Effects of biomass burning on CO, HCN, C2H6, C2H2 and H2CO during long-term FTIR measurements in Hefei, China.

High-resolution solar absorption spectra were continuously collected by a ground-based Fourier transform infrared (FTIR) spectrometer to retrieve the total column of carbon monoxide (CO), hydrogen cyanide (HCN), ethane (C2H6), acetylene (C2H2), and formaldehyde (H2CO). The time series and variation characteristics of these gases were analyzed. The biomass combustion process is identified by using the correlations between the monthly mean deviations of HCN, C2H6, C2H2 and H2CO versus CO and satellite fire point data. The months with high correlation coefficients (R > 0.8) and peaks of fire point number are considered to be with biomass combustion occurrence. The emissions of HCN, C2H6, C2H2 and H2CO in Anhui were estimated using the enhancement ratios of gases to CO in these months when biomass combustion was the main driving factor of gas concentration change. The study proved the ability of FTIR system in inferring the period during biomass combustion and estimating emissions of the trace gases concerning biomass combustion.

Using machine learning approach to reproduce the measured feature and understand the model-to-measurement discrepancy of atmospheric formaldehyde.

The solar absorption spectrometry in the infrared spectral region, using high-resolution Fourier transform infrared (FTIR) spectrometer, has been established as a powerful tool in atmospheric science. These observations cannot be performed continuously, for example, clouds prevent observations. On the other hand, chemical transport models give continuously data. Their results depend on the knowledge of emission inventories, the chemistry involved, and the meteorological fields, yielding to potential biases between measurements and simulations. In our study we concentrated on Formaldehyde (HCHO) and used machine learning approach to fill the gap between the observations, performed on an irregular time scale and having their measurement lacks, and model data, giving continuous data, but having potential variable biases. The proposed machine learning approach is based on the Light Gradient Boosting Machine (LightGBM) algorithm and created by using GEOS-Chem simulations, meteorological fields, emission inventory, and is referred to as the GEOS-Chem-LightGBM model. The results of established GEOS-Chem-LightGBM model have generated consistent HCHO predictions with the ground-based FTIR and satellite (OMI and TROPOMI) observations. In order to understand the GEOS-Chem model to measurement discrepancy, we have investigated the contribution of each input variable to GEOS-Chem-LightGBM model HCHO predictions through the SHapely Additive exPlanations (SHAP) approach. We found that the GEOS-Chem model underestimates the sensitivities of HCHO total column to most photochemical variables, contributing to lower amplitudes of diurnal cycle and seasonal cycle by the GEOS-Chem model. By correcting the model-to-measurement discrepancy, the sensitivities of HCHO total column to all variables by the GEOS-Chem-LightGBM became to be in good agreement with the FTIR observations. As a result, GEOS-Chem-LightGBM model has significantly improved the performance of HCHO predictions compared to the GEOS-Chem alone. The proposed GEOS-Chem-LightGBM model can be extendible to other atmospheric constituents obtained by various measurement techniques and platforms, and is expected to have wide applications.

Observations of atmospheric CO2 and CO based on in-situ and ground-based remote sensing measurements at Hefei site, China.

The characteristics of long time series of CO2 and CO surface concentrations, tropospheric and total column dry-air mole fractions (DMF) from May 2015 to December 2019 were investigated. Both CO2 and CO show different seasonality for the three datasets. The annual increasing trend of CO2 is similar for all three datasets. However, the annual decreasing trend of CO for surface concentration is high compared to the other two measurements, mainly due to the improved combustion efficiency from power generation in recent years. The correlation between the tropospheric and total atmospheric CO2 and CO is higher than that between the surface concentration and tropospheric CO2 and CO. This is because the tropospheric and total atmospheric results both have common vertical profiles for CO2 and CO respective mole fractions that were observed in troposphere. Furthermore, the enhancement ratios of CO2 to CO derived from the three datasets during the period from 2016 to 2019 were compared. The ratio of ∆CO2 to ∆CO has an obvious increase with altitude each year, which means that the combustion efficiencies obtained from the three datasets are different. All ratios for the three datasets showed a slight increasing trend in recent years, which is attributed to increased combustion efficiency due to governmental measures for energy savings and emission reductions.

