The permafrost region is warming at an unprecedented pace. Changing climate increases the vulnerability of permafrost. Borehole observations and ground temperature modelling across the permafrost zone indicate a global rise in permafrost temperatures. Feedback mechanisms in a warming Arctic, such as increases in wildfires, sea-ice loss or infrastructure development will further contribute to permafrost destabilization on various scales. So far, observations of permafrost region disturbances (PRD) typically focus on regional scales and holistic, pan-arctic observations are lacking. Here we present the current advances of a pan-arctic analysis of rapid permafrost landscape dynamics from 2000-2020 based on globally available satellite data (Landsat, Sentinel-2) and machine-learning methods. We expand upon previous continental-scale studies towards a pan-arctic detection and mapping of typical PRD. We analyze the spatial distribution and dynamics of lakes, wildfires and retrogressive thaw slumps (RTS) across the entire pan-arctic permafrost region. Preliminary results reveal that each PRD type has typical occurrence hot-spots. RTS are often found in regions with preserved buried glacier ice. Lake dynamics are most pronounced in ice-rich permafrost terrains along the margin of continuous permafrost. Wildfires are most dominant in arid, boreal regions. Long-term dynamics are superimposed by short-term weather conditions, such as heatwaves and precipitation events, which recently have been shown to trigger or enhance disturbances significantly. The analysis of current disturbance hot-spots and the understanding of PRD dynamics and trajectories will enhance quantifying carbon fluxes and projecting future landscape dynamics. Rapidly growing streams of remote sensing data with high spatial and temporal resolution together with new Artificial-Intelligence techniques will help to improve the monitoring of PRD across the entire pan-arctic with more detail and higher accuracy.