Environmental and Sustainability Indicators 8 (2020) 100056

Contents lists available at ScienceDirect

Environmental and Sustainability Indicators

journal homepage: www.journals.elsevier.com/environmental-and-sustainability-indicators/

Assessing the hidden impacts of hypothetical eruption events at Mount Etna

Giovanni Signorello a,b,*, Alessia Marzo a, Carlo Prato a, Giovanni Sturiale a, Maria De Salvo a,b

a University of Catania, Centre for the Conservation and Management of Nature and Agroecosystems, Cutgana, Italyb University of Catania, Department of Agricultural, Food and Environment (Di3A), Italy

A R T I C L E I N F O

Keywords:Mount etnaVolcanic eruptionsInVESTCarbon storageWater yieldPollination services

* Corresponding author. University of Catania, Ccultural, Food and Environment (Di3A), Italy.

E-mail addresses: g.signorello@unict.it, giovanniit (G. Sturiale), mdesalvo@unict.it (M. De Salvo).

https://doi.org/10.1016/j.indic.2020.100056Received 8 June 2020; Received in revised form 7Available online 14 August 20202665-9727/© 2020 The Author(s). Published by Elsnc-nd/4.0/).

A B S T R A C T

We estimate how the value and the spatial distribution of carbon storage, water yield and wild pollination servicesare expected to change with potential expansion of lava flow inundation at Mount Etna volcano (Sicily, Italy). Werely on a hazard map lava by flow inundation to simulate a set of three future land use/land cover (LU/LC)scenarios related to different hazard levels with a specific probability of occurrence. Our assessment used theIntegrated Valuation of Environmental Services and Tradeoffs (InVEST) tool and GIS spatial analysis and indicatesthat changes in the delivery of all three ecosystem services are biophysically and economically sizeable. Thevariation between scenarios demonstrates that the carbon storage and wild pollination services will decreasebecause of the loss of woods and natural habitats. In contrast, the water yield capacity will increase for thecreation of new naked land surfaces characterised by high permeability values. In the worst-case scenario, weestimate a loss of 17% and 10% for carbon storage and wild pollination services respectively, and an increase ofapproximately 10% for water yield.

1. Introduction

Estimating consequences of volcanic eruptions is a challenging task.Impacts depend on the style, magnitude, dynamicity and intensity oferuptions and vulnerability of surrounding areas, and can vary in size andtemporal and spatial scale (Doocy et al., 2013; Loughlin et al., 2015;Molist, 2017; National Academies of Sciences, 2017; Neild et al., 1999;Papale and Shroder, 2014). Several attempts have economically assessedthe impacts of volcanic eruptions (Daniell et al., 2015; Halliday et al.,2019; McDonald et al., 2017; Scaini et al., 2013; Tuck et al., 1992; Tuckand Huskey, 1994; Wilson et al., 2014; Zuccaro et al., 2013). Despitebeing based on complex computational and sophisticated empiricalmodels, these efforts usually only captured fractions of the full spectrumof eruption effects. They addressed only priced assets (e.g. human-madeand productive capital), marketed economic activities (e.g. tourism,agriculture and forestry) and healthcare expenses. Additionally, whenaimed at estimating indirect and long-term impacts, they relied onmarket macro-economic indicators (Botzen et al., 2019). Consequenceson natural stocks were generally excluded unless effects of volcaniceruption degraded priced goods and services (Neild et al., 1999). Eco-nomic analysis should go beyond market boundaries and indicators by

entre for the Conservation and M

.signorello@unict.it (G. Signorello

August 2020; Accepted 11 Augus

evier Inc. This is an open access a

accounting for the full range of ecosystem services delivered to society bynatural habitats to avoid misleading pictures of impacts. Additionally,the analysis should consider how the flow of these ecosystem servicesvaries with LU/LC changes caused by volcanic eruptions (Burkhard et al.,2009; Cunningham, 2018; Geneletti, 2011; Daily, 1997; Daily et al.,2009; Marske et al., 2007; Marti and Ernst, 2009; Missemer, 2018; Ochoaand Urbina-Cardona, 2017; Polasky et al., 2011). Now, practitioners havean extensive array of tools and models to value and map ecosystem ser-vices (de Groot et al., 2012; Kareiva et al., 2011; Burkhard and Maes,2017; Dennedy-Frank et al., 2016; Neugarten et al., 2018; Villa et al.,2014). Among them, a universally adopted tool is Integrated Valuation ofEnvironmental Services and Tradeoffs (InVEST), which was developedinside the Natural Capital Project (Mandle and TallisGeneletti, 2016;Posner et al., 2016; Sharp et al., 2018; Mandle, 2019; Tallis and Polasky,2009). In comparison to other available instruments, InVEST does notrequire particular expertise, provides relatively accurate simulationswith feasible and low demand for input parameters and is suitable forcontexts in which ecological processes are well understood (Bagstadet al., 2013; Bangash et al., 2013; Neugarten et al., 2018).

This research used InVEST and GIS analysis to value and map carbonstorage, water yield and wild pollination services at the Mount (Mt) Etna

anagement of Nature and Agroecosystems (Cutgana) and Department of Agri-

), alessia.marzo@unict.it (A. Marzo), cprato@unict.it (C. Prato), sturiale@unict.

t 2020

rticle under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

volcano, and assessed how hypothetical eruption events could changethese ecosystem services.

Mt Etna is the largest and highest active volcano in Europe, charac-terised by the occurrence of frequent eruptions that affect the naturalenvironment (Branca et al., 2004). We estimate how carbon storage,water yield and wild pollination services are expected to change in valueand space with the occurrence of potential lava flow inundation. In thesimulation, we use the hazard map by lava flow inundation built by DelNegro et al. (Del Negro et al., 2013) to estimate the probability of certainareas being inundated by future lava flows over the next 50 years. Weimagine three future LU/LC scenarios related to different probabilities ofhazard occurrences: high probability (Scenario 1), medium probability(Scenario 2) and low probability (Scenario 3). The volcanic activity of MtEtna can be divided into two principal categories—summit activityeruptions and flank eruptions. Both types of eruptions produce lava flow,which represents the main volcanic hazard; however, geological recordssuggest that summit eruptions are more frequent than flank ones (DelNegro et al., 2013). In this article, we specifically focus on the effects offlank eruptions because they involve the middle and lower slopes of thevolcanic edifice (densely populated areas that are abundant in wood orcrops). Despite Mt Etna providing a wide array of ecosystem services, inline with their relevance, our analysis is restricted to carbon storage,water yield and pollination (Alcamo et al., 2003). Carbon storage is theamount of carbon stored in a terrestrial ecosystem (He et al., 2016).Forests, grasslands and other terrestrial ecosystems collectively storemore carbon than the atmosphere (Lal, 2004). By storing this carbon inbiomass such as wood or soil, CO2 is prevented from accumulating in theatmosphere and contributing to climate change. Water yield contributesto water accessibility for consumptive use or in situ water supply (e.g.hydropower) (Vigerstol and Aukema, 2011). Pollination is a significantbiotic regulating and maintenance service—more than a third of globalcrop yield depends on the activity of animal pollinators (Bailes et al.,2015; Breeze et al., 2011; Klein et al., 2007). Moreover, 75% of cropspecies depend on animal pollinators to produce fruits or seeds (Hanleyet al., 2015). In particular, Garibaldi et al. (2013) found that the role ofwild pollinators are more effective than honey bees in enhancing cropproductions. Our research hypothesis is that potential volcanic eruptionsat Mt Etna would induce substantial changes in the value and spatialdistribution of investigated ecosystem services.

