By This Hour Science Desk
A NASA-supported group of high school interns has built computational models aimed at a problem that can carry outsized consequences for a lunar landing: determining how much propellant remains in a spacecraft tank when the usual cues of gravity no longer apply. The students’ work centered on a proposed acoustic technique, Modal Propellant Gauging, and on Intuitive Machines’ planned IM-3 mission, which NASA says is scheduled to send the Trinity lander to the Moon’s Reiner Gamma region in 2027.
The effort does not amount to a flight validation of the technology or an operational prediction for the mission. It is a student analysis built from public information, models and simulations. But it addresses a practical question in lunar mission planning: whether a tank can be measured reliably during critical engine burns, and how much time may be available for that measurement. NASA said the team’s results indicated that the method could work during lunar-orbit insertion and powered descent, with possible use during descent-orbit insertion in some scenarios.
The project was carried out during the summer of 2026 by the Mission Architecture Tool for Research, Integration, and eXploration, or MATRIX, group in NASA’s STEM Enhancement in Earth Science internship program. The program, based at the University of Texas Center for Space Research, puts high school students into small teams focused on defined research problems with support from NASA, academic and industry mentors. In this case, the assignment joined an educational program to an active, though still planned, commercial lunar mission.
Fuel behaves differently when gravity no longer settles it
On Earth, the amount of liquid in a vessel can often be inferred from where the liquid settles. In microgravity, propellant can adhere to tank walls rather than collect at a predictable bottom. That complicates the task of knowing how much fuel remains, particularly when spacecraft maneuvers depend on accurate accounting of the propellant available.
Modal Propellant Gauging, abbreviated MPG, is intended to approach the problem through a tank’s acoustic behavior. NASA describes the technique as using the tank’s response to sound to estimate propellant mass. The underlying premise is that the contents of a tank affect the way it responds acoustically. Rather than treating fuel measurement as a simple reading from a settled liquid surface, the method seeks information from the tank-and-propellant system itself.
NASA says MPG has been used in several spaceflight experiments and is scheduled for a demonstration on IM-3. That planned demonstration is central to the significance of the students’ analysis, but it also marks a boundary on what can be concluded. A model of potential measurement periods is different from evidence that the system will perform as expected in flight, and a scheduled demonstration is not the same as a completed one.
The questions examined by the MATRIX team were therefore operational in character. It sought to estimate when the approach might obtain dependable readings during mission events, while also considering the fuel behavior that can interfere with a clean signal. The work was connected to wider lunar-exploration needs because spacecraft that undertake extended operations, transfer propellant or support sustained activity away from Earth need credible ways to track fuel reserves.
NASA has framed propellant management as relevant to concepts including in-space refueling, fuel depots and longer-term exploration beyond a single landing. That does not mean the students’ project establishes those capabilities. It means the project focused on one technical problem that would matter to them: measuring liquid propellant when microgravity makes conventional assumptions less useful.
Public records stood in for an unavailable IM-3 trajectory
A notable constraint shaped the work from the outset. NASA said a complete trajectory for IM-3 was not publicly available. Rather than claiming access to nonpublic planning data, the students constructed a trajectory-reconstruction algorithm using published information from Intuitive Machines’ earlier IM-1 and IM-2 missions.
The approach involved comparing a reconstructed path for the IM-2 spacecraft with a version of its motion modeled without engine thrust. The difference was used to identify burns and estimate the associated changes in velocity. That reconstruction gave the team a basis for generating a prospective framework for IM-3, not a replacement for a final mission trajectory or a statement of the vehicle’s eventual flight plan.
This distinction is consequential. Launch opportunities, trajectory design, engine activity and landing operations can change as a mission develops. Results based on earlier missions and public records are necessarily conditional on the assumptions used to translate those records into a model. The team’s work appears designed to explore sensitivity and possible operating ranges, rather than to present an authoritative set of instructions for a future lander.
To represent fluid behavior inside the tanks, the students used computational-fluid-dynamics simulations to estimate propellant sloshing. They also ran Monte Carlo simulations to explore a range of possible propellant states. The two techniques serve different purposes within the reported analysis: fluid-dynamics modeling addresses the movement of the propellant, while Monte Carlo methods examine outcomes across varied possible inputs or conditions.
