Enhanced methods for analysing landscape structure

Landscape metrics for characterising three-dimensional patterns and ecological gradients

Sebastian Hoechstetter

Kurzübersicht

Landscape metrics for characterising three-dimensional patterns and ecological gradients
ISBN: 978-3-941216-13-6.
Veröffentlicht: Oktober 2009, 1. Auflage, Einband: Broschur, Seiten 180, Format DIN B5, Gewicht 0.35 kg
Lieferzeit: 2 - 6 Werktage
Verfügbarkeit: Auf Lager
32,80 €

Enhanced methods for analysing landscape structure

Mehr Ansichten

Details

One of the central goals of landscape ecology is relating spatial patterns to ecological processes. Therefore, effective methods for measuring landscape structure are needed. One of the most widely used approaches in this context is the patch-corridor-matrix model, which regards landscapes as being mosaics made up of different units or elements. But this concept is connected with a number of problems. For example, the third spatial dimension (elevation) and ecological gradients are largely neglected and cannot be analysed using standard landscape metrics. In this research study, these deficiencies are dealt with in detail and possible methodical solutions are presented. This way, this work is meant to contribute to an improved concept of landscape structure analysis in order to meet the challenges posed by the complexity of landscape ecological problems. It addresses the needs of both scientists and practitioners in this field.

The author

Sebastian Hoechstetter (born in 1979) studied Geoecology at the University of Karlsruhe and the Eberhard Karls University Tübingen from 1999 to 2004 with study and research stays in Ireland (University College Cork) and Switzerland (Agroscope Reckenholz-Tänikon Research Station). He also completed an internship at the Helmholtz Centre for Environmental Research – UFZ in Leipzig. Since 2005, he has been working as a research associate at the Leibniz Institute of Ecological and Regional Development in Dresden in the fields of landscape ecology, biodiversity and climate research. This thesis was part of his PhD degree received from the Dresden University of Technology in 2009.

Kontakt: Leibniz-Institut für ökologische Raumentwicklung e. V., Dresden, http://www.ioer.de/

Der Herausgeber:
Prof. Dr. techn. habil. Elmar Csaplovics leitet den Lehrstuhl Geofernerkundung am Institut für Photogrammetrie und Fernerkundung der Technischen Universitaet Dresden

Kontakt:
TU Dresden, Institut für Photogrammetrie und Fernerkundung, Helmholtzstraße 10, 01062 Dresden
http://www.tu-dresden.de/ipf/

Prof. Dr. Elmar Csaplovics: "In loser Folge sollen in dieser Schriftenreihe wissenschaftliche Arbeiten, die am Lehrstuhl für Fernerkundung bearbeitet und betreut werden, in ansprechender Form veröffentlicht werden. Wir glauben, dass dadurch das Spektrum von Literatur zum Themenkreis projektorientierter angewandter fernerkundlicher Forschung nachhaltig bereichert werden kann und wünschen uns demgemäß eine kritikfreudige Schar von Leserinnen und Lesern." >Professur Geofernerkundung

Editorial

Landschaft, sagen wir, entsteht, indem ein auf dem Erdboden ausgebreitetes Nebeneinander natürlicher Erscheinungen zu einer besonderen Art von Einheit zusammengefaßt wird, einer anderen, als zu der ein kausal denkender Gelehrte, ein religiös empfindender Naturanbeter, ein teleologisch gerichteter Ackerbauer oder ein Stratege eben dieses Blickfeld umgreift. (Simmel 1913, p.638)
Landscape is a vague term which represents a subjective perception of a holistic structure of natural and artificial elements of an environment more or less mixed with a variety of quantitative definitions based on the vocabulary of natural sciences. Therefore landscape never coincides with the analytically defined ecosystem (Csaplovics 2005).
In the German meaning landschaft may describe the site and natural character of a coherent landstrich, ohg. lantscaf, mhg. lantschaft, a gegend, with regard to the sensual impression of the setting on the one hand and to the social context of the surrounding area as a whole on the other, a regio, but also with regard to the aesthetically driven artistic painting (Grimm & Grimm 1865).
The English term landscape is related to the Dutch landschap, from landscap, originally a region, with scap meaning the condition or shape of the respective area. Landscape was used for the first time in 1598 for the description of a painting (Klein 1971).

