Ship emissions are important contributor to air pollution and the climate through interacting with clouds. They are the dominant anthropogenic source over the oceans. However, their magnitudes still have large uncertainty. Here we develop a deep learning model to detect ship-emitted NO2 plumes at the pixel level in TROPOMI tropospheric NO2 data. The trained model performs well and, when applied to a year of data, it finds major shipping routes, but misses several other routes. We show that high cloudiness in these shipping routes is the culprit because clouds block signals from reach the sensor. Indeed, detected shipping routes in this study complements shipping routes detected using ship-tracks that is only available in cloudy regions. Our method can find application in several areas such as improving ship emission estimates and compliance verifications. Our method will benefit from improved tropospheric NO2 retrievals since their quality is critical for plume detection.

The Satellite Coastal and Oceanic Atmospheric Pollution Experiment (SCOAPE) cruise in the Gulf of Mexico (GOM) was conducted in May 2019 by NASA and the Bureau of Ocean Energy Management to determine the feasibility of using satellite data to measure air quality (AQ) in a region of concentrated oil and natural gas (ONG) operations. SCOAPE featured nitrogen dioxide (NO2) instrumentation (Pandora, Teledyne API analyzer) at Cocodrie, LA (29.26°, -90.66°), and on the Research Vessel Point Sur operating off the Louisiana coast with measurements of ozone, carbon monoxide (CO) and volatile organic compounds (VOC). The findings: (1) both satellite and Pandora NO2 observations revealed two AQ regimes over the GOM, the first influenced by tropical air in 10-14 May, the second influenced by flow from urban areas on 15-17 May; (2) Comparisons of OMI v4 and TROPOMI v1.3 TC (total column) NO2 data with all Pandora NO2 column observations on the Point Sur averaged 13% agreement with the largest difference during 15-17 May (~20%). At Cocodrie, LA, at the same time, the satellite-Pandora agreement was ~5%. (3) Three new-model Pandora instruments displayed a TC NO2 precision of 0.01 Dobson Units (~5%); (4) Regions of smaller and older operations displayed high methane (CH4) readings, presumably from leakage; VOC were also detected at high concentrations. Given an absence of regular AQ data in and near the GOM, SCOAPE data constitute a baseline against which future observations can be compared.