Restoration of badlands and natural capital: an application in Saldaña (Palencia, northern Spain)

Abstract

The environment provides many necessary services for a plethora of human activities related to that are being acknowledged as our ecosystems degrade; therefore, ecosystem-based solutions are becoming increasingly more important. Hydrologic restorations (HRs) apply this type of ecosystem approach; there are multiple examples around the Mediterranean area executed since the beginning of the twentieth century. One good example of these restoration-type solutions can be seen in the badlands of Saldaña (Palencia, Spain). This case has been particularly successful in stopping gully erosion from badlands. Despite its importance for the well-being of the inhabitants of Saldaña and surroundings, the economic effects of these services which were put into place by this HR have never been assessed. In the present study, four different and compatible ecosystem restored services have been valued in monetary terms. For their aggregation, an innovative approach has been taken using the analytic hierarchy process methodology, so the weight of each ecosystem service (ES) can be determined. The economic flow of these ESs has been calculated in terms of annual income and updated values, from which natural capital increase in the population of Saldaña can be assessed. After analysing the results, it can be concluded that the village has earned more than seven million euros, which, in turn, generates important benefits to its inhabitants.

Keywords:

• ecosystem services
• hydrologic restoration
• environment valuation
• badlands

1. Introduction

Technological development through reason has satisfied human necessities and expectations throughout history, searching a relative ‘emancipation from nature’. This current peak moment might be modern age during illustration, characterized, among others, for this type of disengagement through reason (Rodríguez Rial, 2001). However, in the contemporary age in which we live now, goods and services taken from the environment are being recognized, suggesting their vital importance to human development and its urgent need for conservation. Humans depend on all different ecosystems and the services they provide, such as food, water, disease management, climate regulation and even spiritual fulfilment or aesthetic enjoyment (Millennium Ecosystem Assessment [MEA], 2005). Ecosystem services (ESs) are the benefits that people obtain from ecosystems (Liu, Costanza, Farber, & Troy, 2010; MEA, 2005). The first step towards a comprehensive assessment of ecosystem goods and services involves the translation of environmental complexity (structures and processes) into a more limited number of ecosystem functions, defined as ‘the capacity of natural processes and components to provide goods and services that satisfy human needs, directly or indirectly’ (de Groot, Wilson, & Boumans, 2002, p. 394). This proper identification of the ES will prevent double counting (Morse-Jones, Luisetti, Kerry Turner, & Fisher, 2011) by selecting only ESs that are compatible with each other.

There are several examples of public policies that have considered ES in decision processes (Turner & Daily, 2008). In 1991, the American Environmental Protection Agency (EPA) ordered New York authorities to construct a new water filtration plant, unless they could prove the availability of drinking water which in the past had even been bottled for commercial use without the need for filtering. The initial cost for this water plant would require US$ 8 billion and annual conservation costs rising to US$ 300 million/year. Instead of building the plant, the Environmental Protection Department of New York executed a land acquisition plan to reforest and restore the reservoir system watershed with an investment of US$ 2 billion. This approach saved US$ 6 billion, giving an internal rate of return of 90–170% and a payback period of 4 to 7 years. This return is one order of magnitude higher than usually available, particularly on relative risk-free investments (Liu et al., 2010).

Hydrologic restorations (HRs) are another solution based on an ES approach at a large scale (Trabucchi, Ntshotsho, O’Farrell, & Comín, 2012) with demonstrated positive effects (Mintegui, 1989). Soil degradation, due to ineffective watershed management, has important consequences on hydroelectric power production, irrigation systems, agriculture and fishing (Aylward, Echeverría, & Barbier, 1995; Pimentel et al., 1995), and HRs have proved to be effective in controlling soil loss. For example, in 1918, after floods had damaged parts of Málaga (Spain), and after fixing several structural solutions, it was necessary to reforest 4192 ha in the Guadalmedina river watershed after suffering severe degradation due to continuous land-use change and forest thinning since the seventeenth century. Other famous cases from the same period include the Jiloca watercourse small dams construction for the railway, the restoration of torrents in the Pyrenees to protect the Canfranc International Railway Station or the creation of the Public Utility Forests Catalogue to exclude some forests from the government seizure process during nineteenth century, in order to avoid their forest clearing and loss of the ecologic and social functions they performed.

Pimentel et al. (1995) estimated that soil conservation practices in the USA could yield an erosion rate reduction up to 100 times less, reducing hydraulic resource loss up to 21.7 times. In their analysis, they concluded that for each dollar invested in erosion control, US$ 5.27 could be saved in relation to the harm caused by sediments.

Another example is the case of Napa in California, where multiple floods of the Napa river generated US$ 500 million of damage in the past years. By the late 1990s, some citizens proposed a ‘Living River’ approach; instead of investing in protection infrastructure (reinforcing levees and concrete barriers), they proposed an ecosystem services framework (ESF) hoping to better guide flood control investments. This meant moving 9 bridges and over 100 buildings and restoring 250 ha of floodplain. This approach costs US$ 50 million more than the structural solution, but the local population preferred it expecting certain fringe benefits such as fish recovery, biodiversity, landscape scenery, recreation, tourism and commerce. Indeed, Napa was revitalized predominantly by this innovative solution; according to the City of Napa’s Economic Development Office, private investment increased in US$ 193 million from 1999 to 2005 (Brauman, 2006).