The Dutch school of landscape painting represented the beginning of a development of the denotation, anyhow without a direct relation to the theoretical background and without the explicit use of the term landscape painting. The implementation of an artistic connotation transformed the original meaning of a regio, a specific part of a territory, into its aesthetically exaggerated image, often with a strong inclination to its picturesque aspects.
Only in 1632 in the English language the term landscape is for the first time used for the description of an image of natural scenery. Thus it served originally for the description of the image of landscape, and only later for the denotation of the specific environments themselves.

The image/impression of landscape develops from a holistic, not only visual, but in a broader understanding sensual subjective a priori perception of landscape (gestalt). The image of landscape is perceived by the observer in an intuitive way. The natural conditions of the environment are described by the analysis of the geological, pedological, hydrological and biological potentials of a section of terrain.

Out of the initial understanding of landscape as an image of nature, which was shaped by mankind more or less unknowingly, and the perception of the character of landscape as an aesthetically influenced image, from the late 18th century onwards a scientifically driven definition gained influence. It was among others Alexander von Humboldt, who intended to merge the holistic and aesthetic interpretation with a structural approach which focused on the splitting of landscape into its components (Humboldt 1849).

In contrary to the perception of landscape as a conglomerate of sensual subjective impression of the whole and realistic structural analysis since the early 20th century quantitative approaches successively prevail. Landscape is understood as a mosaic complex of ecotopes (landschaftszellen) joined together to a consistent entity (Troll 1939).
Since the seventies the term landscape is more and more under discussion in the geographical scientific community, as it is a matter of fact that its meaning is mainly derived from non-scientific connotations, which are related to notions as beauty, harmony, variety, separability and so on (Hard 1975).
There is discordance about the status of landscape in the sense of a geographical term, with special regard to the question whether a specific landscape is dominantly defined by its geomorphological, cultural or aesthetic values, additionally influenced by the subjectivity of individual perception.
Modern landscape ecology puts landscape into a heterogeneous spatial setting, which is characterised by an aggregation of land forms and/or by an association of habitats and which may extend over an area of some hectares up to several hundred square kilometres (Turner & Gardner 1991)
Characteristics which function as descriptive landscape parameters are selected, such as structure, function and change. Structure covers spatial interrelations of ecosystems described by size and extent, form, populations (abundance) and configuration of components (Forman & Godron 1986)

Sebastian Hoechstetter in his study focuses on the critical analysis of structural parameters of landscape metrics as adapted during the last twenty years under the socalled patch-corridor-matrix model. Extraction of significant weaknesses of the model should lead to an adaptation and to the extension of structural measures with special regard to the systematic integration of the third dimension of the coordinative determination of position, shape, extension and heterogeneity of landscape elements and their respective degree of correlation.
It is surprising that quantitative landscape ecology up to now did not pay sufficient attention to this matter. The scientific vacuum is thus significant and needs to be filled with reliable research results. Sebastian Hoechstetter provides a wide spectrum of relevant research work in order to minimise this deficit. Consideration of parameters of relief energy firstly relies to simple components of structural measures like distance, area, circumference, but also to more complex entities such as heterogeneity of relief (roughness) and related changes in texture of landscape elements. In this regard the textural measure of lacunarity, which was implied by fractal geometry (Mandelbrot 1983), is specifically highlighted. This topic is not at all new (e.g. Plotnick et al. 1993), but extended applications for a better description of the three-dimensionality of ecological gradients towards consideration of the fact that transitions between ecological, geographical and environmental units are continua are lacking till now. This fact does not only concern the relations between landscape elements and land use classes (fuzzy logics), but also the variations in multi-scale levels of analysis to the point of the dispersion of populations in respective distribution areas. Statistically simplified the existence of an ecological gradient can be described by the decrease of variance in function of increasing distance to the respective reference entity (landscape element, habitat).
Generally advanced technologies of very high resolution three-dimensional spatial data acquisition like airborne laser scanning (LiDaR) and extended storage, maintenance,
retrieval and analysis capacities of 3D-geodata in a GIS environment push the intensification of research towards the integration of these new dimensions of data analysis and data visualisation. The presented study satisfies these needs to a large extent. It serves as a most valuable and scientifically relevant add-on in research on the development and application of landscape metrics in very high resolution three-dimensional landscape analysis. It is a matter of fact that the publication of this study will be welcomed by the respective research community.

