Detecting Landfill Methane Using Satellite Products

Australia's waste sector contributes approximately 13 million tonnes of carbon dioxide equivalent (CO₂-e) emissions annually, accounting for about 2.5% of the nation's total greenhouse gas emissions [1, 2]. The majority of these emissions are methane resulting from the anaerobic decomposition of organic waste in landfills.

Estimates suggest there are approximately 600 officially registered landfills in Australia, and up to 2,000 unregulated ones, most of which are small [3]. A 2013 report identified 1,168 operational landfills across the country. Despite the large number of landfills, around 75% of Australia's landfill waste is managed by just 38 major sites [4] (Fig.1).

Fig.1 Distribution of Australian landfills according to Digital Atlas of Australia [5].

One way to screen for landfill methane is to use Sentinel-5P TROPOMI. We have discussed the wind rotation, and aggregation, of TROPOMI CH4 products in other blog posts, but will briefly review here. Published by Maasakkers et al. [6], the process has 3 stages. Firstly, TROPOMI CH4 products are spatially binned (re-gridded). Secondly, each product is rotated to true north, around a fixed point, based on historical wind data. Lastly, the stack of TROPOMI products are reduced into a single, wind-rotated mean product over the target period (Fig.2).

Fig.2. TROPOMI CH4 over a landfill in Buenos Aires, Argentina. Left: mean 2018–2019 TROPOMI CH4. Middle: example single TROPOMI CH4 product. Right: wind-rotated 2018–2019  TROPOMI CH4 [6].

To identify an approximate origin, this process can be repeated over a grid of rotation centers. The point of rotation, which corresponds to the best plume enhancement, corresponds to the suggested origin [6] (Fig.3).

Fig.3. Wind-rotated 2018-2019 TROPOMI CH4 over a grid of rotation centers. The central image indicates the best plume enhancement, and thus the suggested origin [6].

GeoSynergy has successfully utilized this method to identify large methane emissions from Australian landfills (and coal mines) (Fig.4/5). 

Fig.4. Wind-rotated TROPOMI CH4 for a landfill in Queensland.

Fig.5. Wind-rotated TROPOMI CH4 for a landfill in Queensland (2018).

Note that this process can be automated, allowing for large scale screening of sites. The limiting factor, in the screening process, is the identification and evaluation of plume enhancements. However, this process can be automated by training a CNN, on a large corpus of wind-rotated examples, to segment and “score” plume quality. The score may be based on plume shape statistics, such as (eccentricity, area, circularity, length along the major axis, etc), or it can be outputted directly, if the training set incorporates labels. The development of AI-mediated screening is an area of active development at GeoSynergy.

Some example plume segmentation masks are shown in Fig.6. Note the model has been trained to ignore plumes that occur outside of the AOI. TROPOMI products cover large swaths, and we are only interested in those that occur at the point of rotation.

Fig.6. Example wind-rotated TROPOMI CH4 (rows 1 and 3) and AI segmentation masks (rows 2 and 4). These examples derive from analyses of Australian coal mines, but the concept directly transfers to landflls.

Below we discuss the emergence of new methane-sensing products for landfill assessment. However, TROPOMI is open source, offers daily global coverage, and has a product back-catalogue dating back to 2018 [7]. As such, innovative modalities, which leverage this product, are likely to remain relevant, even as new methane-sensing satellite products emerge. 

Talking of emerging satellite products, four are particularly relevant to landfill evaluation. Firstly, ISS EMIT can detect emissions as low as 50–100 kg CH₄/hour at 60 m resolution [8]. However, it is limited by its limited coverage and revisit times (3–16 days) [8] (Fig.7). 

Fig.7. ISS EMIT products over international landfills (locations and quantification in yellow) [8,9].

Secondly, Tanager-1 combines 30m resolution with a detection threshold of ~50 kg CH₄/hour [10]. It is open source and focuses on point sources, although coverage will initially target O&G [10] (Fig.8). 

