Hubble to the rescue! The last-ditch effort to discover a Kuiper belt target for New Horizons

Will New Horizons have a mission after Pluto? Back when New Horizons launched in 2006, Alan Stern did not expect to be asking that question in the summer of 2014. The plan for New Horizons has always been to fly it reasonably close past one or two other, smaller Kuiper belt objects after the July 2015 encounter with Pluto. But that plan is in jeopardy because, despite ten years of searching with the world's largest telescopes, no reachable Kuiper belt objects have been found. "When we started, we thought there would be more faint Kuiper belt objects than there are," New Horizons principal investigator Alan Stern told me. "Our knowledge of the Kuiper belt has evolved." We've learned quite a lot about the Kuiper belt, but haven't made the discovery that we needed for New Horizons.

A search team led by New Horizons co-investigator John Spencer learned on Friday that the Space Telescope Science Institute, the organization that awards time on the Hubble Space Telescope, has granted a reprieve. Hubble will begin to search the part of the sky that New Horizons can reach -- in fact, Hubble has already begun to do so -- and if a pilot program of 40 Hubble orbits turns up at least two previously unknown Kuiper belt objects, the search team will get another 160 orbits over the next two months. "A lot of people on New Horizons are very relieved, because from the ground this was taking crazy amounts of large telescope time and not getting there fast enough. Our spacecraft is leaving the solar system; I can't slow it down," Stern said.

"This is fantastic news for New Horizons," Planetary Society CEO Bill Nye told me yesterday. "Here's hoping that in forty trips around the Earth, the Hubble Space Telescope can find a couple of ancient Kuiper belt objects to make this mission's journey even more extraordinary. Whatever we find out there, it will tell us more about the primordial solar system, where we came from, and how we all got here, a worthy investment if there ever was one."

The first raw image from the pilot program arrived on scientists' computers yesterday. Alex Parker -- an astronomer now working for the New Horizons team on the search effort -- shared the photo in a tweet:

There are a lot of stars in the photo -- that's one reason the Earth-based search has been so challenging. "Because of confusion from the very dense Milky Way star background, which cannot be perfectly subtracted due to variations in the point-spread function, search depth is a very strong function of seeing," the proposal reads. But the Hubble Space Telescope does not suffer weather problems. "We expect to reach about 1.8 magnitudes deeper with the Hubble Space Telescope than from the ground, due to Hubble’s much smaller and more stable point-spread function, which virtually eliminates the background confusion problem. This has been demonstrated by the recent discovery of very faint Pluto satellites, with minimal interference from the Milky Way background, by members of this team."

Hubble to the Rescue

New Horizons team members have spent years searching for Kuiper belt targets for New Horizons. They have discovered 52 faint objects, but none that New Horizons can get to. The closest-found object, 2011 HZ102, would require New Horizons to change its speed by 210 meters per second. It has only enough fuel to achieve 130 meters per second.

The mission is running out of time to discover a targetable object. To make the most of extremely limited fuel, the spacecraft must steer toward the post-Pluto target soon after the Pluto flyby. In order to navigate accurately enough to fly close by an object, we have to know its orbit. To know its orbit, we need to have two images of it taken close to a year apart. So we need to discover the object this summer. Only two months remain for the search to succeed. Based on what we now know about the Kuiper belt, the continuing search with such fabulous telescopes as Subaru, Magellan, and Keck -- on which the mission has been awarded 84 hours of search time in 2014 -- have only a 38% chance of finding at least one object.

Only the Hubble Space Telescope can save New Horizons' Kuiper belt mission now. The search team has calculated that using Hubble's Wide Field Planetary Camera 3 for the search would give the team a 94% chance of discovering at least one targetable object. The summer of 2014 is actually the best possible moment to use Hubble for the search, because the region of space that contains objects that New Horizons could reach is at its most compact this year, and the objects will be at opposition (and therefore brightest) in July. Even so, it would take close to 200 precious Hubble orbits. A typical Hubble year includes 3400 orbits. That's a lot of Hubble time.

The search team seems to have made their proposal for a lot of Hubble time more palatable by proposing a pilot search. Space Telescope Science Institute director Matt Mountain awarded 40 orbits of his "director's discretionary time" on Hubble to the New Horizons team. Those 40 orbits will supply two images each of 20 fields on the sky. The pilot program, which has already begun, is designed to check whether the search team's predictions for the structure of the Kuiper belt are correct, implying that using another 160 orbits of Hubble time will very likely yield a targetable object for New Horizons. "The more low-inclination objects we find in the pilot program....the more likely we are to find an accessible object in the full search....if we exactly reach our trigger threshold (2 cold Kuiper belt objects discovered in the pilot), we have an 85% probability of finding at least one accessible object in the full search. We estimate a 78% chance of reaching or exceeding this threshold in the pilot survey." They will find out within a couple of weeks whether the pilot search will pay out.

How were they able to get to work so fast after the acceptance of the proposal? Preparation, Alan Stern told me. Ordinarily, teams that propose to Hubble wait until after they have heard if their proposal has been accepted to do their "Phase 2" studies -- this is where they develop the specific technical instructions for the telescope. But with time being so limited, John Spencer and his team went ahead and performed the Phase 2 work in the hopes that their proposal would be accepted, taking the risk that all the work might be for nothing. For his part, Matt Mountain had instructed telescope schedulers to prepare two plans for Hubble in June, one including the New Horizons team's plans, and one without them, while the peer review of the proposal proceeded. Once the proposal was accepted, the New Horizons target search orbits went straight into the queue, and the team is already analyzing data.