Dresden, August 2009
Prof. Dr. Elmar Csaplovics


References
Csaplovics, E. (2005): Zur Topochronologie der Landschaft um den Neusiedler See bis zum Ende des 16.Jahrhunderts. Burgenländisches Landesarchiv, Eisenstadt
(Burgenländische Forschungen 91).
Forman, R.T.T.; Godron, M. (1986): Landscape ecology. Wiley, New York. 595 pages.
Grimm, J; Grimm, W. (1885): Deutsches Wörterbuch, Bd.12, bearb. v. Moriz Heyne. Hirzel, Leipzig, Sp.131.
Hard, G. (1975): Die 'Landschaft' der Sprache und die 'Landschaft' der Geographen. Colloquium Geographicum Bd. 11, Bonn.
Humboldt, A.v. (1849): Ansichten der Natur mit wissenschaftlichen Erläuterungen, dritte verbesserte und vermehrte Auflage, 2 Bde. Cotta, Stuttgart Tübingen.
Klein, E. (1971): A comprehensive etymological dictionary of the English language, 2 vols. Elsevier, Amsterdam.
Mandelbrot, B. (1983): The fractal geometry of nature. Macmillan, New York. 468 pages.
Plotnick, R.E.; Gardner, R.H.; O'Neill, R.V. (1993): Lacunarity indices as measures of landscape texture. Landscape Ecology 8 (3), 201-211.
Simmel, G. (1913): Philosophie der Landschaft. In: Gallwitz, D.; Hartlaub, G.F.; Smidt, H. (eds.): Die Güldenkammer. Eine Bremische Monatsschrift, 3.Jg., H.2, Bremen, 635-644.
Troll, C. (1939): Luftbildplan und ökologische Bodenforschung. In: Zeitschrift der Gesellschaft für Erdkunde, Berlin 1939, Heft 7/8: 241-298.
Turner, M.G.; Gardner, R.H. (1991): Quantitative methods in landscape ecology. Ecological studies 82, New York. 536 pages.

Foreword

This research work presents innovative methods for advancing the science of “landscape structure analysis”. The study was carried out under the Leibniz Institute of Ecological and Regional Development project “Landscape Metrics for Analysing Spatio-Temporal Dimensions (4-D Indices)” financed by the German Research Foundation (DFG). The background is an approach widely used in landscape ecology under which the structure of landscapes, in other words the "pattern" or "mosaic" of landscapes, is treated as a composition and configuration of discrete landscape elements (“patches”) such as individual land use units. “Landscape metrics” enable these spatial patterns to be quantified. The aim is to determine the structure of a landscape, for instance to document it for monitoring purposes or to provide input parameters for simulation models in landscape ecology.

