A supra-annual census interval distorts the interpretation of population dynamics of tropical trees

Abstract Once every few years, tropical rain forests in Southeast Asia experience severe droughts, which kill trees and shrubs. However, the census intervals for forest monitoring are often supra-annual for many study sites. When an episodic drought is included in a multi-year census interval, the substantial drop in survival rates in the drought year can be rarefied by the census interval years. Here, we demonstrated an important problem caused by conventional forest monitoring using a supra-annual census interval when conducting population dynamics studies for tropical trees. We took a population transition matrix approach using data derived from an equatorial rain forest in Lambir Hills National Park, Malaysia. We found that the population growth rate (lambda) of tree species was exclusively influenced by survival rates, while the effects of growth and reproduction rates on lambda were quite limited. This suggested that the rarefaction of a substantial reduction in the survival rate by supra-annual census intervals could mask the effects of drought on lambda. This may lead to an erroneous interpretation that the tree population remains in equilibrium even after a severe drought, and thus that drought has little effect on population dynamics. Our matrix analyses showed that the impacts of drought on lambda were hard to identify when census intervals were longer than a year. This study highlighted the complexities of designing studies on the population dynamics of tropical trees. Having shown that drought reduces tree survival rates and thus has a major impact on the population dynamics of tropical forests, we need to implement a tree census to appropriately evaluate survival rates in drought years whenever they occur.