Ground-based high-resolution remote sensing of sulphur hexafluoride (SF6) over Hefei, China: characterization, optical misalignment, influence, and variability.

It is a challenge to retrieve atmospheric sulphur hexafluoride (SF6) with high resolution solar spectra because it has only one single retrieval micro window and has interfered with many factors in the retrieval. Optical misalignment is one of the key factors that affect the accuracy of SF6 retrieval. In this study, we first present a long term time series of the SF6 total column over Hefei, China, between January 2017 and December 2020, retrieved by mid-infrared (MIR) solar spectra recorded by ground-based high-resolution Fourier transform infrared spectroscopy (FTIR). The sensitivities of the total column, root mean square of fitting residual (RMS), total error budgets, degrees of freedom for signal (DOFs), and vertical mixing ratio (VMR) profile with respect to different levels of optical misalignment for SF6 retrieval were assessed. The SF6 total column is sensitive to optical misalignment. In order to avoid inconsistencies in the total column due to optical misalignment, we use the true instrumental line shape (ILS) derived from regular low-pressure HBr cell measurements to retrieve the time series of SF6. The total column of SF6 over Hefei presents strong seasonal dependent features. The maximum monthly average value of (3.57 ± 0.21) × 1014 molecules*cm-2 in summer is (7.60 ± 3.50) × 1013 molecules*cm-2 (21.29 ± 9.80) % higher than the minimum monthly average value of (2.81 ± 0.14) × 1014 molecules*cm-2 in winter. The annual average SF6 total columns in 2017-2020 are (3.02 ± 0.17), (3.50 ± 0.18), (3.25 ± 0.18), and (3.08 ± 0.16) × 1014 molecules*cm-2, respectively, which are close to each other. It indicates that SF6 total column over Hefei is stable in the past four years. Our study can improve the current understanding for ground-based high-resolution remote sensing of SF6 and also contribute to generate new reliable remote sensing data in this sparsely monitored region for investigations of climate change, global warming, and air pollution.

Retrieval of vertical profiles and tropospheric CO2 columns based on high-resolution FTIR over Hefei, China.

High-resolution solar absorption spectra, observed by ground-based Fourier Transform Infrared spectroscopy (FTIR), are used to retrieve vertical profiles and partial or total column concentrations of many trace gases. In this study, we present the tropospheric CO2 columns retrieved by mid-infrared solar spectra over Hefei, China. To reduce the influence of stratospheric CO2 cross-dependencies on tropospheric CO2, an a posteriori optimization method based on a simple matrix multiplication is used to correct the tropospheric CO2 profiles and columns. The corrected tropospheric CO2 time series show an obvious annual increase and seasonal variation. The tropospheric CO2 annual increase rate is 2.71 ± 0.36 ppm yr-1, with the annual peak value in January, and CO2 decreases to a minimum in August. Further, the corrected tropospheric CO2 from GEOS-Chem simulations are in good agreement with the coincident FTIR data, with a correlation coefficient between GEOS-chem model and FTS of 0.89. The annual increase rate of XCO2 observed from near-infrared solar absorption spectra is in good agreement with the tropospheric CO2 but the annual seasonal amplitude of XCO2 is only about 1/3 of dry-air averaged mole fractions (DMF) of tropospheric CO2. This is mostly attributed to the seasonal variation of CO2 being mainly dominated by sources near the surface.

Ground-based FTIR observation of hydrogen chloride (HCl) over Hefei, China, and comparisons with GEOS-Chem model data and other ground-based FTIR stations data.

