Economic vulnerability of fire-prone landscapes in protected natural areas: application in a Mediterranean Natural Park

Abstract

Environmental services and landscape goods are rarely incorporated into economic valuation of natural resources, even though these resources may constitute a large proportion of the total ecosystem value, mainly in natural protected areas. The frequent occurrence of wildfires in Spanish protected areas requires a tool, which allows for the comprehensive management of landscape resource and mitigating the potential economic impacts caused by fire. In this paper, we extend the economic valuation theory to the concept of economic vulnerability. Geographic Information System was used to develop a framework for landscape impact assessment using annual landscape value, vegetation resilience and fire behavior. An economic approach of landscape susceptibility was provided by the integration of these three components. Once landscape susceptibility had been spatially characterized, landscape vulnerability was analyzed from criteria associated with landscape susceptibility and burn probability. There was a notable variation in landscape vulnerability ranging from 412,000 Euros to 1,200,000 Euros in Aracena Natural Park (186,828 ha) according to the selected contingent valuation scenario. The availability of cartography of landscape vulnerability could play a critical role in budget optimization and the decision-making process. In this sense, this approach helps managers to identify different efforts according to spatial distribution of the risk and fuel management strategies to increase economic efficiency of fire prevention activities.

Appendices

Appendix 1: Landscape quality characterization

We have developed a GIS dataset due to a quick and easy way to vegetation characterization. The landscape unit cartography was created by the integration of information from these three sources: the Map of Andalusia Land Use, the Spanish National Forest Inventory and the Andalusia Forest Map. Field trips and itineraries were used to validate and improve this GIS characterization. Landscape characterization includes four landscape unit groups: anthropic, agricultural, treeless and forest landscapes (Table 7).

Table 7 Landscape unit area and landscape quality category based on landscape rating (Molina et al. 2016)

Landscape quality assessment was based on social preferences method (Molina et al. 2016) or the contingent rating in which respondents were asked to rate landscapes individually on a numeric scale of 1–10. The landscape quality was identified by analyzing the answers given to 22 scale rating photographs involving the most representative landscapes (main or most extend landscapes that comprise the number of landscape units). Landscape rating was collected by means of 120 personal interviews with random tourists from Huelva province.

Statistical analysis allowed us to classify landscapes rating on five categories. We selected natural breaks classification method (Jenks method) in relation to other classification methods such as equal interval, defined interval and geometrical interval. This method is a data clustering method designed to determine the best arrangement of values into different classes. Jenks optimization method seeks to reduce the variance within classes and maximize the variance between landscape classes.

Appendix 2: Practical example of landscape vulnerability from one spatial unit (100 ha)

We have developed an application for the valuation method to provide to each theoretical step a practical example of value assignments. In this sense, we could find four landscape units in one selected spatial unit (Fig. 7).

Fig. 7

Landscape units on the selected spatial unit

The operational process involved in obtaining a landscape vulnerability model comprises the following steps:

• Step 1 Conversion of the landscape quality to the form of monetary units through contingent valuation method.

Firstly, we classify “Aracena and Picos de Aroche Natural Park” landscape rating on five categories according to social preferences (Table 8). Each landscape quality was converted to form of monetary units through proportional allocation of the Natural Park value according to its extend and contingent rating (Appendix 1 and Molina et al. 2016).

Table 8 Annual value for each landscape unit estimated based on landscape quality category

• Step 2 Landscape susceptibility based on economic landscape valuation and fire intensity level.

Knowing the annual landscape value and vegetation resilience, landscape impacts could be represented by updating the economic value over the years necessary for restoring the original landscape quality (Eq. 6).

$$L = V[(1 + r)^{n} - 1]/[r(1 + r)^{n} ]$$

(6)

where “L” is the economic landscape impacts caused by the wildfire (€/ha), “V” is the annual landscape value (€/ha), “n” is the number of years for landscape restoring based on its vegetation resilience, and “r” is the interest rate. Knowing the rotation age, the interest rate can be represented by updating timber volume over the years necessary for achieving optimal harvesting:

$$r = \left[ {(V_{T} /V_{j} )^{((1/(T - j))} } \right] - 1$$

(7)

where “V _T ” is the timber volume in the optimal rotation age (m³/ha), “V _j ” is the existing stock volume at the moment of economic valuation (m³/ha), “T” is the optimal rotation age (years), and “j” is the estimated stand age (years) (Table 9).

Table 9 Landscape impacts according to resilience and interest rate

The potential fire intensity level (FIL) was included into a GIS database in similar way to the maximum landscape impacts providing a tool for landscape susceptibility assessment according to the following expression:

$${\text{LS}} = {\text{L}} \times {\text{DR}}$$

(8)

where “LS” is the landscape susceptibility (€/ha), “L” is the maximum landscape impacts caused by the wildfire (€/ha), and “DR” is the depreciation rate (%) based on fire intensity (Tables 10, 11).

Table 10 Landscape susceptibility according to the maximum landscape impacts and potential fire behavior

Table 11 Landscape unit area and landscape quality category based on landscape rating

• Step 3 Landscape vulnerability according to the integration of landscape susceptibility and burn probability.

Burn probability (BP) combined with landscape susceptibility (LS) provides a framework for landscape vulnerability (LV) assessment:

$$\text{LV} = \text{BP}\left( \% \right) \times \text{LS}.$$

(9)