A detailed assessment of the radiative balance and its controlling factors in polar regions is a critical prerequisite for understanding and predicting the polar amplification of climate change. Accordingly, we investigate the role of infrared surface emissivity in polar regions as a potential feedback mechanism following Feldman et al, 2014. In this work, we investigate the climatic response of the Community Earth System Model (CESM) with spectral emissivity values that are implemented in a physically consistent manner for non-vegetated surfaces. In a control model run where 1850 CO2 volume mixing ratio (vmr) is fixed, the updated spectral emissivity values are imposed for modified surface boundary conditions in the atmospheric model component. Climatic stability in the emergent globally averaged surface temperature is observed on decadal scales for an unforced (control) run. Analytic kernels representing the change in top of the atmosphere OLR given changes in emissivity are calculated on-line during the model runs, incorporating spatially and temporally varied humidity profiles impactful to transmission. Globally averaged kernels of the sensitivity of OLR to surface emissivity calculated for control and ramped CO2 runs exhibit temporal evolution with statistically significant differences in shape. Additionally, kernel and spectrally-averaged emissivity differences between monthly-averaged maps of control and ramped runs demonstrate a seasonal cycle. Similar to the treatment of cryosphere radiative forcing in Flanner et al, 2011, we define emissivity response as the product of the emissivity kernel and the change in month-to-month emissivity. At the end of 20th century, the 10-year emissivity forcing averaged at latitudes > 60°, is found to be negative (positive) in January (July), due to increasing (decreasing) sea-ice. These findings indicate that differences in surface emissivity between frozen and unfrozen surfaces decrease wintertime and increase summertime Arctic temperatures under sustained CO2 forcing.