Data from: Transient dynamics reveal the importance of early life survival to the response of a tropical tree to harvest

Exploitation of non-timber forest products can contribute to the alleviation of poverty. However, overexploitation can also lead to species decline. Studies on the sustainability of harvest often use stationary population growth rates to assess harvesting effects. For such frequently harvested systems, transient analysis can provide new insights into the response of populations to harvest. To test the prediction that the effect of harvest differs between transient and asymptotic phases, I used four years of demographic data to compare the impact of harvesting bark and foliage on the short- and long-term dynamics of African dry zone Mahogany Khaya senegalensis in West Africa. The effect of harvest was stronger in the short term than in the long term, particularly in the moist region. Patterns of transient elasticities also differed from the elasticities of long-term growth rates to perturbation of vital rates. Survival at early life stages was more important for short-term dynamics than for long-term population dynamics. Synthesis and applications. This study illustrates that using the asymptotic growth rates to assess the impact of harvest may underestimate the short-term effects of harvest. Short-term management plans based on the elasticity analysis of long-term growth rates will be suboptimal given the relative importance of the survival of non-reproductive stages for population dynamics in the near term. Managers should use results from elasticity analysis of both long-term and short-term population dynamics to develop more realistic management plans.Khaya_MeanMatrices_2004_2007This file includes mean matrices for 12 populations of Khaya senegalensis surveyed in Benin (West Africa) from 2004 to 2007. For each population, I averaged three yearly transition matrices (2004-2005, 2005-2006, 2006-2007). These matrices were built using demographic data was collected from the 12 populations using 1-ha plots for each population (two adjacent subplots of 0.5 ha, separated by the river bed when the population was in a gallery forest