All these cases are examples of a cost–benefit analysis comparing structural solutions and environmental ones (Liu et al., 2010). In order to perform this analysis, it is necessary to assess the economic results of the ESF; therefore, it implies an environmental valuation exercise.

The adoption of this new approach is crucial considering that since the twentieth century conservation efforts have been solely justified on scientific and moral/ethical grounds; however, continued and increasing pressure on natural resources suggests that these arguments may no longer be enough (Morse-Jones et al., 2011). The reigning economic paradigm of the last century, neoclassical economics, is becoming inadequate as human societies transition from an empty to a full world (Daly, 2005); it does not achieve to value the finite, scarce and nonsubstitable resources in the economy that make up natural capital, once considered plentiful (Day, Moerschbaecher, Pimentel, Hall, & Yáñez-Arancibia, 2014). Human activities depend on the flow of energy and matter of the earth, considered as a closed system in which only energy enters and exits. Neoclassical economy views the environment as a subsystem of the economy when, in fact, the economy is a subsystem of the environment and is subject to the same biophysical laws that govern the planet (Daly, 1993, 2005; Day et al., 2014). Therefore, economy needs to recognize its limits and learn to manage natural resources and services on which it depends, as these become increasingly scarcer. However, without measurements, it becomes hardly effective to manage earth’s natural capital (Wackernagel, Kitzes, Moran, Goldfinger, & Thomas, 2006). Due to the serious and urgent nature of this problem, considerable effort has recently been directed towards more systematic approaches to measuring, modelling and mapping of ESs, governance analysis and valuation (Murray-Rust et al., 2011; Turner & Daily, 2008).

One of the best known cases is by Costanza et al. (1997), which valued the natural capital of the planet at US$ 33 trillion/year. In Spain, an equivalent study was made by the Ministry of Environment titled ‘Valuation of Spain’s Natural Assets’ (Esteban Moratilla, 2010) that has come to valuate 17 ESs existing in the country at 732 million euros in 2005 current prices. Apart from these global-scale studies, there are other more precise studies at local or regional scales (i.e. Samos Juárez & Bernabeu Cañete, 2011; Voces González, Díaz Balteiro, & López-Peredo Martínez, 2010).

There are several methodologies to valuate these ESs (de Frutos Madrazo, 2008) which are classified as direct methods, when values are obtained by estimating the willingness to pay (WTP) of the population, and indirect methods, when natural assets are combined with productive inputs to develop ESs that generate a utility flow. On the other hand, the benefit transfer method is much less expensive to apply; this is defined as the monetary transposition of environmental values, estimated with valuation techniques based (or not) in the market, from one site (original study site) to a different situation where an action is carried out (policy site) (Brouwer, 2000).

Despite the existing numerous examples, few studies have assessed the ES stemming from HRs, apart from other technical assessments that surely exist. The present work attempts to fill this gap by the monetary valuation of the ES gained by an exemplary HR executed in the Duero basin: the HR of Saldaña’s badlands. This was carried out by the Duero Basin Authority in the early twentieth century to stop gully erosion in the Carrión river hillsides located in Saldaña’s village. Before the HR, mud flows and massive movements from gully heads and river margins were common. These processes filled the streets of Saldaña with mud, contaminating the Carrión river with high suspended sediment loads, collapsing irrigation channels and stopping traffic in the Saldaña – Osorno road. All these affected intensively the common life of Saldaña and other villages downstream the Carrión river. Restoration activities were undertaken from 1932 to 1936, and some inhabitants were involved with their development. They did manual tree planting, with 3000 plants/ha (Pinus, Ulmus and Robinia genera) and built some hundreds of wattle fences and check dams; 8 years after the end of the HR, erosion has been almost totally controlled and the landscape has changed drastically; where it was all badlands, now there is a forest covering the hill slopes, making vegetation the main landscape component together with geomorphology (Navarro, de Araújo, & Mongil, 2014).

The four ESs considered in this study are: wood products, water quality improvement, landslide risk mitigation and landscape scenery improvement. For each of them, the best available valuation method has been used; they are, respectively, wood market price method, avoided cost method, adjusted benefit transfer by income and risk valuation and, finally, hedonic pricing method. The principal aim of this study was to be as rigorous as possible when applying each method; in this way, the results obtained can contribute to valuation databases in the Mediterranean regions and ultimately to be transferred to another policy site.

All of these ESs are directly or indirectly related to the beneficiary population; therefore, they can be considered direct or indirect use values (de Frutos Madrazo, 2008). It is not the scope of this study to assess option, bequest or existence values. The dependence of inhabitants on ES is a function of geographical conditions, cultural values and interests that vary within each region. Therefore, in order to determine the economic value (EV) of these ESs, an aggregation of results has been performed using an analytic hierarchy process (AHP) methodology (Saaty, 1980, 1990). This is an innovative approach, as it has allowed us to derive each assessed ES’s effect on the welfare of the local and regional population, taking into account their preferences and importance of each ES in their well-being. Only then, the aggregated EV has been calculated, considered as the minimum EV of gains obtained by the HR.

