Fall NEARC 2024

As communities and states modernize their electrical grids to overcome power outages due to energy sources or extreme weather events, emphasis should be placed on equity and resilience. In collaboration with the Vermont Electric Power Company and the University of Vermont College of Engineering and Mathematical Sciences, the Spatial Analysis Lab is supporting the mission of energy resiliency and equity focused on rural Vermont. This project is in collaboration with multiple utility distributions across the state. The wider-project's intent is to support resilience and equity of Vermont’s electrical grid by using geospatial data to fill pressing informational gaps by developing models that utility providers can use to improve energy resiliency in the face of climate change and increasing extreme weather. Specifically, the lab is mapping utility electrical infrastructure information from 2023 statewide Q1 LiDAR data. This presentation will cover the series of techniques used to classify the LiDAR point clouds and custom surface model development. Techniques included such as ERSI's pretrained GEOAI models and active learning to further train LiDAR classification models. With this research, the lab will provide utilities with highly accurate vector data that can be used for utility maintenance and preparedness. This project is funded by the Leahy Institute for Rural Partnerships.

Each fall, Western Arctic caribou migrate across frozen Northern Alaskan rivers to reach wintering grounds. Regional climate changes have altered ice and vegetation phenology dynamics throughout caribou habitats, preventing caribou from crossing rivers and successfully calving. Caribou cannot cross partially frozen rivers and require healthy vegetation in the spring to calve. The Massachusetts NASA DEVELOP Team partnered with the National Park Service in Alaska to analyze changes in river ice formation and terrestrial vegetation. Landsat 8 and 9 Operational Land Imagery, Terra and Aqua Moderate Resolution Imaging Spectroradiometer (MODIS), Sentinel 1-Synthetic Aperture Radar (SAR) C-band, and Sentinel 2A/B Multispectral Instrument imagery were used to assess ecological conditions relevant to caribou survival. The timing and duration of ice formation was assessed using the Normalized Difference Infrared Index (NDII), Relative Difference River Ice (RDRI), and Vertical-Vertical/Vertical-Horizontal (VV/VH) backscatter values based on the spectral and surface characteristics of rivers. Similarly, phenology cycles were quantified using the Nadir Bidirectional Reflectance Distribution Function Adjusted Reflectance (NBAR) for Normalized Difference Vegetation Index (NDVI) time-series data. The team produced annual freezing timelines, time series plots and maps, and composite imagery at various caribou locations to analyze annual changes. This research compared various optical and radar remote sensing methods in detecting high ice extent and high-quality vegetation over remote Alaska regions where in situ measurements are often infeasible. These study results can help evaluate spatiotemporal migratory shifts and contextualize recent shifting population dynamics.

Current, accurate land cover information is a common foundational data set that can be used to address a wide range of management issues, from flooding risk and natural infrastructure to policy evaluation and land use planning. Knowing what exists on the ground (and how it has changed through time) gives planners more information, and the better that data, the better our understanding.

For more than two decades, NOAA’s Office for Coastal Management has been producing consistent, accurate land cover and change information for the coastal U.S through its Coastal Change Analysis Program (C-CAP), with the goal of continually updating these maps every 5 years. In recent years, NOAA has been working to establish an operational higher resolution land cover product line, allowing for local and site-specific applications. This work has been possible because of the wealth of available imagery and lidar data, improved software and hardware capabilities, and artificial intelligence classification techniques.

This presentation will review the products that have been released for most coastal areas, nationally, plans for additional data and updates, as well as focus on our most recent products (just released) for the state of Maine. We will speak to the methods, accuracy, and potential applications of that data, as well as the partnership NOAA had with the state to fund this mapping.