Speaking: Katie Braun & Christian Andresen, University of Wisconsin - Madison
The Permafrost Discovery Gateway hosts a monthly webinar series on a Thursday at 09:00 Alaska time. The webinar aims to 1) connect the international science community interested in big data remote sensing of permafrost landscapes, and 2) provide the Permafrost Discovery Gateway development team with end-user stories (by the presenter and webinar participants), such as exploring tools the community needs to create and explore big data.
Abstract
Common methods of mapping ice wedge degradation use surface water in remotely sensed imagery as a proxy for ice-wedge degradation; this method consistently underestimates total degradation as surface hydrology in ice-wedge troughs is complex and only a portion of degrading ice wedges are flooded. More accurate remote sensing methods for detecting ice- wedge degradation stages – that depict ice wedges in undegraded, degraded, and stabilized states – will better allow us to monitor and predict Arctic landscape change. We characterized ice wedge degradation stages near Prudhoe Bay, Alaska using a novel approach that combines spectral (e.g., NDVI and NIR) and geometric properties of thermokarst pits and troughs with high-resolution (0.5 m) WorldView-2 imagery. Spatial patterns in ice-wedge degradation were identified through clustering: areas with significantly similar trough widths and flooding stage were grouped into hydrogeomorphic units. We can associate these hydrogeomorphic units with ice-wedge degradation stages based on both fieldwork and high-resolution temporal analysis of the evolution of these ice wedge landscapes. The resulting maps of ice-wedge trough networks and degradation estimates have been validated with field observations, drone LIDAR, multi- spectral and photogrammetry surveys. These improved maps of ice-wedge networks reveal that ice wedge degradation is heterogeneous across both meter and kilometer scales. In addition, this approach can provide high resolution baseline datasets for training deep-learning AI mapping efforts of ice-wedge degradation stages across greater spatial and temporal scales.