UCLA

The largest source of uncertainty in global climate models is terrestrial carbon cycle feed- backs. One of the most important but most poorly understood vegetation types in the global carbon cycle is evergreen needleleaf forests (ENFs). To address this challenge, a growing appreciation for the stress physiology of photosynthesis has inspired emerging techniques to detect ENF photosynthetic activity with optical signals. This includes the use of solar- induced chlorophyll fluorescence (SIF), a small light signal emitted by plants during the photosynthetic process. SIF has shown a marked improvement over traditional reflectance- based vegetation indices in tracking ENF photosynthesis. However, SIF, as well as other optical signals, in ENF are complicated by photon-plant interactions over complex canopy structures (physical) and unique adaptations to deal with the seasonal stress of winter while retaining their needles (biological).

In this dissertation, we identify the physical and biological drivers of optical signals in ENF and connect remote sensing observations with physiological processes to improve monitoring of evergreen needleleaf photosynthesis. In Chapter 2, we provide a broad overview for non-specialists of the biological basis for using optical signals to track evergreen needleleaf photosynthesis. We then explore these topics in more detail by using tower-based remote sensing data across four ENF sites which span the climatic gradient experienced by ENF (details in Chapter 3). In Chapters 4 and 5 we zoom in to a single site in Canada and explore the temporal dynamics of different optical metrics and their biological underpinnings. In Chapter 6 we then show how to combine multiple metrics across multiple sites to improve predictions of forest carbon uptake.

Ultimately this work advances our understanding of ENF photosynthesis and our ability to predict the fate of ENFs in a changing climate. Future work will help scale and integrate the understandings gleaned in this dissertation to satellite and modeling frameworks.