Regresar

Global patterns of drought deciduous phenology in semi-arid and savanna-type ecosystems

Abstract:

Seasonal variations in vegetation have direct and indirect impacts on biogeochemical cycling, net primary productivity, and the global climate. While leaf phenology in many high latitude ecosystems can be pbkp_redicted based on a four-season cycle, semi-arid and savanna-type (SAST) ecosystems are thought to be much less consistent and interannual variability can be quite large. While leaf phenology in SAST ecosystems is often attributed to soil moisture availability, different plants appear to respond to different signals, and pbkp_redicting drought deciduous phenology, especially at large scales, as in land surface models (LSMs), continues to be a challenge. Here we undertook a meta-analysis of published research on leaf phenology in SAST ecosystems to assess whether any consistent patterns would be revealed across the SAST regions of the world. After reviewing 188 papers and correlating green-up and brown-down dates with a number of possible drivers, including latitude, temperature, precipitation, continent, and plant functional type (PFT), we found that latitude, then continent and PFT, were the best pbkp_redictors of phenological timing. This result suggests that solar radiation may be an important driver of phenology even in the low latitudes, and that biogeographic differences among continents and PFTs may be critical to improving our pbkp_redictions of SAST phenology. All global LSMs lump plant species into functional groups, and these groups typically span continents despite known physiological differences between species on different continents. Our review suggests that more research is needed into SAST phenology at both field and remotely sensed scales, but that LSM pbkp_redictions could be improved by simply developing continent- and latitude-specific model parameterizations. With a large fraction of the world's human population living in and relying on services from SAST ecosystems, it is critical that we focus more attention on these understudied systems and improve their representations in global models.