The effect of disturbances of various intensities on tree species diversity was measured on forest regeneration. Researchers carried out an inventory of over 17 000 trees on a set of small forest plots, either subjected to different levels of commercial logging, or leaved untouched. Strong disturbances (such as intensive commercial logging), appear to favour pioneer species. If there are very little disturbances (fallen branches or only few and small canopy light gaps), the community is dominated by shade-loving species. Intermediate disturbances (more common or more extensive light gaps in untouched forests, only limited-scale logging) between the two extremes, induce peak diversity owing to the creation of a greater variety of ecological niches, some favourable to heliophilous trees, others better for shade-tolerant species. These results bring partly into question the findings that an American research team published in 1999 after investigations in Panama (2). This is the only other attempt to date to check the intermediate disturbance hypothesis at the same scale of observation. From the study of a 50-ha plot of natural forest, these researchers concluded that disturbance level had no effect on tree species diversity. They proposed instead a "recruitment limitation" hypothesis to explain how high diversity might be maintained. It states that the presence of a species in a given place is mainly due to the combination of largely chance circumstances. In other words, for a seed to have germinated in a given place, a "seed source" of the same species must have been present nearby. Then the seed should have been carried up to that point, and found suitable conditions for germination. In turn, the plantlet, then the sapling should have survived the high number of accidents that befall individual plants, such as bouts of drought, falling branches or trees, trampling, attack from predators or disease, and so on. These massive and seemingly random losses within young populations would effectively cancel out most of any "niche" effect, in other words the connection between a given species and particular environmental conditions, the basis of the intermediate disturbance hypothesis. The findings of this IRD investigation do not preclude the effect of chance. That certainly is an important factor. But it provides the first confirmation in a tropical rain forest that moderately intense disturbances contribute to the persistence of a high level of diversity. Moreover, the study highlights the importance of the heliophilous group for investigating tropical rain forests. The presence of a tree of this type signifies that there has been a relatively recent canopy light gap -even though that has left no other trace. The proportion of heliophilous trees in a forest community can therefore be taken as an indicator of past disturbance regimes. They are especially good indices in that the most light-loving species, which have been studied extensively, are relatively easy to identify. Although the intermediate disturbance hypothesis remains to be demonstrated at a scale larger than that studied to date, especially in situations of true-scale commercial logging, the results of the IRD researchers give some better indications of the dynamics of biodiversity in tropical rain forest.
1SOFT: Sols et Forêts tropicaux (Tropical Soils and Forest) programme, financed by the French Ministry of the Environment and coordinated by the GIP ECOFOR. 2S. P. Hubbell et al., Science 283, 554 (1999).