Land surface phenology from remote sensing
Phenology is nature’s calendar. Changes in phenology are among the most sensitive biological responses to climate change. Across the world, many spring phenological events are occurring earlier, and fall events are happening later, than they did in the past. Land surface phenology is the seasonal pattern of variation in vegetated land surfaces observed from remote sensing.
In this project, we first assess the potential uncertainties and errors in detecting land surface phenology using current remote sensing vegetation indexes, such as NDVI (Normalized Difference Vegetation Index) across the Northern Hemisphere (Wang et al. 2017). The novel method helps to remove snow melting-induced fluctuation for vegetation phenology monitoring (Wang et al. 2018). We further find a new, underappreciated source of errors in constructing vegetation index time series, which can lead to misestimation of growing season length (Wang and Zhu 2019).
Phenology can also serve as a gateway to understanding human impacts. Phenological shifts are expected to match the pace of climate change, but human land uses might alter phenology and contribute to the climate-phenology mismatch. Our analysis shows that across the Northern Hemisphere, land surface phenology has had a spatial mismatch with climate change over the past three decades. Importantly, the mismatch is much more significant in human-dominated landscapes and increases with population density. This study reveals a widespread risk of climate-phenology mismatch exacerbated by anthropogenic land use at a global scale (Song et al. in press). We are also exploring phenology as a tool for climate change education. This project is supported by an NSF CAREER grant.