13.5 References
The first family of landscape metrics includes
simple indices that describe composition (e.g., proportion or number of patch classes, also known as
richness), a nonspatial characteristic. The second
family describes configuration, a spatial property,
using patch-based metrics (e.g., area, size, perimeter, shape, fractal dimension). These landscape
metrics quantify four phenomena (Table 13.5; Hargis et al., 1997): proportional representation of each
patch class, aggregation of each class into patches,
frequency distribution of patch sizes, and spatial
distribution of patches. Hargis et a1. (1997, pp.
232-233) summarize the influence of these phenomena on landscape patterns as follows. The proportional representation of each class determines
which class comprises the landscape matrix. The
degree of aggregation affects the size, shape, and
perimeter of each patch. The frequency distribution
of patch sizes creates landscape texture. The spatial distribution of patches determines their pattern
of clumpiness or dispersion. Landscape metrics are
generally applied to thematic (categorical) maps,
which assume within-class homogeneity, an assumption that may not always be met. Reviews of
properties, applications, and limitations of landscape metrics can be found in Riitters et a1. (1995),
Hargis et a1. (1997, 1998), and Gustafson (1998).
13.4 Conclusions
An important emphasis of this chapter is the need
for identifying methods of spatial analysis that:
1. Make appropriate assumptions for addressing
the objectives of the analysis.
2. Can identify and/or be resistant to data characteristics (e.g., extreme values, outliers, deviation
from strict distributional assumptions).
3. Are appropriate for the spatial attributes of the
study area (e.g., distribution of observations, the
type of environmental stratification, influence of
within- and between-region boundaries).
In practice, it is likely that arbitrary decisions
may be made about how to represent spatial characteristics or about some properties of the data.
Therefore, alternative methods, variables, and the
like, should be considered in an analysis and the
effects of such changes interpreted (Haining, 1990).
However, EA spatial analysts should be aware of
the fact that, given the number of possible analyses that can be conducted on a data set, the synthesis of multiple interpretations may be a very difficult task. A final word of caution concerns the
volume of spatial data that is common in EAs.
197
Haining (1990) suggests that the critical issue is not
how many analyses are conducted on a given data
set, but how much information is provided by the
data.
13.5 References
Anselin, L. 1988. Spatial econometrics: methods and
models. Dordrecht, The Netherlands: Kluwer Academic Publishers.
Austin, M. P.; Nicholls, A 0.; Margules, C. R 1990.
Measurement of the realized qualitative niche: environmental niches of five Eucalyptus species. Ecol.
Monogr.60:161-177.
Bailey, T. C.; Gatrell, A. C. 1995. Interactive spatial data
analysis. Essex: Longman Scientific & Technical.
Baker, W. L.; Cai, Y. 1992. The r.le programs for multiscale analysis of landscape structure using the
GRASS geographical information system. Landscape
Ecol. 7:291-302.
Balling, R. C. 1984. Classification in climatology. In:
Gaile, G. L.; Willmott, C. J., eds. Spatial statistics and
models. Dordrecht, The Netherlands: Reidel: 81-108.
Bell, G.; Lechowicz, M. F.; Appenzeler, A; Chandler,
M.; DeBlois, E.; Jackson, L.; Mackenzie, B.; Preziosi,
R; Schallenberg, M.; Tinker, N. 1993. The spatial
structure of the physical environment. Oecologia 96:
114-121.
Bjornstad, O. N.; Falck, W. 1997. An extension of the
spatial correlogram and the x-intercept for genetic
data. In: Bjornstad, O. N. Statistical models for fluctuating populations-patterns and processes in time
and space. Ph.D. dissertation, University of Oslo. 9 p.
Borcard, D.; Legendre, P. 1994. Environmental control
and spatial structure in ecological communities: an example using oribatid mites (Acari, Oribatei). Environ.
Ecol. Stat. 1:37-61.
Borcard, D.; Legendre, P.; Drapeau, P. 1992. Partialling
out the spatial component of ecological variation.
Ecology 73:1045-1055.
Bradshaw, G. A. 1998. Defining ecologically relevant
change in the process of scaling up: implications for
monitoring at the "landscape" level. In: Peterson,
D. L.; Parker, V. T., eds. Ecological scale: theory and
applications. New York: Columbia University Press:
227-249.
Burrough, P. A. 1995. Spatial aspects of ecological data.
In: Jongman, R H. G.; ter Braak, C. J. F.; van Tongeren, O. F. R, eds. Data analysis in community and
landscape ecology. Cambridge, UK: Cambridge University Press: 213-251.
Casetti, E.; Jones, J. P., III. 1987. Spatial aspects of the
productivity slow down: an analysis of U.S. manufacturing data. Ann. Assoc. Amer. Geogr. 77:76-88.
Cliff, A D.; Ord, J. K. 1981. Spatial processes: models
and applications. London: Pion.
Cliff, A D.; Haggett, P.; Ord, J. K.; Bassett, K.; Davies,
R B. 1975. Elements of spatial structure. Cambridge,
UK: Cambridge University Press.
