A Reconstruction of Top-of-the-Atmosphere Radiation Fields During the Instrumental Era

Abstract

Earth’s energy budget is a fundamental aspect of the climate system, relating to climate sensitivity,poleward energy transport, climate dynamics, and the forced versus internal components to climate variability. Satellites provide direct observations of the top-of-atmosphere (TOA) energy budget, but global, continuous observations are limited to the past few decades—a period of strong anthropogenic radiative forcing. Longer-term estimates of Earth’s energy budget are available from climate-model simulations, but these simulations are not constrained to follow the actual evolution of the climate system. While reanalyses approaches provide this constraint, studies show that existing reanalysis are biased in their TOA radiation estimates during the satellite era. Here we assess the ability of data assimilation to constrain the TOA radiation budget based solely on observations of surface air temperature. We use offline data assimilation to reconstruct annual TOA radiation fields from 1850 to present, by combining surface temperature observations with physical constraints from climate models. To address uncertainty in the climate-model constraints, we use four different climate models in our reconstructions. Results are validated against CERES TOA outgoing longwave and shortwave radiation measurements and Earth’s energy imbalance measurements. We find that data assimilation based reconstructions track well with CERES observations during the satellite era, but results are sensitive to the climate model prior.