The terrestrial biosphere carbon sink is a major source of uncertainty in projections of 21^st-century climate. Enhanced primary production under rising atmospheric CO₂ concentrations would increase the storage of carbon as soil organic matter (SOM). But the short-term efficacy depends on rates of SOM decomposition, which are highly uncertain in Earth System Models. The radionuclide carbon-14 (¹⁴C) provides a powerful constraint on the temporal dynamics of soil decomposition. We used the International Soil Radiocarbon Database (ISRaD), an extensive database of ¹⁴C dated, vertically-resolved soil samples, to calibrate key parameters governing these processes in version 1 of the Energy Exascale Earth System Model (E3SMv1). The default parameters in the E3SMv1 Land Model (ELM) produce reasonable SOM stocks, but underestimate the ¹⁴C ages at depth. This suggests the model may be cycling SOM too quickly, which would lead to an overestimate of the short-term carbon sink. Using ISRaD, we optimized the parameter that reduces decomposition rates with depth while introducing a new parameter that reduces carbon use efficiency at depth. An initial goal is to enable comparison of the optimized parameters with observationally-inferred values for depth-dependent controls on carbon stabilization and microbial activity. The fundamental goal is to use optimized ELM parameters to increase our confidence in the rate of terrestrial carbon uptake in E3SM simulations with increased atmospheric CO₂. This will improve projections of 21^st-century climate change and indicate how much the terrestrial carbon sink can offset anthropogenic CO₂ emissions.