A neutral vs. non-neutral parametrizations of a physiological forest gap model

Abstract

Tropical forests are vital ecosystems which provide numerous ecological functions and forest models can be used to simulate their dynamics. However, due to the high species diversity of tropical forests and the common lack of detailed knowledge about these species, simulating the behavior of each species separately is not feasible. Therefore, species with common characteristics are usually aggregated into species groups.Although the number of species groups is likely to be an important characteristic of forest models, little research has been done on its effect on the results of the model. In this article, we compare the effect of the number of species groups using a physiological forest gap model and review our results using 28 years of field data from a 50ha forest plot in Panama as well as a chronosequence over four centuries from the same area. The number of simulated groups ranged from 1 to 4, 9 and 16. The parameterization with a single species group is a “neutral” parameterization with all simulated trees being ecophysiologically identical.The number of groups turned out to be an important characteristic of the model which influences its results on a fundamental level. Steady-states characteristics of forests like stem-numbers and stem-size distributions can be successfully simulated with even a single species group while modeling the long-term dynamics requires a higher number of groups. On the downside, a higher number of groups makes parameterization and fitting of the model more difficult. We conclude that the number of species groups is a vital characteristic of a forest model which has far-reaching consequences for its results and needs to be chosen with care.