Using Ground- And Space-Based Observations to Quantify Within-Day to Synoptic-scale Variance Budgets of Total Column Averaged Carbon Dioxide (XCO2) Measurements

Abstract

Carbon dioxide (CO2) is the single most important greenhouse gas contributor to increases in global temperatures. Since 1850, humans have emitted approximately 700 Gt of carbon. About half of that CO2 has remained in the atmosphere with the rest being stored in the land and ocean. While global human emissions of CO$_2$ from fossil fuels are well-understood, regional-scale exchanges between the land and atmosphere and ocean and atmosphere remain highly uncertain and are a very active area of study. Inferences of carbon fluxes can be made by modeling carbon cycle processes, statistical upscaling of local fluxes, human emissions, or accurately partitioning drivers of spatiotemporal changes in atmospheric CO2 observations.

The launch of space-based CO$_2$ monitoring systems in the 21st century has provided unprecedented global coverage of CO2 observations. Instruments, such as NASA’s Orbiting Carbon Observatory-2 (OCO-2), which launched in 2014, provide approximately 1 million total column-averaged CO2 (XCO2) soundings per day at a horizontal spatial resolution of 1.29 × 2.25 km2, with a repeat cycle of every 16 days. OCO-3, which was installed on the International Space Station in 2019 provides the first space-based XCO2 observations at different times of day, potentially allowing for the direct observation of the diurnal cycle of XCO2 from space.

This thesis leverages the abundance of XCO$_2$ data available to quantify a subseasonal variance budget of XCO$_2$. At these timescales, we expect XCO2 to vary as a function of local fluxes and advection via atmospheric transport. Here, we first quantify climatological variations in XCO2 driven by fine mesoscale atmospheric transport, larger synoptic scale atmospheric transport, and local diurnal fluxes using ground-based XCO2 observations from the Total Carbon Column Observing Network (TCCON). We then use the one dimensional tracer conservation equation to provide a statistical framework for estimating along-track mesoscale variations of XCO2 detected by OCO-2 data. We also provide a framework to detect the diurnal cycle of XCO2 from space-based missions, such as OCO-3.

This research contributes an early attempt at quantifying "XCO2 meteorology," analogous to our understanding of subseasonal variations of atmospheric water vapor. These studies provide alternative methods to quantify transport-driven variability and random and spatially-coherent errors in XCO2 retrievals, which can be applied to inverse models or evaluating the performance retrieval algorithms. Most importantly, these studies move us closer to intuitively quantifying the local-to-regional scale fluxes globally using XCO2 observations.