Search

Satellite retrievals of carbon monoxide (CO) are routinely assimilated in atmospheric chemistry models to improve air quality forecasts, produce reanalyzes and to estimate emissions. This study applies the quantile‐conserving ensemble filter framework, a novel assimilation algorithm that can deal with non‐Gaussian and modestly nonlinear distributions. Instead of assuming normal distributions like the Ensemble Adjustments Kalman Filter (EAKF), we now apply a bounded normal rank histogram (BNRH) distribution for the prior. The goal is to efficiently estimate bounded quantities such as CO atmospheric mixing ratios and emission fluxes while maintaining the good performance achieved by the EAKF. We contrast assimilating meteorological and MOPITT (Measurement of Pollution in the Troposphere) observations for May 2018. We evaluate the results with the fourth deployment of the NASA Atmospheric Tomography Mission (ATom‐4) airborne field campaign. We also compare simulations with CO tropospheric columns from the network for the detection of atmospheric composition change and surface in‐situ observations from NOAA carbon cycle greenhouse gases. While the differences remain small, the BNRH approach clearly works better than the EAKF in comparison to all observation data sets. Plain Language Summary: The MOPITT instrument on the NASA/Terra satellite can detect carbon monoxide (CO) pollution in the lower and mid‐tropospheric atmosphere but cannot accurately differentiate small changes in the altitude of pollution plumes. Such satellite observations are assimilated in numerical model predictions to improve the spatial and temporal distribution of CO in the atmosphere and to estimate emission fluxes. We present a novel method that does not require assumptions about the model and the observations, leading to a more efficient and accurate assimilation of the satellite observations. Key Points: A novel non Gaussian and nonlinear ensemble data assimilation (DA) framework is applied to MOPITT joint state/flux optimizationThe new method performs better than the Ensemble Adjustment Kalman Filter in comparison to independent observationsMOPITT observations indicate that CAMS‐GLOB‐ANT_v5.3 emission fluxes are underestimated across the mid‐latitudes in May 2018 [ABSTRACT FROM AUTHOR]

This study evaluates tropospheric columns of methane, carbon monoxide, and ozone in the Arctic simulated by 11 models. The Arctic is warming at nearly 4 times the global average rate, and with changing emissions in and near the region, it is important to understand Arctic atmospheric composition and how it is changing. Both measurements and modelling of air pollution in the Arctic are difficult, making model validation with local measurements valuable. Evaluations are performed using data from five high-latitude ground-based Fourier transform infrared (FTIR) spectrometers in the Network for the Detection of Atmospheric Composition Change (NDACC). The models were selected as part of the 2021 Arctic Monitoring and Assessment Programme (AMAP) report on short-lived climate forcers. This work augments the model–measurement comparisons presented in that report by including a new data source: column-integrated FTIR measurements, whose spatial and temporal footprint is more representative of the free troposphere than in situ and satellite measurements. Mixing ratios of trace gases are modelled at 3-hourly intervals by CESM, CMAM, DEHM, EMEP MSC-W, GEM-MACH, GEOS-Chem, MATCH, MATCH-SALSA, MRI-ESM2, UKESM1, and WRF-Chem for the years 2008, 2009, 2014, and 2015. The comparisons focus on the troposphere (0–7 km partial columns) at Eureka, Canada; Thule, Greenland; Ny Ålesund, Norway; Kiruna, Sweden; and Harestua, Norway. Overall, the models are biased low in the tropospheric column, on average by - 9.7 % for CH 4 , - 21 % for CO, and - 18 % for O 3. Results for CH 4 are relatively consistent across the 4 years, whereas CO has a maximum negative bias in the spring and minimum in the summer and O 3 has a maximum difference centered around the summer. The average differences for the models are within the FTIR uncertainties for approximately 15 % of the model–location comparisons. [ABSTRACT FROM AUTHOR]

