This work stands apart: there's almost nothing nearby, with the nearest neighbor at a distance of 0.588. Adjacent studies in the archive concern muon tomography, underground cosmic-ray counters, and the response of the muon flux to thunderstorms and stratospheric weather. All of them use muons as an atmospheric probe in one way or another, but none have taken the idea as far as assimilating muon data into a full numerical weather prediction system. This paper makes exactly that leap: it embeds the muon flux into an ensemble Kalman filter pipeline and tests the effect on a realistic tropical cyclone scenario.
The authors designed an idealized observing system simulation experiment (OSSE): a reference weather simulation using the WRF model reproduces Tropical Cyclone Freddie, and synthetic muon flux observations are generated from it. The muon flux is computed with the MCEq program, modified to read density profiles directly from the atmospheric model. Observations are assimilated with an ensemble Kalman filter with 50 members and localization radius of 640 km. Different detector exposures (from $10^3$ to $10^6$ m²·s) and assimilation of point surface pressure are compared.
Assimilating the muon flux reduces the root-mean-square error of surface pressure by up to 10% or more compared to a free forecast, outperforming the effect of point pressure measurements. Improvements for wind and potential temperature are more modest but statistically noticeable; for humidity, the effect is small. The key difference from point pressure: muons sense density over a large volume around the detector, so correlations with pressure away from the measurement point are stronger. A detector with an effective area of 0.27 m² with hourly accumulation already provides an advantage over point pressure.
This work differs from its neighbors by being the first to demonstrate the practical benefit of the muon flux specifically in the context of data assimilation for numerical weather prediction, not just by noting correlations. Neighboring studies showed that muons respond to thunderstorms and typhoons, but did not quantify the forecast improvement. Here, it is shown that even a modest detector yields a gain comparable to traditional measurements, and this gain is partly unique due to the volumetric nature of the probing.
A natural extension is to use not only the total integrated flux but also its angular distribution: muon anisotropy carries information about the direction of atmospheric inhomogeneities. Another direction is optimizing localization and assimilation frequency for specific meteorological tasks. It is also promising to deploy networks of inexpensive scintillator detectors in regions where traditional observations are sparse, such as over oceans or in polar areas.
The result could influence the design of meteorological observation networks, adding cheap muon counters to them. Astrophysical observatories such as IceCube or Pierre Auger could become incidental suppliers of data for weather models.
Test the approach on real data from an operating muon detector and compare with operational forecasts. Investigate how assimilating directional muon flux improves the reconstruction of the three-dimensional atmospheric structure.
For meteorology, the unresolved problem of the lack of direct measurements of air density in the free atmosphere, especially over oceans, remains. This work shows that the muon flux can serve as a cheap and passive probe, partially filling this gap.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The authors built an idealized observing system simulation experiment (OSSE) for tropical cyclone Freddy: a WRF reference simulation, synthetic muon flux observations through MCEq, and assimilation with an ensemble Kalman filter with 50 members and localization. They compared different detector exposures and point pressure assimilation. Assimilating the muon flux reduced surface pressure error by 10% or more, outperforming point pressure; for wind and temperature the improvements were more modest, and for humidity they were small. A detector with an area of 0.27 m² with hourly accumulation already provides an advantage.
Where its strength is
This work is the first to bring muon flux into assimilation within a full numerical weather prediction system, demonstrating the practical value of volumetric density sensing. It shows that even a modest detector yields gains comparable to traditional measurements, and part of that gain is unique thanks to correlations at a distance.
Where it could go next
Worth trying
Why this may matter
If muon detectors become part of the observation network, they could provide cheap, continuous density sensing of the atmosphere in regions where traditional measurements are sparse, potentially improving forecasts of tropical cyclones and other hazardous phenomena.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.56, against a usual 0.64. It looks like few have drilled nearby.
Tropical Cyclone Freddie, used in the simulations, was the longest-lived tropical cyclone on record, lasting 36 days.