Challenges and next steps

Integrating satellite estimates as a key tool for verification

Estimating methane emissions using satellites requires selecting diverse data sources and modelling methods, taking into consideration spatial and temporal coverage.

The preliminary findings presented in this research highlight the potential value of incorporating satellite estimations within both national and subnational emissions verification processes. This is a necessary addition to Australia’s existing emissions quantification approach, which still relies primarily on bottom-up emissions estimates for most of the country’s coal production.

This is an especially pertinent consideration for Australia’s upcoming Expert Panel on Atmospheric Measurement of Fugitive Methane Emissions, as they consider the potential role and utility of satellite emissions estimations.

Incorporating satellite-based estimates as a top-down verification tool, without attributing emissions to individual facilities, presents significant challenges. However, it is not an unprecedented approach, and has proven to be a valuable addition to Canada’s emissions inventory. Canada has progressively integrated this approach into its emission reporting inventory for its resources sector, following the principle of “continuous improvement”. 

This principle allows for incremental advancements in accuracy and effectiveness, especially when estimating fugitive emissions from an individual sector. As such, Canada has adopted a hybrid emission reporting framework, incorporating an ensemble inversion top-down method with tower-based atmospheric measurements to independently verify bottom-up facility reporting.

Despite its challenges, satellite data provides a broader, more comprehensive view of emissions, helping to identify discrepancies, monitor trends and support policy decisions aimed at reducing greenhouse gas emissions. By enhancing the overall emission measurement system, this approach complements other methods, contributing to more robust and transparent environmental oversight.

There is a large amount of freely available satellite data on open platforms such as the IMEO’s Methane Alert and Response System, Carbon Mapper and Kayrros Methane Watch. These platforms provide emissions data at the precise moment a satellite passes overhead. While useful for identifying methane hot spots and leaks, these collective datasets still lack the frequency of coverage needed for a direct comparison or quantification of annual emissions reports. 

Assessing coal mine methane emissions at a national or basin-level requires satellite coverage over large regions with frequent and repetitive overpasses. The TROPOMI satellite instrument used in this study provides unique, global daily coverage of methane concentrations. As such, it is widely used in similar studies aiming to quantify and verify global and country level methane inventories. 

The TROPOMI satellite is an area concentration mapper, measuring methane over large regions at high temporal frequency. This is in contrast to point-source imagers which detect methane at the facility level, and offer much finer spatial resolution but require targeted observations. This presents a critical trade-off between point-source imagers with higher-resolution and infrequent data in comparison to area mappers, with lower resolution with frequent coverage. 

TROPOMI’s spatial resolution is 7 km × 5.5 km, which is limited in its ability to detect methane emissions from individual coal mines, but can be used to effectively verify emissions across clustered areas, as has been shown in this study. Despite these limitations, we believe that the TROPOMI satellite is an effective tool for comparing satellite-derived methane estimates with reported emissions data, and could be readily incorporated within Australia’s fugitive emissions inventory. 

Fortunately, Australia offers some of the most favourable conditions for satellite methane monitoring, with relatively clear skies and bright, flat surfaces. Data coverage from TROPOMI can be affected by environmental conditions, such as cloud cover, mountains and surface brightness, but a recent study by Ember estimated that 91% of Australia’s annual coal production occurs in regions well-suited for methane monitoring.

Detailed location-based emission data is useful when comparing satellite measurements to reported emissions. This is because many studies, like the Kayrros analysis in this report, focus on specific areas rather than the entire state or country. 

In Australia, emissions are reported as methane at the state level, but at the mine level, they are reported as CO2-equivalent (CO2e) under the Safeguard Mechanism. CO2e includes a combination of carbon dioxide (CO2) and methane (CH4), so assumptions are needed to estimate the methane fraction from the total CO2e value. 

This challenge is the motivation to create spatial maps of reported coal mine methane emissions specifically for comparison to satellite estimations. Fortunately, starting in April 2024, the Safeguard Mechanism will release reported methane emissions as methane quantities rather than CO2e.  This type of spatial information on reported emissions will make it easier to compare satellite estimates, though significant uncertainties will likely remain.