Environmental Engineering Reference
In-Depth Information
components. This can be compensated through the employment of a cellular
automaton as illustrated by Langmead and Sheppard (2004). It allows for spatially
explicit analyses through disaggregating populations into single interacting coral
polyps. On the downside, because the rules of this CA model do not address cell
aggregations (i.e. the whole coral colony) they cannot change in relation to individ-
ual colony attributes.
The grid based approach by Mumby (2006) allows to integrate a suite of
important components of a coral reef system, which makes it possible to describe
the complex characteristic processes for coral reef communities. This application
constitutes a novel approach to the analyses of resilience and phase shifts of coral
reefs. It builds on the concept of a cellular automaton by implementing distinct
procedures within one cell and allows for dynamic changes of certain rules, e.g.
larger colonies are less likely to be overgrown by algae than smaller ones. The
combination of different modelling techniques does not only improve model
performance but also helps to identify some of the deficits in our current research
and may reveal how future experiments could be adjusted in order to fill the gaps.
Yet, the structure with distinct spatial entities - the cell - limits this approach in its
flexibility. The formulation of rules for several cells or across cells becomes very
complicated and could be easier accomplished by utilizing a continuous area.
In contrast, individual based modelling (IBM) is free from such limitations
because the model area does not have to comprise spatial aggregation of the acting
units. Yniguez et al. (2008) give a good example for an applied IBM. In their model
the environment is organized as a grid which holds different states for environmen-
tal variables. Interactions either between algal modules and/or algal modules with
their environment are possible in all directions with dynamic changes of rules in
relation to the component's attributes. The utilisation of IBM offers several useful
tools to study resilience as object-oriented programming (Chap. 4) provides the
possibility for a detailed description of organisms (as objects) in separate subpro-
grams (see also Chap. 12). This constitutes a very fine-tuned approach to model
detailed interactions on small scales. In addition, an IBM allows to integrate all
earlier developed modelling techniques, like equation based sub-models or CAs,
wherever intended or needed to create an application with highly dynamic perfor-
mance and realistic behaviour.
Understanding the factors supporting resilience and the characteristics of phase
shifts is imperative if we want to understand current and future coral reef dynamics.
Both resilience and phase shifts comprise highly complex processes that are not yet
fully understood. Over the past few years modelling has become a prominent tool to
tackle ecological questions in coral reef science. Modelling has not only contrib-
uted a great deal to advance our understanding of potential driving forces pertaining
to reef resilience, but also helped to identify the current scientific gaps and research
deficits in this discipline. Future modelling approaches that merge past and present
information derived from previous models, with data of specific sites will substan-
tially enhance our abilities to identify local driving forces of reef dynamics. This
may be employed in management programs that can help to improve the sustainable
utilisation of resources.
