| Abstract: | Large‐scale forest loss and degradation have dire consequences for biodiversity maintenance and provision of vital ecosystem services. Despite recent increasing efforts for forest restoration and sustainable management, there have been no comparative studies of biological taxonomy and multiple ecosystem functions to assess the effectiveness of forest restoration programmes, and how they vary through space and time. Here, we provided a quantitative assessment of the recovery of biodiversity and ecosystem functions by forest restoration in China using a meta‐analysis of 172 studies. We found that biodiversity and ecosystem functions were substantially increased in restored forests comparing with the degraded states. However, these restoration effects varied considerably by degradation origin, restoration approach, restoration age, ecological domains, taxonomic group and ecosystem function that is measured. Results also revealed that forest restoration from degraded states could not lead to full recovery of biodiversity and ecosystem functions, highlighting the irreplaceability of primary forests. We advocate allowing for natural or passive recovery, especially where biophysical conditions are favourable for spontaneous succession, or too harsh for human‐aided restoration, and choosing a combination of passive and active restoration measures based on adaptive management strategies. Our meta‐analysis provided fundamental insights into bridging the gap between small‐scale experiments and broad‐scale management needs towards highly effective and sustainable forest ecosystem restoration. Copyright © 2017 John Wiley & Sons, Ltd. |