This research work presents innovative methods for advancing the science of “landscape structure analysis”. The study was carried out under the Leibniz Institute of Ecological and Regional Development project “Landscape Metrics for Analysing Spatio-Temporal Dimensions (4-D Indices)” financed by the German Research Foundation (DFG). The background is an approach widely used in landscape ecology under which the structure of landscapes, in other words the "pattern" or "mosaic" of landscapes, is treated as a composition and configuration of discrete landscape elements (“patches”) such as individual land use units. “Landscape metrics” enable these spatial patterns to be quantified. The aim is to determine the structure of a landscape, for instance to document it for monitoring purposes or to provide input parameters for simulation models in landscape ecology.
Landscape metrics are now used in much scientific experimental work in Europe and North America. In recent years they have also found practical application, for example in assessment procedures for planning. Landscape metrics are currently making the transition from scientific use to use in practical fields. Since these parameters can in many cases complement or supplement existing methods, any obstacles or uncertainties need to be eliminated.
Against this background, Sebastian Hoechstetter discusses the shortcomings of the landscape structure analysis approach and its potential for expansion. On the basis of comprehensive research, he clearly demonstrates the problems posed by the need to include the third dimension and by the hitherto sharp definition of boundaries between patches. For instance, landscape metrics mostly take account only of the twodimensional surface. Ecologically significant three-dimensional structures, like elevation or differences in height to neighbouring patches are more or less ignored; nor is height structure (texture) within a use class taken into account.
In considering the further development of landscape metrics, Sebastian Hoechstetter therefore focuses on capturing three-dimensional patterns and ecological gradients. In the first place, this involves integrating true surface areas and distances into common landscape metrics. Metrics hitherto used only in the material sciences, namely in “surface metrology” are also drawn on at the landscape level. A new development is the measurement of lacunarity. In describing the methodological state of the art in this field, Hoechstetter includes geomorphology, which, so far relatively independently of the landscape metrics of landscape ecology, has developed its own parameters for describing relief.
More or less “on the side”, Hoechstetter provides an overview of the national and international state of the art in landscape structure analysis, of how the subject matter relates to landscape ecology and landscape research, and of the roots of the German and North American “schools.”
The LandMetrics-3D tool should also be mentioned, which enable the indices developed to be used. The author collaborated on its development in the context of the project financed by the German Research Foundation. With the aid of this tool, the metrics could be used for neutral, simulated landscapes and subsequently for test areas in Saxony and Baden-Württemberg and their behaviour tested in landscapes of differing relief.

In the concluding section Sebastian Hoechstetter critically discusses the results obtained and the subject matter of landscape metrics.
Hoechstetter’s study makes an important contribution to progress in this area of research – also in the international context. This is evidenced by the innovative approaches he has developed and which he presents in exemplary fashion. He has thus provided not only a comprehensive treatment of the current status of research but has also led the way in a topical line of research in the analysis of landscape structure. In particular, Hoechstetter points to concrete potential for further research into ecological gradients.
Over and above this, the work shows the need and potential for application in the fields of planning, landscape ecology, and nature conservation. It is clear that there are practical applications for landscape metrics and that the methods he has developed are not merely a theoretical construct.
I am certain that the work will gain widespread acclaim.

Dr. Ulrich Walz

Acting Head of the Research Area “Development and Management of Landscapes” Leibniz Institute of Ecological and Regional Development, Dresden 

Abstract

Relating spatial patterns to ecological processes is one of the central goals of landscape ecology. Thus, effective methods for measuring landscape structure are needed. A common approach is based on the so-called patch-corridor-matrix model, which regards landscapes as being mosaics made up of different units or elements (“patches”). Landscape metrics are used for quantifying the composition and configuration of these mosaics.
However, this widely used approach is connected with a number of problems. In this work, two main deficiencies of the concept are identified and dealt with in detail. The first issue refers to the fact that the third spatial dimension (“elevation”) is largely neglected in the patch-corridor-matrix model. The landscape is perceived as being a flat surface regardless of the underlying relief. Thus, information about ecologically relevant terrain features is lost. Moreover, it is shown that in landscapes with a variable relief, the calculation of common landscape metrics can lead to erroneous results, since the planimetric projection of landscape elements leads to the underestimation of their area and perimeter as well as of the distances between them.
The second problem associated with the patch-corridor-matrix model is the fact that it requires clearly defined landscape elements with sharp boundaries as a basis for calculating landscape metrics. However, in many cases environmental parameters appear in the form of ecological gradients rather than as categorical patterns. Thus, the concept can be regarded as oversimplifying in certain situations.
Existing approaches to these issues are summarised and compared. More importantly, some new solutions are introduced. A technique that corrects the area and perimeter of patches for effects of the underlying relief is applied to a number of landscape metrics and is tested in different study areas. For certain groups of metrics (e.g. area and perimeter metrics, fragmentation metrics, edge metrics), this method proves to provide more realistic results, especially when it is used in regions exhibiting a very variable relief. The so-called “A*-algorithm” serves as an effective approach for estimating the true surface distance between patches and can thus be regarded as being beneficial in many respects, for example in analyses dealing with the migration of species. Certain surface metrology indices derived from materials science prove to be useful for integrating terrain features such as surface roughness into analyses of landscape structure.
Lacunarity analysis is suggested as a method for analysing ecological gradients and is tested on the basis of different sample data. Especially in combination with other techniques (e.g. landform analysis, surface metrology indices), lacunarity analysis can be a suitable supplement to the standard spectrum of methods for landscape structure analysis.
In summary, it can be stated that methods for quantifying landscape structure need to be enhanced and adapted in a suitable manner in order to meet the challenges posed by the complexity of landscape ecological problems. Besides that, the availability of new data sources offers a potential which has not yet been tapped in landscape ecology. The approaches presented in this work serve as possible methodical solutions for the problems mentioned and can help to stimulate further technical and conceptual developments in the near future.

