Successful launch of the Tropospheric Monitoring Instrument (TROPOMI) onboard the Copernicus Sentinel-5 Precursor satellite to support the services and science related to air quality, climate and stratospheric ozone opens up a new possibility to provide the daily global concentrations of trace gases and aerosols with high spatio-temporal resolution.
As a part of international efforts to better calibrate and validate the TROPOMI along with the possibilities of application to the Geostationary Environment Monitoring Spectrometer (GEMS) program, the environmental monitoring mission of the Korea’s next generation of geostationary earth observing satellite program, we try to cross-validate the TROPOMI Level 2 products by utilizing algorithms prepared for the GEMS program and the data obtained from ground based network.
For an example, aerosol index, layer height, and UV aerosol index of Sentinel-5p will be compared with the product of GEMS algorithm. Aerosol information including aerosol optical depth, single scattering albedo at 443 nm, aerosol index, and layer height can be retrieved from the Level 1B data of Sentinel-5p by using GEMS algorithm. In addition, trace-gases such as nitrogen dioxide, sulfur dioxide and ozone will be retrieved on April 19th, 2018 over GEMS FOV from the Level 1B data of Sentinel-5p by using GEMS algorithm. These products will be compared qualitatively with Level 2 products of Sentinel-5p products. On April 19th, 2018, GOCI RGB and AOD showed high aerosol concentration in Korea with transportation from Eastern China. Additional retrieval results of trace-gases in terms of spatial distribution will provide interesting case study over East Asia. Algorithm difference between TROPOMI and GEMS will be analyzed too.
Another activity is validation of the Level 2 products using the ground-based network over East Asia. In-situ remote sensing network such as AERONET, SKYNET, Brewer, Pandora, MAX-DOAS, and Ozone sonde are consistently maintained over East Asia. By using these measurements, we can provide the quantitative assessments of Sentinel-5p Level 2 products.

During May - October 2018, a localized integrated observing system was built to study the emissions from New York City (NYC) and their impacts to the air quality locally and downwind as part of the Long Island Sound Tropospheric Ozone Study (LISTOS). This observing system included a network of a dozen Pandora spectrometers (10 ground-based and 2 ship-based) and high-spatial resolution observations from two NASA airborne mapping UV-VIS spectrometers—GeoCape Airborne Spectrometer (GCAS) and GEOstationary Trace gas and Aerosol Sensor Optimization (GeoTASO). The airborne spectrometers were flown in regular grid patterns during the daylight hours on 17 days spanning a wide variety of meteorological and air quality conditions in the LISTOS domain. Flight opportunities were prioritized for forecasted clear sky days and were timed to provide measurements coincident with S5P/TROPOMI–overpasses spanning from ~17:00-18:50 UTC locally. Spectra from the airborne mappers were used to retrieve NO₂ at spatial resolutions as fine as 250 x 250 m across the 7 km swath provided by the instruments’ 45° across-track field of view. Parallel flight lines spaced to ensure overlap between adjacent swaths allowed gapless NO₂ maps to be built over the NYC region over a 2-4 hour period. Pandora has been used as a validation tool for instruments like TROPOMI due to its precise and accurate measurements of NO2 columns from the surface using direct-sun DOAS. However, in areas subject to extreme NO₂ spatial heterogeneity (e.g. the NYC metropolitan area), the local NO₂ column from Pandora may differ from the NO₂ column retrieved from S5P/TROPOMI due to sub-pixel variation within the space-based observation footprint. This work explores using the airborne spectrometer retrievals as a transfer standard between Pandora and TROPOMI to assess the TROPOMI NO₂ retrieval in this complex coastal/urban environment. The airborne spectrometers observed NO₂ varying by two orders of magnitude over the LISTOS domain over multiple flight days. Preliminary comparisons show high correlation coefficients between all three instrument NO₂ retrievals (r² > 0.75). In addition to providing correlative statistics between the three instruments to help evaluate S5P/TROPOMI NO₂ products in this region, we will characterize factors, such as spatiotemporal mismatches, spatial heterogeneity, and retrieval biases, that could cause differences between the measurements.

In collaboration with NASA GSFC and LuftBlick, the US EPA and our state partners are seeking to address both air quality management and satellite validation needs via deployment of nine Pandora UV/Visible spectrometers in and around New York City. The addition of NO2 and HCHO column measurements to existing air quality sites supports state monitoring efforts by addressing the Enhanced Monitoring Requirement included in the 2015 updates to the Photochemical Assessment Monitoring Station Program. We demonstrate how the addition of vertical column measurements enhances the conventional monitoring system for air quality decision making, and present initial TROPOMI formaldehyde and NO2 total column validation results for Summer 2018 through Spring 2019. We briefly discuss the capabilities of the measurement system to address the time-of-day validation needs that will be required for geostationary orbiting spectrometers.

