Monitoring and understanding fine-scale phenological variability in temperate forests

Abstract

Leaf phenology, referring to the study of the timing of periodic and recurring events in plant leaves, importantly regulate photosynthetic seasonality, as well as carbon and water cycling in forest ecosystems. It also serves as a sensitive indicator of climate change due to its close relationship with climate. While abiotic factors such as climate and topography have been identified as primary drivers of leaf phenology variations in temperate forests, significant phenological variations have also been observed among individuals and species within localized regions that share similar environmental conditions. This indicates that, in addition to abiotic factors, biotic factors also play a crucial role in regulating phenological variations, although the underlying mechanisms remain understudied. This thesis aims to address this knowledge gap by improving the monitoring of fine-scale phenology and investigating the effects of biotic factors on the spatial and temporal variability in phenology. This thesis begins by assessing the capacity and scalability of PlanetScope data, which have a 3 m spatial resolution and near-daily revisiting frequency, for individual- and species-scale phenology monitoring by comparing PlanetScope-derived phenological metrics with corresponding ground-based metrics. The results demonstrate that PlanetScope data can effectively capture significant variations in phenology among individual trees and species, with greater potential for species-scale phenology monitoring. Compared to ground observations and satellite data with coarser spatial resolutions, PlanetScope data proves to be more efficient in capturing fine-scale phenological variability across large spatial areas. Subsequently, the thesis investigates the role of biotic factors in regulating the spatial variability of land surface phenology within four temperate forest sites. This investigation involves the integration of plant functional type (PFT) maps, functional traits, and start and end of season (SOS and EOS) derived from harmonized Landsat and Sentinel-2 satellite data. The results indicate that PFTs alone explain only a limited portion of the variance in phenological metrics, and functional traits exhibit stronger explanatory power for spatial variability in SOS and EOS. Functional traits related to competitive ability and productivity emerge as particularly important variables in explaining phenological variations. Lastly, the thesis further explores how functional traits and diversity influence the temporal variability of phenology. The findings suggest that functional traits, especially those traits related to resource use and productivity, play a crucial role in driving interannual phenological variations, while the influence of functional diversity is less pronounced. Additionally, the results reveal that there are strong negative associations between interannual phenological variability and the stability of ecosystem productivity, and interannual phenological variability acts as a mediator in the relationship between functional traits, functional diversity, and ecosystem productivity stability. Collectively, this thesis underscores the presence of significant fine-scale variations in phenology and demonstrates the potential of using satellite data with high spatial and temporal resolutions for fine-scale phenology monitoring. Moreover, it emphasizes the crucial role of biotic factors in regulating both spatial and temporal phenological variability. The findings contribute to enhancing our understanding of the mechanisms driving phenological variability and have implications for predicting plant phenological responses to climate change.