Fig. 1. Geographical cont

2

2. Material and methods

2.1. Study area

Mt Etna is the largest active volcano in Europe. In 2013, it wasincluded in the UNESCO World Heritage List. It was edified by theproducts of different eruptive centres that were active at various timesduring the past 600,000 years. The youngest of these centres, Mongibellovolcano, is prevalently effusive in style eruptions (Branca et al., 2011).Mt Etna is a stratovolcano localised along the east coast of Sicily(southern Italy) and has a basal diameter of approximately 40 km, asurface of 1340 km2 and an elevation of 3340 m above mean sea level(AMSL).

The Mt Etna volcanic body represents an important aquifer in Sicilybecause of its position, high elevation and internal structure. Indeed, it ishighly exposed to winter precipitation from being surrounded by lowermountain chains (Hyblean and Peloritani), the Ionian Sea and the Cata-nia plain (see Fig. 1). In addition, the volcanic edifice is constructed byoverlapped lava flows (permeable rocks) and volcanoclastites (imperme-able volcanic layers), which create a large ‘multiple aquifer’. The highpermeability of Mt Etna’s volcanic rocks is reflected by an infiltrationcoefficient for rainfall of approximately 75% and a runoff coefficient ofapproximately 25% (Ferrara, 1975). These features present Mt Etna as animportant groundwater reservoir; an approximate medium volume of500 � 106 m3 is withdrawn per year (Ferrara and Pappalardo, 2008) tosatisfy the water demand of 42 urban centres (a combined population ofapproximately 857,000) and 20,000 ha equipped for irrigation. Thegroundwaters of Mt Etna’s many aquifers are naturally enriched in ele-ments such as magnesium, iron andmanganese. Similarly, Veschetti et al.(2007) highlighted a possible low presence of fluorine and Vanadium.Regarding fluorine, the small concentration relates to the adsorption andretention capacity of the Etnean soils (D’Alessandro et al., 2012). In MtEtna’s aquifer, Vanadium is widely present in the less-active reducingform; however, the effects of this chemical contaminant on health are notsignificant; therefore, water fromMt Etna is drinkable (Giammanco et al.,1995).

In the current (or baseline) scenario, natural and semi-natural com-munities (especially forests and shrublands) cover approximately 35% ofthe study area and prevail above 1000 m AMSL (see Fig. 2). Artificial

ext of the study area.

Fig. 2. LU/LC of Mt Etna in the current or baseline scenario.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

areas cover 10% of the surface and are located predominantly in thesouth-eastern sector below 700 m AMSL, and the agricultural regionsoccupy approximately 43% and prevail below 1000 m AMSL. These areascover a surface of 56,853 ha and contain both intensive crops and severalcombinations of closed agroforestry systems. The Mt Etna flanks hostvarious crops, including field crops, citrus, orchards, prickly pear, olivetrees, vineyards, hazelnuts and pistachios. In 2019, the estimated totalgross annual value of agricultural production amounted to €376 million(Consiglio per la ricerca, 2020). Insect-dependent crops, including veg-etables, citrus, orchards, prickly pear, olive trees and vineyards, coverapproximately 45,000 ha and yield a total gross annual value equal to€323 million. Open spaces with minimal or non-existent vegetationdominate areas above 2500 m of altitude. Since 1985, the majority of MtEtna’s territory—approximately 59,000 ha—is included in the regionalprotected areas network. Altitude, temperature, rainfall and frequenteruptions are the predominant force for landscape and biodiversity in MtEtna (Signorello et al., 2018) (more than 70 flank eruptions haveoccurred in the last 420 years (Branca et al., 2004)).

2.2. LU/LC change scenarios

To create hypothetical LU/LC change scenarios we relied on thevolcanic hazard map proposed by Del Negro et al. (Del Negro et al.,2013), which linked the morphological setting and historical data of theMt Etna volcano. This map simulates lava flow paths and indicates theprobability of a certain area being inundated by future lava flows over thenext 50 years. The map is produced by a digital elevation model, prob-ability analyses and MAGFLOW—a quantitative simulation model based

Fig. 3. Lava flow paths related to different probabilities of lava inundation occu

3

on historical eruptions, rheological behaviour of flowing lava and effu-sion rate values. To produce the hazard map, MAGFLOW followed afour-stage methodology based on ‘(i) assessment of the spatiotemporalprobability of future vent opening, (ii) estimation of the occurrenceprobability associated with classes of expected eruptions, (iii) simulationof a large number of eruptive scenarios with the MAGFLOW model, and(iv) computation of the long-term probability that a lava flow willinundate a certain area’ [40, p. 2].

Three occurrence probabilities of lava inundation (high, medium andlow) were considered to create maps of the three LU/LC change scenarios(see Fig. 3). These maps were created by combining the three differentclasses of probability from MAGFLOW with the current LU/LC mapthrough a GIS overlay analysis.

Scenario 1 considers a 40% probability of inundation by lava flows.This scenario represents a realistic situation and only affects a verynarrow area (approximately 1160 ha). The map, displayed in Fig. 2, in-dicates that the eastern sector of Mt Etna is affected by a higher rate of avolcanic hazard than the western sector. Specifically, the zones with thehighest probability of lava flow inundation are the Valle del Bove areaand the north-east sector of the volcano. Scenario 2 describes an inter-mediate case referring to 20% probability of inundation. Various areasaffected by lava flow are the same as Scenario 1; however, they are wider(approximately 4800 ha) because of larger emission rates and durationvalues. Scenario 3 depicts further severity to changes in LU/LC byconsidering a 5% probability of lava flow inundation. It represents a verydramatic scenario (approximately 17,800 ha covered by lava) partiallyreflects conditions of the first two scenarios. However, it considers veryhigh effusion rate values (over 100 m3/s) over a few days, which

rrences (P): high (Scenario 1), medium (Scenario 2) and low (Scenario 3).

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

represents a rare event. In the third scenario, several densely populatedareas that rest on the lower south-eastern flank of the volcano will beaffected by lava flow invasion.

It is important to highlight that, despite the three hypothesised sce-narios indicating that approximately 1160 ha, 4800 ha and 17,800 ha ofthe landscape will be invaded by lava, the changes in LU/LC will beconsiderably lower because most of the lava flow would cover pre-existent volcanic bare soils. The predominant changes in LU/LC regardareas located in medium or high belts that are close to eruption vents.When considering how this would influence each scenario, Scenario 1would only be minorly changed, but Scenario 2 would be greatly affec-ted—approximately 1937 ha of natural areas (forest, scrub and herba-ceous vegetation associations) would be lost. Scenario 3 would suffer themost change: natural areas would decrease by 6870 ha, forests woulddecrease by 3411 ha, approximately 1224 ha of artificial areas would bedestroyed and agricultural regions would reduce by 10% (see Table 1).