Both measurements and modelling of air pollution in the Arctic are difficult. Yet with the Arctic warming at nearly four times the global average rate, and changing emissions in and near the region, it is important to understand Arctic atmospheric composition and how it is changing. This study examines the simulations of atmospheric concentrations of methane, carbon monoxide and ozone in the Arctic by 11 models. Evaluations are performed using data from five high-latitude ground-based Fourier transform infrared (FTIR) spectrometers in the Network for the Detection of Atmospheric Composition Change (NDACC). Mixing ratios of trace gases are modelled at three-hourly intervals by CESM, CMAM, DEHM, EMEP MSC-W, GEM-MACH, GEOS-Chem, MATCH, MATCH-SALSA, MRI-ESM2, UKESM1 and WRF-Chem for the years 2008, 2009, 2014, and 2015. The comparisons focus on the troposphere (0–7 km partial columns) at Eureka, Canada; Thule, Greenland; Ny Ålesund, Norway; Kiruna, Sweden; and Harestua, Norway. Overall, the models are biased low in the tropospheric column, on average by -9.6 % for CH4, -21 % for CO and -18 % for O3. Results for CH4 are relatively consistent across the four years, whereas CO has a maximum negative bias in the spring and minimum in the summer, and O3 has a maximum difference centred around the summer. The average differences for the models are within the FTIR uncertainties for approximately 15 % of the model-location comparisons. [ABSTRACT FROM AUTHOR]

When retrieving geophysical parameters, it is advantageous to have an estimate of prior information that is based on observations with associated uncertainties, but this is often not possible. Long-term ground-based remote sensing measurements and the ozonesonde program at Summit Station, Greenland, provide an opportunity to create a unique framework to retrieve atmospheric ozone using observationally based prior information in the Arctic. This study investigates the potential of using the ground-based polar atmospheric emitted radiance interferometer (P-AERI) to estimate ozone below 10 km. Downlooking or limb-viewing sensors, such as those from satellites, have limited sensitivity to the lower atmosphere; however, uplooking, ground-based instruments provide complementary information to satellite observations to improve trace gas estimates at lower atmospheric levels. Modern-Era Retrospective Analysis for Research and Application, version 2 (MERRA-2) is a reanalysis product that integrates satellite information but also inherits their higher uncertainties at lower atmospheric levels. An observation-based climatology of the uncertainty in the MERRA-2 ozone dataset is estimated using ozonesondes launched at Summit Station. MERRA-2 shows high accuracy in the middle stratosphere but larger uncertainties below 10 km. Retrievals that use spectral radiance measurements from the P-AERI improve the estimates of ozone concentrations in the troposphere and lower stratosphere by using prior information from MERRA-2 and our climatology of MERRA-2 uncertainties as the covariance of the prior. Using ozonesonde observations from 2012 to 2017 at Summit Station, Greenland, the quality of the retrieved results is assessed. Comparisons show that retrieved partial columns reduce the bias of MERRA-2 ozone estimation below 10 km, and the average tropospheric ozone concentration is improved significantly. [ABSTRACT FROM AUTHOR]

Stratospheric circulation is a critical part of the Arctic ozone cycle. Sudden stratospheric warming events (SSWs) manifest the strongest alteration of stratospheric dynamics. Changes in planetary wave propagation vigorously influence zonal mean zonal wind, temperature, and tracer concentrations in the stratosphere over the high latitudes. In this study, we examine six major SSWs from 2004 to 2020 using the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2). Using the unique density of observations around the Greenland sector at high latitudes, we perform comprehensive comparisons of high latitude observations with the MERRA-2 ozone dataset during the six major SSWs. Our results show that MERRA-2 captures the high variability of mid stratospheric ozone fluctuations during SSWs over high latitudes. However, larger uncertainties are observed in the lower stratosphere and troposphere. The zonally averaged stratospheric ozone shows a dramatic increase of 9-29% in total column ozone (TCO) near the time of each SSW, which lasts up to two months. The SSWs exhibit a more significant impact on ozone over high northern latitudes when the polar vortex is mostly elongated as seen in 2009 and 2018 compared to the events in which the polar vortex is displaced towards Europe. The regional impact of SSWs over Greenland has a similar structure as the zona1 l average, however, exhibits more intense ozone anomalies which is reflected by 15-37% increase in TCO. The influence of SSW on mid stratospheric ozone levels persists longer than their impact on temperature. This paper is focused on the increased (suppressed) wave activity before (after) the SSWs and their impact on ozone variability at high latitudes. This includes an investigation of the different terms of tracer continuity using MERRA-2 parameters, which emphasizes the key role of vertical advection on mid-stratospheric ozone during the SSWs. [ABSTRACT FROM AUTHOR]