Given these results it is apparent that by considering an economic point of view, environmental conservation and restoration are vital to the development of society (TEEB, 2010), since its degradation would require the application of expensive technologies to supply lost ES, raw materials production would be diminished, and society would lack the possibility to satisfy other indirect values, such as landscape and its enjoyment.

In the same way, this study is an example of a practical and rigorous application of the European Water Framework Directive (OJ, 2000), as it responds to the needs of member states to valuate, in an integral way, measures taken to improve water quality. Thus, those that are most cost-effective relating to their social benefit can be selected.

2. Study site: forest restoration of Saldaña’s slopes

The study site is located in the province of Palencia, between Saldaña and Lobera municipalities (Figure 1) occupying 3.17 km². This is part of the original 9 km² area restored in 1930 by Forest Engineer Ayerbe from the Duero Basin Authority. The mean annual temperature (T) is 9.3°C, and the mean maximum and mean minimum temperatures are 15°C and 3.5°C, respectively. The potential evapotranspiration (PET) is 610.6 mm, and mean annual precipitation (P) is 577 mm. The rainfall erosivity R factor (Wischmeier & Smith, 1978) is 720 MJ·mm·ha⁻¹·h⁻¹. This zone is characterized by a continental Mediterranean climate with long winters, short summers and low erosivity rainfall. The United Nations Environment Programme (UNEP, 1997) aridity index (P/PET) is 0.94 (> 0.65); although it corresponds to a humid zone, the mean annual rainfall can be considered low. The Lang (1915) pluviosity index (P/T) is 62, which corresponds to a semi-arid/warm temperate climate; however, according to the Bagnouls and Gaussen (1957) bioclimatic classification, the study area is considered to be temperate Mediterranean.

Figure 1. Situation of Saldaña (Spain).

Saldaña’s hillsides lie within a morphologic unit called detritic Paramo (Ceñal, Glaría, & Blanco, 1988) of miocen-plioquaternary age. They form a group of alluvial fans developed by mountains in northern Palencia, which date prior to the actual fluvial network, which is now responsible for its morphology. In this way, Carrión river has excavated on the left margin some hundred meters long slope among these materials, covered by a variable thickness of pebbles coming from original alluvial fans (Hernández Pacheco, 1928). This layer is formed by a clay sandy matrix with semiquartzitic decimetric pebbles, as thick as 30 m. It is connected through gullied slopes to former deposits made of pebbles and quartzitic sands, located in terraces of Carrión river. Gullies are generally orientated to the southwest which provides the best sun exposure.

Before the HR, it is probable that existing soils were classified as entisol order, orthent suborder (NRCS, 2003) or leptosol (FAO, 1998). As a consequence of inadequate previous human activities (ancient wars, forest clearing, cattle raising, etc.), a progressive desertification process began from the fifteenth century until the beginning of the twentieth century, after which the aforementioned materials were exposed, so no real edaphic soil existed before the HR. For more information about the history of Saldaña’s hillsides and its HR, see Navarro Hevia, Mongil Manso, and Araújo, 2013, 2014).

After the disappearance of natural forest, the soil layer was easily removed as a consequence of the naked land, intensive erosive processes derived from high slopes, soil crust and continuous action of ice and rainfall. Therefore, original clay sandy matrix with pebbles from the terrain was exposed to severe weather conditions. Nevertheless, under the restored forest canopy, soils are developing slowly. In general, soil texture is diverse due to its alluvial origin, ranging from loamy to clayey soils, with a predominantly massive structure.

Gully network ranges from an altitude of 900 to 990 m, with a mean altitude of 950 m. Mean slopes stand at 43% and in some cases can reach up to 60%. Hydrographically, it lies within the left margin of the Carrión river, which is the main river in the region and a right-bank tributary of the Pisuerga river. They are located between 40 and 200 m apart from the Carrión river, which is representative of the importance of these slopes for the sediment load to the watercourse.

Potential vegetation corresponds to an oakwood of Pyrenean oak (Quercus pyrenaica Willd.). Nevertheless, continuous human presence has produced different vegetation covers. Nowadays, and after eight decades since the beginning of the HR, a Pinus sylvestris L. forest covers a zone formerly characterized by an intensely desertified landscape (Figures 2 and 3), with minimal prairie vegetation as a consequence of wars, overgrazing, wood coaling, forest fires, etc. (Navarro et al., 2014).

Figure 2. Slope state before the HR (Duero Basin Authority, 1930).

Figure 3. Slope state nowadays (JMB, April 2012).

3. Materials and methods

3.1. Wood products valuation

Forest wood products can be considered a direct extractive value (de Frutos Madrazo, 2008). Sawmill importance in the Castile and Leon regions (Figure 1) has been diminishing throughout the years, passing from a 1148 million euro invoicing in 2002 to 785 million euro in 2009 (Cear-Madera, 2010). However, it still employed 7843 locals in the autonomous region in 2009 (Cear-Madera, 2010) and is supplier of wood to important sectors such as carpentry, furniture and construction. Its valuation has been performed using the market price method, determining the value of infinite cuttings for each thinning rotation of T years according to the following equation (Ortuño Pérez, Madrigal Collazo, & González Doncel, 2007):

(1)

In this equation, R_T is the capitalized income over year T, and i is the discount rate considered.

