Assessing resiliency in Maine’s coastal spruce forests using remote sensing and tree-level data
This collaborative, USDA-funded project combines remotely sensed data like hyperspectral UAV imagery, ECOSTRESS thermal data, and LiDAR, with field-based physiological and dendrochronological measurements to assess vegetation stress in coastal red spruce forests in Maine. We examine how indicators such as land surface temperature, spectral vegetation indices, and forest structure relate to tree-level measures of water deficit and growth. Exploring relationships with environmental variables like drought, vapor pressure deficit, soil moisture, and temperature also inform our analysis of spruce forest responses to changing climate regimes. These relationships enable scaling across spatial scales, from individual stands to broader forested landscapes. Our goal is to develop a potential framework for detecting signs of forest stress, understand coastal forest resiliency in the context of a rapidly warming climate, and inform adaptive management strategies in these vulnerable ecosystems.
MAPPING VEGETATION-PERMAFROST DYNAMICS ACROSS ARCTIC LANDSCAPES
This collaborative work, conducted as part of DOE’s Next Generation Ecosystem Experiment (NGEE-Arctic) with Brookhaven and Oak Ridge National Laboratories, integrates field measurements with advanced remote sensing techniques to understand how permafrost thaw drives vegetation change across Arctic landscapes. We combine ground-based spectral measurements, hyperspectral UAV imagery, and airborne AVIRIS-NG data with historical Landsat time-series imagery to map “thaw functional types” and track vegetation dynamics along permafrost gradients on Alaska’s Seward Peninsula. Our approach links plot-scale measurements of individual plants and canopies to landscape-scale patterns of shrub-tussock tundra, sedges, and other key species groups, enabling us to scale from local thermokarst features to regional vegetation trajectories. By analyzing decades of satellite imagery through time-series segmentation algorithms, we characterize how disturbance and recovery patterns relate to different permafrost thaw processes. This scaling framework provides critical insights into Arctic ecosystem responses to climate change and supports improved representation of vegetation-permafrost feedbacks in Earth System Models used for climate projections.