You know how if you stir chocolate syrup into cold milk instead of hot, you get streaks and lumps that never fully dissolve? The milk looks mixed from the outside, but take a sip from the wrong spot and you get a mouthful of pure syrup. Mars, it turns out, is that glass of cold milk. The planet took enormous impacts during its formation — collisions that should have delivered rocky material deep into the interior — but instead of blending smoothly into a uniform mantle, that impactor material persists as giant discrete lumps scattered throughout. The planet never got hot enough, or stayed molten long enough, to fully homogenise. The committed claim: NASA marsquake data reveals that Mars's mantle is heterogeneous, with rocky material from ancient impacts preserved as giant lumps rather than being mixed into a uniform interior. This overturns the standard assumption that rocky planet mantles are compositionally well-mixed. The category of result is observational planetary seismology — using seismic waves from marsquakes to image internal structure, the same way we use earthquakes to map Earth's interior. This is not the first use of InSight seismic data to study Mars's interior (that work has been ongoing since InSight landed in 2018), but the specific finding of persistent impactor heterogeneity in the mantle appears to be new. The ladder here is against the prior model of Mars's interior. Previous seismic analyses of InSight data established basic mantle structure — layering, core size, crustal thickness. The baseline assumption in planetary science has been that rocky planet mantles are reasonably well-mixed after billions of years of convection. This finding says: no, the mixing was incomplete, and we can detect the lumps seismically. The comparison is not against a competing method but against a competing model of planetary interiors. The "well-mixed mantle" model is the incumbent, and this data challenges it directly. The architecture is observational seismology — passive detection of natural marsquakes by InSight's SEIS (Seismic Experiment for Interior Structure) instrument, followed by inversion modelling to map velocity anomalies in the mantle that correspond to compositional heterogeneities. This is the same methodological family as terrestrial seismic tomography, adapted for a single-station setup (InSight was the only seismometer on Mars). The single-station constraint is severe — on Earth, we use networks of thousands of seismometers. Everything here depends on clever signal processing to extract 3D structural information from one measurement point. Integrity: this is real observational data from a physical instrument on another planet — about as far from simulation-grading-simulation as you can get. The marsquakes are natural events, not controlled experiments, which means the researchers work with whatever seismic sources Mars provides. The main integrity question is model-dependence: converting seismic waveforms into statements about mantle composition requires assumptions about mineral physics, temperature profiles, and wave propagation that introduce uncertainty. The finding of "lumps" is an interpretation of velocity anomalies — the data shows the anomalies; the "impactor material" explanation is the most parsimonious interpretation but not the only possible one. The milestone to track: InSight's mission has ended (the lander lost power in late 2022), so no new marsquake data is coming from this instrument. The next concrete milestone is a multi-station seismic network on Mars — even 3-4 stations would allow triangulation that transforms what we can resolve. NASA and ESA have discussed concepts, but no funded mission is on the books. Realistically, a Mars seismic network is 15-20 years away. In the meantime, the milestone is whether this heterogeneous-mantle finding survives re-analysis by independent groups using the same InSight dataset. The obvious experiment not run: seismic tomography with multiple stations, which would resolve the 3D geometry of the lumps rather than just detecting their existence. Why not? Because there's only one seismometer on Mars. This isn't a choice the researchers made — it's a hardware constraint imposed by the fact that landing instruments on Mars costs billions. The successor mission that would validate and extend this work is a Mars geophysical network, and the reason it hasn't happened is budget, not science.