The income has been calculated using wood market prices (Forest Products Commercial Bulletin [FPCB], 2008), the wood volume has been determined by an angular sampling performed in 2012 in three stations, and to determine wood volume for the thinning rotation (established in 120 years) Madrigal et al.’s (1999) yield tables have been used. Considering the soil and meteorology of the study area, very low site quality yield tables have been used.

3.2. Water quality improvement valuation

Water quality improvement can be considered an indirect extractive value (de Frutos Madrazo, 2008). This ES has been assessed using the avoided cost method. First, it has been determined the suspended load Carrión river would carry in the case that the HR had not been executed; and second, the cost of eliminating this suspended load has been assessed using investment and physical treatment exploitation costs in the wastewater treatment plant (WWTP) already executed in Saldaña (López Herrero, 1997). To calculate the suspended sediment load before the HR, we have recurred to Ayerbe’s reference (1930, p. 9) that states that ‘normal floods in the Carrión river present a suspended load of 250 g/m³ before the works are being performed’ (Figure 4). To apply this statement, it became necessary to calculate bankfull discharge. To do so, water discharge data from the closest gauging station have been adjusted to a Gumbel-type probability distribution function, and a 3-year return period discharge has been calculated for the normal flood (Martín Vide, 2002). From the gauging station database, it has been determined that the minimum duration of this flood is 2 days (Figures 5 and 6).

Figure 4. Mud flows before the HR (Duero Basin Authority, 1930).

Figure 5. Actual state of the Carrión river and its shore; today it is a fishing reserve (Joaquín Navarro Hevia).

Figure 6. An example of a landslide occurring nowadays (June 2011).

On the other hand, when applying the avoided cost method, it is necessary to calculate the annual equivalent cost (AEC) using Equation (2) (CIS, 2003), in which I is the investment made, AOC are the ordinary exploitation costs, and the WWTP lifespan (n) is 35 years.

(2)

3.3. Valuation of landslide risk mitigation

Landslide risk mitigation can be considered a direct nonextractive value (de Frutos Madrazo, 2008). It is an ES that the inhabitants of Saldaña cannot recognize as current generations have not suffered any devastating effects of massive movements in the slopes. This is the reason why, in this case, the benefit transfer method has been used, using the valuation work (Olschewski, Bebi, Teich, Wissen Hayek, & Grêt- Regamey, 2012) of a study site in which the local population can perceive this type of risk. It is a small town located in the Swiss Alps. This study determines the WTP of its 1250 inhabitants and 1500 visitors to conserve forest canopy that prevents avalanches, using choice experiment methodology. This method is based on surveying the local population in order to obtain their preferences for goods and services, describing them in terms of their attributes and levels that can be satisfied. Olschewski et al. (2012) conclude that the WTP for a family from this town for the protection of forests is US$ 56. This value has been transferred to the policy site of Saldaña and adjusted considering two effects: risk effect, comparing damage made in each of the sites, and income effect, comparing differences in purchasing power between Swiss and Spanish populations.

3.4. Landscape scenery valuation

To better understand the valuation of the landscape from the study area, it is wise to do a little review of the HR’s role in the economic development of the region. Saldaña is the most important town in the region bordering the mountainous area in northern Palencia. The economy of the village is based on agriculture, highlighting the importance of irrigation agriculture in which Saldaña’s bean is of great gastronomic value. The area is not a tourism attraction pole, but in recent years the number of visitors has increased due to the opening of the visitor centre in ‘La Olmeda’ roman village (around 250,000 tourists a year). Compared to other municipalities in the province, second homes are abundant (40%) and they are located in the greenest area of the municipality, close to the HR or the restored river. Economic activity and population have remained stable over the last 10 years (La Caixa economic annual directory, 2014), about 3000 residents.

The landscape scenery improvement due to the HR can be considered an indirect nonextractive value (de Frutos Madrazo, 2008). In order to assess this value, hedonic pricing methodology has been used, by determining the average surplus derived from landscape for single-family housing prices, which are those overlooking the HR in Saldaña. Used demand function for house prices is the one developed by de Frutos Madrazo and Esteban Laleona (2006) (Equation (3)). This was originally developed to determine the effect of green spaces view in the city of Soria, which belongs to the same autonomous community as Saldaña, Castile and Leon. In this study, it was determined that average house prices in the city of Soria were a function of house area, the existence of heating facilities, solar orientation and view of green spaces.

