Representing Model Uncertainty for Global Atmospheric CO<sub>2</sub> Flux Inversions Using ECMWF-IFS-46R1

Abstract

Abstract. Atmospheric flux inversions use observations of atmospheric CO2 to provide anthropogenic and biogenic CO2 flux estimates at a range of spatiotemporal scales. Inversions require prior flux, forward model and observation errors to estimate posterior fluxes and uncertainties. We use a numerical weather prediction model to diagnose the global forward model error associated with uncertainties in the initial meteorological state, physical parameterisations and in-model biogenic response to meteorological uncertainty. We then compare the error with the atmospheric response to uncertainty in the prior anthropogenic emissions. Although transport errors are variable, average total column CO2 (XCO2) transport errors over anthropogenic emission hotspots (0.1–0.8 ppm) are comparable to, and often exceed prior monthly anthropogenic flux uncertainties project onto the same space (0.1–1.4 ppm). Average near-surface transport error at 3 sites (Paris, Caltech and Tsukuba) range from 1.7–7.2 ppm. The global average XCO2 transport error standard deviation plateaus at ~0.1 ppm after 2–3 days, after which atmospheric mixing significantly dampens the concentration gradients. Error correlations are found to be highly flow-dependent, with XCO2 spatiotemporal correlation length scales ranging from 0 km to 700 km and 0 to 260 minutes. Globally, the average model error caused by the biogenic response to atmospheric meteorological uncertainties is small (

Global nature run data with realistic high-resolution carbon weather for the year of the Paris Agreement

“…The CHE nature run aims to support scientific studies that will shed light on the challenges of estimating CO 2 emissions with the goal to build a CO 2 monitoring and verification support capacity 3 . These challenges span a wide range of aspects from sparse observing systems, consistency between ocean/land observations from different satellite-view modes 16 , large variability in the biogenic signal 17 , large representativity errors in anthropogenic emissions 13 , transport errors 18 and stringent requirements of high accuracy observations to estimate small signal with respect to large background values 16 , 19 . This global high-resolution dataset can provide a reference for testing different approaches to address those challenges.…”

WOMBAT: A fully Bayesian global flux-inversion framework

“…where v 2 (• , • , •) is the residual mole-fraction process arising from the use of an approximate initial molefraction field, imperfect meteorology inside the transport model, imperfect transport-model parameters and physics, and potentially sub-grid-scale variation in the mole-fraction when Ĥ is a numerical model evaluated at a coarse resolution. It is difficult to place prior beliefs on the structure of v 2 (• , • , •), which we model as statistical error, but it is known that using the approximation Ĥ introduces errors that could span hundreds of kilometres and several days (Lauvaux et al, 2019;McNorton et al, 2020). Transport-model implementations tend to differ considerably in their vertical and inter-hemispheric mixing behaviour, and flux-inversion estimates are known to be particularly sensitive to transport-model choice (Gurney et al, 2002;Schuh et al, 2019).…”

“…The causes of model misspecification are numerous; for a comprehensive discussion, see Engelen et al (2002). The main ones are (i) flux-process dimension-reduction error (e.g., Kaminski et al, 2001), which is a consequence of using a spatio-temporal model for the flux field that is low-dimensional and inflexible; (ii) an inaccurate prior flux mean, variance, and covariance (e.g., Philip et al, 2019); (iii) transport-model errors (e.g., Houweling et al, 2010;Basu et al, 2018;Schuh et al, 2019) arising from the underlying assumed physics, meteorology, and discretisation schemes used (e.g., Lauvaux et al, 2019;McNorton et al, 2020); (iv) retrieval biases (e.g., O'Dell et al, 2018) and incorrect associated measurement-error statistics (e.g., Worden et al, 2017); and (v) measurement-error spatio-temporal correlations that are not fully accounted for (e.g., Chevallier, 2007;Ciais et al, 2010). Two other causes of model misspecification worth noting are an incorrectly specified initial global mole-fraction field, and flux components assumed known in the inversion (i.e., assumed degenerate at their prior mean), such as anthropogenic emissions (e.g., Feng et al, 2019).…”

Discussion on A high‐resolution bilevel skew‐tstochastic generator for assessing Saudi Arabia's wind energy resources

“…WRF has a good track record of being fairly accurate but, like any numerical model, is imperfect (e.g., Carvalho, Rocha, Gómez‐Gesteira, & Santos, 2014). In global carbon dioxide flux inversion, for example, the lack of consensus between research groups worldwide is largely due to the disparity between atmospheric transport models (McNorton et al, 2020; Schuh et al, 2019). A related issue is that output from atmospheric transport models are generally not provided with a measure of uncertainty, which is often substituted with the notion of variability.…”