Search

Different spatial interpolation techniques have been applied to construct objective bioclimatic maps of La Palma, Canary Islands. Interpolation of climatic data on this topographically complex island with strong elevation and climatic gradients represents a challenge. Furthermore, meteorological stations are not evenly distributed over the island, with few stations at high elevations. We carried out spatial interpolations of the compensated thermicity index (Itc) and the annual ombrothermic Index (Io), in order to obtain appropriate bioclimatic maps by using automatic interpolation procedures, and to establish their relation to potential vegetation units for constructing a climatophilous potential natural vegetation map (CPNV). For this purpose, we used five interpolation techniques implemented in a GIS: inverse distance weighting (IDW), ordinary kriging (OK), ordinary cokriging (OCK), multiple linear regression (MLR) and MLR followed by ordinary kriging of the regression residuals. Two topographic variables (elevation and aspect), derived from a high-resolution digital elevation model (DEM), were included in OCK and MLR. The accuracy of the interpolation techniques was examined by the results of the error statistics of test data derived from comparison of the predicted and measured values. Best results for both bioclimatic indices were obtained with the MLR method with interpolation of the residuals showing the highest R of the regression between observed and predicted values and lowest values of root mean square errors. MLR with correction of interpolated residuals is an attractive interpolation method for bioclimatic mapping on this oceanic island since it permits one to fully account for easily available geographic information but also takes into account local variation of climatic data. [ABSTRACT FROM AUTHOR]

The Canary Islands belong to Mediterranean basin hotspot, especially because their rich endemic flora. Nearly 40% of the total archipelago area is protected through protected natural area network. However, some plant communities show high degree of disturbance or they develop in unprotected zones. So, multivariate analysis combined with vegetation mapping can be useful for identify, describe and protect the main plant communities. These tools have been applied in Caldera de Taburiente national park (La Palma. Canary Islands). An actual vegetation map of this park is showed and its major natural vegetation units are described. These tools could be applied to other protected and unprotected areas, especially for planning, management and conservation purposes. [ABSTRACT FROM AUTHOR]