(3)

From Equation (3), it can be determined that the consumer surplus (considered as the consumer’s WTP to satisfy this variable) E_i for any independent variable X_i and house price P_i, is:

(4)

4. Analytic hierarchy process (AHP methodology)

This methodology was created by Saaty (1977, 1980) as a structured technique for decision-making in multicriteria contexts and to determine the relative importance, or weight, of different criteria considered by any decision-making unit. AHP can be used to valuate goods and noncommercial services as it can determine social preferences for goods or services taking into account the relative importance given to each service. The advantage of this method is that significant results can be obtained using just a group of individuals (experts or not) in the investigation subject (Ananda & Herath, 2003; Dwivedi & Alavalapati, 2009), over other statistical methods that require big samples to obtain relevant results. In AHP methodology, first there must be a fixed main objective; in this case, it is to obtain the EV of the HR. The next step is to define the attributes or criteria which will fulfil this objective. In the present study, these attributes are the EVs obtained for each ES (Figure 7).

Figure 7. Attributes to determine economic value of the HR.

AHP has the capability to obtain weights assigned to each ES value. The weighted sum of these values will allow us to quantify the aggregated EV of the HR. To do so, the AHP states that attributes must be compared pairwise in a way that can determine the intensity of preference among each paired comparison, that is, establishing ratios for the relative importance of each attribute in every comparison. As Saaty (1980) proposes, the preference intensity can be established through a linear scale ranging from 1 (same importance among measurements) to 9 (absolute importance of one measurement over the other). Table 1 elaborates on the scale for each pair comparison used in this study.

Table 1. Definition of the preference scale for attributes pairwise comparisons.

Importance index	Scale definition
1	Both attributes are equally important
3	Preferred attribute is moderately more important than the other
5	Preferred attribute is strongly more important than the other
7	Preferred attribute is very strongly more important than the other
9	Preferred attribute is absolutely more important than the other

The Saaty matrix is obtained where each element a_ijk represents the value obtained from the comparison between attribute i and attribute j for each k individual; that is, the number of times that the attribute i is preferred to the attribute j. Once Saaty matrices are obtained for each surveyed k individual, the weight (W_i) of each attribute is obtained. The process presented by Saaty (1980) detects inconsistencies provoked by contradictory responses of the surveyed individuals by calculating the consistency ratio (CR). Therefore, following Saaty’s criterium (1980), it has been checked that the CR values in each matrix are always below 15%, which is the limit that is considered adequate to validate obtained results. In the case of perfect consistency, weights of all n attributes could easily be determined by the n·(n−1)/2 values of a_ijk that have been declared. Since perfect consistency is quite rare due to the subjectivity of the surveyed person, different methods have been proposed to estimate the priority vector; the most common methods are eigenvalues and the row geometric mean.

As there is no documented evidence in the literature as to the benefit or advantage of any of these methods, in the present study the aggregated vector for each attribute (ŵ_i) has been calculated by the row geometric mean for each individual (Equation (5)).

(5)

in which:

is the attribute weight for individual k.

is the comparing value for attribute i.

Once ŵ_ik weights are obtained, they must be normalized so that:

(6)

In our study, to develop the AHP method, a selection of stakeholders was previously made, representing a broader group of agents directly or indirectly related to the HR, either because they work in areas related to water, forest or environment management, or because they are agents that directly perceive the effects of the HR as they live in the area (see Table 2). Interviews included seven technicians responsible for the management and direction of river works in the catchment area, three experts in forestry and environmental management and three Saldaña residents, one of whom participates in municipal management. Therefore, a total of 13 surveys were conducted.

Table 2. Stakeholders involved in the surveys.

Stakeholder profile	Number
1. Technicians in the catchment area	7
2. Forestry and environmental management experts	3
3. Saldaña inhabitants	3
Total	13

These surveys supposed a pairwise comparison of the importance of each ES according the each expert. All surveys presented valid CR (Saaty, 1980, 1990).

5. Total value and annual surplus

A recurring aspect in environmental valuation studies is the relation between natural capital (V₀) and its generated income (r). ESs are the externalities the environment produces continuously by the simple operation of physical and ecological processes on which they are based. Updating an infinite flow of these incomes for a given discount rate i, it can be demonstrated that (Equation (7)):

(7)

6. Results

6.1. Wood products

From the angular sampling performed in the policy site and recurring to most suitable yield tables (Madrigal Collazo et al., 1999), a wood volume of 882.2 m³/ha for a rotation age of 120 years has been calculated. Interspersed pruning and clearing costs were omitted due to their low values. Considering that wood sells for 22.85 €/m³ (FPCB, 2008) after the final harvest and that reforestation costs 1500 €/ha (in 2012 prices), and keeping in mind that the actual forest mass is 80 years old, the income for the final harvest in the rotation age (120-year-old mass) was calculated to be 54,160 €/ha. This involved a total income R_T (Equation (1)) of 17 million euros for the entire reforested area which measures 317 ha. It must be noted that this datum refers to its value at its harvesting age, in 40 years. Applying Equation (1) for infinite rotation harvests of 120 years and a 3% discount rate, the actual value of wood products was 509.290 euros.

