We've been working with some conceptional framework of the simulation models which describe the processes of "desertification". The objective of our team is to construct the appropriate guideline to construct the simulation model and to propose a new paradigm of ecological models.
Environmental conditions of desertification effective to
plants
The term "desertification" should be used with the view
point of the effectiveness of its environmental conditions to
plants. Because the strict meaning of "desertification"
is the decline or elimination of the vegetation caused by drought,
the major effective environmental conditions can be easily determined.
That is, water shortage, salt accumulation and soil erosion.
How many plant responses should be considered ?
Plant responses to these three environmental conditions must be
divided in to two levels; physiological responses and reproductive
responses. Physiological responses are examined mainly on the
level below the individual plants. Reproductive responses, which
has been often called as ecological responses, are studied on
the level higher than the individual plants, and mainly on the
population-level. Thus, we must consider 3x2=6 components to
describe the plant responses to drought and the process of "desertification".
Temporal and spacial nature - heterogeneity
Spatial distributions of the three environmental conditions are
characterized by its heavy heterogeneity. It is now strongly
proposed that the spatial heterogeneity of the environmental conditions
plays a key role to maintain plant species survival. We must
include the spatial heterogeneity of the environmental conditions
into our model.
Temporal distribution of the environmental conditions are characterized
by its deep unpredictability. Suppose that the rainfall pattern
at a certain desert. Such an deep unpredictability is quite critical
and often lethal to all plant species. Thus, we can't exclude
the unpredictability of the environmental conditions from our
model.
Importance of individual-based models (IBMs)
Individual-based models (IBMs) are categorized as analytical
mathematics model and Monte-Caro-type simulation models. The
most prominent difference of IBMs from the previous models is
its structure.
Unlikely to the old models, the IBMs are constructed based on
the characteristics of individuals. For example, most of the old-styled
forest dynamics models consisted of a series of size or age classes.
In these models, the informations about positions of each trees
and/or those about structure of crown were neglected. IBMs do,
however, consist of individual trees with specific positions and
canopy structures.
The IBMs have three useful features. The first is their ability
to make connections between phenomena at different levels of biological
organization. It can elucidate the consequence of individual
responses to population and community responses to environmental
conditions. The second is that they can examine the effects of
high heterogeneity and high stochasticity (or, say, uncertainty)
of environments on biological communities. The third is their
usage as correction methods for general theories. For example,
much of competition theory in ecology is at the population level,
and is not tested at the individual level. The IBMs must give
some corrections to the theories based on population.
Additional viewpoint - chaos
It is well-known that some complicated systems often show chaotic
behavior. The IBMs also sometimes show the chaotic behavior.
They often return to regular or random behavior after a certain
period of chaos, and they may go to chaos again. This is because
some chaotic systems have "window". In this window,
some regular or random pattern can be seen. As the desertification
is a kind of simplification of vegetation, it must be useful to
consider the "window" of the chaotic system.