(iii) Ecological complexity of the study area.
Environmental conditions are quite variable (see
Table 3) but the study area is not large. Environmental
heterogeneity within relatively narrow
spaces is a typical characteristic of Mediterranean
countries, especially on hills and mountains. Given
such heterogeneity and the connected highly
variable synergic interactions among the factors
involved, it may be difficult to extract a single set
of predictors that are strongly related to a potential
productivity measure such as site index.
(iv) Failure to adequately sample the variability.
Connected to environmental heterogeneity is the
problem of the adequate sampling the variation of
site index associated to the variations of the
environmental factors. From a theoretical point of
view, besides the accurate prestratification of the
study area, an effective strategy for gaining insight
into the workings of a highly variable multifactor
system is to increase sample size drastically.