Date:
Location:
Title: Perspectives on plant phenology in deciduous forest ecosystems at multiple spatial scales
Abstract: Deciduous forest phenology is a sensitive indicator of the ecological effects of climate change, and mediates the seasonality of key ecosystem functions including photosynthesis and transpiration. A process-based understanding of phenology relies on studies of the leaf life cycles of individual plants, which together make up ecosystem level phenology. This dissertation links the organismic scale description of plant phenology to the ecosystem scale, at which the functional effects of phenological processes are measured and modeled.
The second chapter uses records of digital repeat photography of deciduous forest phenology from 13 sites in Eastern North America to establish the connection between phenology metrics calculated from image analysis and events identified by human observers. We then link these phenology metrics from near-surface imagery to the larger scale of satellite remote sensing. We find that the difference between remote sensing and near-surface phenology depends on how representative the near-surface observations are of the surrounding landscape: less representativeness leads to a larger difference.
The third chapter uses unmanned aerial vehicles (UAVs) to obtain near-surface observations with a thorough spatial representation of canopy phenology at the ecosystem scale. We use UAV photography to observe phenology in a remote sensing pixel (250 m) at Harvard Forest with organismic scale spatial resolution (10 m). Analysis of spatial variance reveals that species variability explains most of the variance in phenology, although microclimates of land surface temperature have a significant effect. By synthesizing UAV imagery with high (30 m) and medium (250 m) resolution remote sensing data, we find a logarithmic relationship between spatial variance in phenology dates and spatial resolution of observation.
Chapter 4 examines spatial and interannual trends in deciduous forest phenology, using multiple growing seasons of UAV imagery. We link UAV image metrics to leaf life cycle events using in situ observations of trees, finding high correlation of the seasonal progression of UAV metrics with increasing leaf area in spring, and senescence in fall. Across the landscape, we find a linear relationship between years for the rates of green-up as well as senescence, with the same rate in most but not all cases. Satellite remote sensing proves to be an effective way to determine interannual variation of phenology, but not spatial variation between nearby pixels.
This dissertation uses novel observational methods and analyses to advance understanding of the connection between the ecosystem and organismic scales of phenology. We quantify the effects of overall ecosystem composition and spatial resolution of sampling on sample representativeness, and show how organism level variability relates to ecosystem phenology within and across years. Our results reveal how the seasonal activities of plants combine to create a key ecological trait.
Committee: Andrew Richardson (Advisor), Mark Friedl, Missy Holbrook, Elena Kramer and Paul Moorcroft