Academic journal article Estonian Journal of Ecology

Analysing the Spatial Structure of the Estonian Landscapes: Which Landscape Metrics Are the Most Suitable for Comparing Different landscapes?/Eesti Maastikumustrite Analuus: Millised Maastikuindeksid Tootavad Ja Millised Mitte?

Academic journal article Estonian Journal of Ecology

Analysing the Spatial Structure of the Estonian Landscapes: Which Landscape Metrics Are the Most Suitable for Comparing Different landscapes?/Eesti Maastikumustrite Analuus: Millised Maastikuindeksid Tootavad Ja Millised Mitte?

Article excerpt

INTRODUCTION

Spatial structure of landscapes is a central object of investigation in landscape ecology. This structure finds its expression in landscape pattern, which integrates both complex conditions of the natural environment (Quaternary cover, soil, topography, vegetation, local climate) and human-induced changes, first of all land use, the main object of human impact. Describing landscape pattern is essential in order to understand the relationships between landscape pattern and ecological processes. Therefore measurement of different parameters of spatial structure is an essential part of landscape ecology.

Within the past 30 years hundreds of landscape metrics indices have been proposed by various researchers to analyse the composition and configuration of landscape structure. Most of them are covered by the FRAGSTATS computer program (McGarigal et al., 2002). Since its emergence the measures and methods incorporated into the FRAGSTATS software have been very widely used.

In Estonia landscape metrics have been used in various studies: Roosaare (1982) applied a system of indices to analyse the landscape structure of the small island of Vormsi; Uuemaa et al. (2005) studied scale issues; Palang et al. (1998) and Aunap et al. (2006) used landscape metrics for studying landscape changes in Estonia; Uuemaa et al. (2007) examined how landscape pattern influences water quality in catchments; Mander et al. (2010) investigated the coherence of landscapes using landscape metrics; Oja et al. (2005) analysed the relationships between bird diversity and spatial heterogeneity of Estonian landscapes at different scales. Uuemaa et al. (2009) also gave an extensive overview of various uses of different landscape metrics.

There are many studies on the scale dependence of landscape metrics (Wickham & Riitters, 1995; Wu et al., 2002), correlations between landscape metrics (Cain et al., 1997; Riitters et al., 1995) and also several papers on how to interpret landscape metrics (Haines-Young & Chopping, 1996; Turner et al., 2001). However, it is not rare that landscape metrics are misused. Some metrics have theoretically a reasonable range but give actually almost constant values for all landscapes. The selection of the metrics depends first of all on the purpose of the study (mostly ecological process) and also on the landscape character.

A large number of metrics can be considered for the analysis of landscape pattern and structure. Ideally, there is a small set of metrics that span the important dimensions of pattern and structure, but which are not redundant (U.S. Environmental Protection Agency, 1994). Previous studies have attempted to determine if the major components of landscape structure can be represented by a parsimonious suite of independent metrics (e.g., McGarigal & McComb, 1995; Riitters et al., 1995; Cain et al., 1997; Griffith et al., 2000; Lausch & Herzog, 2002; Linke & Franklin, 2006; Cushman et al., 2008; Schindler et al., 2008). All these studies found that landscapes can be described by a few components, but these components were different among studies. Cushman et al. (2008) argued that different studies did not use the same pool of metrics and applied different methods to identify components but also it is possible that there are no fundamentally important aspects of landscape structure and instead structure patterns are peculiar to specific landscapes. It is more likely that different landscapes have different spatial aspects and therefore also the suitable landscape metrics vary. Besides, the maps used for analysis are different (scale, resolution, classification). Therefore it is useful to find a core set of metrics suitable for each region. Moreover, often landscape metrics are used on categorical maps of land cover/use that only reflect one theme. However, there are also some studies performed on a combination of several data sets. For example, Van Eetvelde & Antrop (2009) used DTM, CORINE Land Cover, a soil map, and a satellite image for landscape classification in Belgium, and Fasona & Omojola (2009) combined topographic data and land cover data for determining the landscape changes in Nigeria. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.