Such modeling cannot eliminate uncertainty inherent in a future mission. It can, however, expose which assumptions matter most and indicate where a measurement method may be more or less robust. In the team’s analysis, spacecraft starting mass had the largest effect on the duration of the proposed gauging windows. That finding directs attention to a variable that is basic to vehicle performance but can influence how long an acoustic measurement might have useful conditions.
The group also built an interactive dashboard. NASA said users could alter engine, spacecraft and orbital parameters to see the effects on propellant margins and the availability of landing zones. A tool of that kind is valuable chiefly as a way to inspect dependencies: changing one assumption can alter fuel use, timing and the feasible set of landing outcomes. The available account does not specify the full model inputs, uncertainty ranges, validation tests or dashboard outputs, so its precision and limits cannot be assessed from the reported description alone.
Critical burns emerged as the leading measurement opportunities
The students’ simulations indicated that MPG could provide reliable readings during Lunar Orbit Insertion and Powered Descent, the two primary burns identified in NASA’s account. In some cases, the analysis also found an opportunity during Descent Orbit Insertion. The wording matters. The conclusion concerns modeled windows and includes an explicit qualification for the latter maneuver; it is not a blanket finding that readings would be reliable throughout every phase of descent.
Lunar Orbit Insertion and Powered Descent are consequential moments in the mission profile described by NASA. The former concerns the transition into lunar orbit, while the latter concerns the burn leading toward the surface. A useful propellant estimate during such operations could inform an understanding of available margin. Yet the report does not say that the interns’ work would be used by mission controllers or incorporated into the lander’s final operational system.
Alongside the gauging analysis, the team identified candidate IM-3 launch opportunities for 2027 and examined the distribution of possible landing locations near Reiner Gamma. NASA says Trinity is scheduled to land in that region. The location analysis appears to have linked orbital and mission assumptions to where a landing might occur near the target area, rather than identifying a fixed final landing point.
That linkage between fuel accounting and landing geography helps explain why the project extends beyond a narrowly defined tank-measurement exercise. Propellant margin, engine choices, orbital conditions and a desired landing region can all interact in a mission model. The dashboard and simulations reportedly allowed the team to examine those interactions. But the public description does not establish how closely the modeled scenarios match the configuration that ultimately flies, or whether IM-3’s schedule and target profile will remain unchanged.
Student work moved from simulations to technical briefings
NASA said the interns briefed Intuitive Machines’ propulsion group on their methods and results. The group also presented its research at the SEES Annual Symposium and at the Wisconsin Space Grant Consortium Annual Conference. A further presentation is scheduled for the American Geophysical Union Annual Meeting in San Francisco in December 2026, NASA said.
Those presentations show that the work was shared with technical audiences, including engineers associated with the company named in the modeled mission. They do not, by themselves, establish external peer review, adoption by the company or independent confirmation of the analysis. Nor does the available material describe a published technical paper, a validation campaign or data from an IM-3 flight demonstration.
For the SEES program, the assignment illustrates the kind of training NASA says it is designed to provide: direct use of aerospace data, analytical tools and mission-related questions by students in the 11th and 12th grades. The work also connects with broader discussions of lunar infrastructure, propulsion and workforce development that NASA has placed on the agenda for its exploration community, including at a planned 2026 space-exploration symposium.
There are clear limits to the available evidence. NASA’s account supports that the interns conducted the modeling, reconstructed elements of a mission from public data and reported the stated findings. It does not provide enough detail to independently reproduce the simulations, determine their error bounds, assess the acoustic system’s flight performance or verify the planned mission timetable. The reported analysis has not been independently corroborated.
The next meaningful test of the underlying technology, on the account provided, would be the planned MPG demonstration on IM-3. Until then, the internship project should be read as a model-based examination of potential measurement opportunities, not as proof of performance during a lunar landing. Its strongest contribution may be to make assumptions visible: showing how fuel state, starting mass, trajectory choices and maneuver timing can shape the narrow intervals in which a spacecraft may be able to measure what it has left.
Reporting notes
What is confirmed: NASA says the analysis identified potential gauging windows during key burns and found starting mass most affected their duration.
Why this matters: Measuring fuel in microgravity could affect propellant-margin assessments during lunar operations.
What remains unclear: The public account does not provide full model details, flight validation or confirmation that the results will be used operationally. This report is based on one source and has not been independently corroborated.
