What is it about?
We propose a method to quantify pixel adjacency with a bivariate summary statistics that is not adversely influenced by pixel resolution
Featured Image
Why is it important?
Landscape ecology starts from the assumption that diversity and spatial arrangement of ecosystem mosaics has ecological implications and tries to understand the interactions between diversity and structure of large spatially heterogeneous areas and its ecological functions. This approach implies effective use of earth observation techniques and geographic information systems, enabling a global view of the landscape mosaics. Consequently, a large number of indices has been used to quantify the structure of categorical maps as a surrogate of actual landscapes and correlate them to ecological processes. In particular, the entropy-based contagion index has been extensively used to summarize the amount of clumping or fragmentation of patches on raster categorical maps. However, despite its widespread application, the contagion index is very dependent on pixel resolution. This effect may render it inadequate as a meaningful measure of landscape structure.
Perspectives
We proposed a bivariate measure of edge-type evenness and landscape fragmentation to overcome the dependency of the contagion index on pixel resolution. However, it should be clear that there is no ideal function capable of collapsing all aspects of pixel adjacency into few summary statistics. Moreover, in analyzing the ecological implications of large spatially heterogeneous regions, a landscape classification scheme is needed that provides ecologically meaningful units for quantifying different aspects of landscape mosaics and correlate them to ecological processes.
Piermaria Corona
CREA Research Centre for Forestry and Wood
Read the Original
This page is a summary of: Beware of contagion!, Landscape and Urban Planning, February 2003, Elsevier,
DOI: 10.1016/s0169-2046(02)00146-9.
You can read the full text:
Contributors
The following have contributed to this page
