Sensitivity of HCHO retrievals to air mass factor (AMF) for the environmental geostationary application

Abstract

Formaldehyde (HCHO) is produced by the oxidation of non-methane volatile organic compounds (NMVOCs) by hydroxyl radical (OH) in the atmosphere. NMVOCs are precursors of tropospheric O3 and organic aerosols and play a significant role in air pollution. Therefore, an accurate estimate of NMVOCs emission is critical for air quality model simulations, but a large uncertainty still remains with the bottom-up emission inventory of NMVOCs in chemical transport models. In particular, the biogenic emission of NMVOCs is poorly quantified and is one of determining factors for O3 and aerosol simulations. Previous studies have used HCHO column abundance measurements from the instruments onboard sun-synchronized satellites such as GOME, SCIAMACHY, OMI, and GOME-2 to improve the bottom-up emission inventory of NMVOCs especially from the biogenic sources. However, the measurements onboard the polar orbit satellites are limited temporally by the frequency of the satellite overpass time, which is at best once a day at the same local time. This is further challenged by the presence of cloud. In order to overcome the limitations and monitor air quality changes in East Asia in near-real time, the Ministry of Environment in Korea plans to launch GEOstationary-Korean MultiPurpose SATellite (GEO-KOMPSAT) with Geostationary Environment Monitoring Spectrometer (GEMS) in 2018. Here, I examine the HCHO retrieval algorithm for GEMS by conducting a sensitivity test with respect to air mass factor (AMF) and attempt to improve the retrieval accuracy by providing better a priori information. AMF is used to convert satellite measured slant column to vertical column density (VCD), and it depends on interfering gases, aerosols, and clouds as well as solar zenith angle and the line-of-sight zenith angle. Monthly values of AMF are typically used for the measurements onboard the polar orbit satellites. Because GEMS measurements are more frequent than those of the polar orbit satellites, high temporal variations should be reflected in the AMF calculation. Here, I examine the sensitivity of the GEMS HCHO retrieval to the temporal variation of AMF. Since the GEMS measurements are not available yet, I use instead a synthetic calculation by integrating the retrieval algorithm with a 3-D chemical transport model (GEOS-Chem) and a radiative transfer model (VLIDORT). I first generate GEMS measured radiances using the VLIDORT with profiles of trace gases and aerosols simulated by the GEOS-Chem. I then apply the retrieval algorithm to the GEMS radiances to retrieve HCHO VCDs. For the HCHO retrieval, I use hourly and monthly AMF values, resulting in the two sets of HCHO column abundances. The comparison of the two retrievals with the true HCHO VCDs from the GEOS-Chem, allows me to understand the retrieval sensitivity to the AMF values with the different temporal variability. My analysis shows that the retrieved HCHO VCDs with the hourly AMF are significantly correlated with the true values (R2=0.99), whereas the VCDs with the monthly AMF show relatively lower correlation (R2=0.87), indicating that the use of hourly AMF in the retrieval captures the GEMS measured spatial pattern of HCHO better than that with the monthly AMF over East Asia (105E-135E, 15N-45N). The discrepancy between the two retrievals is due to not only the temporal variation of aerosol loading but also that of the chemical composition. Absorbing aerosols such as black carbon (BC) and soil dust can reduce AMF while scattering aerosols such as sulfate and nitrate increase AMF. My results have also an important implication for the low orbit satellite measurements in East Asia. For example, when I compare my calculation with AMF values used in the OMI retrieval, a range of discrepancy from -61% to 36% is found over China primarily because of the consideration of aerosols in the AMF values. A priori information with high temporal resolution is needed to reduce the existing uncertainty with the satellite measurements for chemically complex regions such as East Asia, and the need is even higher for the high temporal measurements of the geostationary satellite.