Field observations and model simulations confirm the presence of unfrozen permafrost in shrub-dominated landscapes.

Light-harvesting processes involved in photosynthesis by tundra plants are unexpectedly sensitive to growth at low temperatures.

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.