Drivers of permafrost degradation along the Inuvik to Tuktoyaktuk Highway (ITH)

Infrastructure construction on permafrost is challenging. Not only are northern regions undergoing a faster and more intense global warming than the rest of the world, inducing thawing of the permafrost at a worldwide scale. In addition, linear infrastructures such as gravel highways, built on embankments to protect the underlying permafrost, change environmental conditions in various ways, enhancing permafrost degradation. This work aims to utilize remote sensing data and explore the physical parameters that drive permafrost degradation in the regions adjacent to the Inuvik to Tuktoyaktuk Highway (ITH) in Northwest Territories, Canada. Within the work, snow accumulation along the embankment toe, vegetation moisture increase, surface water increase in poorly drained areas, earlier snowmelt and vegetation increase along the road are defined as factors that (I) enhance permafrost degradation and (II) are observable using remote sensing techniques. The analysis is conducted using cloud computing services, open-source software packages, and primarily freely available datasets. Snow accumulation conditions are derived using Digital Elevation Models (DEM) as baseline data. The cardinal direction of the road and the predominating wind direction significantly impact the snow accumulation. Moreover, the results indicate that the enhanced snow accumulation generally reaches further distances from the road than previous studies suggest. The impact from the road on vegetation moisture and vegetation conditions, indicated by the Normalized Difference Moisture Index (NDMI) and the Normalized Difference Vegetation Index (NDVI), respectively, demonstrated significant decreases within the first 25 m from the road edge. This is in line with previous studies. However, whether the observed effect reflects the field conditions or if the spectral signal is affected by other factors like dust is critically discussed. Furthermore, my study revealed that by normalizing the median NDMI and NDVI values on an undisturbed reference area, an additional effect is observed reaching up to 200 m from the road. The analysis of the NIR band indicates that the downstream side became wetter throughout the years compared to the upstream side. The snowmelt pattern indicated by the Normalized Difference Snow Index (NDSI), derived from Landsat images, shows that the areas next to the road are snow-free earlier in spring than the areas further away. The result indicates that the road affects the snowmelt up to 600 m from the road. The findings of this work highlight the importance of future research into the impact of dust on satellite-derived indices. Furthermore, the findings contribute to a better understanding of the spatial scale of altered permafrost drivers following the construction of the ITH