Climate change signals of radiative forcing and feedback unveiled from long-term trends of spectrally resolved surface longwave radiation

Surface warming is directly associated with the surface energy balance, where downwelling longwave radiation is a critical factor influencing and reflecting surface temperature variations. Accurately identifying various forcing and feedback mechanisms is essential to making more realistic predictions about future climate change. Spectrally-resolved radiance measurements play an important role in this pursuit by leveraging the distinctive absorption features of atmospheric compositions. Only recently, the availability of comprehensive, long-term records of spectrally resolved radiation and atmospheric properties has enabled us to observe and quantify the forcing and feedback factors, such as the cloud feedback characterized by its high uncertainty. This study initiated by homogenizing the 23-year record of downwelling longwave radiance (DLR) observed by the Atmospheric Emitted Radiance Interferometer (AERI) at the Southern Great Plains site. A detailed DLR record for diverse sky conditions was obtained, enabling the determination of long-term trends in both clear-sky and all-sky scenarios. These trends reveal distinct spectral signals associated with various meteorological variables, forming the basis for further climate change signal attribution analysis. Subsequently, we develop and validate a novel spectral fingerprinting method tailored to constrain surface forcings and feedbacks from long-term DLR trends. Our analysis identifies positive CO2 and negative O3 surface forcings in both clear-sky and all-sky conditions. Moreover, we observe that changes in temperature and water vapor concentration over the 23-year period contribute to an increase in downwelling longwave radiation. Significantly, our study discovers a negative cloud feedback that offsets the increase in downwelling longwave radiation resulting from elevated CO2, water vapor, and atmospheric temperature. These attributions of radiation changes, derived from AERI observations using the fingerprinting method, are validated against the kernel method and compared with the simulations of Global Climate Models.