Deutsche Kurzfassung

 „Erweiterte Methoden für die Landschaftsstrukturanalyse – Landschaftsstrukturmaße zur Erfassung dreidimensionaler Muster und ökologischer Gradienten”
Der Zusammenhang zwischen räumlichen Landschaftsmustern und ökologischen Prozessen ist ein zentraler Forschungsgegenstand der Landschaftsökologie. Daher werden effektive Methoden zur Erfassung der Landschaftsstruktur benötigt. Das sogenannte Patch-Korridor-Matrix Modell ist ein weit verbreitetes Konzept, bei dem Landschaften als aus einzelnen Elementen bzw. Einheiten („Patches“) zusammengesetzte Mosaike betrachtet werden. Landschaftsstrukturmaße dienen zur Beschreibung der Komposition und Konfiguration solcher Mosaike.
Die vorliegende Arbeit beschäftigt sich mit den Mängeln dieses Konzeptes. Zwei wesentliche Probleme werden dabei identifiziert und detailliert behandelt. Zum einen wird die dritte räumliche Dimension (die „Höhe“) hier bislang weitgehend vernachlässigt, denn die Landschaft wird auf eine zweidimensionale Ebene reduziert. Dadurch geht ökologisch bedeutsame Information zu den Geländeeigenschaften verloren. Es kann gezeigt werden, dass in Landschaften mit bewegtem Relief die Berechnung von Landschaftsstrukturmaßen zu Fehlern führen kann, da die Projektion der Landschaftselemente in eine Ebene in einer systematischen Unterschätzung der Oberflächen und Umfänge von Patches sowie der Distanzen zwischen ihnen resultiert.
Als ein zweites wesentliches Problem des Patch-Korridor-Matrix Modells kann die Tatsache betrachtet werden, dass zur Berechnung gängiger Landschaftsstrukturmaße die entsprechenden Landschaftselemente eindeutig voneinander abgrenzbar sein müssen. In der Realität können jedoch häufig eher graduelle Verläufe von Umweltparametern („ökologische Gradienten“) beobachtet werden. In solchen Fällen liegt also eine übermäßige Vereinfachung durch das Modell vor.
Bestehende Lösungsansätze für diese Probleme werden vorgestellt und verglichen. Darüber hinaus werden einige neue Methoden eingeführt. Ein Algorithmus zur Einbeziehung der Höhenstruktur in die Flächen-und Umfangsberechnung wird in mehreren Untersuchungsgebieten getestet. Dabei zeigt sich, dass für bestimmte Gruppen von Strukturmaßen (z. B. Flächenmaße, Fragmentierungsmaße, Kantenmaße) auf diese Weise in Regionen mit ausgeprägtem Relief realistischere Ergebnisse erzielt werden können. Weiterhin wird der „A*-Algorithmus“ als ein methodischer Ansatz zur Abschätzung der realen Oberflächendistanz zwischen Patches eingeführt. Besonders für Analysen zum Ausbreitungsverhalten von Arten eignet sich diese Methode. Den Materialwissenschaften entlehnte Oberflächenindizes erweisen sich als nützlich zur Erfassung von Eigenschaften wie der Oberflächenrauigkeit.
Die Lakunaritätsanalyse wird als eine Methode zur Untersuchung ökologischer Gradienten vorgeschlagen und wird ebenfalls auf Grundlage verschiedener Testdatensätze erprobt. Auch in Kombination mit anderen Ansätzen (z. B. Reliefanalyse, Oberflächenindizes) ist sie eine geeignete Ergänzung des gängigen Methodenspektrums.
Zusammenfassend kann gesagt werden, dass Methoden zur Landschaftsstrukturanalyse einer Erweiterung und einer Anpassung an die Komplexität ökologischer Fragestellungen bedürfen. Außerdem bieten neue Datenquellen Möglichkeiten, die noch nicht vollständig ausgeschöpft erscheinen. Die hier vorgestellten Methoden können als mögliche Lösungsansätze für die genannten Probleme verstanden werden und sollen als Anreiz für künftige technische und konzeptionelle Entwicklungen dienen.

