Characterizing Sub-Seasonal Coupling between Surface Temperature and Top-of-Atmosphere Radiative Imbalance through Graphs of 3D Information Flow

Causal discovery seeks to discover potential cause-effect relationships from observational data. Here we adopt the idea of interpreting large-scale atmospheric dynamical processes and local convection as information flow around the globe, which can then be calculated using causal discovery methods.  We apply a well-established causal discovery algorithm - based on constraint-based structure learning of probabilistic graphical models - toward 10 years of daily NASA CERES Top-of-Atmosphere (TOA) radiative flux and MERRA surface temperature data to   construct graphs of information flow within and between the two fields. These graphs are created globally for different seasons as well as the entire year and their connections to horizontal/vertical mixing and radiatively active ingredients in the atmosphere (such as clouds, water vapor and aerosols) will be discussed. Specifically, we will examine how the overall structure of the information flow discovered here affects the magnitude of the climate sensitivity of the Earth.