The Sentinel-5 Precursor (S5P) mission with the TROPOspheric Monitoring Instrument (TROPOMI) onboard has been measuring solar radiation backscattered by the Earth's atmosphere and its surface since its launch on 13 October 2017. Methane (CH4) and carbon monoxide (CO) data with a spatial resolution (initially 7 x 7 km2, upgraded to 5.5 x 7 km2 on 6th of August 2019) have been retrieved from shortwave infrared (SWIR) and near-infrared (NIR) measurements since the end of November 2017 and made available to the experts for early validation and quality checks before the official product release. In this paper, we present for the first time the S5P CH4 and CO validation results (covering a period from November 2017 to September 2020) using global Total Carbon Column Observing Network (TCCON) and Infrared Working Group of the Network for the Detection of Atmospheric Composition Change (NDACC-IRWG) network data, accounting for a priori alignment and smoothing uncertainties in the validation, and testing the sensitivity of validation results towards the application of advanced co-location criteria.We found that the required bias (systematic error) of 1.5 % and random error of 1 % for the S5P standard and bias-corrected methane data are met for measurements over land surfaces with pixels having quality assurance (QA) value > 0.5. The systematic difference between the S5P standard XCH4 and TCCON data is on average -0.69 ± 0.73 %. The systematic difference changes to a value of -0.25 ± 0.57 % for the S5P bias-corrected XCH4 data. We found a correlation of above 0.6 for most stations, which is mostly dominated by the seasonal cycle. The contributions of smoothing uncertainty at the individual stations are estimated and found to be dependent on the location. The highest contribution of the smoothing uncertainty is observed for mid-latitude TCCON stations and high latitude stations for NDACC. A seasonal dependency of the relative bias is seen. We observe a high bias during the springtime measurements at high SZA and a decreasing bias with increasing SZA for the rest of the year.We found that the required bias (systematic error) of 15 % and random error of < 10 % for the S5P carbon monoxide data are met in general for measurements over all surfaces with pixels having quality assurance value of >0.5. There are a few stations where this is not the case, mostly due to co-location mismatches and the limited availability of co-located data. We compared the S5P XCO data with respect to standard TCCON XCO and unscaled TCCON XCO (without application of the empirical scaling factor) data sets. The systematic difference between the S5P XCO and the TCCON data is on average 9.14 ± 3.33 % (standard TCCON XCO data) and 2.36 ± 3.22 % (unscaled TCCON XCO data). We found that the systematic difference between the S5P CO column and NDACC CO column data (excluding two stations that were obvious outliers) is on average 6.44 ± 3.79 %. We found a correlation of above 0.9 for most TCCON and NDACC stations indicating that the temporal variations in CO column captured by the ground-based instruments are reproduced very similarly by the S5P CO column. The contribution of smoothing uncertainty at the individual stations is estimated and found to be significant. They are found to be dependent on the location with large changes seen for stations located in the Southern Hemisphere as compared to the Northern Hemisphere and at highly polluted stations. A cone co-location criterion, which gives a better match between the ground- based instrument's line-of-sight and satellite pixels, seems to give better results for high latitude stations and stations located close to emission sources. The validation results for the clear-sky and cloud cases of S5P pixels are comparable to the validation results including all pixels with quality assurance value of > 0.5. We observe that the relative bias increases with increasing SZA. We estimated this increase is about 10 % over the complete range of measurement SZAs.The study shows the high quality of S5P CH4 and CO data by validating the products against reference global TCCON and NDACC stations covering a wide range of latitudinal bands, atmospheric conditions, and surface conditions. [ABSTRACT FROM AUTHOR]