The first family of landscape metrics includes
simple indices that describe composition (e.g., proportion or number of patch classes, also known as
richness), a nonspatial characteristic. The second
family describes configuration, a spatial property,
using patch-based metrics (e.g., area, size, perimeter, shape, fractal dimension). These landscape
metrics quantify four phenomena (Table 13.5; Hargis et al., 1997): proportional representation of each
patch class, aggregation of each class into patches,
frequency distribution of patch sizes, and spatial
distribution of patches. Hargis et a1. (1997, pp.
232-233) summarize the influence of these phenomena on landscape patterns as follows. The proportional representation of each class determines
which class comprises the landscape matrix. The
degree of aggregation affects the size, shape, and
perimeter of each patch. The frequency distribution
of patch sizes creates landscape texture. The spatial distribution of patches determines their pattern
of clumpiness or dispersion. Landscape metrics are
generally applied to thematic (categorical) maps,
which assume within-class homogeneity, an assumption that may not always be met. Reviews of
properties, applications, and limitations of landscape metrics can be found in Riitters et a1. (1995),
Hargis et a1. (1997, 1998), and Gustafson (1998).
13.4 Conclusions
An important emphasis of this chapter is the need
for identifying methods of spatial analysis that:
1. Make appropriate assumptions for addressing
the objectives of the analysis.
2. Can identify and/or be resistant to data characteristics (e.g., extreme values, outliers, deviation
from strict distributional assumptions).
3. Are appropriate for the spatial attributes of the
study area (e.g., distribution of observations, the
type of environmental stratification, influence of
within- and between-region boundaries).
In practice, it is likely that arbitrary decisions
may be made about how to represent spatial characteristics or about some properties of the data.
Therefore, alternative methods, variables, and the
like, should be considered in an analysis and the
effects of such changes interpreted (Haining, 1990).
However, EA spatial analysts should be aware of
the fact that, given the number of possible analyses that can be conducted on a data set, the synthesis of multiple interpretations may be a very difficult task. A final word of caution concerns the
volume of spatial data that is common in EAs.
197
Haining (1990) suggests that the critical issue is not
how many analyses are conducted on a given data
set, but how much information is provided by the
data.
13.5 References
Anselin, L. 1988. Spatial econometrics: methods and
models. Dordrecht, The Netherlands: Kluwer Academic Publishers.
Austin, M. P.; Nicholls, A 0.; Margules, C. R 1990.
Measurement of the realized qualitative niche: environmental niches of five Eucalyptus species. Ecol.
Monogr.60:161-177.
Bailey, T. C.; Gatrell, A. C. 1995. Interactive spatial data
analysis. Essex: Longman Scientific & Technical.
Baker, W. L.; Cai, Y. 1992. The r.le programs for multiscale analysis of landscape structure using the
GRASS geographical information system. Landscape
Ecol. 7:291-302.
Balling, R. C. 1984. Classification in climatology. In:
Gaile, G. L.; Willmott, C. J., eds. Spatial statistics and
models. Dordrecht, The Netherlands: Reidel: 81-108.
Bell, G.; Lechowicz, M. F.; Appenzeler, A; Chandler,
M.; DeBlois, E.; Jackson, L.; Mackenzie, B.; Preziosi,
R; Schallenberg, M.; Tinker, N. 1993. The spatial
structure of the physical environment. Oecologia 96:
114-121.
Bjornstad, O. N.; Falck, W. 1997. An extension of the
spatial correlogram and the x-intercept for genetic
data. In: Bjornstad, O. N. Statistical models for fluctuating populations-patterns and processes in time
and space. Ph.D. dissertation, University of Oslo. 9 p.
Borcard, D.; Legendre, P. 1994. Environmental control
and spatial structure in ecological communities: an example using oribatid mites (Acari, Oribatei). Environ.
Ecol. Stat. 1:37-61.
Borcard, D.; Legendre, P.; Drapeau, P. 1992. Partialling
out the spatial component of ecological variation.
Ecology 73:1045-1055.
Bradshaw, G. A. 1998. Defining ecologically relevant
change in the process of scaling up: implications for
monitoring at the "landscape" level. In: Peterson,
D. L.; Parker, V. T., eds. Ecological scale: theory and
applications. New York: Columbia University Press:
227-249.
Burrough, P. A. 1995. Spatial aspects of ecological data.
In: Jongman, R H. G.; ter Braak, C. J. F.; van Tongeren, O. F. R, eds. Data analysis in community and
landscape ecology. Cambridge, UK: Cambridge University Press: 213-251.
Casetti, E.; Jones, J. P., III. 1987. Spatial aspects of the
productivity slow down: an analysis of U.S. manufacturing data. Ann. Assoc. Amer. Geogr. 77:76-88.
Cliff, A D.; Ord, J. K. 1981. Spatial processes: models
and applications. London: Pion.
Cliff, A D.; Haggett, P.; Ord, J. K.; Bassett, K.; Davies,
R B. 1975. Elements of spatial structure. Cambridge,
UK: Cambridge University Press.