Environmental Engineering Reference
In-Depth Information
water bodies may exist as gradients, or as interacting gradients. In this case, a mosaic is
better represented as a surface or mathematical field, analogous to a gravitational or radia-
tion field, having continuous spatial variation.
FIRST PRINCIPLES FOR ASSESSING
HETEROGENEITY
Recognizing the kinds of fluxes and kinds of structures of heterogeneity prepares us to
review the basic principles that suggest when heterogeneity would be important for ecosys-
tems. Heterogeneity is now recognized to be a pervasive feature of ecosystems and the
larger contexts in which they occur (
Kolasa and Pickett 1991; Lovett et al. 2005
). However,
when might heterogeneity not be functionally important?
Strayer (2005)
suggests some first
principles for making the decision about when to ignore heterogeneity.
Heterogeneity may be unimportant if its spatial size or temporal duration is much smal-
ler than the extent in space or time over which the measurements of interest are taken. In
such situations, the measurement integrates the finer-scale heterogeneity, and it becomes
essentially invisible. Similarly, if the grain of heterogeneity is much larger than the study
area, hierarchy theory assumes that heterogeneity is so inclusive or slowly changing relative
to the size of the measurements that it can be taken to be a constant.
The second set of reasons to ignore heterogeneity deals with the functional connections it
sets up (
Strayer 2005
). If the vectors across the patches or gradients are small relative to the
spatial extent, or slow relative to the temporal duration of the study, then that heterogeneity
is not likely to matter. Likewise, if the process dynamics across the heterogeneity are linear,
then it can be ignored. A third functional exception is in situations where one patch type
dominates the dynamics, allowing focus on that type to summarize the dynamics of the sys-
tem. Note that understanding when to ignore heterogeneity still requires an empirical
assessment of structures and transfers across the system of interest.
There are two situations in which it is especially unwise to ignore spatial heterogeneity:
human origin of pattern and regional scope of study (
Strayer 2005
). Many authors have
noted that anthropogenic heterogeneity has two conspicuous features (
Forman and
Godron 1986
). First, anthropogenically generated edges in both built and managed land-
scapes tend to be linear, as compared to curvilinear or fractal patterns in less manipulated
or natural systems. In nonurban systems that are subject to management, edges are often
made more distinct and linear by fire, timber harvest, or construction of access roads.
The second practical situation in which heterogeneity is especially likely to be function-
ally significant is modeling regional and global scales (
Strayer 2005
). Management typi-
cally involves multiple patches or the interaction of different patches at these coarser
scales (
Knight and Landres 1998
). Social processes may appear as a distinct matrix of pat-
tern over a regional landscape. In addition, each kind of system is likely to have mixtures
of attributes, sometimes even including built and biological components. For example,
understanding the structure of urban systems requires assessing the combined covers of
buildings, vegetation, and various kinds of surfaces (
Cadenasso et al. 2007
). The three spa-
tial dimensions of each kind of attribute add additional complexity to the heterogeneity