6.2. Water quality improvement

Following Martín Vide (2002), it was estimated that bankfull discharge in the Carrión river corresponded to a return period (T) of 3 years. Using discharge data from the closest gauging station (Confederación Hidrográfica del Duero [CHD], 2012), located just 12 km upstream from the study site and having data for 9 hydrologic years, the following Gumbel equation was adjusted (Equation (8)):

(8)

Solving this function for a return period of 3 years, it was determined that the bankfull discharge in this part of the Carrión river should be around 10 m³/s. When comparing the dates in which the maximum mean flows and peak flows were recorded, it could be determined that these floods lasted at least 2 days. On the other hand, knowing that the suspended sediment load for these events before the HR was 250 g/m³ (Ayerbe, 1930), it was calculated that sediments coming from the slopes in this area of the river should be around 444.11 t/flood.

In the case that the HR had not been completed, the suspended load should be eliminated by oversizing the WWTP in the municipality of Saldaña. In order to know the AEC of this measure (Equation (2)), the WWTP project was consulted (López Herrero, 1997) resulting investment costs for the physical treatment of 858,648 euros and running costs of 43,406 €/year. Therefore, for a WWTP with a lifespan of 35 years, the calculated AEC was 83,336 €/year for the physical treatment.

The next step was to determine the cost overrun for the eliminated suspended sediment load that would enter into the river if the HR had not been completed. Knowing that the WWTP was designed to eliminate 96 solid t/year (263 kg/day), a cost overrun of 868 €/t was obtained; multiplying it by the former value of 444.11 t/flood event and with a return period of 3 years, it was finally determined that the avoided cost should be 128,496 €/year.

6.3. Landslide risk mitigation

The risk for landslides was determined by the cost of road cleaning affected by slope erosion and from the damage caused by massive movements to houses. In the first case, the roads likely to be affected by landslides were digitized from an orthophotography (Figure 8), showing a total length of 5700 m. As it costs 1.08 €/m to clean this roads (Government of Cantabria [GC], 2011), this would mean a total cost of 6177 €.

Figure 8. Roads affected by landslides.

As for the houses, single-family house prices were checked on the Internet and an average, yet conservative, cost was set at 879 €/m². Damage to houses due to massive movements was estimated to occur every 5 years before the HR. In the case of a 100 m² house, this would suppose an annual cost of 17,572 €/year. After adding up both values, a total risk of 23,749 €/year has been calculated. By comparing this value with the risk calculated by Olschewski et al. (2012) of US$ 30,000, it was finally determined that the risk effect for the benefit transfer should be 1.03 for an exchange rate of US$ 1.3/1 €.

On the other hand, according to the Eurostat (http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home/), it was established that the purchase power standard (PPS) index of the Swiss population was 147, whilst the Spanish population had an index of 100. After comparing both values, it was therefore obtained an income effect factor of 0.68.

Olschewski et al. (2012) determined in their study that the WTP at the study site in the Swiss Alps for the conservation of forest defence is US$ 56 per family. Transferring this value to the population of Saldaña and adjusting it by the risk and income effects, it was determined that the WTP for a family in Saldaña to prevent landslides should be 30 €/family/year for an exchange rate of US$ 1.3/1 €. As there are 3183 inhabitants in Saldaña (a total of 796 families), it was finally obtained that the HR generates a consumer-averaged surplus to the population of Saldaña of 23,880 €/year.

6.4. Landscape scenery

After performing an analysis of the effects of the HR in the landscape, it could be concluded that scenery improvement is due, mainly, to the increase of the forest component in the surrounding areas; in turn, this supposes a chromatic enrichment with bare lands and emphasizes terrain forms coming from former gully erosion processes. To evaluate the landscape, the Bureau of Land Management (BLM, 1980) methodology was employed. It uses morphology, vegetation, water, colour, scenic surroundings, rarity and human action as criteria to characterize the landscape into three categories: type A belongs to areas with outstanding characteristics, type B to areas with exceptional characteristics for some criteria and common for others, whilst type C refers to areas with very common characteristics in the physiographic region where it lies. Saldaña and surroundings were classified as type B.

The landscape effect is remarkable when the distribution of the most expensive houses is analysed, as all of them are located very close to the HR. From analysing aerial photography, a total of 51 houses overlooking the woods have been observed. Using the demand function of de Frutos Madrazo and Esteban Laleona (2006) (Equation (3)), it was determined that visual effect had a regression coefficient of 0.232. After a Web search, it could be determined that single-family housing prices in Saldaña and surroundings had a mean value of 137,000 €. Multiplying this value by the regression coefficient (Equation (4)), a consumer surplus of 31,784 €/household was obtained because of the restored slopes, which provides 1,620,984 € for the population of Saldaña considering the 51 single-family houses that overlook the HR.

6.5. Aggregation of ES values

Table 3 presents the updated value and the annual income derived from each ES from Equation (5). The chosen discount rate is 3%; this value is considered adequate for these types of ES (Arrojo Agudo, Bernal Cuenca, Fernández Comuñas, & López Gracia, 1998). The only exception is that of the landscape scenery improvement, where a different discount rate was applied for housing because capitalization rate for this sector in Spain is 1.75% (Martínez Pagés & Maza, 2003).

Table 3. Updated values and annual income for assessed ecosystem services (own source).

