Determination of global Earth outgoing radiation at high temporal resolution using a theoretical constellation of satellites

New, viable, and sustainable observation strategies from a constellation of satellites have attracted great attention across many scientific communities. Yet the potential for monitoring global Earth outgoing radiation using such a strategy has not been explored. To evaluate the potential of such a constellation concept and to investigate the configuration requirement for measuring radiation at a time resolution sufficient to resolve the diurnal cycle for weather and climate studies, we have developed a new recovery method and conducted a series of simulation experiments. Using idealized wide field‐of‐view broadband radiometers as an example, we find that a baseline constellation of 36 satellites can monitor global Earth outgoing radiation reliably to a spatial resolution of 1000 km at an hourly time scale. The error in recovered daily global mean irradiance is 0.16 W m−2 and −0.13 W m−2, and the estimated uncertainty in recovered hourly global mean irradiance from this day is 0.45 W m−2 and 0.15 W m−2, in the shortwave and longwave spectral regions, respectively. Sensitivity tests show that addressing instrument‐related issues that lead to systematic measurement error remains of central importance to achieving similar accuracies in reality. The presented error statistics therefore likely represent the lower bounds of what could currently be achieved with the constellation approach, but this study demonstrates the promise of an unprecedented sampling capability for better observing the Earth's radiation budget. Key Points A large constellation of satellites provides unprecedented sampling that shows promise for improving Earth outgoing radiation observations Hourly synoptic‐scale global outgoing irradiance fields are recovered from idealized synthetic measurements after 1 h of sampling Future engineering should focus on limiting systematic instrument error which remains of central importance to the observational problem