Temperature variance measures how much temperatures fluctuate over time. Higher variance increases the risk of extreme weather, threatening health, agriculture, and infrastructure. Studies suggest that temperature variance increases when atmospheric circulation moves air parcels across regions with strong background temperature gradients and allows them to travel long distances before mixing. How far these parcels travel depends on how much energy they lose when interacting with the land surface. This work introduces a framework to quantify how land–atmosphere interactions influence temperature variance by modulating that travel distance. Using numerical experiments that perturb atmospheric circulation and land processes, we show that land-driven changes in surface longwave radiation dissipate most of the energy as air parcels move over winter midlatitude land, while atmospheric circulation has a smaller effect. Our findings highlight the importance of accurately representing land processes in climate models to improve projections of future temperature variance.