How to Fuse Moonquake and Marsquake Data With Local Echo Maps
The Problem
A lunar analog team mapping a Hawaiian lava tube at Mauna Loa had a local passive acoustic quilt showing tube geometry out to roughly 80 meters from the entrance, but needed to extrapolate to a kilometer-scale Mare Tranquillitatis target. Active seismic surveys in the analog environment took weeks to plan and produced results that did not clearly transfer to the lunar regime. Meanwhile, the Apollo Passive Seismic Experiment catalog sat available for free download, containing ~13,000 lunar events across 8 years that carry precisely the wavelength information needed to extend a local quilt to kilometer scales. The team had no workflow to fuse the two.
This is the gap between long-baseline planetary seismology and local cave-scale passive acoustic mapping. The two domains sample completely different wavelengths, instruments, and reference frames. Four moonquake classes provide distinct source signatures, and InSight observed over 1,300 marsquakes with travel paths that penetrate deeper than pre-mission models predicted. Without a fusion workflow, analog teams cannot use either catalog as a constraint on their local maps, so they reinvent long-wavelength structure using active surveys that waste mission budget.
The problem is not data availability. Horizon Magazine's overview of moonquakes and marsquakes documents how P-wave and S-wave arrivals constrain interior structure. The problem is alignment. An InSight arrival is timestamped in UTC, referenced to the lander, and expressed as a teleseismic waveform. A local EchoQuilt patch is timestamped in LMST, referenced to a rover or EVA sensor, and expressed as a microseismic correlation field. Fusion requires a map layer that speaks both.
A complementary technique addresses the surface-noise problem that often confounds fusion. Our wind-noise separation work shows how to disambiguate surface-driven acoustic energy from subsurface seismic energy, and the separation step is essentially a precondition for credible fusion outputs. Without it, surface-derived noise gets folded into the cross-correlation against catalog arrivals and produces spurious velocity contrasts that the fusion layer cannot easily reject downstream.
The reference-frame mismatch compounds at every step. Apollo PSE catalog entries use a coordinate system rooted in the Apollo 12, 14, 15, and 16 station array, with arrival times expressed against an early lunar UTC convention that has been reanalyzed multiple times since 1977. InSight catalog entries use the SEIS-frame coordinate system, with magnitudes expressed in a Mars-specific scale that does not map directly to lunar conventions. A local quilt patch carries its own rover-relative coordinate system, with positions accurate to centimeters in that frame but uncertain to meters in any planet-centered frame. Without an explicit transform layer, fusing these reference frames produces silent shifts that propagate into the geometry conclusions. Analog teams that try to do the alignment manually almost always introduce a constant offset somewhere in the chain, which then shows up as a systematic bias in the final map that nobody can trace.
The Solution: A Stitched Seismic Fusion Layer
EchoQuilt's seismic fusion layer treats the Apollo PSE catalog, the InSight marsquake catalog, and any federated lunar/Martian seismic stream as a parallel quilt that sits above the local echo map. Each cataloged event becomes a patch with metadata including origin time, estimated location, magnitude, and wavelength regime. The fusion layer projects ray paths from each event into the local geometry and highlights where those paths intersect the current echo-map patches. Where they overlap, operators see a combined confidence view showing both scales of evidence.
The stitching is strongest at the boundaries. The local quilt carries high-resolution geometry out to tens of meters with near-real-time updates; the seismic catalog carries long-wavelength structural constraints out to kilometers with decades of statistical weight. Fusion preserves both. Where Apollo-era PSE arrivals suggest a subsurface velocity contrast at the edge of the local map, EchoQuilt flags that contrast as a boundary prior, which nudges the echo-cross-correlation algorithm to look harder at that zone. The result is a quilt that extends beyond the direct passive-imaging range using catalog priors rather than fresh active surveys.
The approach draws directly from methods that have worked on Apollo-era data. The re-evaluated Apollo 17 Lunar Seismic Profiling Experiment showed how a small geophone array can resolve three velocity layers to 5 km depth when the traveltime analysis is done carefully. EchoQuilt's fusion layer treats each moonquake or marsquake as a natural-source event in an implicit array geometry formed by the rover-carried sensors over a multi-sol campaign. Cross-correlating catalog arrivals against local passive recordings reveals velocity structure that the local quilt alone cannot reach.
