Estimating power plant CO₂ emission using OCO-2 XCO₂ and high resolution WRF-Chem simulations

Abstract

Anthropogenic CO 2 emission from fossil fuel combustion has major impacts on the global climate. The Orbiting Carbon Observatory 2 (OCO-2) observations have previously been used to estimate individual power plant emissions with a Gaussian plume model assuming constant wind fields. The present work assesses the feasibility of estimating power plant CO 2 emission using high resolution chemistry transport model simulations with OCO-2 observation data. In the new framework, 1.33 km Weather Research and Forecasting-Chem (WRF)-Chem simulation results are used to calculate the Jacobian matrix, which is then used with the OCO-2 XCO 2 data to obtain power plant daily mean emission rates, through a maximum likelihood estimation. We applied the framework to the seven OCO-2 observations of near mid-to-large coal burning power plants identified in Nassar et al (2017 Geophys. Res. Lett. 44, 10045-53). Our estimation results closely match the reported emission rates at the Westar power plant (Kansas, USA), with a reported value of 26.67 ktCO 2 /day, and our estimated value at 25.82-26.47 ktCO 2 /day using OCO-2 v8 data, and 22.09-22.80 ktCO 2 /day using v9 data. At Ghent, KY, USA, our estimations using three versions (v7, v8, and v9) range from 9.84-20.40 ktCO 2 /day, which are substantially lower than the reported value (29.17 ktCO 2 /day). We attribute this difference to diminished WRF-Chem wind field simulation accuracy. The results from the seven cases indicate that accurate estimation requires accurate meteorological simulations and high quality XCO 2 data. In addition, the strength and orientation (relative to the OCO-2 ground track) of the XCO 2 enhancement are important for accurate and reliable estimation. Compared with the Gaussian plume model based approach, the high resolution WRF-Chem simulation based approach provides a framework for addressing varying wind fields, and possible expansion to city level emission estimation.

Review of amt-2019-257

“…Further considerations about the data version are that some limited v8 validation (i.e., comparison to TCCON) was already presented in O'Dell et al, 2018, and v9 is already being used in scientific studies (Zheng et al, 2019, Reuter et al, 2019.…”

Observing carbon dioxide emissions over China's cities with the Orbiting Carbon Observatory-2

“…This non-negligible noise in the XCO 2 retrievals is hardly balanced by the amount of data sampled near emission sources with a narrow swath, which hampers the detection of emission plumes and the precision of emission quantification. Only under rare occasions, the OCO-2 tracks cross CO 2 plumes downwind large cities (Labzovskii et al, 2019;Reuter et al, 2019) or power plants (Schwandner et al, 2017;Nassar et al, 2017;Zheng et al, 2019), limiting the possibility to quantify the corresponding CO 2 emissions to few cases within a year. So far, studies on the potential of spaceborne CO 2 observation to infer CO 2 emissions from large cities or power plants have relied on the Observing System Simulation Experiments (OSSE) (Bovensmann et al, 2010;O'Brien et al, 2016;Broquet et al, 2018;Kuhlmann et al, 2019;Wang et al, 2020) and on several well-chosen cases with real OCO-2 retrievals (Nassar et al, 2017;Reuter et al, 2019;Zheng et al, 2019;Wu et al, 2020).…”

Development and evaluation of CO<sub>2</sub> transport in MPAS-A v6.3

“…T. Zheng et al: Development and evaluation of CO 2 transport in MPAS-A v6.3 have been developed and applied to estimate carbon fluxes at higher resolution (Gerbig et al, 2009;Pillai et al, 2012;Lauvaux et al, 2012;Hu et al, 2019;Zheng et al, 2018Zheng et al, , 2019.…”

“…Global high-resolution CO 2 simulations require large computational resources. Regional (limited-area) models, which have lower computational cost than their global model counterpart at the same horizontal resolution, are often used for high-resolution CO 2 transport (Feng et al, 2016;Diaz-Isaac et al, 2019 and inverse modeling (Sarrat et al, 2007;Gerbig et al, 2008;Lauvaux et al, 2012;Zheng et al, 2019). However, a regional model requires CO 2 transported from outside its model domain to be prescribed.…”

“…For a CO 2 inversion system, having lateral boundaries increases the size of the control vector to be optimized (Rayner et al, 2019). A number of approaches have been applied to the CO 2 lateral boundary problem, such as assuming the boundary inflow is perfectly known (Gockede et al, 2010), correcting the lateral boundary condition using observation prior to inversion (Lauvaux et al, 2012;Schuh et al, 2013), or jointly optimizing flux and lateral boundary condition (Zheng et al, 2018). When the CO 2 lateral boundary is optimized, an inversion system adjusts its CO 2 fields at the boundary prescribed by a parent global model in addition to adjusting surface fluxes.…”