There has been much interest in ecological processes, including maintenance of species diversity, turnover rates, growth rates and carbon fluxes, in tropical forests, especially in the light of climate change. Tropical forests encompass a wide range of climatic regimes from semi-arid woodland to aseasonal rainforests. Comparative studies of tropical forest types have generally used data from plots in different regions of the globe. We have been studying the comparative structure and dynamics of a tropical forest in a series of 19 one-ha permanent plots established during 1993-94 along a strong east-west rainfall gradient (c.600-1800 mm annual average) within a distance of <50 km in Mudumalai, southern India. Forest types vary correspondingly from dry thorn forest through dry deciduous and moist deciduous forest to semi-evergreen forest. All woody stems >1cm DBH were tagged, identified and mapped during the first enumeration; our plots recorded 21564 individuals from 198 species of 56 families of Angiosperms. The plots have been subsequently monitored for survival and diameter growth once about every four years (1998, 2003, 2007).
Results/Conclusions
Species composition changes strongly along the gradient as expected, with no species shared between the two extreme plots. Species richness shows an overall increase with rainfall but declines in the region of medium rainfall possibly as a result of higher fire frequency. Overall mortality rates at the community level did not show significant relationship with rainfall across the rainfall gradient during any of the three census intervals; one exception to this was a negative relationship between mortality and rainfall in mid-sized stems (10-30 cm dbh) during the third census interval. On the other hand, growth rates of mid-sized and large (>30 cm dbh) stems showed consistent positive relationship with rainfall in all the three census intervals. The negative effects of prolonged drought (during 2000-03) on mortality and growth rates of trees could be seen not only during the second census interval, but also seem to be carried over into the third census interval. During the drought in the second census interval, trees in the driest eastern sites experienced shrinkage in their girths, resulting in negative growth. Over all plots over the census intervals showed significant reduction in species density and diversity. Comparative analyses across environmental gradients would help us better understand the sensitivity of various tropical forest communities to climate change and variability.
