The old adage that “space is hard” is usually said whenever a spacecraft is lost because of a miscalculation, malfunction or fatal encounter with the harsh extremes of an off-world environment. But the villain putting a thorn in NASA’s side right now is nothing more than some seemingly innocuous Martian soil.

When the space agency’s InSight mission landed on Mars in November 2018, it carried instruments meant to create a detailed picture of the planet’s innards for the very first time. One of the probes on this robot geophysicist was the Heat Flow and Physical Properties Package, or HP3. This instrument was designed to measure how much of Mars’s internal heat is escaping outward—a key metric for learning how much geologic “life” a world ever had and how much it has left. But HP3 never managed to dig deep enough into the soil to get proper readings. After two years of noble but futile attempts to force it in, NASA officially threw in the towel last week.

“Success was not guaranteed,” says Tilman Spohn of the German Aerospace Center, who is principal investigator of InSight’s HP3 instrument. But “it’s a little hard to accept that this is the end.”

There is no doubt that by the time InSight’s missions ends, scientists will have a better idea about the state, size and composition of the Martian interior than ever before. HP3‘s failure, however, means the overall picture will be fuzzier than many were expecting. And as NASA’s focus shifts away from interplanetary geophysics to returning pristine rock samples from the surface, questions about the way terrestrial worlds work could remain unanswered for a generation.

HP3’s 40-centimeter-long pile driver, affectionately dubbed the “mole,” needed to dig just three meters into the ground to begin its scientific operations. Geologists suspected the Martian soil would be a bit like basic sand. But to be on the safe side, they tested the mole on trickier soil types, too. “We were actually quite optimistic that it would work on Mars,” Spohn says. Instead, when the scientists first tried to dig and failed, all they could ask was, Wwhy does this Goddamn thing not penetrate?” he adds. Even after co-opting the lander’s equipment-moving robotic arm to rearrange soil and push the mole down, the team remained stymied.

That something as simple as soil could prove to be such a problem may sound bizarre, but it has always been an irritant for planetary geophysicists. Similar heat probes were deployed on the moon in the early 1970s. “During the Apollo missions, when you talk to the astronauts, deploying the heat probes, having to drill in, was the single most difficult thing they always identified,” says Lauren Jozwiak, a planetary geologist at the Johns Hopkins University Applied Physics Laboratory.

Back then, the lunar soil kept locking up the drill. In the case of the mole, Mars’s soil proved oddly adhesive, preventing the device from obtaining enough friction to dig. Jozwiak’s former adviser once told her that, when planning a mission, you have options A, B, C and D that you think could happen. “And invariably, the planet will be E: none of the above,” she says.

“Mars threw us a curveball with this soil,” says Paul Byrne, a planetary scientist at North Carolina State University, who is unaffiliated with the InSight mission. Consequently, “we’re still in the dark about the amount of heat coming from Mars’s deep interior to the surface.”

That is a frustrating ignorance to dwell in. The way a planet loses heat can profoundly shape its surface, governing everything from lava-spewing, atmosphere-changing volcanism to the tectonic heaves that thrust up tall mountains and carve out deep basins. The faster a planet loses its internal heat—which comes from its initial fiery formation, as well as thermal energy from radioactive decay—the sooner the planet’s geologic activity will cease. Without data from HP3, it is hard to say whether the Martian interior is hot or cold or how quickly it has cooled over the past 4.5 billion years.

InSight’s temblor-monitoring seismometer and other instruments are working as planned, and they should keep gathering data for at least two more years. That means planetary scientists will not be entirely clueless about the hot-or-not nature of Mars. For example, scientists are attempting to use proxy measurements from the seismometer to estimate the heat flow of the planet’s uppermost geologic layers. But Spohn says that technique will yield results far less precise than the readings a fully buried mole would have provided.

Getting a precise heat-flow measurement for a place where one knows the local thickness of the crust—such as InSight’s landing zone—is an incredibly valuable constraint, says Sue Smrekar, a planetary geophysicist at NASA’s Jet Propulsion Laboratory and InSight’s deputy principal investigator. Without those data, it is hard to say how much radioactive material Mars has and whether the preponderance of those heat-producing elements is in the crust or the underlying mantle.

That knowledge would help scientists unravel several key puzzles about the region the InSight lander calls home. Elysium Planitia, where the lander touched down, has some of the youngest volcanic deposits on Mars. Cerberus Fossae, a series of deep riftlike crevasses some 1,600 kilometers east of InSight, is producing “Marsquakes” today. Some are tectonic, but others may be coming from magma moving about deep below.

Scientists are keen to know whether Mars, which has not yet been caught erupting, could be capable of modern-day volcanic pyrotechnics. If magma still exists below Cerberus Fossae (or even if the region’s depths are magma-free but still sufficiently toasty), any groundwater down there could be positively balmy—providing the sort of setting Earth’s microbial life is known to favor. These are all big ifs, however, and the existence of present-day volcanism is an open question. “Getting the heat flow at the InSight landing site, fairly close by, would have been a really important starting point,” Smrekar says.

Our grasp of the onionlike layers of Mars will also be less well constrained. All of InSight’s instruments provide idiosyncratic windows into the subsurface, yet they were “designed to work in concert,” Jozwiak says. Seismic waves provide information about the locations, structures and compositions of the crust, mantle and core. The planet’s heat flow is affected by the same variables, so it can be used to interrogate the seismometer’s data, and vice versa. Losing the mole means scientists now lack an independent way of determining the properties of the Martian underworld.

There are some silver linings. Future missions attempting similar soil-penetrating feats will certainly be savvier. And Smrekar says that new code was written to allow the robotic arm to pound the mole into its hole. The arm’s recently gained dexterity will now be put to an unexpected use: getting some dirt onto the tether that connects the lander and the seismometer. That makeshift insulation should reduce some of the thermal expansion and contraction of the tether throughout the Martian day-night cycle, an activity that generates unwanted noise.

The heat probe’s unavailing endeavor is unquestionably regrettable. Yet being ambitious and trying something novel, such as taking another planet’s internal temperature, is what space exploration is about. “Sometimes you fail,” Smrekar says. But “if we only did things that we knew we could accomplish, it would be boring.”