2.3. Carbon storage

The InVEST carbon storage model uses the sum of carbon storedaboveground (Ca), belowground (Cb), in the soil (Cs), in dead organicmatter (Cd) and harvested wood products to estimate the average amountof carbon stored in each LU/LC class (Sharp et al., 2018). We used theNational Land Cover Database at a 20 m resolution to spatially identifyrelevant LU/LC classes, which included forests, woods, orchards, rowcrop and barren (see Fig. 4). Estimates for carbon storage were based onthe carbon storage values available in InVEST. The total amount of car-bon stored (Ci,m) in each grid cell (i) with land use type (m) was calcu-lated by

Ci;m ¼ A� ðCai;m þCbi;m þCsi;m þCdi;mÞ (1)

where A is the actual area of each grid cell equal to 0.04 ha. Woodharvesting was not considered because it was not a significant activity onMt Etna. Therefore, the total carbon storage to the current landscape(Ccurrent) and for each scenario (Cscen1, Cscen2 and Cscen3) was calculated inunits of tonnes of carbon by

Cs¼X

Ci;m; s ¼ current; scen1; scen2 and scen3 (2)

Finally, the change in carbon storage (ΔCEtna) from Ccurrent, whichrepresents the baseline for future scenarios, was calculated by

ΔCEtna ¼Ci � Ccurrent; I ¼ scen1; scen2 and scen3 (3)

To convert carbon storage in monetary terms, we followed van denBerg and Botzen (van den Berg and Botzen, 2015), who assigned a valueof US$125 per tonne of emitted CO2, which corresponded to US$34 pertonne of C (or €26.6 per tonne of C in 2019).

2.4. Water yield

The InVESTWater yield model estimates the annual water yield (Yi,m)(mm/year) using

Yi;m ¼�1�AETi;m

Pi

�� Pi (4)

Table 1Predominant LU/LC changes in Mount Etna between scenarios. P: occurrence probab

Land use classes (LULC) Current Scenario Scenario

area (ha) area (ha)

Open spaces with little or no vegetation 16,371 17,036Forest 20,904 20,702Agricultural areas 56,853 56,707Scrub and/or herbaceous vegetation associations 24,641 24,364Artificial areas 12.991 12.972

4

where Pi is the annual precipitation (expressed in mm) on grid cell I,AETi,m is the annual evapotranspiration (expressed in mm) on grid cell ifor LU/LC m.

The InVEST model relates AET to potential evapotranspiration. It isbased on a formula by Budyko (1974) and an algorithm by Zhang et al.(2004), in which soil depth, plant available water content, plant rootdepths and land use characteristics were considered. Figures regardingroot restricting layer depth and available water content (AWC), bothmeasured in mm, were obtained from the European Soil Database (ESDB)version 2.0 (European and Soil Data Centre [ESDAC]). Similarly, LU/LCdata were extracted from this source to run the carbon model (i.e. theItalian land cover map) at a 20 m resolution. From the period of1990–2010, the mean precipitation data (P) and mean potential evapo-transpiration data (PET), both measured in mm, were obtained from theSicilian meteorological database. P and PET data were spatially inter-polated using the Kriging method provided by the ESRI ArcGIS® soft-ware. The total water yield in the current landscape (Ycurrent) and for eachscenario (Yscen1, Yscen2 and Yscen3) was calculated in m3 by

Ys¼AxX

Ymi (5)

where s indicates the current and other scenarios (scen1, scen2 and scen3)and A is the total surface of Mt Etna. Because the InVEST water modeldoes not discern between surface and sub-surface water flow, the wateryield value was multiplied by the infiltration coefficient (0.75) (Ferrara,1975) to estimate groundwater recharge. The change of water yield(ΔYEtna) from Ycurrent, which represents the baseline, to future scenarioswas calculated by

ΔYEtna ¼Yi � Ycurrent; I ¼ scen1; scen2 and scen3 (6)

We quantified volumes of water for agricultural and domestic use.Water volume for the agricultural sector (WVagriculture) was calculated bymultiplying the crop water requirement (m3/ha) by the irrigated area(approximately 20,000 ha). Water volume for the domestic sector(WVdomestic) was calculated multiplying the area’s population (approxi-mately 850,000 people) by the average household water consumption(200 l/day per person). To assess the monetary value associated withwater yield, we employed €0.90/m3 as unit prices of domestic water and€0.40/m3 as unit prices of water supplied to agriculture. The monetaryvalue of total water yield was calculated as the sum of total agricultureand domestic water values.

2.5. Wild pollination service

The InVEST pollination model maps where the wild bee pollinationservice is provided (i.e. map of pollination supply) as well as the cropareas that receive biotic pollination for optimal production (i.e. map ofpollination delivery). An index of agricultural production attributable towild pollinator insects was also evaluated (index of a pollination service).The map of pollinator supply indicates where wild pollinator insectsoriginate and elaborated on the available nesting sites, the floral re-sources across the landscape and the foraging range of pollinators. Thisspatially explicit map can be understood as an index of pollination supplyfor each unit area in the landscape. It represents the potential sources of

ilities of lava inundation.

1 (P ¼ 40%) Scenario 2 (P ¼ 20%) Scenario 3 (P ¼ 5%)

Δ (%) area (ha) Δ (%) area (ha) Δ (%)

4.1 19,541 19.4 30,196 84.4-1.0 19,887 -4.9 17,493 -16.3-0.3 55,903 -1.7 51,426 -9.5-1.1 23,721 -3.7 21,181 -14.0-0,2 12.765 -1,7 11.767 -9,4

Fig. 4. Spatial distribution of Etnean land-use classes (20 m resolution).

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

pollination services but does not incorporate any demand information.To connect the areas of ‘supply’ and ‘demand’, an index of potential wildbee abundance (PWA) was developed for each agricultural cell. PWAwascalculated by considering the suitability of nesting, relative amount offloral resources in the land use types, distance between habitats (areasthat supply the service) and crops (areas that demand the service) and themaximum foraging flight distance of wild pollinators. We generated a setof maps—one for each pollinator-dependent crop. These maps conveyedthe potential sources of pollination services to the relative degree ofpollination service at the demand points, or at points in which the serviceis ‘delivered’.

To estimate the agricultural production attributable to wild pollina-tors (Ywp) (i.e. index of a pollination service), the InVEST tool measuresthe PWA by using a simple saturating crop yield function, which assumesthat yield increases as wild pollinator visitation increases. However,there are diminishing returns (Greenleaf and Kremen, 2006), asdemonstrated by

Ywp ¼DcPWA

PWAþ Kc(7)

where Dc represents the portion of yield that is dependent pollinators(wild and managed) and Kc is a half-saturation constant that representsthe abundance of pollinators required to reach 50% of pollinator-dependent yield. In this study, Kc ¼ 0.125 for all crops, whereas Dc

Table 2Area (ha) and dependency on wild insect pollinator (D) of each pollinator-dependent crop.

Crops Area(ha)

Dependence on insectpollination (Dc)a

Vineyard (Vitis vinifera L.) 4,688 0,10Vegetable (Lycopersicon esculentum Mill.,Brassica chinensis L. …)

998 0,83

Orchards (Pyrus malus L./Pyrus communisL./Prunus avium L.)

7,741 0,86

Prickly pear (Opuntia ficus-indica) 1,378 0,80Citrus (Citrus aurantium L.,/C. limon L.) 20,672 0,10Prickly pear and olive (Opuntia ficus-indica, Olea europaea L.)

