報告摘要：Both light and water are important environmental factors governing tree growth. Responses of tropical forests to their changes are complicated and can vary substantially across different spatial and temporal scales. Of particular interest is the dry-season greening-up of Amazon forests, a phenomenon undergoing considerable debates whether it is real or a “l(fā)ight illusion” caused by artifacts of passive optical remote sensing techniques. Here we analyze seasonal dynamic patterns of vertical canopy structure in the Amazon forests using lidar observations from NASA’s Ice, Cloud, and and land Elevation Satellite (ICESat). We found that the net greening of canopy layer coincides with the wet-to-dry transition period, and the net browning occurs mostly at the late dry season. The understory also shows a seasonal cycle, but with an opposite variation to canopy and minimal correlation to seasonal variations in rainfall or radiation. Our results further suggest a potential interaction between canopy layers in the light regime that can optimize the growth of Amazon forests during the dry season. This light regime variability that exists in both spatial and temporal domains can better reveal the dry-season greening-up phenomenon, which appears less obvious when treating the Amazon forests as a whole.