Table of contents

List of figures   XVI
List of tables    XVIII
List of abbreviations     XIX
Abstract    XXI
Kurzfassung    XXIII

1 Introduction     1
1.1     Background and research questions: Modern landscape ecology and its shortcomings   1
1.2     Structure of the present work     3

2 Theoretical background   
5
2.1 “Landscape” as a research subject – History, terms and definitions    5
2.1.1 A brief history of landscape ecology    5
2.1.2 Landscape definitions    7
2.1.3 Landscape ecology – two fundamental approaches    9
2.1.4 General principles in landscape ecology    11
2.2 Models and concepts in landscape ecology    12
2.2.1     The “North American approach” and the patch-corridor-matrix model   12
2.2.1.1 Landscape elements    13
2.2.1.2 Spatial heterogeneity and landscape structure  17
2.2.1.3 Landscape metrics – Theory, software, application   19
2.2.1.4 Scale and landscape structure – Effects and implications    24
2.2.2 An alternative approach: the “Leipzig-Dresden School”    28

3 Motivation: reasons for enhancing the patch-corridor-matrix model
    31
3.1 The “3D”-aspect: the third spatial dimension in landscape ecology    32
3.1.1     The necessity of integrating the third dimension into landscape ecological analyses   32
3.1.2     An overview of existing methodical approaches    35
3.1.2.1 Relief parameters and relief classification    35
3.1.2.2 Landform indices     38
3.1.2.3 Texture analysis    42
3.2 The “gradient”-aspect: the assessment of ecological gradients    43
3.2.1 Relevance of gradients and distance decay in ecological systems    43
3.2.2 Existing approaches for the analysis of ecological gradients    46
3.2.2.1 Usage of “moving windows” in landscape ecology    46
3.2.2.2 Multi-scale analyses     48
3.2.2.3 Fuzzy approaches    50
3.2.2.4 Spectral and wavelet analysis    51
3.3     The “technology”-aspect: Advancing software performance and new base data as a foundation for new analysis approaches    53
3.3.1 Geographic Information Systems: On the way to 3D-GIS    53
3.3.2 High-resolution elevation data    55

4 A proposal of new methodical approaches    59
4.1 Metrics for the assessment of the third dimension     59
4.1.1 Adjusted calculation formulas for standard landscape metrics    59
4.1.1.1 General correction approach for patch area and perimeter    60
4.1.1.2 Area and perimeter metrics    63
4.1.1.3 Shape metrics     65
4.1.1.4 Edge metrics     67
4.1.1.5 Fragmentation metrics    68
4.1.1.6 Diversity metrics    69
4.1.1.7 Contrast metrics    70
4.1.1.8 Distance Metrics – A correction approach using the A*-algorithm     72
4.1.2 Surface metrology   77
4.2 Ecological gradients    79
4.2.1 Lacunarity analysis     79
4.2.2 Landform indices and lacunarity analysis combined    83
4.3 Technical implementation – The ArcGIS-extension LandMetrics-3D    83

