Identifying, Quantifying and Predicting of Landscape Patterns in Space and Time

In landscape ecology, approaches to identify and quantify landscape patterns are well developed for discrete landscape representations. Discretization is often seen as a form of generalization and simplification. Landscape patterns however are shaped by complex dynamic processes acting at various spatial and temporal scales. Thus, standard landscape metrics that quantify static, discrete overall landscape pattern or individual patch properties may not suffice when viewing landscapes as gradients or when quantifying spatially dynamic response surfaces resulting from model simulations of landscapes. As such, this study aimed to identify and provide an understanding of the applied approaches for describing spatial heterogeneity of ecosystems in landscape ecology. Different approaches were evaluated including landscape metrics, surface metrics, wavelength and spectral analysis, geostatic indices and Markova chains in quantifying changes of temporal– spatial patterns. Existence of different approaches and methods in identifying, quantifying and predicting of landscape patterns shows that landscape patterns may not be quantified using only one index or metric, rather, quantifications methods should be selected with regards to the aim when revealing the desired landscape patterns and ecological processes.