Environmental Engineering Reference
In-Depth Information
and the state of other individuals (e.g. if predated then a predator will influence the
“ALIVE” variable of a caught prey object).
Because the interaction structure becomes successively more complex with an
increasing number of concurrent objects, it is usually not reasonable or possible to
specify it directly. Instead, a self-scheduling mechanism is required. How to set-up
such a mechanism is described in Sect.
12.2.3
.
12.2.2 Representation of the Environment
One of the important potentials of IBMs is to facilitate an easy way for simulating
spatial (and temporal) variability and heterogeneity. In the following, we focus on
how to simulate heterogeneous environmental states. When starting on the organis-
mic level, a heterogeneous spatial organization will, to some extent, already be
reached by specifying location variables for the represented organisms and installing
an activity procedure “movement” to change these variables adequately (see
Sect.
12.3
., the IPP example). When organisms interact (e.g. as predators or prey,
or as schooling organisms), the presence/absence of other individuals structures an
otherwise homogeneous environment. In addition, various other data structures can
be used to represent spatial heterogeneities. Frequently, grid-based representations
are used. Grid maps with the relevant information attached to each cell of the grid can
store e.g. (water) depth and currents in aquatic surroundings, or altitude and habitat
types in terrestrial environments. It is possible to include spatially heterogeneous
resource levels, physical structures, local light intensity, eventually in relation to
slope orientation. Often, this information is read into the programme from external
sources at the beginning of a simulation run. Also, it can be generated and modified in
the programme itself with particular updating routines. In more complex computer
models, the environmental information is frequently generated by external modules
or by other programmes. In these cases, the simulation requires the coordinated
employment of an overall system of coupled models. For simulations in marine
environments, an IBM can be coupled with a regional oceanography model
(ROM), which provides regularly updated information on currents and physical
water conditions (Penven et al. 2006). In a similar manner, it is possible to read-in
weather data in order to specify seasonal changes. Sometimes, Cellular Automata (see
Chap. 8) are employed to generate environmental structures (Breckling et al.
2006
).
The resource density and the way it is influenced by the organisms is a frequent topic
considered in IBM (Reuter 2005; Charnell 2008).
12.2.3 Scheduling Programme Processes
After discussing the update of individuals, and aspects of the data management
to provide dynamically changing environmental structures we now consider the