4,817 0,40

Olive (Olea europaea L.) 4,742 0,10

a Mean value from Klein et al. (2007), Giannini et al. (2015) and Lo Verde et al.(Lo Verde and La Mantia, 2011).

5

was calculated using data from Klein et al. (2007), Giannini et al. (2015)and Lo Verde et al. (Lo Verde and La Mantia, 2011) (see Table 2).Following this method, Ywp was not calculated as a fixed proportion oftotal production attributed to insect pollination; instead, it related to thepotential spatial distribution of wild bees.

To use the InVEST pollination model, we created a table of pollinatorspecies and a LU/LC map. The table concerned only bees (Apis spp.) andbumbles bees (Bombus spp.) to represent all pollinator insects and con-tained information on flight season, nesting requirements and flightdistance. The LU/LC map (see Fig. 5) with a 20 m resolution was elab-orated in a GIS environment. The map on ‘Landscape Protection Plan ofCatania’, which was created by the Regional Department for Cultural andEnvironmental Heritage and Public Education, was combined with LaVia’s (La Via, 2009) map of orchards in the Etna.

The LU/LC attribute table includes data regarding the suitability fornesting and the amount of floral resources. The InVEST pollinationmodelproduces the index of a pollination service (Ywp) as final output. Tomeasure the economic value of pollination, we followed Morse andCalderone’s (Morse and Calderone, 2000) study in which the economicvalue is obtained by equating the value of pollination service to theproportion of total production value dependent on insect pollination. Weassumed that the pollination index (Ywp) was the proportion of thepollination service provided by wild bee. Consequently, the economicvalues were estimated by

Vw ¼P*Ywp (8)

where Vw is the pollination service value provided by wild bees and P isthe market value of each crop retrieved from a recent Italian agriculturereport by the Italian Council for Agricultural Research and Economics(CREA) (Consiglio per la ricerca, 2020). To compare pollination servicevalues across the three scenarios, we assumed that there was no variationin P, farmers were not compensated for the loss in pollinators, honeybeeswere not added, and there was no change in crop species or farm man-agement systems.

3. Results and discussion

3.1. Carbon storage

Currently, Mt Etna stores approximately 16.4 million tonnes of carbon.This amount was calculated as the sum of carbon density of various LU/LC

Fig. 5. Spatial distribution of Etnean land-use classes (20 m resolution) used to run the InVEST pollination module.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

types. The amount is equivalent to the amount of carbon emitted from theentire Sicilian population over two years, which is based on average percapita emission rates of 5.9 tonnes of CO2 per person each year (Munteanet al., 2018). The amount of carbon for each unit of LU/LC changes ac-cording to variations in vegetation density and species composition (seeFig. 6). Berries and vineyard have the lowest carbon storage capacitybecause of their low surface and low vegetation density, whereas orchardsplay an important role in carbon storage because of their large surface.Natural shrub, which covers approximately 18,257 ha, is the third-largestLU/LC class and stores approximately 803,348 tC, which corresponds to5% of Mt Etna’s total carbon storage. Wooded areas are the largest sourceof carbon storage with approximately 7.9 million tonnes, which is equal to48.7% of the total. The carbon store capacity of wood correlates to surfaceextension and vegetation density, as well as belonging to temperate ev-ergreens, which is a plant category that demonstrates the greatest carbonstorage capacity (Papale and Shroder, 2014).

Fig. 7 presents the distribution of carbon storage across the threescenarios. The changes result from the differences in vegetation struc-ture. The highest carbon storage is within the supra-Mediterranean belt(1000–1500 m AMSL) because of the large presence of forests (mixedforest, conifer, woodlot and oak savanna), which is characterised by ahigh carbon storage capacity (a mean value of 400 tC/ha).

Fig. 6. Area and relative percentage of carbon storage for different LU/LC types.

6

The maximum carbon storage coincides with the biggest forests,including the pine forest in the south sector, the chestnut and the oakforests in the eastern sector, the birch and pine forests in the north-eastsector, and the mixed forest (beech and pine) in the north sector. For-ests have the lowest carbon storage capacity and are often interrupted bybarren areas that follow the contours of post–18th century lava flow (seeFig. 7). Mt Etna’s higher slopes are homogeneously characterised byexceptionally low values of carbon storage because of the presence ofvolcanic desert. This loss results in a decreased amount of orchard andforest surfaces and an increase in fresh lava surfaces where no vegetation(which is responsible for carbon store activity) will be present for a longtime. On the lower altitudes (below 1000 m AMSL), Fig. 7 depicts acluster of areas with different values of carbon storage. This is evident onthe east flank where different LU/LC types are present, including urbanareas, relicts of fresh lava surfaces, uncultivated areas, vineyards, or-chards, oak savannas and small mixed forests. Fig. 7 overlaps the newlava flow fields that were hypothesised in the third scenario. Mt Etnastores approximately 13.6 million tonnes of carbon in the third scenario,which is 2.8 million tonnes of carbon less than the present day. Thisresults from the increase of fresh lava surface and the reduction of or-chards and forest surfaces.

Fig. 8 demonstrates the decrement of carbon storage between thescenarios. The decrease is strictly linked with an increase in barren sur-face but it is not directly proportional to the reduction of wooded areas.Indeed, Scenarios 1 and 2 indicate that the invasion of a new lava flowfield is confined on the high slopes where woods are scarcely present.Additionally, the carbon storage values are similar to those in the currentsituation. Further, on the medium slopes, new lava will cover only barrensurfaces and a limited section of wooded surfaces. However, the thirdscenario depicts a large portion of territory below 1000 m AMSL will beaffected by the invasion of lava flow, which would destroy of largeportion of orchards and woods and, consequently, reduce carbon storagecapacity. Unfortunately, similar to Scenario 3, a large amount of orchardsare already reducing; a large portion of territory previously used foragricultural purposes are now abandoned and planned as building areas.If this trend does not change, the reduction in carbon storage asdemonstrated in Scenario 3 may become likely to occur. Crucially, thiswould not be caused by natural phenomena, but by the careless actions ofhumans.

Fig. 7. Amounts of carbon stored within the different scenarios** Values are expressed in tonnes/grid cell (20 m � 20 m). The loss of carbon from present day to Scenario 3 is emphasised by the decrease of dark green areas. . (Forinterpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 8. Amount of carbon storage (tonne C/year) and value of carbon storage(€) within the different scenarios.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

Economically, the value of carbon storage services provided byEtnean natural capital in the baseline scenario is approximately equal to€435 million. It is expected that carbon storage will decrease in futurescenarios because of changes in LU/LC by potential lava eruption eventsin the future. Specifically, a loss equal to €2 million and €11 millionwould occur in the first and second scenarios, respectively. Dramatically,a loss of €74 million would occur in the third scenario.

3.2. Water yield

The InVEST water yield model is founded on the difference betweenrainfall and the rate of evapotranspiration in each land parcel. The wateryield of Mt Etna in the baseline scenario is approximately 418 � 106 m3

7

and the groundwater recharge is estimated to be equal to 310 � 106 m3.These volumes are lower than the amount of water that is availableannually (700 � 106 m3/year), as reported in Federico et al. (2017), andlower than the medium volume annually withdrawn (500 � 106

m3/year), as published in Ferrara and Pappalardo (2008) and based onthe monitoring of wells and springs. Two reasons can explain the dif-ferences between InVEST estimates and available data. First, the InVESTwater yield module is not a hydrological model; therefore, it containsseveral limitations and simplifications such as lacking consideration ofthe effects of complex land use patterns or underlying geology (Sharpet al., 2018). Second, a more accurate definition of model parameters andinput data is needed, including AWC, root restricting layer depth and theplant evapotranspiration coefficient (Kc), which is used to obtain PET towhich the model is most sensitive (Redhead et al., 2016).

