This documnetiation is designed to be a technical document with relation to the example of calculating TROPOMI Observational Coverage (TOC) and TOC-reducing days by using observations and vairables included in the TROPOMI Level-2 data product.
Global observational coverage of oil and gas methane sources with TROPOMI
Satellite observations have provided important insights about methane (CH4) emissions from the oil and gas (O&G) industry, particularly by revealing previously undocumented, very large (ultra) emission events and estimating basin-level emissions. They are expected to play a growing role in global CH4 monitoring, especially with many launches of new systems. However, most satellite systems use passive remote sensing to retrieve CH4 mixing ratios, which is sensitive to sunlight, earth surface properties, and atmospheric conditions. Accordingly, the reliability of satellites for routine CH4 emissions monitoring varies across the globe. To better understand the potentials and limitations of routine monitoring of CH4 emissions, we investigated the global observational coverage of the TROPOMI instrument onboard the Sentinel-5P satellite – the only satellite system with daily global coverage. Based on the recommended measurement retrieval quality-assurance threshold of ≥ 0.5, we mapped the average percentage of days in a year with valid observations on a 0.1° x 0.1° grid from 2019 to 2021. The average annual TROPOMI observational coverage (TOC) was higher over dryland regions (maximum: 58.6%) and lower over tropical regions and high latitudes (minimum: 0%). Cloud cover and solar zenith angle were the primary impeding factors at high latitudes, while aerosol optical thickness was the primary impeding factor over dryland regions. To assess the reliability of satellites for monitoring CH4 emissions from the O&G sector, we extracted the average annual TOC over O&G infrastructure for 160 countries. Seven of the top-10 O&G-producing countries had average annual TOC < 10% (< 36 days per year). Low TOC limits the ability to routinely identify large emissions events, track their duration, and quantify emissions rates using inverse modelling. We further assessed the accuracy of the latter by identifying overlapping grid cells containing O&G infrastructure with low TOC and high O&G CH4 inventory uncertainty. Results indicate atmospheric inversion modelling may only be accurate for quantifying O&G targets in countries located in dryland or mid-latitude regions. We conclude that current passive-sensing satellites have low potential for frequent monitoring of large methane emissions from O&G sources in countries located in tropical and high latitudes (e.g., Canada, Russia, China, Brazil, Norway, Columbia, Venezuela). Therefore, other methods should be considered for routine emissions monitoring in these and similar regions.