Recent InSight work has demonstrated how tightly seismic and acoustic data can couple. The combined seismic+acoustic analysis of newly formed Mars craters used InSight data alongside orbital imagery to triangulate impact locations with sub-km accuracy. That same triangulation template works for skylight identification inside a lava tube: an event with both an orbital optical signature and a characteristic seismic fingerprint can be pinned to a specific void in the local quilt, tightening its geometry.
These fusion mechanics scale into advanced seismic fusion for mission-scale campaigns, where the same per-arrival weighting logic operates across larger event catalogs and longer time horizons than a single analog campaign supports.
Advanced Tactics
Weight catalog priors by event class. Apollo's shallow moonquakes produce different path effects than the deep moonquakes or meteoroid impacts in the same catalog. EchoQuilt's fusion layer lets you filter the prior set to only the classes that match your expected geometry regime, which avoids pulling unrelated deep-source signals into a shallow cave map.
Use the fusion confidence field as a targeting tool. Where catalog priors strongly agree with local echo cross-correlation, the quilt geometry is locked and needs no further recording. Where they disagree by more than the combined uncertainty, the zone is a target for additional passive acoustic patches in the next supervisory command window. This lets mission planners triage recording time toward the map regions that will improve most, rather than tiling uniformly across a tube.
For teams running multi-agency campaigns, adopt a shared seismic metadata schema up front. NASA, JAXA, and ESA catalogs use slightly different time standards and coordinate frames. EchoQuilt ships with importers for the Apollo PSE archive, the InSight SEIS bundles, and SELENE/Kaguya-derived products; adding a new catalog takes a schema definition and a reference-frame transform, not a codebase fork. This federation approach is what lets mission planning tools stay useful as new catalogs come online from Artemis-era landers.
A cross-domain analogue extends the fusion concept beyond planetary work. Our seismic overlay rescue work shows how the same fusion principle applies to terrestrial incident response, where rescue teams overlay structural seismic monitoring against acoustic mapping to triangulate void location and stability inside collapsed mines. The lessons from terrestrial rescue applications often surface fusion-engine bugs that planetary deployments would not encounter for years, and the cross-domain feedback loop has been valuable for both the planetary and rescue applications.
Cross-validate fusion outputs against any independent geophysical constraint available for the target site. For a lunar tube candidate, that means checking the fused velocity model against GRAIL gravity gradients in the same area, even when the GRAIL resolution is coarser than the local quilt. Where GRAIL identifies a low-density anomaly that overlaps the fused velocity contrast, both methods agree, and the fusion confidence rises. Where they disagree, the disagreement itself is informative — it usually means either the fusion is overweighting a single arrival or the GRAIL signal is being blurred by topography. EchoQuilt's fusion layer surfaces this kind of cross-method check as an optional overlay rather than a hard constraint, because mission scientists are the right audience to interpret disagreements rather than have an algorithm silently smooth them away.
When fusion sources span Apollo-era data and modern InSight-class instruments, weight the priors by instrument SNR rather than treating all catalog entries equally. A 1972 Apollo PSE arrival has known noise characteristics that differ substantially from a 2019 InSight SEIS arrival; folding both into the same fusion without weighting biases the result toward the noisier instrument. The fusion layer accepts an SNR vector per catalog and applies it during the cross-correlation step, which improves resolution in the regions where both catalogs contribute and falls back gracefully where only one catalog has coverage.
Ready to Fuse Catalog Priors With Your Local Quilt?
Planetary analog researchers and NIAC flight concept teams gain most when a local passive acoustic map sits inside a larger seismic context rather than in isolation. EchoQuilt's fusion layer is built for precisely that use case, with native importers for Apollo PSE and InSight catalogs, SELENE/Kaguya-derived products, and a roadmap for Artemis III seismometer streams and Farside Seismic Suite arrivals. Each pilot ships with a per-arrival SNR weighting vector tuned to differentiate 1972 Apollo PSE noise characteristics from 2019 InSight SEIS arrivals, a GRAIL gravity gradient cross-validation overlay sized to candidate lunar tube targets, and a moonquake event-class filter pre-configured for shallow versus deep moonquake signal separation. Pilot teams shape the per-event-class threshold defaults and the cross-method disagreement reporting format that the 2027 reference release will adopt.
Priority goes to NIAC PIs targeting Marius Hills or Mare Tranquillitatis pit concepts, MatISSE proposers evaluating Apollo-station reoccupation strategies, and ESA PANGAEA campaign coordinators integrating cave seismometer streams with surface SEIS-class instrument heritage. Join the Waitlist for Planetary Analog Researchers to test fusion against your current campaign datasets and help us scope the next catalog integrations.