Temperature and water vapor profile singular value decomposition scaling to improve OCO-2 XCO2 errors

The NASA Orbiting Carbon Observatory-2 (OCO-2) was launched in 2014 with the goal of accurately and precisely measuring column-averaged dry-air mole fractions of carbon dioxide (XCO2). In order to fit the measured near-infrared radiances properly, many physical parameters besides CO2 are included in the optimal estimation state vector, including atmospheric water vapor and temperature. The current operational NASA Atmospheric Carbon Observations from Space (ACOS) XCO2 retrieval algorithm (Version 11) solves for a multiplicative scaling factor on an a priori water vapor profile and an additive offset on an a priori temperature profile. However, simulations have indicated that both water vapor and temperature have 1.5-3 degrees of freedom in the vertical atmospheric column of OCO-2 observations. This means that the retrieval is limited in its ability to fit the true profiles of temperature and water vapor. Simulations using synthetic data and retrievals performed on real measurements indicate that solving for properly constrained full profiles of temperature and water vapor can lead to improved XCO2 retrievals. However, due to operational constraints on the allowed size of the state vector and corresponding output files, a more simplistic approach was desired. In this work, we use singular value decomposition (SVD) to determine the three most common vertical profile ``shapes'' of atmospheric water vapor and temperature error, then retrieve a single scaling factor applied to each of the shapes. We assess retrieval errors statistics by comparing to the Total Carbon Column Observing Network (TCCON) and a collection of CO2 models. We find that the radiance residuals are reduced, especially in the strong CO2 band. We find that after applying XCO2 quality filtering from Data Ordering Genetic Optimization (DOGO) and a custom three-parameter bias correction, the scatter of the overpass-mean XCO2 error versus TCCON for a DOGO throughput of 70% is reduced from 1.02 to 1.01 ppm (2.3% reduction in variance) for land glint observations, 1.04 to 0.96 ppm (14.5% reduction in variance) for land nadir observations, and 0.68 to 0.66 ppm (4.7% reduction in variance) for ocean glint observations. We also see a small improvement in the agreement between OCO-2 XCO2 and the collection of CO2 models over the global oceans and over the Amazon rainforest.