A Quantitative Graph-Based Approach to Monitoring Ice-Wedge Trough Dynamics in Polygonal Permafrost Landscapes

In response to increasing Arctic temperatures, ice-rich permafrost landscapes are undergoing rapid changes. lowlands, polygonal ice wedges especially prone degradation. Melting of results in deepening troughs and the transition from low-centered high-centered ice-wedge polygons. This process has important implications for surface hydrology, as connectivity such determines rate drainage these lowland landscapes. this study, we present a comprehensive, modular, highly automated workflow extract, represent, analyze remotely sensed trough networks graph (i.e., network structure). With computer vision methods, efficiently extract locations well their geomorphometric information on depth width high-resolution digital elevation models link data within graph. Further, discuss benefits analysis algorithms characterizing erosional development thaw-affected Based our analysis, show how thaw subsidence progressed between 2009 2019 following burning at Anaktuvuk River fire scar northern Alaska, USA. We observed considerable increase number discernible study area, while simultaneously disconnected decreased 54 small only six considerably larger 2019. On average, increased by 13.86%, average slightly 10.31%. Overall, new approach allows monitoring dynamics unprecedented spatial detail, reducing quantifiable geometric measures relationships.