Linking Disturbance History to Current Forest Structure to Understand How Disturbance Affects Tropical Dry Forests
40 Pages Posted: 17 Nov 2022
Abstract
Tropical dry forests are widespread, harbour vast amounts of carbon and unique biodiversity, and underpin the livelihoods of millions. A variety of natural and anthropogenic disturbances affect tropical dry forest canopy, yet our understanding of how these disturbances impact on forest structure and ecosystem functioning, and how forests develop after different disturbances, is partial. This translates into knowledge gaps regarding long-term outcomes of disturbances on forest structure as well as which of these outcomes signify recovery vs. forest degradation. Here, we use a rich dataset of remotely-sensed, high-resolution forest indicators in a multilevel Bayesian regression framework to understand the effect of different disturbance agents (partial clearing, fire, logging, drought and riparian changes) on aboveground biomass, and woody cover in the Argentine Dry Chaco. Our models show that post-disturbance trajectories of forest structural indicators differ markedly among different disturbance agents. For example, riparian changes affected biomass most strongly but had the fastest recovery, whereas logging had a generally lower impact and mostly affected tree cover, but recovery was slow or never occurred. Importantly, even three decades after the disturbance event, woody cover and biomass exhibited higher values for natural disturbances compared to anthropogenic disturbances. Furthermore, anthropogenic disturbances had slower recovery rates than natural disturbances. Overall, our approach shows the potential of remote-sensing indicators and space-for-time substitution to unravel the diverse vegetation response of different disturbance agents. Given the high and rising human pressure on dry forests in the Chaco and globally, our findings also show the long-lasting effects that anthropogenic disturbances have on these valuable forests.
Keywords: Above-ground biomass, Woody cover, Bayesian multilevel models, Dry Chaco, Forest degradation
Suggested Citation: Suggested Citation