Abstract:
|
Earth-orbiting satellites that monitor atmospheric greenhouse gases, such as NASA's Orbiting Carbon Observatory-2 (OCO-2), collect measurements of reflected sunlight at fine spatial and temporal resolution. The atmospheric constituent of interest, such as carbon dioxide (CO2) concentration, is estimated from these observations using a retrieval algorithm, which typically involves a mathematical representation of the transfer of radiation through the atmosphere and its interaction with gas molecules and particles in the atmosphere and Earth's surface. These additional nuisance constituents are not perfectly known in this retrieval problem, but errors in their representation can be correlated with errors in CO2. We illustrate the impact of these interference errors and their relationship with the nonlinearity of the physical model and with uncertainty in retrieval algorithm inputs through a simulation framework.
|