A Model of Stand Dynamics for Holm Oak-Aleppo Pine Forests
107
changes in species abundance in terms of environmental variation and competitive interactions (see Kobe 1996; Pacala et al. 1996). The mathematical
detail required to characterize individual variability, however, increases
model complexity and hinders on the mathematical and biological interpretability of the model. Therefore, the challenge for developing explanatory and
predictive models is to achieve an adequate balance between biological realism and model complexity.
8.2.2 The Case of Mediterranean Forests
Models of forest dynamics developed to explain successional patterns in
temperate forests have provided important insights into the mechanisms
structuring plant communities (Horn 1975; Shugart 1984; Pacala et al. 1996).
These models have been implemented specifically for temperate systems and
some of their underlying assumptions are clearly inadequate in the context
of other forest types.
In the case of Mediterranean forests, topographic factors, disturbances
and agro-silvicultural systems provide recruitment opportunities at very
different spatial and temporal scales from the forest gaps that characterize
temperate forest dynamics. This contrasts with the inherent gap phase
structure common to most JABOWA-FORET-type models and raises doubts
about the possibility of using these models for Mediterranean systems. The
main factors affecting species recruitment in Mediterranean forests are also
very different from those typically considered in existing models of forest
dynamics. In Mediterranean forests, the ability to tolerate drought stress
during the seedling stage and the differential ability of species to re-establish
after disturbance are considered major determinants of species distributions
(Retana et al. 1992; Pigott and Pigott 1993; Espelta et al. 1995). Although the
effects of water stress on the physiology of these species have received considerable attention (e.g. Terradas and Save 1992; Castell et al. 1994; Sala and
Tenhunen 1994; Chaps. 10 and 11) the mechanisms linking water limitation
and whole tree performance are not well understood and have not been explicitly incorporated in any model of stand dynamics.
With the purpose of developing models of stand dynamics specifically
suited to the specific conditions of Mediterranean plant communities I initiated a series of studies on mixed holm oak (Quercus ilex L.)-Aleppo pine
(Pinus halepensis Mill.) forests. These two species dominate the overstory of
extensive areas of the western Mediterranean Basin and their population
ecology has been the subject of extensive research during recent years (e.g.
Retana et al. 1992; Espelta 1996; Chaps. 5-7). These studies have shown that
the structure of mixed holm oak-Aleppo pine forests throughout the landscape consists of a compositional gradient of mixed and monospecific stands
related to physiographic factors and stage of forest development after disturbance (Retana et al. 1996). Through time, the relative abundances of the two
species within a stand are expected to change according to the local thinning
107
changes in species abundance in terms of environmental variation and competitive interactions (see Kobe 1996; Pacala et al. 1996). The mathematical
detail required to characterize individual variability, however, increases
model complexity and hinders on the mathematical and biological interpretability of the model. Therefore, the challenge for developing explanatory and
predictive models is to achieve an adequate balance between biological realism and model complexity.
8.2.2 The Case of Mediterranean Forests
Models of forest dynamics developed to explain successional patterns in
temperate forests have provided important insights into the mechanisms
structuring plant communities (Horn 1975; Shugart 1984; Pacala et al. 1996).
These models have been implemented specifically for temperate systems and
some of their underlying assumptions are clearly inadequate in the context
of other forest types.
In the case of Mediterranean forests, topographic factors, disturbances
and agro-silvicultural systems provide recruitment opportunities at very
different spatial and temporal scales from the forest gaps that characterize
temperate forest dynamics. This contrasts with the inherent gap phase
structure common to most JABOWA-FORET-type models and raises doubts
about the possibility of using these models for Mediterranean systems. The
main factors affecting species recruitment in Mediterranean forests are also
very different from those typically considered in existing models of forest
dynamics. In Mediterranean forests, the ability to tolerate drought stress
during the seedling stage and the differential ability of species to re-establish
after disturbance are considered major determinants of species distributions
(Retana et al. 1992; Pigott and Pigott 1993; Espelta et al. 1995). Although the
effects of water stress on the physiology of these species have received considerable attention (e.g. Terradas and Save 1992; Castell et al. 1994; Sala and
Tenhunen 1994; Chaps. 10 and 11) the mechanisms linking water limitation
and whole tree performance are not well understood and have not been explicitly incorporated in any model of stand dynamics.
With the purpose of developing models of stand dynamics specifically
suited to the specific conditions of Mediterranean plant communities I initiated a series of studies on mixed holm oak (Quercus ilex L.)-Aleppo pine
(Pinus halepensis Mill.) forests. These two species dominate the overstory of
extensive areas of the western Mediterranean Basin and their population
ecology has been the subject of extensive research during recent years (e.g.
Retana et al. 1992; Espelta 1996; Chaps. 5-7). These studies have shown that
the structure of mixed holm oak-Aleppo pine forests throughout the landscape consists of a compositional gradient of mixed and monospecific stands
related to physiographic factors and stage of forest development after disturbance (Retana et al. 1996). Through time, the relative abundances of the two
species within a stand are expected to change according to the local thinning