In October 2017, the Sentinel-5 Precursor (S5P) mission was launched, carrying the TROPOspheric Monitoring Instrument, TROPOMI, which provides a daily global coverage at a spatial resolution as high as 7 km x 3.5 km and is expected to extend the atmospheric composition record initiated with GOME/ERS-2 in 1995, bringing up significant new components to the scientific knowledge of atmospheric processes. Due to the ongoing need to understand and monitor the recovery of the ozone layer as well as the evolution of the tropospheric pollution, total ozone will remain one of the leading species of interest during this mission.
The Laboratory of Atmospheric Physics of the Aristotle University of Thessaloniki, Greece, is the co-ordinator of the TROPOMI’s Total Ozone Column Validation (VALTOZ) team and in this work we present the validation results for the TROPOMI NRTI and OFFL total ozone column (TOC) products, which are compared to ground-based quality-assured Brewer and Dobson TOC measurements deposited in the World Ozone and Ultraviolet Radiation Data Centre (WOUDC). Additional comparisons to Brewer measurements from the Canadian Network and the European Brewer Network (EUBREWNET) are performed. Furthermore, also twilight zenith-sky measurements obtained with ZSL-DOAS (Zenith Scattered Light Differential Optical Absorption Spectroscopy) instruments, that form part of the SAOZ network (Système d'Analyse par Observation Zénitale) of the Network for the Detection of Atmospheric Composition Change (NDACC), are used for the validation.
Through the comparison of the TROPOMI TOC data to the ground-based measurements from stations that are distributed globally as the background truth, the dependence of the new instrument’s measurements on latitude and various other influence parameters, such as cloud properties, solar zenith and viewing angles etc. is examined and the results are presented and commented. The overall statistical analysis of the comparison shows that the mean and the standard deviation of the percentage difference between TROPOMI and ground-based TOC is within 1.5 – 2 % and 1.6 – 2.5 %, respectively, well within the S5P mission requirements.
Finally, the inter-sensor mutual consistency is investigated, comparing (a) the OFFL TROPOMI TOC product to OMPS/Suomi-NPP and OMI/Aura processed with the GODFITv4 algorithm, and (b) the NRTI product to GOME-2/Metop-B processed with the GDP4.8 algorithm. This analysis shows a very good agreement for both products of TROPOMI to well established instruments, which assures the scientific community of the good quality of the TROPOMI TOC products.

TROPOMI XCO accuracy and daily variability evaluated at a high-latitude site in Sodankylä, Finland

Tomi Karppinen, Anu-Maija Sundström, Hannakaisa Lindqvist, Iolanda Ialongo, Rigel Kivi, Juha Hatakka, Johanna Tamminen

Sentinel-5 precursor satellite was launched in October 2017. The TROPOMI instrument on-board is the first instrument that measures CO concentrations daily and globally using SWIR wavelength region. It is producing us global information of air pollutants with an unprecedented accuracy and spatial resolution.

In this study, we concentrate on observations of carbon monoxide (CO) at high Northern latitudes. CO is emitted in the combustion processes of fossil fuels in the partial oxidation of carbon compounds. The main sources of CO are traffic, industry and wildfires. CO is an important factor of air quality as it can disrupt the transport of oxygen by blood, and contributes to the formation of ground-level ozone, another air pollutant with unhealthy effects. As its lifetime in the atmosphere is rather long, it can be used to follow the transportation patterns of pollutants from their sources.

We study the accuracy of the TROPOMI column-averaged CO (XCO) against the XCO retrieved by the Fourier Transform Spectrometer (FTS) at the Arctic Space Centre in Sodankylä, Northern Finland. The Sodankylä FTS participates in the Total Carbon Column Observing Network (TCCON) and has been operational since 2009. We evaluate co-located satellite and ground-based retrievals for year 2018.

Due to its location at 67°22' N latitude, Sodankylä TCCON has several TROPOMI overpasses during a single day. Therefore, particular attention is given to assess if TROPOMI will be able to reliably reproduce the short-term, diurnal variability shown by the TCCON retrievals. We will also compare TROPOMI XCO to in-situ CO profile measurements when available. Ground level measurements are used to check how the variability of total column measurements mirror the ground level mixing ratios.