Fig. 9 depicts Mt Etna’s water yield by using different shades of blue.Values range from 0 mm to 990 mm. When considering the entire Etneanedifice, the water yield values are bigger in the eastern sector than thewestern sector. This evidence is likely to be closely linked with thevarying rate of rainfall between the two sectors, which depends on thevolcano’s geographic and geological setting. Mt Etna’s eastern side isgenerally more humid than its western side because the former overlooksthe sea and the latter faces dry inner-land zones. Further, the easternslopes are steeper and humid currents coming from the sea climb thevolcano’s flank rapidly, which forms dense clouds and results in heavyrainfall. Importantly, the difference in water yield values between thevolcano’s eastern and western sides is because of the larger fresh lavasurface that covers the eastern side, which results in a higher infiltrationrate. Additionally, the eastern side is generally more affected by theoccurrence of eruptions.

Fig. 9 depicts the higher and medium slopes of the volcanic edifice inthe current or baseline scenario presenting the highest values of wateryield (indicated by dark blue colouring), which generally follows the

Fig. 9. Water yield in Mt Etna within the different scenarios**. The increase of water yield value from current to Scenario 3 is emphasised by the increase of dark blue areas. . (For interpretation of the references to colour in thisfigure legend, the reader is referred to the Web version of this article.)

Fig. 10. Water yield (1,000,000 m3/year) and economic estimation within thedifferent scenarios.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

post–19th century lava flow field contours. This results from the lack ofvegetation on the recent lava volcanic surfaces. These areas do not pre-sent the evapotranspiration process and the lack of soil favours the rate ofwater infiltration. This feature is mostly evident in the volcano’s high andmedium slopes that are within Mt Etna Park and thus, less affected byhuman activity. Conversely, the areas with lower values of water yieldare covered by forestry, which implies a high ratio of evapotranspirationprocesses and thickness of soil. Interpreting the map generated byInVEST becomes more complicated when focusing on the volcano’slower zones (below the altitude of 1000 m AMSL). Because these areasare outside Mt Etna Park’s perimeter, the high degree of anthropisationcan occasionally supply false information. Additionally, the areas withthe highest values of water yield on this belt overlap the urbanised areas(town and villages). This likely occurred because the input of evapo-transpiration and the presence of soil in these areas are equal to zero;therefore, the InVEST data depended solely on the precipitation level.The increase of water yield value increasing with the extension of freshvolcanic surfaces is easily explained when considering the loss ofevapotranspiration rate and the corresponding increase of permeablesurfaces.

Fig. 10 displays the water yield values relative to the different sce-narios. The LUI/LC changes are primarily represented by the formationsof new naked lava surfaces (barren surfaces). The consequent decrease offorested areas (see Table 1) relate to the varying water yields across therespective scenarios. The economic values relative to the baseline sce-nario (418,000,000 m3 and €278 million) and the Scenario 1(420,000,000 m3 and €279 million) are very similar, as it is speculatedthat the new lava flow predominantly overlapped the naked areas.Conversely, the water yield values in Scenario 3 (the worst but leastlikely scenario) is higher than previous values (465,000,000m3 and €308million) because this scenario involves lava flow invasion on thevolcano’s medium and lower vegetated sectors.

Nevertheless, it is important to note that the values of water yield inthe various scenarios share some similarities. This pattern emphasises the

8

significant role of the rate of precipitation in respect to other inputconsidered by the InVEST model.

3.3. Pollination

Fig. 11 describes the index of pollination supply map, which dem-onstrates potential capabilities of specific LU/LC to provide crop polli-nation service on a relative scale ranging from 0 to 1. Zero indicates thatthere is no capacity to supply pollination services, whereas 1 refers tomaximum capacity to supply pollination services by wild pollinators(represented by bees and bumbles).

The general map pattern in the baseline scenario indicates highpollination supply between 500 m and 1500 m AMSL (depicted by greencolouring) in which the mixed forests, often interrupted by natural pas-tures and cultivated crops, prevail. The areas with low pollination supply

Fig. 11. PWA on landscape within the different scenarios.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

(depicted by the purple colouring) are characterised by volcanic desert(over 2000 m AMSL) and coastal and sub-coastal plains (lower than 500m AMSL). These areas have relatively low nesting suitability and offerlimited resources for foraging because of a low abundance of flowerscarrying nectar. Future scenarios (particularly Scenario 3) observe agenerally widespread reduction of areas with high pollination supply andan increase of areas with low capability to provide the service. This canbe explained by the decrease in natural and extensive agricultural areasand the increase of naked lava surfaces that are unfit to support wild

Fig. 12. PWA of pollinator-dependent

9

pollinators. Such a trend is particularly evident in the eastern andsouthern sectors of Mt Etna where the lava flow is more abundant.

Mapping and assessing where pollination services are delivered canbe useful to locate crops, optimise conservation investments that benefitbiodiversity and farmers, and predict consequences of different policieson pollination services. An area could be abundant in pollinator; how-ever, if there are no bee-pollinated crops in that area, pollinator insectscould not provide a crop pollination service. Fig. 12 maps natural areas(indicated in green) and agricultural areas (indicated in orange and blue)

crops within different scenarios.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

where pollination services are delivered, to demonstrate the exchangebetween PWA and pollinator-dependent crops. The present-day scenariodisplays a tremendous pollination service in the eastern sector of Mt Etna,where the landscape presents a high level of heterogeneity with amixture of LU/LC types, including pollinator-dependent crops. Thepollination service is low in the northern and southern areas where cropsare minimal in natural and urban areas, as well as low in the easternsector because insects do not depend on these crops for pollination.

The blue square in Fig. 12 demonstrates an example of decreasedPWA and increased new lava surfaces. In this area, crop isolation and theconsequent reduction of pollination services is evident. Table 3 reportsthe economic value provided by wild bees for crop yield dependencywithin different scenarios. Overall, the pollination service value isapproximately €45 million and represents 12% of the gross value of farmproduction of the entire Etnean crop sector, which is approximately €376million. A progressive reduction of economic value across scenarios wasobserved with a loss of approximately €4million between the current andthird scenario. This reduction is mostly imputable on the lava flow’sdestruction of approximately 7000 ha of natural habitat. Resultsdemonstrate that orchards are the predominant relevant crop for polli-nation services on Mt Etna because of its extension (7,700 ha), hectarevalue (€7000) and strict dependence (Dc ¼ 0.7–0.8) on pollination ser-vice. Despite irrigated crops holding high Dc value (0.7) and averageoutput, they are less important because of their low extension (900 ha),and their citrus causes their very low Dc value of (0.1).

