Observational Requirements for Quantifying the Diurnal Cycle of XCO ₂ From Space

Abstract Spatiotemporal variations in atmospheric carbon dioxide (CO 2 ) provide a means to quantify surface fluxes of carbon over a range of space and timescales. NASA's Orbiting Carbon Observatory‐3 (OCO‐3), aboard the International Space Station, is the first CO 2 ‐monitoring mission to observe the sunlit portion of the diurnal cycle of total column‐averaged CO 2 ( X CO2 ) from space, since OCO‐3 collects data at various times of the day. Previous analysis of the climatological diurnal cycle in X CO2 measured from a ground‐based spectrometer in the Total Carbon Column Observing Network (TCCON) suggests that the X CO2 diurnal cycle provides information about local fluxes. Here, we examine the diurnal signal at four TCCON sites spanning the tropics through midlatitudes. The signal is typically less than 1 ppm even at the peak of the growing season. Because relatively sparse OCO‐3 data observes a diurnal cycle at a given location only across multiple days, mesoscale transport variations complicate detection of the diurnal signal from the space‐based record. We bootstrap the long‐term records of TCCON X CO2 to quantify the minimum number of OCO‐3 observations necessary to infer the climatological diurnal cycle of X CO2 , and find that even during the peak growing season, of order 10 observations per daylight hour for each month are required for robust detection. The number of required observations increases outside the growing season when fluxes are weaker. Our results show that dense and long‐term observation are required to infer the diurnal cycle from OCO‐3 or future CO 2 ‐monitoring satellite missions.