In addition, we show a few case studies of wildfires to demonstrate the ability of TROPOMI to detect pollution sources and to follow the transport at high Northern latitudes during summer 2018. The cases include a small scale forest fires in Russia close to the Finnish border near Raja-Jooseppi in late July 2018. As an example of long distance transport we follow the CO emissions from wildfires in Canada reaching Finland in August 2018.

As part of the S5P Validation Team, we will present the results of the NIDFORVal project (S5P NItrogen Dioxide and FORmaldehyde VALidation using NDACC and complementary FTIR and UV-Vis DOAS ground-based remote sensing data), focusing on the formaldehyde (HCHO) S5P products.

Fourier Transform Infrared (FTIR) and Multi-Axis Differential Optical Absorption Spectroscopy (MAXDOAS) ground-based HCHO measurements at a few stations have been used in the past for satellite validation. In the NIDFORVal project, about 20 FTIR and 40 MAXDOAS stations are involved to provide a wide range of observation conditions including high, mid, and low latitudes, as well as remote, sub-urban, and urban polluted sites. The HCHO retrieval settings have been recently optimized and applied consistently at most of the FTIR stations part of the Network for the Detection of Atmospheric Composition Change (NDACC), creating a unique harmonized data set ensuring a consistent validation among the sites. For the MAXDOAS stations, a best-effort basis harmonized dataset is used so far, which will be improved and enlarged within another S5P Validation Team project (FRM4DOAS).

Data from 16 FTIR sites and 10 MAXDOAS stations were already used for the preliminary validation of the HCHO S5P columns (commissioning and pre-operational phase). The comparison results showed usually negative biases (TROPOMI being smaller) that were well within the accuracy requirements (40-80%). While the standard deviations were within the precision requirements for clean sites (1.2x1016 molec/cm2 for a single pixel), polluted sites showed larger deviations attributed to higher spatial/temporal variability. This preliminary validation exercise needs further investigation, which becomes now feasible with the availability of more TROPOMI measurements.

While only four to six months of data were available for the preliminary S5P HCHO validation, we will show in this presentation the validation results of one complete year of measurements, allowing the seasonal cycles to be compared at more than 30 sites around the globe. We will also be able to compare validation results where both FTIR and MAXDOAS techniques are available, such as at the polluted site Mexico City, where the validation is more tricky, or the suburban Xianghe station.

The Sentinel 5 Precursor Tropospheric Ozone Monitoring Instrument (S5P TROPOMI) instrument, launched on November 17, 2017, provides a suite of trace gas and aerosol products relevant to air quality. Key products include measurements of carbon monoxide (CO), nitrogen dioxide (NO2), aerosol index (AI), and aerosol optical depth (AOD). These products, when made available in near real time, can be valuable to air quality forecasters, particularly in situations when emissions plumes from large wildfires enhances ozone (O3) production. In many parts of the United States (U.S.), ground-level O3 formation is limited by regional emissions of nitrogen oxides (NOx; sum of NO and NO2). Fires release large amounts of trace gases, including NOx, as well as smoke aerosols. When fire emissions plumes are both optically thin enough to allow for photochemistry and contain high levels of NOx, production of O3 can be greatly increased, resulting in violations of the U.S. daily national ambient air quality standard (NAAQS) for O3. These are the cases that are most critical for air quality forecasters to accurately predict, so alerts and health warnings can be issued to the public. For example, forecasters in the Mid-Atlantic region of the U.S. are often faced with this challenge due to fire emissions plumes transported from Mexico, Canada, Alaska, and the western U.S. The issue for forecasters is identifying these plumes, since they are often not well captured by U.S. operational numerical air quality models. The TROPOMI products are expected to help with forecasting challenges for O3 daily NAAQS violations when and where smoke enhances O3 production by allowing forecasters to locate and track NOx-rich fire emissions plumes. As a part of NOAA contributions to the calibration and validation of TROPOMI products, one proposed activity is to work with its focus group of air quality forecasters in the U.S. regarding the usefulness of the TROPOMI products in daily operational air quality forecasting. Air quality products from NOAA operational geostationary and polar-orbiting satellites are currently provided to forecasters through the web-based interactive tools AerosolWatch (https://www.star.nesdis.noaa.gov/smcd/spb/aq/AerosolWatch/) and eIDEA (https://www.star.nesdis.noaa.gov/smcd/spb/aq/eidea/). The NOAA S5P TROPOMI AI, NO2, and CO are proposed to be added to these websites for use by forecasters. We will present results from summer 2018, when multiple large wildfires in Mexico, Canada, and the western U.S. impacted O3 air quality in the U.S. Our study probed whether forecasters are able to come to more definitive conclusions about issuing O3 air quality alerts if TROPOMI data are available on a day to day basis in near real time.