A convolutional neural network (CNN) approach produced highly accurate vegetation classifications. Hyper-spectral datasets (e.g., AVIRIS) were most useful for our machine learning approaches. Accurate and high-resolution datasets generated using our approach are needed for Arctic models.
Biochemical composition is proposed to improve process-based models of SOM degradation and climate feedbacks
Several scaling strategies based on geomorphology were evaluated for complex polygonal landscapes.
Progress and Accomplishments
Snapshot of simulated snowpack surface (cyan line) and ground temperature in a low centered polygon on December 24, 2014, demonstrating that rims function as preferential outlets of subsurface heat in winter, becoming the coldest zone of the polygon.
Investigations of topographic control on thermokarst development and the ground thermal regime in ice wedge polygons using the Advanced Terrestrial Simulator
NGEE Arctic researchers found that rim height and trough depth in ice wedge polygons considerably influence the vulnerability of the underlying permafrost, shaping feedbacks which ultimately control topographic deformation and increased soil aeration in t
NASA ABoVE group photo including NGEE Arctic scientists, Shawn Serbin (top left) and Eugenie Euskirchen (bottom left).
Shawn Serbin represented NGEE Arctic at the NASA ABoVE meeting, showcasing our science and identifying opportunities for synergy with NASA ABoVE.
Examples of the collected data (e.g. LiDAR and aerial imagery) and products (Digital Elevation model) using UAS at NGEE Teller research site.
Characterization of permafrost landscapes using an unmanned aerial system: LiDAR mapping the NGEE Arctic Teller field site
A state-of-the-art unmanned aerial system was deployed to collect centimeter-resolution imagery and topographic data of the NGEE Teller research site.