Evaluation and Potential Impacts of OCO-2 XCO2 Variability at Regional and Synoptic Scales, using Lidar and In Situ Observations from the ACT-America Campaign

The Orbiting Carbon Observatory 2 (OCO-2) presents exciting possibilities for monitoring the global carbon cycle with its nearly 1 million observations of column-mean carbon dioxide concentration (XCO₂) per day. The XCO₂ dataset has been shown to achieve target precision and accuracy on a single-sounding level, but the validation of XCO₂ spatial gradients on sub-continental scales remains challenging. In this work, we investigate the use of an integrated path differential absorption (IPDA) lidar for the evaluation of OCO-2 observations via NASA's ACT-America project. The project has provided a precisely colocated, high-resolution validation dataset for OCO-2, spanning nearly 3800 kilometers across four seasons and eight clear-sky underflights, with both the Multi-Functional Fiber Laser Lidar (MFLL) and a suite of in situ instruments recording meteorological state and CO₂ data along the OCO-2 ground track. Our work investigates the challenges and opportunities involved in comparing the MFLL and OCO-2 datasets, and evaluates their agreement on synoptic and regional scales. We find that OCO-2 synoptic-scale gradients generally agree with those derived from the lidar within ±0.1 ppm per degree latitude, indicating a suitable averaging scale for this data when used to inform transport inversions. We also find that CO₂ reanalysis products typically agree to ±0.25 ppm per degree latitude when compared to a novel in situ-informed CO₂ "curtain." Real XCO₂ features at regional scales remain challenging to observe and validate from space, but ACT-America data have helped to investigate interesting local XCO₂ patterns, and to identify systematic spurious cloud-related effects in the OCO-2 dataset. As a consequence, there has been work to develop a more effective, MODIS-derived cloud flag for OCO-2 data processing. We attempt to use this new flag to evaluate the impact of spurious cloud-related XCO₂ features on regional flux estimates from a 4DVar-based flux inversion model.