4. Conclusion

Sicily’s Mt Etna volcano is continuously transforming; frequenteruptions create new lava flow fields that can radically and suddenlychange ecosystem services and natural and semi-natural environments.In this paper, we used InVEST models and GIS spatial analysis to assessMt Etna’s carbon storage, water yield and pollination. These ecosystemservices were valued in the present-day environmental setting and inthree possible future scenarios of LU/LC changes, which were created byfollowing Del Negro et al.‘s (Del Negro et al., 2013) hazard map of lavaflow inundation. In the hypothetical scenarios, the ecosystem servicevalues can be affected over time by variations linked to the volcano’s rateof activity. InVEST analysed the landscape’s spatial distribution ofecosystem services and accounted for changes in ecosystem servicesunder various scenarios.

Biophysical carbon storage assessment required few data as it de-pends on LU/LC map accuracy and the availability of carbon pools data.The evaluation of water yield required more data; the relative InVESTmodel is sensitive to precipitation and evapotranspiration values anddoes not consider the effects of complex land use patterns or underlyinggeology. Pollination services evaluation required a small-scale approachbecause its provision is reliant on the distribution of floral resources and

Table 3Economic value of pollination service.

Pollinationdependentcrops

Gross farmproduction(million eurosper year)

Value of crop pollination service provided bywild bees (€ million per year)

Current Scenario1

Scenario2

Scenario3

Vineyard 42,42 2,16 2,14 2,10 1,90Vegetable 10,77 3,75 3,75 3,74 3,68Orchards 53,75 22,73 22,63 22,06 19,67Prickly pear 19,27 3,70 3,70 3,70 3,54Citrus 9,22 3,31 3,31 3,30 3,24Prickly pearand olive

6,08 0,29 0,29 0,28 0,26

Olive 181,67 8,95 8,92 8,87 8,40TOT 323,19 44,88 44,74 44,05 40,69Natural area(ha)

45,545 45,066 43,628 38,674

10

distance between habitats. Mapping pollination services permitted vis-ualising and understanding the connection between habitats where in-sects live and areas where pollination services are delivered.

Improvements to the InVEST pollination service model shouldconsider landscape elements such as riparian margins, water bodies andsecondary roads that offer ideal foraging and nest conditions for wild bee.Additionally, the service model could integrate climate conditions suchas solar radiation or atmospheric temperature. Our analysis examines therole pollinators play in the Etnean agriculture and highlights how thiseconomic activity requires stable pollinator populations. These issues areparticularly relevant in Mt Etna’s east sector, which is characterised as ahigh volcanic hazard, where traditional crops (e.g. citrus) are progres-sively replaced by tropical crops that strongly depend on insect pollina-tion (e.g. mango, avocado and lychee).

The results demonstrate a different trend regarding variations to theecosystem services that are linked to an increase of eruptive activity, ashypothesised in the three different scenarios. The carbon storage andpollination service will be reduced by the loss of woods and naturalhabitats resulting from new volcanic surfaces. Conversely, water yieldincreases with the rise of the volcanic activity because the rate of infil-tration water is positively influenced by the creation of new naked landsurfaces.

Overall, the findings indicate that the volcano’s east sector will beaffected by a substantial decrement in ecological functions. Despite thisstudy only considering a section of Mt Etna’s full range of ecosystemservices, it provides useful insights that can be integrated into thetraditional impact assessment of Mt Etna’s eruptions. Additionally, thisstudy can assist policymakers in designing superior land use and con-servation plans in areas subject to volcanic events.

Declaration of competing interest

The authors declare that they have no known competing financialinterests or personal relationships that could have appeared to influencethe work reported in this paper.

References

Alcamo, J., Ash, N.J., Butler, C.D., Callicot, J.B., Capistrano, D., Carpenter, S.R.,Castilla, J.C., Chambers, R., Chopra, K., Cropper, A., Daily, G.C., Dasgupta, P., deGroot, R., Dietz, T., Duraiappah, A.K., Gadgil, M., Hamilton, K., 2003. Ecosystems andHuman Well-Being: A Framework for Assessment. Island Press, Washington.

Bagstad, K.J., Semmens, D.J., Waage, S., Winthrop, R., 2013. A comparative assessment ofdecision-support tools for ecosystem services quantification and valuation. EcosystServ 5, 27–39. https://doi.org/10.1016/j.ecoser.2013.07.004.

Bailes, E.J., Ollerton, J., Pattrick, J.G., Glover, B.J., 2015. How can an understanding ofplant–pollinator interactions contribute to global food security? Plant Biol. 26,72–79. https://doi.org/10.1016/j.pbi.2015.06.002.

Bangash, R.F., Passuello, A., Sanchez-Canales, M., Terrado, M., L�opez, A., Elorza, F.J.,Ziv, G., Acu~na, V., Schuhmacher, M., 2013. Ecosystem services in Mediterraneanriver basin: climate change impact on water provisioning and erosion control. Sci.Total Environ. 458–460C, 246–255. https://doi.org/10.1016/j.scitotenv.2013.04.025.

Botzen, W.J.W., Deschenes, O., Sanders, M., 2019. The economic impacts of naturaldisasters: a review of models and empirical studies. Rev. Environ. Econ. Pol. 13,167–188. https://doi.org/10.1093/reep/rez004.

Branca, S., Del Carlo, P., 2004. Eruptions of Mt. Etna during the past 3,200 years: arevised compilation integrating the historical and stratigraphic records. In:Bonaccorso, A., Calvari, S., Del Negro, C., Falsaperla, S. (Eds.), GeophysicalMonograph Series. American Geophysical Union, Washington, pp. 1–28.

Branca, S., Coltelli, M., Groppelli, G., 2011. Geological evolution of a complex basalticstratovolcano: Mount Etna, Italy. Ital J Geosci 130, 306–317. https://doi.org/10.3301/IJG.2011.13.

Breeze, T.D., Bailey, A.P., Balcombe, K.G., Potts, S.G., 2011. Pollination services in theUK: how important are honeybees? Agric. Ecosyst. Environ. 142, 137–143. https://doi.org/10.1016/j.agee.2011.03.020.

Budyko, M., 1974. Climate and Life. Academic Press, New York.Burkhard, B., Maes, J. (Eds.), 2017. Mapping Ecosystem Services. Pensoft Publishers,

Sofia, Bulgaria.Burkhard, B., Kroll, F., Müller, F.D., Windhorst, W., 2009. Landscapes’ capacities to

provide ecosystem services—a concept for land-cover based assessments. LOON 15,1–22. https://doi.org/10.3097/LO.200915.

Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria (CREA), 2020.Annuario dell’Agricoltura Italiana, Rome, Italy. LXXII.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

Cunningham, E.J., 2018. Nature interrupted: affect and ecology in the wake of volcaniceruption in Japan. Conserv. Soc. 16, 41–51. https://doi.org/10.4103/cs.cs_16_50.

Daily, G.C., 1997. Nature’s Services: Societal Dependence on Natural Ecosystems. IslandPress, Washington.

Daily, G.C., Polasky, S., Goldstein, J., Kareiva, P.M., Mooney, H.A., Pejchar, L.,Ricketts, T.H., Salzman, J., Shallenberger, R., 2009. Ecosystem services in decisionmaking: time to deliver. Front. Ecol. Environ. 7, 21–28. https://doi.org/10.1890/080025.

