Terrestrial processes influence the atmosphere by controlling land surface fluxes of energy, water, and carbon. A growing body of research has demonstrated that parameter uncertainty is a major driver of uncertainty in land surface fluxes. However, the impact of land parameter uncertainty on atmospheric processes and land-atmosphere interactions remains underexplored. Here, we quantify how land parameter uncertainty affects the atmosphere and how atmospheric responses modulate parameters’ impact on land surface fluxes. We ran an ensemble of perturbed parameter experiments for 18 land parameters in the Community Earth System Model (CESM2) under preindustrial CO2 concentrations. For each parameter, we ran one-at-a-time simulations with observationally-informed high and low values in both a partially coupled configuration of CESM2 (with active atmosphere and land models, and a slab ocean model) and an uncoupled, land-only configuration.

We find that land surface parameters generate biophysical feedbacks that significantly influence the mean climate state. Global mean land surface temperatures range by 2.2°C across our coupled parameter perturbation ensemble, and parameter perturbations drive spatially variable changes in humidity, cloud cover, and precipitation. We also find that atmospheric feedbacks modulate land parameters' impact on surface fluxes of energy, water, and carbon. For example, atmospheric feedbacks dampen land parameters' impact on global evapotranspiration. This poses a challenge to the widespread practice of developing and evaluating land models in an uncoupled configuration and then deploying them to understand and predict terrestrial processes in a coupled context. Our analysis demonstrates that land parameter uncertainty propagates to the whole Earth system, and provides insights into where and how atmospheric feedbacks modulate terrestrial processes.