Development of a Mesoscale Inversion System for Estimating Continental‐Scale CO₂ Fluxes

Abstract

Computational requirements often impose limitations on the spatial and temporal resolutions of atmospheric CO2 inversions, increasing aggregation and representation errors. This study enables higher spatial and temporal resolution inversions with spatial and temporal error structures similar to those used in other published inversions by representing the prior flux error covariances as a Kronecker product of spatial and temporal covariances and by using spectral methods for the spatial correlations. Compared to existing inversion systems that are forced to degrade the resolution of the problem in order to bring the dimensionality down to computationally tractable levels, this inversion framework is able to take advantage of mesoscale transport simulations and more of the complexity of spatial and temporal covariances in the surface CO2 fluxes. This approach was successfully implemented over one month with an identical‐twin observing system simulation experiment (OSSE) using a set of assumptions about the prior flux uncertainties compatible both with continental‐scale uncertainty estimates and with comparisons of vegetation models to flux towers. The demonstration illustrates the potential of the newly developed inversion system to use high‐temporal‐resolution information from the North American tower network, to extract high‐resolution information about CO2 fluxes that is inaccessible to coarser resolution inversion systems, and to simultaneously optimize an ensemble of prior estimates. This demonstration sets the stage for regional flux inversions that can take full advantage of the high‐resolution data available in tower CO2 records and mesoscale atmospheric transport reanalyses, include more realistic prior error structures, and explore specifying prior fluxes with ensembles.

Characterizing Average Seasonal, Synoptic, and Finer Variability in Orbiting Carbon Observatory‐2 XCO2 Across North America and Adjacent Ocean Basins

“…For instance, clouds that obscure OCO‐2 measurements are often present in mesoscale weather systems. Some inverse frameworks have improved the spatial resolution of transport models to simulate mesoscale atmospheric transport despite the great required computational expense (Wesloh et al., 2020 ), but inversions on this scale require accurate representations of subgrid‐scale spatially coherent variability in assimilated XCO 2 .…”

Temporal Error Correlations in a Terrestrial Carbon Cycle Model Derived by Comparison to Carbon Dioxide Eddy Covariance Flux Tower Measurements

“…We also indicate that much of the signal for improving CASA's prediction of AmeriFlux data is in the daily cycle and may be obscured by looking only at daily or monthly averages. These errors in the daily cycle could either be corrected in the prior flux model (CASA and the (Olsen & Randerson, 2004) downscaling), or in an inversion that allows for adjustment of the daily cycle, such as CarbonTracker-Lagrange (ESRL, 2017;Hu et al, 2019), the geostatistical modeling framework used by S. Gourdji et al (2012), the inversion frameworks used for the Mid-Continent Intensive regional inversions (Lauvaux, Schuh, Uliasz, et al, 2012;Schuh et al, 2013), or the framework of Wesloh et al (2020) (Wesloh, 2020).…”

“…We chose to increase the uncertainty of the 3‐hourly flux estimates above the monthly uncertainty estimates to allow for day‐to‐day variation and errors in the downscaling. We chose a factor of two so that the uncertainty estimates would not be much larger the flux estimates while helping with the problem of too‐small monthly and annual uncertainty estimates (Kountouris et al., 2015; Wesloh et al., 2020). The standard deviation values are a constant within each month, and undergo a step change at month boundaries.…”

“…The Olsen‐Randerson downscaling code is available at Wesloh (2022) under a three‐clause BSD license. The AmeriFlux eddy‐covariance data filtering and formatting code is available at Wesloh (2023a) under a three‐clause BSD license.…”

Regional CO2 Inversion Through Ensemble‐Based Simultaneous State and Parameter Estimation: TRACE Framework and Controlled Experiments

“…Subsequent studies found a wide spread in inversion results on continental to subcontinental scales (Gurney et al., 2002; Peylin et al., 2013), which was partly attributed to differences in the modeling of atmospheric CO 2 transport. The expansion of the in situ CO 2 network and availability of new remote sensing observations of atmospheric CO 2 have motivated the development of regional inversion systems (e.g., Gourdji et al., 2010; Lauvaux et al., 2012; Monteil & Scholze, 2021; Schuh et al., 2010; Wesloh et al., 2020). Compared with global systems, regional inversions can represent the CO 2 fluxes and atmospheric transport at higher resolutions at feasible computational costs.…”