Ecosystem service (ES)	Discount rate (i)	Annual income ri (€/year)	Updated value Vi (€)
1. Wood products	3.00%	15,279	509,300
2. Water quality	3.00%	128,496	4,283,200
3. Risk mitigation	3.00%	23,880	796,000
4. Landscape scenery improvement	1.75%	28,367	1,620,984
Total		196,022	7,209,484

On the other hand, according to the AHP surveys, it was concluded that wood products generation and landscape scenery improvement had a similar importance (respectively 11% and 13%), and the most valued services were landslide risk mitigation (47%) and water quality improvement (29%). Table 4 shows the economic valuation of each ES weighted by these values.

Table 4. Economic values of weighted ecosystem services (own source).

Ecosystem service (ES)	Wi (AHP)	Weighted annual income ri (€/year)	Weighted updated value Vi (€)
1. Wood products	0.127	1940	64,681
2. Water quality	0.286	36,750	1,224,995
3. Risk mitigation	0.473	11,295	376,508
4. Landscape scenery improvement	0.114	3234	184,792
Total	1.000	53,219	1,850,976

Therefore, for m attributes, the aggregated EV would be (Equation (9)):

(9)

From Equation (9), it was determined that the aggregated EV of Saldaña’s HR, only considering the four ESs analysed in this study, is 7,403,904 €, which supposes an annual surplus of 212,876 €/year for the population of Saldaña.

The results obtained through AHP analysis are directly conditioned by the selected sample. Variables involved in the valuation of generated ES are well known by all stakeholders considered. From the results of the survey, it could also be concluded that environmental and forestry experts, who are well aware of the problems before the HR, showed a tendency to discriminate positively those ES-considered priorities in their value scale, such as water quality improvement and risk mitigation. On the other hand, local population constitutes a different group that valuated at the same rate all ESs. This type of behaviour, in which environmental awareness leads to a wider acknowledgement of ES, has already been documented in other studies (e.g. Jobstvogt, Watson, & Kenter, 2014).

However, this does not imply the expert judgment being more valid than the one of the local population, as the latter live daily with these ESs; their ratings just belong to a different dimension. It must also be noted that one of the selected experts is also resident in the area, so his opinion stated this dual dimension. Thus, this aspect has also been considered in the sample design.

7. Discussion

As a consequence of inadequate previous human activities (ancient wars, forest clearing, cattle raising, etc.), a progressive desertification process began from the fifteenth century until the beginning of the twentieth century. Saldaña’s surroundings were not resilient enough (Folke, 2006) to adapt to these new conditions, reaching a new environmentally poor equilibrium state (desertified land), in which mud flows and mass movements from the gully heads and the river margins were common. These processes filled Saldaña’s streets with mud, contaminated the Carrión river, collapsed irrigation channels and even road traffic. After the execution of Saldaña’s HR, gully erosion processes have almost completely stopped, accruing several ESs with remarkable environmental and social benefits for the population of Saldaña. ES derived from ecosystem functions (de Groot et al., 2002) did not exist in the previous state, before the execution of the HR; therefore, the value of the ES constitutes an economic surplus to the population of Saldaña. Simply by producing more wood products, improving the water quality, mitigating the risk of landslides and improving the landscape scenery, Saldaña has earned additional 213,000 €/year. This means that the village’s natural capital has been increased by almost 7.5 million euros, which is quite a very significant amount considering the surface area (3.17 km²) and the small population (3183 inhabitants). This remarkable impact is a consequence of the effectiveness of the HR in initiating ecosystem processes that have solved environmental problems with significant effects on the general well-being of the population. In addition to these aforementioned benefits, the downstream population has also benefited from water quality improvement, although these effects have not been studied in this article.

Adopting an ecosystem approach has proven to be beneficial. Excluding scale, just as New York city avoided building an expensive water filtration plant (Turner & Daily, 2008), Saldaña has avoided oversizing the WWTP in order to comply with the Water Framework Directive (OJ, 2000). It has also benefited from other effects, such as landslide risk mitigation and road protection from slope mass movements. There is no documentation on real costs for the execution of the HR, so it is not feasible to compare the invested and saved costs in HR, as Pimentel et al. (1995) do in their work about erosion in the United States. Nevertheless, it is remarkable the effect of the HR in the water quality improvement. If it had not been executed, surrounding municipalities would have to spend as much as 128,496 €/year more in order to comply with the Water Framework Directive relating to water quality.

On the other hand, just like in Napa (California) (Brauman, 2006), other complementary ESs have improved the general well-being of the population, such as increasing the availability of wood products and improving the surrounding area visual quality. The existence of high-quality housing depends directly on the existence of places with desirable views which would not have been possible had erosion continued. The pine forest growth will provide raw material for human activities, in a sustainable manner, in the near future (in approximately 40 years according to yield tables). In this way, the multifaceted ecosystem approach has revitalized economic activity in the municipality.

The studied environmental work is part of the group of HRs which were predominantly executed in the early twentieth century to avoid torrential phenomena, flooding in urban areas and soil loss. These improvements have proven to be highly effective particularly when completed in conjunction with other structural solutions; the HR in Saldaña is a strong example.

