Continuous Monitoring of Osaka Reveals Methane Sources 'Vastly Underestimated' in City Inventories

Methane is, by atmospheric effect, a far more potent greenhouse gas than carbon dioxide over the near term - roughly 80 times more powerful over a 20-year window. Climate policy has focused heavily on large industrial sources: chemical plants, natural gas infrastructure, livestock operations. These are tractable targets. They are visible, registered, and in many jurisdictions subject to regulation and mandatory reporting. But a study from Osaka Metropolitan University suggests that in the dense urban fabric of a major Japanese city, the sources that go uncounted may collectively matter more than the ones being tracked.

A different kind of measurement campaign

Most emissions inventories rely on spot measurements at specific facilities, or on activity-based calculations that estimate emissions from known processes. Associate Professor Masahito Ueyama and an international team took a different approach. They installed sensors on a tall tower for continuous high-altitude readings of methane and ethane concentrations across the Osaka metropolitan area, while also conducting ground-level measurements from a bicycle-mounted sensor that could weave through streets and neighborhoods.

The key difference from standard monitoring is continuity. Rather than periodic checks at specific locations, the measurements integrated emissions across the city center over time, capturing the aggregate signal from all sources - large and small - contributing to the urban atmosphere. Ethane was measured alongside methane because it serves as a chemical tracer for fossil fuel origin: natural gas contains ethane, while biological processes produce methane without it. This ratio allows researchers to separate leakage from city gas infrastructure from biologically generated methane.

What the data showed

When Ueyama's team compared their measurements with government emissions inventories for the Osaka area, they found large discrepancies. Methane concentrations were substantially higher than the inventories predicted. The pattern of when and where emissions appeared was also informative.

Emissions were higher on weekdays than weekends, and they followed a clear day-night cycle. Both patterns point to human activity rather than natural processes. The presence of ethane confirmed that fossil fuel leakage - from city gas pipes serving commercial buildings and residences - contributed significantly to the unaccounted total. Restaurants, commercial facilities, and private homes were identified as categories of small, dispersed emitters that existing monitoring largely misses.

Biological methane sources were also underestimated. The culprits here appear to be small but widespread: sewage manholes releasing methane from decomposing organic matter, and fermentation processes associated with traditional Japanese foods such as miso, soy sauce, and various pickled products that are produced throughout the urban area.

The challenge of counting the uncountable

Large industrial emitters are registered. They have permits, they submit reports, and they can in principle be audited. The tens of thousands of restaurants, the residential gas connections, and the kilometers of aging sewer infrastructure scattered across a metropolitan area like Osaka do not submit emissions reports. They exist below the threshold of conventional monitoring systems, not because they emit nothing but because the cost and complexity of tracking them individually is prohibitive.

The integrated measurement approach the Osaka team used sidesteps this problem by measuring the cumulative atmospheric signal of all sources rather than trying to account for each one. "By clarifying the existence of methane emissions originating from city gas that had previously been overlooked, our research is expected to aid in identifying these unaccounted emission sources within urban areas," Professor Ueyama said.

The ethane-methane ratio analysis adds an additional tool. By distinguishing fossil fuel leakage from biological emissions in real time, the method can potentially track whether infrastructure repairs or policy changes targeting gas distribution networks are actually reducing leakage - something activity-based inventories cannot do with the same temporal resolution.

Policy implications and technical limitations

The study was conducted in one city. Osaka's mix of residential density, food culture, and aging infrastructure may not generalize directly to other urban environments. Cities in different countries, with different building stock ages, different gas distribution systems, and different food production patterns, would need their own campaigns to characterize local emission profiles.

Ueyama and his colleagues are calling for their technique to be expanded to other cities and used as a standard method for urban methane management. The approach - continuous integrated atmospheric measurement paired with chemical fingerprinting - requires specialized equipment and trained personnel, but the cost of that investment may compare favorably against the alternative: systematically undercounting a significant portion of urban greenhouse gas emissions.

The findings were published in Environmental Science and Technology.