶ € for Sc1, Sc2 and Sc3, respectively, which consists of 0.7%, 1.3% and 2.0% of the GDP of Fthiotida regional unit, respectively (Figure 6(a)).

Based on the calculations, the economic return of the crops cultivated in the study area (except pastures) range between 0.06 and 1.46 €/m² for the year 2013 (Table 4). The economic impact of SLR on the agricultural activities range from 0.20 × 10⁶ €, through 1.04 × 10⁶ €, to 1.71 × 10⁶ € for Sc1, Sc2 and Sc3, respectively, which consists of 0.01%, 0.05% and 0.08% of the GDP of Fthiotida regional unit, respectively (Figure 6(a)). In addition, the loss of cultivated areas leads to a corresponding financial loss of 1.7 × 10⁶ €, 8.1 × 10⁶ € and 14.8 × 10⁶ € for scenarios Sc1, Sc2 and Sc3, respectively, which consist of 0.08%, 0.37% and 0.69% of the GDP of Fthiotida regional unit, respectively (Figure 6(a)).

The greatest economic loss is expected due to the inundation of urban areas and damage to infrastructure. More specifically, the economic impact of 0.3 m SLR (Sc1) is expected to reach 51.86 × 10⁶ € and 5.32 × 10⁶ € due to the loss of residential areas and damage to the road network, respectively (2.4% and 0.2% of the GDP, respectively; Table 3). This corresponds to 75% of the total economic loss. In the case of 0.6 m SLR (Sc2), the financial loss rises to 97.14 × 10⁶ € and 24.50 × 10⁶ € due to urban areas lost and damage to infrastructures, respectively (4.5% and 1.1% of the GDP, respectively), which corresponds again to 75% of the total economic loss for this scenario. Finally, in the case of 1.0 m SLR (Sc3), the economic consequences rise to 343.60 × 10⁶ € and 97.15 × 10⁶ € (15.9% and 4.5% of the GDP, respectively) which combined consists of 86% of the total cost (Figure 6(a) and 6(b)).

In the present study, the impact of SLR due to anthropogenic climate change on the Spercheios River and Maliakos Gulf system was estimated. Based on the results, the terrestrial zone to be lost in the study area fluctuates from 6.2 km² to 31.1 km², depending on the SLR scenario. The highest percentage loss was observed in wetland areas which will lose from 39% to 68% of their total extent. This disturbance within the wetland areas is expected to affect their functionality. The areas characterized as wetlands in the coastal zone nowadays are temporarily flooded during winter periods and progressively dry out during spring and summer. The SLR will cause the permanent inundation of these wetlands, which will change completely their natural hydrologic regime and impact the food and habitat availability for many organisms such as birds and aquatic mammals that depend on them (Cahoon et al. 2006).

The total economic impact of SLR was estimated to be 75.4 × 10⁶ €, 161.7 × 10⁶ € and 510.7 × 10⁶ €, which consists of 3.5%, 7.5% and 23.7% of the GDP of Fthiotida regional unit, in scenarios Sc1, Sc2 and Sc3, respectively. The loss of wetlands led to an economic loss of up to 43.0 × 10⁶ €, which, depending on the scenario examined, corresponds to up to 19% of the total SLR cost estimated.

Reviewing the methodologies for the estimation of the value of wetlands and coastal ecosystems, it was noted that there is a variety of approaches, while in many cases also mitigation and protection measures were taken into consideration, leading to corresponding varying results. Additionally, there was a wide spatial variation of the wetland value estimated, due to its direct relevance to the local income, population density and wetland size. Nevertheless, the wetland values estimated in this study were in agreement with the calculations of other studies conducted in Europe. Darwin & Tol (2001) estimated the average wetlands' value due to 0.5 m SLR in 12 member countries of the European Community in 1990 (including Greece) at 4.89 €/m² (5.21 $/m²) or 5.32 €/m² (7.06 $/m²) for reference year 2013. Kontogianni et al. (2012) used a unit value of wetlands of 4.8 $/m², adopted via the value transfer method by Darwin & Tol (2001), in order to estimate the long-term financial losses of SLR in Greece. Martinez et al. (2007) reported that the ecosystem service product (ESP) of terrestrial and aquatic coastal ecosystems in terms of $US per year for Greece is 1,316 million $US and the corresponding surface covered 32,270 km², which can be translated to 0.04 $US/m²/y (or 0.03 €/m²/y) and 3.92 $US/m² (or 2.95 €/m²) by the year 2100.

The consequences of SLR to the ecosystem services provided are indisputable, while adaptation and mitigation planning is required. Natural coastal systems have a capacity to respond autonomously to external pressures such as climate change and are, in principle, able to adjust to SLR by migrating inland, given the space and time (Fankhauser 1995). This capacity largely determines the system's resilience and resistance to pressure. In many places, however, human activities have reduced the natural coastal system's resilience to SLR and the potential for autonomous adaptation has decreased (Klein & Nicholls 1998).

Nevertheless, there is a large potential for coastal adaptation to reduce the worst expected impacts of SLR on coastal resources (Diaz 2016) and given the potential socio-economic impacts SLR countries face, in spite of autonomous adaptation, governments may wish to proceed to measures. In addition to doing nothing and reversing maladaptive trends, three distinct response strategies to SLR can be identified: managed retreat, accommodation and protection. Nevertheless, adaptation options need to be evaluated in the context of a region's coastal management or development objectives, which could determine the evaluation technique to be applied (Klein & Nicholls 1998; Massey et al. 2015). Consequently, adaptation is a social, political and economic process, rather than just a technical exercise, as it is often conceived (Tol et al. 2008). In this area, developing an integrated coastal zone management plan under the prism of SLR could offer solutions and mitigation measures for climate change impacts if appropriate spatial and marine planning is undertaken by the local authorities and stakeholders. The positive factor of this area is that it is not densely populated, most of the areas are either natural or semi-natural and the value of land is still moderate to low. Therefore, there is the opportunity to place new and existing anthropogenic activities and infrastructure in areas of low climate change and SLR risks.

This study was conducted under the research project ‘Development of an integrated management framework of river basins and associated coastal and marine zone’ (MIS 448841), funded by Ministry for Education, Religious Affairs, Culture and Sport of Greece and the European Union – European Regional Development Fund, within the framework of the Action entitled ‘Proposals for Development of Research Bodies-KRIPIS’ NSRF (Operational Programme II, Competitiveness & Entrepreneurship).