Abstract

Promoting natural regeneration can be a viable strategy to achieve ambitious restoration commitments, but predicting where native forest cover is likely to expand is challenging. Different biophysical and socioeconomic factors may determine the chances of native forest regrowth at different spatial and temporal scales, producing complex spatial patterns in the landscape and adding more and deeper information about potential for different forms of forest restoration. On the basis of a systematic literature review including 64 peer-reviewed articles from the global tropics and subtropics published from 1990 to 2017, we assessed the methodologies employed to remotely detect forest cover increase and to identify its biophysical and socioeconomic drivers. Automatic classification of multitemporal images and transition matrices were the most popular methods to detect areas of forest regrowth, whereas regression analysis was the most used analytical approaches to assess drivers. Forest cover increased more often on steeper slopes, close to forest remnants, inside protected areas, and far from population centers. However, the effects of most drivers varied among scales of evaluation and may be further affected by the scale of forest regrowth. The most influential biophysical and socioeconomic drivers of forest cover increase identified here can be used to develop predictive models on the likelihood of native forest regrowth to guide the implementation of cost-effective tropical forest restoration, thus contributing to the mitigation of climate change and species extinction. Current remote-sensing literature evaluating forest cover changes seems to be largely focused on deforestation dynamics, with little attention given to forest cover increase.