By Meine van Noordwijk
Biodiversity encompasses all forms of life, and it is thus nearly impossible to measure. Most of the time we have to rely on “proxies”, or correlates, such as the presence of trees – with the expectation that the larger and more diverse the trees are, the higher the diversity of other forms of life will be. Natural rain forest is the most diverse ecosystem on land, only rivaled by its marine counterpart in the coral reef.
Once in a while, however, scientists get a chance to test these assumptions on correlations and proxies, by sampling many groups of organisms at the same location. A recent paper1 discusses “Plant functional types and traits as biodiversity indicators for tropical forests: two biogeographically separated case studies including birds, mammals and termites.” It draws on the results of an integrated sampling effort in two of the worlds’ biodiversity hotspots: the Brasilian Amazon and insular Southeast Asia, Sumatra. While the overall results on relationship between the canopy height and tree basal area with diversity in other groups come as no surprise, the study added a simple measure of thickness of the litter layer to the arsenal of confirmed proxies. In detail, however, diversity rules: every subdivision of biodiversity (birds, beetles, bats, bees, butterflies, beasts) has its own story on which habitat supports the most diverse array.
What I remember from the fieldwork underlying this paper is the challenge of the incompatible sampling techniques: the bird people had to be first at each plot before others would make noise, while the mammal folks set their life traps and mist nets for an overnight sampling; the next day the vegetation and carbon stock team (to which I belonged) would be there, but if they didn’t finish quick enough the team fogging the canopy to sample insects would come and chase us out. As last in the sequence, the termite crew went through the area, turning every piece of wood and leaving the plot behind as if a herd of wild boar had been there. During meals and evening chat sessions we all learned a lot about the landscape we all sampled as well as the peculiarities of each group studied, as well as the people studying them.
We sampled the last remaining lowland rainforest in the Muara Bungo district in Jambi (Sumatra) – at a research station that is now part of natural history and replaced by an industrial timber plantation. We compared it with the jungle rubber agroforests, rubber plantations, and early stages of secondary forest succession (which were full of ticks). The small-mammal team found some of the life traps that had disappeared from the sampling sites back in the local market, with or without content.
The data, after all compilations and calculations were completed, show that carbon stocks are more sensitive to change in forest condition, e.g. due to logging, than the biodiversity indicators used, while plantations restore carbon stocks before they restore biodiversity. While there is some overall relationship between carbon stocks and biodiversity, the dynamics are quite different when one zooms in. This implies that efforts to maintain and restore forest carbon stocks from the perspective of reducing greenhouse gas emissions align with biodiversity conservation, but are not identical. This finding, aligned with other data, suggests that the climate change agenda of the UNFCCC is only partially parallel to the biodiversity agenda of its system convention, the CBD. Whereas the carbon stock assessments for climate change debates can focus on the forest, biodiversity researchers have to look for the trees and their variation.
The paper now published in the journal Biodiversity and Conservation is focused on the concept of plant functional types: across the continents pioneer trees have large leaves, late-successional species have other traits in common, such as finely divided leaves with drip-tips that allow the leaves to be free of adhering water films after rain. Without knowing the species you can deduce function from form. And indeed this classification was found to add value to the surveys, but especially when combined with taxonomic accounts of plant diversity using the family, genus, species concepts. The rain forest contains the highest diversity of species within a single “functional group” of plants, while the man-made land cover types tend to have only one species per functional type, in the current classification system.
The overall conclusion can be summarized disrespectfully as: the more it looks like a rainforest in terms of tree height, tree basal area and litter layer, the higher it is in biodiversity value, across all major groups assessed. This helps in comparing agroforests and more intensively managed plantations. It may appear to contain no surprises, but empirical evidence of the same level of comprehensiveness is very scarce in the scientific literature.
- Gillison AN, Bignell DE, Brewer KRW, Fernandes ECM, Jones DT, Sheil D, May PH, Watt AD, Constantino R, Couto EG, Hairiah K, Jepson P, Kartono AP, Maryanto I, Neto GC, van Noordwijk M, Silveira EA, Susilo FX, Vosti SA, 2013. Plant functional types and traits as biodiversity indicators for tropical forests: two biogeographically separated case studies including birds, mammals and termites. Biodiversity and Conservation, Published online: 12 July 2013. http://link.springer.com/article/10.1007/s10531-013-0517-1/fulltext.html