Earth’s energy budget is controlled by the amount of incoming and outgoing radiation. Outgoing radiation is comprised of reflected solar radiation and emitted terrestrial radiation from the Earth. The energy emitted by Earth is in part controlled by the emissivity of the surface. Emissivity depends on a variety of factors, notably wavelength and surface composition. Atmospheric energy absorption is also strongly wavelength dependent. However, many Earth system models (ESMs) currently employ a broadband assumption for surface emissivity, approximating emissivity as a constant. The broadband assumption introduces error, or bias, in the models’ representation of outgoing radiation. We have devised and investigated novel methods of representing spectral emissivity that reduce bias and optimize computational resources. We find that eight or fewer bands are necessary to effectively eliminate LW flux and heating biases. In addition, a modified greybody method with updated values for emissivity can reduce total error to a degree comparable to a semi-spectral method. In future work, we plan to implement our novel methods in a fully coupled ESM. This project will improve our understanding of the ways that spectrally structured surface emissivity impacts atmospheric heating and the climate system. It is imperative that the scientific community continue to update and improve our models in order to better predict and constrain the future consequences of anthropogenic climate change.