Abstract. Earth System Models (ESMs) are our primary tool for projecting future climate change, but are currently limited in their ability to represent small-scale land-surface processes. This is especially the case for permafrost landscapes, where melting of excess ground ice and subsequent subsidence affect lateral processes which can substantially alter soil conditions and fluxes of heat, water and carbon to the atmosphere. Here we demonstrate how dynamically changing microtopography and related lateral fluxes of snow, water and heat can be represented with a tiling approach suitable for implementation in large-scale models, and investigate which of these lateral processes are important to reproduce observed landscape evolution. Combining existing methods for representing excess ground ice, snow redistribution and lateral water and energy fluxes in two coupled tiles, we show how the same model approach can simulate known degradation processes in two very different kinds of permafrost landscapes. Applied to polygonal tundra in the cold, continuous permafrost zone, we are able to simulate the transition from low-centered to high-centered polygons, and show how this results in i) more realistic representation of soil conditions through drying of elevated features and wetting of lowered features with related changes in energy fluxes, ii) reduced average permafrost temperatures at 13 m depth with up to 2 °C in current (2000–2009) climate, iii) delayed permafrost degradation in the future RCP4.5 scenario by several decades, and iv) more rapid degradation through snow and soil water feedback mechanisms once subsidence starts. Applied to warm, sporadic permafrost features, this two-tile system can represent an elevated peat plateau underlain by permafrost in a surrounding permafrost-free fen, and how it degrades in the future following a moderate warming scenario. These results show the importance of representing lateral fluxes to realistically simulate both the current permafrost state and its degradation trajectories as the climate continues to warm; both of which are likely to have important implications for simulations of the magnitude and timing of the permafrost carbon feedback.