The Arctic is a historically understudied biome that holds at least twice as much carbon as the atmosphere. Climate change and arctic amplification are warming the Arctic three to four times faster than the rest of the world, and increasingly frequent wildfires and widespread degradation of permafrost are impacting ecosystem structure and function. The Next-Generation Ecosystem Experiments in the Arctic (NGEE Arctic) project has an overarching goal to advance predictive understanding of the Earth system by improving the representation of the evolution of arctic ecosystems in a changing climate at the scale of a high-resolution model grid cell. For more than a decade, a large team of scientists – spanning institutions and scientific disciplines – has made boots-on-the-ground and remotely-sensed observations across a gradient of permafrost landscapes in Arctic Alaska. These observations were used to initialize, parameterize, or evaluate models of varying complexity to improve the representation of tundra processes in the land surface component of the Energy Exascale Earth system model (E3SM, ELM). Our arctic-informed ELM has been used to make novel predictions of processes ranging from permafrost thaw to soil biogeochemical cycling to carbon–climate feedbacks associated with the unique characteristics of tundra plants to the distribution of snow across tundra hillslopes. In the fourth and final phase of the NGEE Arctic project, the team is testing our new predictive understanding under novel conditions across the arctic domain in collaboration with partners at long-term, pan-Arctic research sites. We are examining whether an arctic-informed ELM can faithfully simulate interactions among surface and subsurface processes at site, regional, and pan-Arctic scales. High-fidelity model simulations will lend confidence to global climate predictions, whereas a mismatch with long-term observations may indicate the need for new understanding, helping to guide future work.