5     Examples of use and results: application and functional relevance of the proposed analysis methods  
   87
5.1     Comparing “2D” and “3D” landscape metrics using neutral landscape models    87
5.2     General index behaviour under realistic conditions    90
5.2.1 Study areas and data basis     90
5.2.1.1 Study areas 1 and 2: Rathen and Rosenthal (Saxony)    91
5.2.1.2 Study area 3: Bad Urach (Baden-Wuerttemberg)    92
5.2.1.3 Data basis     92
5.2.2 Basic patch geometries    95
5.2.3 General behaviour of landscape metrics in the study areas   97
5.2.3.1 Shape metrics     97
5.2.3.2 Edge metrics     98
5.2.3.3 Fragmentation metrics    98
5.2.3.4 Diversity metrics   99
5.2.3.5 Isolation/proximity metrics    100
5.2.3.6 Contrast metrics   102
5.2.3.7 Surface metrology indices    104
5.3     Analysis of scale effects connected with the proposed 3D-metrics    106
5.4     3D-metrics in different regions: Case study analysis using additional German study areas     110
5.5     Lacunarity analysis in practice   115
5.5.1     Application to simulated data    115
5.5.2     Lacunarity analysis performed on a normalised digital surface model     116
5.5.3     Application of lacunarity analysis in combination with landform indices   120

6 Discussion and evaluation    123
6.1 Evaluation of the applied methods   123
6.1.1 Applying correction algorithms to standard landscape metrics   123
6.1.2 The value of surface metrology indices     125

6.1.3     Interpretation of the scale effects     126
6.1.4     General applicability in different terrain situations   127
6.1.5     Lacunarity analysis as a tool for the assessment of ecological gradients   128
6.2 Summarising evaluation and comparison with existing methods    129
6.3 Possible fields of application     133

Summary and outlook 
   137

References    141

 

Erschienene Rezensionen

Since the 1980s landscape metrics play a vital role within the landscape ecology scientific community. Especially for fragmentation indices, landscape metrics are even utilized in applied landscape planning. As a contribution to further development of this quantitative discipline, the reviewed book (based on the PhD thesis of Sebastian Hoechstetter) presents new perspectives and enhancements of existing methods.

The book adopts two paradigms of modern geography and ecology, which were neglected in the majority of landscape metrics approaches in the past. Firstly, the third spatial dimension was integrated into the existing methods of landscape metrics and a new algorithm (A*-algorithm) was adopted. Secondly, ecological gradients were accounted for analysis of landscape structure, especially by means of the lacunarity approach, which determines the texture associated with patterns of spatial dispersion.

The book is organized in seven chapters. After giving an introduction, the author is reviewing the history, models, and concepts of landscape metrics. In more details, the third chapter evaluates the existing methods and their shortcomings to propose new methodological approaches for assessing the third dimension and ecological gradients in chapter four. These techniques were tested and evaluated on simulated virtual landscape as well as different on different test areas in Germany (Chapter 5 and 6). In chapter seven, the raised research questions about the deficiencies of common landscape metrics and the integration of the third spatial dimension as well as ecological gradients were answered.

The newly introduced approaches show promising results. For the 3D approach, the metrics which include area and perimeter in their formulas tend to be highly sensitive to the variability of the underlying terrain. For the investigation on ecological gradients, the lacunarity method has the potential of serving as an easily interpretable approach even for applied planning tasks. The methods are integrated into the ArcGIS-extension LandMetrics-3D. Unfortunately, this potentially very helpful tool is not easily applicable up to now. Nevertheless, the proposed methods are a very valuable contribution for further development of landscape metrics as indictors in applied ecology, geography and landscape planning.
(Rezension von Dr. Michael Förster, TU Berlin; erschienen unter http://www.landschaftsplanung.net/)

Zusatzinformation

Gewicht 0.3500
Lieferzeit 2 - 6 Werktage