Quantification and spatial distribution of land-cover types in semiarid environments with hyperspectral imaging, a case study in Cabo de Gata-Níjar Natural Park (Almería)

Abstract

Escribano, P., Palacios-Orueta, A., Oyonarte, C. (2008). Quantification and spatial distribution of land-cover types in semiarid environments with hyperspectral imaging, a case study in Cabo de Gata-Níjar Natural Park (Almería). Ecosistemas 17(3):79-
88.

In  semiarid  environments  the  high  spatial  variability  along  with  the  low  photosynthetic  activity  of  vegetation  results  in  difficulties  of multispectral sensors in retrieving vegetation cover. Vegetation cover is a key variable in ecosystem processes and directly related with primary productivity. We have assessed the use of hyperspectral sensors and spectral mixing analysis to quantify land-cover types (soil, photosynthetic  and  non-photosynthetic  vegetation)  in  Cabo  de  Gata-Níjar Natural Park. The analysis was performed within homogeneous ecosystems which allowed breaking down the high spatial variability of the area making possible the quantification of land-cover types. The aim  of  this  work  is  to  generate  quantitative  information  for  monitoring  programs.  The  results  allowed  establishing  the  range  of  variability within  each  ecosystem  type  while  identifying  and  analysing  land-cover  types  link  with  specific  ecosystem  processes.  Besides,  it  was possible to identify the distinct spatial patterns of land-cover types in each ecosystem studied. These results contribute significantly as the basis for the development of change detection methodologies based on hyperspectral data.