Daniell, J., Skapski, J.-H., Vervaeck, A., Wenzel, F., Schaefer, A., 2015. Global earthquakeand volcanic eruption economic losses and costs from 1900–2014: 115 years of theCATDAT database—trends, normalisation and visualisation. Geophys. Res. Abstr. 17,2015EGUGA.17.8119D.

de Groot, R., Brander, L., van der Ploeg, S., Costanza, R., Bernard, F., Braat, L.,Christie, M., Crossman, N., Ghermandi, A., Hein, L., Hussain, S., Kumar, P.,McVittie, A., Portela, R., Rodriguez, L.C., ten Brink, P., van Beukering, P., 2012.Global estimates of the value of ecosystems and their services in monetary units.Ecosyst Serv 1, 50–61. https://doi.org/10.1016/j.ecoser.2012.07.005.

Del Negro, C., Cappello, A., Neri, M., Bilotta, G., H�erault, A., Ganci, G., 2013. Lava flowhazards at Mount Etna: constraints imposed by eruptive history and numericalsimulations. Sci. Rep. 3, 1–9. https://doi.org/10.1038/srep03493.

Dennedy-Frank, P.J., Muenich, R.L., Chaubey, I., Ziv, G., 2016. Comparing two tools forecosystem service assessments regarding water resources decisions. J. Environ.Manag. 177, 331–340. https://doi.org/10.1016/j.jenvman.2016.03.012.

Doocy, S., Daniels, A., Murray, S., Kirsch, T.D., 2013. The human impact of floods: ahistorical review of events 1980–2009 and systematic literature review. PLoS Curr 5.https://doi.org/10.1371/currents.dis.f4deb457904936b07c09daa98ee8171a.

D’Alessandro, W., Bellomo, S., Parello, F., 2012. Fluorine adsorption by volcanic soils atMt. Etna, Italy. Appl. Geochem. 27, 1179–1188. https://doi.org/10.1016/j.apgeochem.2012.02.028.

Federico, C., Longo, M., D’Alessandro, W., Bellomo, S., Bonfanti, P., Brusca, L., 2017.Hydrological versus volcanic processes affecting fluid circulation at Mt. Etna:inferences from 10 years of observations at the volcanic aquifer. Chem. Geol. 452,71–84. https://doi.org/10.1016/j.chemgeo.2017.01.004.

Ferrara, V., 1975. Idrogeologia del versante orientale dell’Etna. In: Proceedings of the 3rdInternational Conference on Groundwaters, Palermo, Italy, pp. 91–144.

Ferrara, V., Pappalardo, G., 2008. The hydrogeological map of the Etna volcanic massif asuseful tool for groundwater resources management. Ital J Eng Geol Environ (SpecialIssue) 1, 77–89.

Garibaldi, L.A., Steffan-Dewenter, I., Winfree, R., Aizen, M.A., Bommarco, R.,Cunningham, S.A., 2013. Wild pollinators enhance fruit set of crops regardless ofhoney bee abundance. Science 339, 1608–1611. https://doi.org/10.1126/science.1230200.

Geneletti, D., 2011. Reasons and options for integrating ecosystem services in strategicenvironmental assessment of spatial planning. Int J Biodivers Sci Ecosyst Serv Manag7, 143–149. https://doi.org/10.1080/21513732.2011.617711.

Giammanco, G., Giammanco, S., Valenza, M., 1995. Quality of Mount Etna groundwatersutilized for the potable supply. Inquinamento 37, 56–60.

Giannini, T.C., Cordeiro, G.D., Freitas, B.M., Saraiva, A.M., Imperatriz-Fonseca, V.L.,2015. The dependence of crops for pollinators and the economic value of pollinationin Brazil. J. Econ. Entomol. 108, 849–857. https://doi.org/10.1093/jee/tov093.

Greenleaf, S.S., Kremen, C., 2006. Wild bee species increase tomato production andrespond differently to surrounding land use in Northern California. Biol. Conserv.133, 81–87. https://doi.org/10.1016/j.biocon.2006.05.025.

Halliday, T.J., Lynham, J., de Paula, A., 2019. Vog: using volcanic eruptions to estimatethe health costs of particulates. EJournal 129, 1782–1816. https://doi.org/10.1111/ecoj.12609.

Hanley, N., Breeze, T.D., Ellis, C., Goulson, D., 2015. Measuring the economic value ofpollination services: principles, evidence and knowledge gaps. Ecosyst Serv 14,124–132. https://doi.org/10.1016/j.ecoser.2014.09.013.

He, C., Zhang, D., Huang, Q., Zhao, Y., 2016. Assessing the potential impacts of urbanexpansion on regional carbon storage by linking the LUSD-urban and InVEST models.Environ. Model. Software 75, 44–58. https://doi.org/10.1016/j.envsoft.2015.09.015.

Kareiva, P., Tallis, H., Ricketts, T.H., Daily, G.C., Polasky, S., 2011. Natural Capital:Theory and Practice of Mapping Ecosystem Services. Oxford University Press.

Klein, A.M., Vaissiere, B.E., Cane, J.H., Steffan-Dewenter, I., Cunningham, S.A.,Kremen, C., Tscharntke, T., 2007. Importance of pollinators in changing landscapesfor world crops. P Roy Soc B-Biol Sci. 274, 303–313. https://doi.org/10.1098/rspb.2006.3721.

La Via, G., 2009. Analisi economiche delle produzioni e del mercato della frutticolturaetnea e ruolo dell’azione pubblica. Bonanno Editore, Catania.

Lal, R., 2004. Soil carbon sequestration impacts on global climate change and foodsecurity. Science 304, 1623–1627. https://doi.org/10.1126/science.1097396.

Lo Verde, G., La Mantia, T., 2011. The role of native flower visitors in pollinating Opuntiaficus-indica (L.) Mill., naturalized in Sicily. Acta Oecol. 37, 413–417. https://doi.org/10.1016/j.actao.2011.05.006.

Loughlin, S.C., Sparks, R.S.J., Brown, S.K., Jenkins, S.F., Vye-Brown, C. (Eds.), 2015.Global Volcanic Hazards and Risk. Cambridge University Press, UK.

Mandle, L., 2019. Natural Capital Project, Database of Publications Using InVEST andOther Natural Capital Project Software. Available at: https://purl.stanford.edu/bb284rg5424.

Mandle, L., Tallis, H., 2016. Spatial ecosystem service analysis for Environmental ImpactAssessment of projects. In: Geneletti, D. (Ed.), Handbook on Biodiversity andEcosystem Services in Impact Assessment. Edward Elgar Publishing, Cheltenham, UK.https://doi.org/10.4337/9781783478996.

11

Marske, K.A., Ivie, M.A., Hilton, G.M., 2007. Effects of volcanic ash on the forest canopyinsects of Montserrat, West Indies. Ecol. Entomol. 36, 817–825. https://doi.org/10.1093/ee/36.4.817.

Marti, J., Ernst, G. (Eds.), 2009. Volcanoes and the Environment. Cambridge UniversityPress, UK.

McDonald, G.W., Smith, N.J., Kim, J., Cronin, S.J., Proctor, J., 2017. The spatial andtemporal ‘cost’ of volcanic eruptions: assessing economic impact, businessinoperability, and spatial distribution of risk in the Auckland region, New Zealand.Bull. Volcanol. 79, 1–13. https://doi.org/10.1007/s00445-017-1133-9.

Missemer, A., 2018. Natural capital as an economic concept, history and contemporaryissues. Ecol. Econ. 14, 90–96. https://doi.org/10.1016/j.ecolecon.2017.07.011.