Stern told me that they are now preparing the Phase 2 studies for the rest of the Hubble orbits. They won't wait for the completion of the 40 pilot orbits; as soon as they find two objects in the pilot program, they'll forward the discoveries to the Space Telescope Science Institute, and request the remaining 160 search orbits. Here is how those 160 orbits will appear on the sky, compared to a probability distribution of where they can hope to spot targetable objects:

What if the Hubble search fails?

The 94% chance of success is much greater than the 38% chance afforded by the continuing search with ground-based telescopes, but it's not a certainty. So the search team's proposal included some information on the science that will be enabled by the Hubble search even if it winds up not locating a target for a New Horizons flyby. The search team estimates that they'll discover 30 objects. Of those, about half will have useful orbit constraints (meaning that they can be followed up later to establish their orbital paths). Most of those will be too far away for New Horizons to fly past, but nearly all of them will be close enough that New Horizons can perform observations of them that are higher-resolution than Hubble can.

New Horizons can search for binary companions and study them at higher phase angles. And the Hubble discoveries will help outer solar system scientists understand the small end of the size distribution of Kuiper belt objects than they currently do. Why do we care? The fact that there are fewer small Kuiper belt objects than we once thought is an important clue to how the solar system formed in the first place. Here are some relevant passages from the search team's proposal:

The luminosity function (LF) of the Kuiper belt acts as a proxy to the size/frequency distribution (SFD), a strong signature of the collision and accretion physics that are at play in the outer solar system and in other circumstellar disks, providing key information that cannot be obtained for extrasolar disks (Petit et al., 2008). The existence of a break in slope in the size distribution, resulting in a relative paucity of objects smaller than 100 km, is now well established (Fuentes and Holman, 2008; Fraser and Kavelaars, 2008), and may be due to collisional grinding (eg. O’Brien and Greenberg, 2003; Pan and Sari 2005) in which case it traces the strength properties of the material, or may be primordial (Campo Bagatin and Benavidez, 2012 and Morbidelli et al., 2009), reflecting fundamental accretionary processes. Our program will also help distinguish between these alternatives by means of the binary fraction, as described below.

The “cold classical” component of the Kuiper belt may be the only part of the outer solar system that has remained largely intact since the formation of its member objects. As such, the LF, as a proxy for the SFD, provides a unique insight into the process of collisional grinding or accretion in this region of the solar system....

In addition, the proposed observations will be targeting the Cold Classical region of the Kuiper Belt, which has a known high binary fraction (at least 30%, at HST resolution) for large KBOs (Noll et al. 2008). Nesvorny et al. (2011) predict that if the observed break in slope in the KBO size frequency distribution, is due to collisional grinding of smaller objects, there should be relatively fewer binaries for KBOs fainter than this. By obtaining a large sample of high-resolution images of fainter KBOs, the proposed dataset can therefore address the fundamental question of whether the break in slope is a primordial feature of the Kuiper Belt, providing insights into accretional physics, or a result of later collisional evolution.

New Horizons has already had some opportunities to study objects at relatively close range (close, that is, for the outer solar system), but has not taken advantage of them, Stern says. "There was a Trojan in 2013 and a distant KBO early next year that I opted out of, in both cases because they were distracting from Pluto flyby preparations. But after Pluto we should be able to get several where you can get various kinds of information that you can't get from Earth."

All of this science is interesting, but Stern admitted that, on its own, it wasn't enough to justify the Hubble time. "Giving away 200 orbits of Hubble time to do mission support is a big deal, and we wanted the time allocation committee to know that, if we came up with nothing, there was also some science that would come out of it." One other way that the New Horizons team sought to make their request for Hubble time less onerous is to ask for no funds from the Space Telescope Science Institute in their proposal. "New Horizons has the budget; we don't need Hubble money. Not that it wouldn't have helped; we can always use resources for something that was unplanned, like a Hubble search. But I didn't want it to cloud the issue. I just wanted to get the U.S. to the Kuiper Belt." The New Horizons mission currently spends a few hundred thousand dollars a year supporting scientists to perform the Kuiper belt target search; it doesn't matter, he said, whether the images are coming from giant telescopes on Earth or in Earth orbit.

Stern seems sanguine about the Hubble search. He told me a story to explain why: "The first time I talked about satellites at Pluto was about the time we started studying a Pluto mission. I gave a talk at AAS in 1989 or 1990. And we tried, from the ground, to find satellites at Pluto for ten years. We had time on big telescopes, and we never found them. We proposed to Hubble three times. The third time we got turned down, we were pretty dejected. The reason we got Hubble time was because one of the instruments -- STIS -- failed; it had a short circuit. And when it was dead, they looked around for what proposals [for other instruments] had almost made the grade. We got a call that was kind of Christmas in October: 'Hey, we're gonna resurrect your proposal.' And after all those years -- a decade of looking from the ground with the best tools available, and never finding anything -- Hubble found Nix and Hydra in two orbits. Done. Literally 8 minutes of observing time."

So they have two weeks to discover at least two new Kuiper belt objects in Hubble's search field. Keep your fingers crossed for the New Horizons mission!