The economic and environmental value of the HR should be considered in town hall decisions in order to avoid economic and environmental disadjustments. If the nearby forest is destroyed, it is highly probable that gully erosion would start, which, in turn, could create significant environmental damage that Saldaña might not be able to afford. Considering the current state of the HR, the most cost-effective option is to protect the forest; if this is deemed impossible, a thorough cost–benefit analysis should be completed and carefully analysed. In Daly’s (1998, p. 21) words

If we are to avoid uneconomic growth we must be sure that the value of the natural capital services sacrificed as a result of human expansion is not greater than the value of the services gained for the expanded manmade capital.

All the assessed ESs in this study are perfectly compatible, so the ESs have not been double counted; this is a premise in any environmental valuation work (Morse-Jones et al., 2011) that avoids overvaluation. This has allowed us to aggregate obtained results. To differ from the majority of environmental valuation studies, it has been decided to aggregate ES values in a weighted manner to emphasize the relative importance of each ES in the economic, environmental and social situation of Saldaña and its surroundings. This is an innovative aspect and an important strength of the study, because it integrates the concept of consumer surplus from the point of view of experts, using a methodology that can recognize scarcity of defined ES for environmental problems that already exist in Saldaña. Survey results show that hydrological risks are undervalued by the youngest stakeholders. This can be explained because they have never suffered from such disasters. Agents without forestry training value the current situation as normal, as they have not been aware of the situation that had pre-HR. However, any of them recognize landslide risk mitigation as the most important ES, followed by water quality improvement, wood products and landscape scenery enrichment. It can be seen that locals tend to decrease differences among the importance of ES, whilst environmental knowledge tends to increase it.

The data gathered in Saldaña could also be applied to other sites in inland Spain, particularly regarding water quality improvement, landslide risk mitigation and landscape scenery improvement. Therefore, the results of this study could be benefit transferred to another policy site.

Developed methodology is based on data easily available, adapting them to site-specific conditions using AHP analysis, which strongly validates calculated value and becomes an important strength of the methodology. However, ES weights are totally dependent on the sample design, so careful consideration should be given to early stages during the survey. It is also time-consuming in the sense that field work is necessary to obtain some data: wood products inventory is a good example of this.

The aggregated value is the minimum EV of the HR since only four of the multiple ESs have been analysed, whereas in the project ‘Valuation of Spain’s Natural Assets’ (Esteban Moratilla, 2010) 17 ESs were considered. As future research, other ESs that could have been valued in this study include nonwood forest products (commercial mushrooms), genetic resources, climate regulation (extreme temperature decrease, atmospheric humidity increase), atmospheric regulation (suspended particles filtration, CO₂ sequestration), biodiversity, cultural services (recreational, educational and spiritual activities, sense of place), option, bequest and existence values.

On the other hand, valuation has been performed just using monetary units. There are other metrics such as sustainability indices and ecologic footprint that better suit ecological economics and are beyond the scope of this study, but should be worth investigating in the future in order to facilitate a change towards more sustainable policies.

8. Conclusion

Saldaña’s HR is an example of an ecosystem-based solution which developed new ES that did not exist before, which in turn has increased the natural capital by more than 7 million euros. This value has been obtained by aggregating calculated assessments of each ES following the AHP methodology. The assessed ESs are totally compatible (there are no trade-offs among them), so double counting has been avoided. The calculated value is the minimum EV of the HR as other ESs should also be considered to obtain the total EV (TEV). From the four ESs analysed in this study, landslide risk mitigation and water quality improvement are the most important ones according to the AHP methodology, the latter having the highest economic impact among all.

Nevertheless, these results are somehow remarkable because the income produced ascends to more than 200,000 €/year. This increase in and of itself constitutes a natural wealth that should be considered in any public decision affecting this territory; only in this way, a sustainable use of this natural capital that produces aforementioned ES would be assured. This assessment has been made using monetary units. However, there are other biophysical and energy-based indicators related to ecological economics that would make a really interesting assessment for this HR. This is beyond the scope of this study but would surely be worth investigating in future studies.

This natural capital increase has been possible thanks to the adoption of an effective ecosystem-based approach. Furthermore, we should ask ourselves if it could have been possible to obtain the similar results with other structural type measures; earned ES and saved costs make us believe that the HR has been the most suitable solution.

This study could also be considered an example of the application of the European Water Framework Directive (OJ, 2000), as it responds to the need of member states to evaluate various methods to improve water quality, in order to select those methods which are most cost-effective considering the social benefits derived from them.

Monetary valuation helps to drive attention into ES derived from ecosystem-based solutions, such as HR, and allows to consider them into land-use policy and planning, if a shift towards sustainability is desired. This shift could also be benefited by considering other metrics apart from monetary values, such as sustainability analysis, ecological footprint calculations, that lie beyond the scope of this study, but would be worth investigating in the future.

Acknowledgements

We are grateful to the Duero Basin Authority, for facilitating project’s information and technical support, and to Tragsatec, for completing some of the surveys and media facilitation. We are also grateful to a reviewer for his insightful suggestions which greatly improved this article.

Disclosure statement

No potential conflict of interest was reported by the authors.