Spacial analysis al multiple scales of ecosystem services supply of forest and agricutural landscapes in NW Iberian peninsula

Abstract

Ecosystem services (ES) can be defined as the goods and services that ecosystems provide to the society for its well-being. The importance of ES in different fields of environmental sciences, ecology and land planning has been highlighted in recent years. The relevance of spatial aspects of the data describing ecosystems has driven the use of several spatial analysis techniques and methods for ES assessment. However, the extended use of Land Use/Land Cover maps may lead to uncertainty and bias due to oversimplification and generalization of the study area and associated elements. Ecosystem services and the ecological processes that generate them are strongly related. These processes are not constant at all spatial scales and display high levels of spatial and temporal heterogeneity. Thus, appropriate selection of the data sources and analytical methods used to characterize the ecosystems is essential for development of accurate models and minimization of potential bias in the assessments. The general objective of this research was to carry out a multi-scale analysis of the spatial pattern of ES supply and the distribution of some ecosystems that provide the ES. Different data sources were used to characterize agricultural and forest landscapes in the NW Iberian Peninsula. Forests are major components of the landscape in the area, and several authors have highlighted the wide range of ES that these ecosystems provide. We first characterized the ecosystems in order to identify the areas where the main forest tree species in the area are growing. Quercus robur L. and Castanea sativa Mill. forests are mainly located in zones below 500 m where low temperatures do not usually occur. By contrast, Quercus petraea (Matt.) Liebl. and Fagus sylvatica L. forests are associated with lower temperature zones at higher elevations. We then used digital soil cover data to analyze the pattern of supply of six ES (provision of food, materials and energy, regulation of climate and erosion, and cultural services) by spatial statistical methods. We used the lacunarity metric to analyze the regularity of the no supply areas at multiple spatial scales. We found some divergence in the thresholds of spatial clustering of the ES and obtained higher values for food and materials provision. We also identified the spatial scales (extensions) at which the probability of supply of each ES is maximized. In addition, we used a functional approach to analyze the pattern of supply of a similar set of ES. We developed spatial models of these ES on the basis of Net Primary Production, calculated for a phenological year from Landsat 5-TM data and using the ¿NDVI index. The index was then combined with some socio-environmental variables. We used two multi-scale metrics (lacunarity and four term local quadrat variance) to analyze the observed patterns. The analysis revealed clustered patterns for provisioning services and a more extended distribution for regulating services, with different characteristic spatial scales for each type. Finally, we used a high resolution forest thematic map to assess the potential supply of some ES from forest ecosystems. We thus explored the spatial relationships between these ES and found some differences between the ES associated with native forests and those supplied by forest plantations of exotic species. We also determined the zones with a high density of elements involved in ES supply at different spatial scales. This enabled us to identify the locations of the most important areas for some ecosystems in relation to the supply of different ES in the NW Iberian Peninsula. Finally, we also analyzed the most characteristic spatial scales and the effect of different types of data sources on ES assessment.