In this study, the characterization of Hydrogen Chloride (HCl) seasonal variations and inter-annual linear trend are presented for the first time over the polluted region at Hefei (117°10'E, 31°54'N), China. The time series of HCl were retrieved by the mid-infrared (MIR) solar spectra recorded by the ground-based high-resolution Fourier transform infrared spectroscopy (FTIR) between July, 2015 and April, 2019. The magnitude of HCl reaches a peak in January (2.70 ± 0.16) × 1015 molecules*cm-2 and a minimum in September (2.27 ± 0.09) × 1015 molecules*cm-2. The four-year time series of HCl total column show a negative linear trend of (-1.83 ± 0.13) %. The FTIR data are compared with GEOS-Chem data in order to evaluate the performance of the GEOS-Chem model to simulate HCl. In general, total column FTIR data and GEOS-Chem model data are in a good agreement with a correlation coefficient of 0.82. GEOS-Chem model data present a good agreement with FTIR data in seasonal variation and inter-annul trend. The maximum differences occur in January and April with mean differences of 4%-6%. We also present HCl time series observed by 6 NDACC stations (Bremen, Toronto, Rikubetsu, Izana, Reunion.maido, Lauder) in low-middle-latitude sites of the northern and southern hemispheres and Hefei stations in order to investigate the seasonal and annual trends of HCl in low-middle-latitude sites. The HCl total column at the northern hemisphere stations reached the maximum in the late winter or early spring and the minimum in the early winter or late autumn. In general, the seasonal variations of HCl over Hefei is similar to that in other northern hemisphere mid-latitude FTIR stations.

FTIR time series of stratospheric NO2 over Hefei, China, and comparisons with OMI and GEOS-Chem model data.

We present the trend and seasonal variability of stratospheric NO2 column for the first time over the polluted atmosphere at Hefei, China, retrieved using Fourier transform infrared spectroscopy (FTIR) between 2015 and 2018. The FTIR observed stratospheric NO2 columns over Hefei show a peak in June and reach a minimum in January. The mean stratospheric NO2 column concentration in June is (3.49 ± 0.25) × 1015 molecules*cm-2, and is 39.20% ± 8.95% higher than that in January with a mean value of (2.51 ± 0.21) × 1015 molecules*cm-2. We find a negative trend of (-0.34 ± 0.05) %/yr in the FTIR observations of stratospheric NO2 column. The FTIR data are compared to the satellite OMI observations to assess the new data set quality and also applied to evaluate the GEOS-Chem model simulations. We find in general the OMI observations and GEOS-Chem model results are in good agreement with the coincident FTIR data, and they all show similar seasonal cycles with strong correlation coefficients of 0.84-0.86. The annual average OMI minus FTIR difference is (1.48 ± 5.33) × 1014 molecules*cm-2 (4.82% ± 17.37%), and average GEOS-Chem minus FTIR difference is (2.36 ± 2.33) × 1014 molecules*cm -2 (7.66% ± 7.49%). Their maximum differences occur in April and May with mean differences of 12-16%. We also found negative trends in the stratospheric NO2 column over Hefei for 2015-2018 with both OMI observations (-0.91 ± 0.09%/yr) and GEOS-Chem model results (-0.31 ± 0.05%/yr), demonstrating some consistency among them.

CO and CO2 daily time series and time series of wind speed.

The global CO2 has increased rapidly because of the vast use of fossil fuels over past 40 years. CO, co-emitted with CO2, also increased during this period. To understand the CO2 and CO regional emission, it is necessary to monitor the atmospheric CO2 and CO. Ground-based high-resolution Fourier transform infrared spectroscopy (FTIR), an important technique to observe atmospheric trace gases, is used to measure the column-averaged dry-air mole fractions (DMF) of CO2 and CO [1]. The DMF of CO and CO2 are not only insensitive to vertical diffusion, but also insensitive to the variation of surface CO2 and CO concentrations. Therefore, high-resolution Fourier transform spectrometer (FTS) is used to measure atmospheric CO2 and CO and obtain daily variation of CO2 and CO in Hefei site. A weather station was installed near the FTS to record the weather data. And the wind speed is related to turbulence. So the wind speed time series and Cumulative Distribution Function (CDF) are also shown in data article.