Fig.8. Tanager-1 products over Australian landfills. a: Tanager-1, Badgerys Creek, NSW. 13 Dec 2024, 503+/-266 kg/h CH4. b: ISS EMIT, Badgerys Creek, NSW. 25 Mar 25, 505+/-257 kg/h CH4. c: ISS EMIT, Badgerys Creek, NSW. 15 Sept 2023, Not yet quantified. d: Tanager-1, Sunshine Coast, QLD. 20 Feb 2025, 956 +/- 208 kg/h CH4. [9] 

Thirdly, GHGSat is a commercial, tasked satellite with 25m resolution, which can detect emissions as low as 100 kg CH₄/hour [11]. It has been used extensively for evaluating landfill emissions [12], although its reliance on tasking make it expensive, and infrequent revisits (days to weeks) limit scalability (Fig.9).  

Fig.9. GHGSat product for a landfill in Punjab, Pakistasn, 24th May 2021 [12].

Lastly, MethaneSAT is an emerging, regional methane-sensing product, with exceptionally high sensitivity (~20 kg CH₄/hour), and near-daily global coverage [12]. It is currently staging, and will initially prioritize O&G targets. At the time of writing, landfill examples are not available, so we present an O&G example below (Fig.10).

Fig.10. Example of a MethaneSat product over the Permian Basin, demonstrating its capacity for regional methane evaluations [12].

In conclusion, effective evaluation of landfill emissions will incorporate incumbent satellite products, as well as emerging products. Regional products will complement the high resolution products that can detect instantaneous emissions. AI modalities can enable rapid screening for plumes within bulk products.

References

[1] Department of Climate Change, Energy, the Environment and Water (2024) Recovering organic waste. https://www.dcceew.gov.au/environment/protection/waste/food-waste/recovering-organic-waste 

[2] Climate Council (2024) Dangerously Overlooked: Why we need to talk about methane. https://www.climatecouncil.org.au/wp-content/uploads/2024/07/Dangerously-overlooked-why-we-need-to-talk-about-methane-report.pdf

[3] The University of Queensland (2017) Explainer: how much landfill does Australia have? https://www.eait.uq.edu.au/article/2017/06/explainer-how-much-landfill-does-australia-have

[4] Department of Climate Change, Energy, the Environment and Water (2013) National Waste Report 2013 – Waste Infrastructure. https://www.dcceew.gov.au/environment/protection/waste/publications/national-waste-reports/2013/infrastructure

[5] Digital Atlas of Australia. https://digital.atlas.gov.au/

[6] Maasakkers, J.D., Varon, D.J., Elfarsdóttir, A., McKeever, J., Jervis, D., Mahapatra, G., Pandey,S., Lorente, A., Borsdorff, T., Foorthuis, L.R. and Schuit, B.J., 2022. Using satellites to uncover large methane emissions from landfills. Science advances, 8(31), p.eabn9683.

[7]. Hu, H. et al. [2018] ‘Toward global mapping of methane with TROPOMI: First results and intersatellite comparison to GOSAT’, Atmospheric Measurement Techniques, 11[10], pp. 5501–5517.

[8] Thorpe, A.K. et al. [2023] ‘EMIT: New capabilities for methane detection from the International Space Station’, Geophysical Research Letters, 50[4], p. e2023GL103713.  

[9] CarbonMapper open data portal. https://data.carbonmapper.org/#1.33/30.8/50.5

[10]. Carbon Mapper [2023] Tanager-1 Mission Overview: Technical Specifications, Tucson: Carbon Mapper Coalition. 

[11]. GHGSat [2023] GHGSat-D: Advanced Methane Monitoring Satellite, Montreal: GHGSat Inc. 

[12] GHGSat applied to Landfill Gas. https://www.ghgsat.com/en/case-studies/landfill-gas/

[13]. MethaneSAT LLC [2023] MethaneSAT Technical Specifications: Instrument and Mission Design, Boulder: MethaneSAT.