Molist, J.M., 2017. Assessing Volcanic Hazard: A Review. Oxford Handbooks. https://doi.org/10.1093/oxfordhb/9780190699420.013.32 2017.

Morse, R.A., Calderone, N.W., 2000. The Value of Honey Bees as Pollinators of US Cropsin 2000. Bee Culture Magazine, A.I Root Company, Ohio.

Muntean, M., Guizzardi, D., Schaaf, E., Crippa, M., Solazzo, E., Olivier, J.G.J., Vignati, E.,2018. Fossil CO2 Emissions of All World Countries—2018 Report. Publications Officeof the European Union, Luxembourg.

National Academies of Sciences, 2017. Engineering, and Medicine, Volcanic Eruptionsand Their Repose, Unrest, Precursors, and Timing. The National Academies Press,Washington.

Neild, J., O’Flaherty, P., Hedley, P., Underwood, R., Johnston, D., Christenson, D., 1999.Impact of a Volcanic Eruption on Agriculture and Forestry in New Zealand. MAFPolicy Technical Paper 99/2.

Neugarten, R.A., Langhammer, P.F., Osipova, E., Bagstad, K.J., Bhagabati, N.,Butchart, S.H.M., Dudley, N., Elliott, V., Gerber, L.R., Gutierrez Arrellano, C.,Ivani�c, K.-Z., Kettunen, M., Mandle, L., Merriman, J.C., Mulligan, M., Peh, K.S.-H.,Raudsepp-Hearne, C., Semmens, D.J., Stolton, S., Willcock, S., 2018. Tools forMeasuring, Modelling, and Valuing Ecosystem Services: Guidance for KeyBiodiversity Areas, Natural World Heritage Sites, and Protected Areas. IUCN, Gland,Switzerland.

Ochoa, V., Urbina-Cardona, N., 2017. Tools for spatially modeling ecosystem services:publication trends, conceptual reflections and future challenges. Ecosyst Serv 26,155–169. https://doi.org/10.1016/j.ecoser.2017.06.011.

Papale, P., Shroder, J.F., 2014. Volcanic Hazards, Risks and Disasters. Elsevier,Amsterdam.

Polasky, S., Nelson, E., Pennington, D., Johnson, K., 2011. The impact of land-use changeon ecosystem services, biodiversity and returns to landowners: a case study in thestate of Minnesota. Environ. Resour. Econ. 48, 219–242. https://doi.org/10.1007/s10640-010-9407-0.

Posner, S., Verutes, G., Koh, I., Denu, D., Ricketts, T., 2016. Global use of ecosystemservice models. Ecosyst Serv 17, 131–141. https://doi.org/10.1016/j.ecoser.2015.12.003.

Redhead, J.W., Stratford, C., Sharps, K., Jones, L., Ziv, G., Clarke, D., Oliver, T.H.,Bullock, J.M., 2016. Empirical validation of the InVEST water yield ecosystem servicemodel at a national scale. Sci. Total Environ. 569–570, 1418–1426. https://doi.org/10.1016/j.scitotenv.2016.06.227.

Scaini, C., Bolic, T., Castelli, L., Folch, A., 2013. A GIS-based tool for the estimation ofimpacts of volcanic ash dispersal on European air traffic. In: Proceedings of the 3rdSESAR Innovation Days, Barcelona, Spain.

Sharp, R., Tallis, H.T., Ricketts, T., Guerry, A.D., Wood, S.A., Chaplin-Kramer, R.,Nelson, E., Ennaanay, D., Wolny, S., Olwero, N., Vigerstol, K., Pennington, D.,Mendoza, G., Aukema, J., Foster, J., Forrest, J., Cameron, D., Arkema, K.,Lonsdorf, E., Kennedy, C., Verutes, G., Kim, C.K., Guannel, G., Papenfus, M., Toft, J.,Marsik, M., Bernhardt, J., Griffin, R., Glowinski, K., Chaumont, N., Perelman, A.,Lacayo, M., Mandle, L., Hamel, P., Vogl, A.L., Rogers, L., Bierbower, W., Denu, D.,Douglass, J., 2018. InVEST 3.6.0 User’s Guide. The natural capital project, Stanford.

Signorello, G., Prato, C., Marzo, A., Ientile, R., Cucuzza, G., Sciandrello, S., Martínez-L�opez, J., Balbi, S., Villa, F., 2018. Are protected areas covering importantbiodiversity sites? An assessment of the nature protection network in Sicily (Italy).Land Use Pol. 78, 593–602. https://doi.org/10.1016/j.landusepol.2018.07.032.

Tallis, H., Polasky, S., 2009. Mapping and valuing ecosystem services as an approach forconservation and natural-resource management. Ann. N. Y. Acad. Sci. 1162,265–283. https://doi.org/10.1111/j.1749-6632.2009.04152.x.

Tuck, B.H., Huskey, L., 1994. Economic disruptions by Redoubt volcano: assessmentmethodology and anecdotal empirical evidence. US Geol. Surv. Bull. 2047, 137–140.

Tuck, B.H., Huskey, L., Talbot, L., 1992. The Economic Consequences of the 1989–1990Mt. Redoubt Eruptions. University of Alaska, Anchorage, Institute of Social andEconomic Research. Unpublished report prepared for the United States GeologicalSurvey Alaska Volcano Observatory.

van den Berg, J.C.J.M.A., Botzen, W.J.W., 2015. Monetary valuation of the social cost ofCO2 emissions: a critical survey. Ecol. Econ. 114, 33–46. https://doi.org/10.1016/j.ecolecon.2015.03.015.

Veschetti, E., Maresca, D., Lucentini, L., Ferretti, E., Citti, G., Ottaviani, M., 2007.Monitoring of V (IV) and V (V) in Etnean drinking-water distribution systems by solidphase extraction and electrothermal atomic absorption spectrometry. Microchem. J.85, 80–87. https://doi.org/10.1016/j.microc.2006.05.005.

Vigerstol, K.L., Aukema, J.E., 2011. A comparison of tools for modeling freshwaterecosystem services. J. Environ. Manag. 92, 2403–2409. https://doi.org/10.1016/j.jenvman.2011.06.040.

Villa, F., Bagstad, K.J., Voigt, B., Johnson, G.W., Portela, R., Honz�ak, M., Batker, D., 2014.A methodology for adaptable and robust ecosystem services assessment. PloS One 9,e91001. https://doi.org/10.1371/journal.pone.0091001.

G. Signorello et al. Environmental and Sustainability Indicators 8 (2020) 100056

Wilson, G., Wilson, T.M., Deligne, N.I., Cole, J.W., 2014. Volcanic hazard impacts tocritical infrastructure: a review. J. Volcanol. Geoth. Res. 286, 148–182. https://doi.org/10.1016/j.jvolgeores.2014.08.030.

Zhang, L., Dawes, W.R., Walker, G.R., 2004. Response of mean annual evapotranspirationto vegetation changes at catchment scale. Water Resour. 37, 701–708. https://doi.org/10.1029/2000WR900325.

12

Zuccaro, G., Leone, M.F., Del Cogliano, D., Sgroi, A., 2013. Economic impact of explosivevolcanic eruptions: a simulation-based assessment model applied to Campania regionvolcanoes. J. Volcanol. Geoth. Res. 266, 1–15. https://doi.org/10.1016/j.jvolgeores.2013.09.002.