Several of the presentations from the past MPEG meeting were posted recently

http://mepag.jpl.nasa.gov/reports/index.html

These include Draft priorities for Mars Sample Return, Mars Science Orbiter (MSO) Science Definition (nothing new, but a nice summary), and the Mars Strategic Science Assessment Group.

I found the latter document extremely interesting. It defines the rational for MSO as the mission to follow Scout 2013. It also defined 8 distinctive classes of deposits (of which Opportunity is exploring one) with emphasis that six have been identified in just the last three years showing that "we are still in an active phase of discovery". Understanding these deposits "are critical to understanding the history of water near Mars' surface." Ideally, it would seem to me, MSL and ExoMars would each explore a different type.

Most interesting for me, was the rational for a twenty year roadmap of exploration based on the assumption that sample return #1 begins at the end of the next decade:

2013: Mars Scout aeronomy
2016: MSO - determine presence and source of key trace gases, continue detailed monitoring of climate, continue to characterize surface with Hi-Rise class camera (and not mentioned, be available as a communications relay)
2018: MSR - lander/rover
2020: MSR - orbiter
2022: Mars network. Not chosen for 2016 in part to allow lessons learned from ExoMars geophysical station to be incorporated
2024: Mid-range rover (MER sized, but with updated instruments and tighter landing ellipse) - explore new class of deposit
2026: Scout
2028: MRS #2 orbiter
2030: MSR #2 lander/rover

Speculation: if budgets don't allow for a sample return in the 2018/20 opportunity, I wonder if the sequence of other missions would simply move forward, and perhaps a second mid-range rover would be added. Also, could these mid-range rover be powered by Sterling-engine plutonium systems?

If we are to delay Mars Sample Return, then I would like to see a few more mid-size rovers
sent to check out deposits at any one of the 20 unique geologic units on Mars. Also, whenever
MSR does get a definite go-ahead, I would like to see it put up for competitive bid, by way
of a Flagship AO. I think that we should stop assuming that JPL is the only place where Mars
rovers or the MSR can be built. Phoenix was built by Lockheed-Martin, as well as Mars Odyssey.
An AO for the MSR mission would allow other capable parties, such as APL, to propose
a design. Recall that Viking, the mega-Flagship project in Mars exploration, was led by
NASA-Langley, with the Orbiter built at JPL and the Lander built by Martin Marietta.

Another Phil

I agree with all of that, Phil. In fact, I think we'll end up seeing MSR restricted to locations where we have established ground truth and know we'll get rocks that are worth the huge cost.

I also agree that we should delay MSR until we have several more landers down. You want to know that you are sampling the *one* best spot.

Phil, I like your idea of competition, but I worry that enabling MSR may take a decade of technology development. Whoever does the technology development will have an edge that will be hard to match outside. I'd like to hope that I'm wrong...

I think a decade is over-optimistic for MSR. Here we are in 2008 with a grand total of one successful rover mission to date. (OK, two vehicles, but one design.) Assuming Phoenix and MSL are roaring successes, I can conceive of a bigger more powerful rover design, better targeted with the benefit of another decade or so of HiRISE class and other advanced remote sensing, that's capable of collecting multiple samples and returning them to a central site being feasible in ten years. I can just about imagine a Mars orbiter with enough fuel capacity, and reliable enough engines, to get itself back to Earth from Mars orbit becoming feasible in 10 to 15 years' time. But I suspect getting the samples off the surface and accelerated to 5 km/s (compared with the moon at 2.4 km/s) [source - wikipedia] will take a lot more trial and error.

I know there are design concepts, I just think a remote or autonomously controlled take-off from Mars to orbit, by a vehicle that will already have had to get there from here is going to take a lot of blood sweat and tears and probably some failed attempts along the way. Flight ready hardware within a decade just looks like too big a stretch to me, even on $650m / yr.

Too pessimistic? I hope so.

For the really big missions -- Mars sample return, Europa orbiter, Titan missions, Venus long lived surface missions -- there is a lot of technology development that needs to be done. I personally think that Titan is more interesting than Europa, but the next Flagship mission is likely to go to Europa because NASA has invested well over a decade in the technology development and mission definition. That's why the recent assessment of Flagship concepts rated both sites as excellent for science, but Europa medium risk and Titan high risk.

JWST and MSL's cost overruns are likely, in my opinion, to be due to the problems of having to develop technologies as part of the mission development. (Caveat: I don't follow the implementation problems of either closely enough to know this is true.) You can get around this problem by vastly over resourcing the development. My former employer, which had development programs that were a meaningful fraction of a Flagship mission, did this for some projects. It worked, but it was expensive.

I like the idea of open competition for mission ideas and implementations that is done in the Discovery and New Frontiers lines. However, these missions can draw only on the technology that is already developed. (Fortunately, there's a lot of interesting places to go to with developed technology.)

The planetary community needs to decide the order of Flagship missions about 10 years before their formal mission New Starts. That gives NASA the lead time to work through the development ahead of time.

Space News just posted an article about MSL's cost overruns. The best defense against these problems is to thoroughly define and partially predesign missions up front so that these problems get worked out before full development begins. You can still be surprised, but the number and costs of surprises goes down.

Perhaps what NASA needs to do is to add a formal Flagship project line to its budget to fly one every 7-10 years. The budget line would include any mission, Mars or otherwise, that exceeds $1B. About 5 years before the time for a new start, NASA could have a competition of concepts, much as it has just done for the next Flagship mission, and select the two best for 5 years of predefinition and technology development work. Then when it comes time to select the new start, NASA has two well defined missions to chose from.

Here is a portion of the article regarding the source of MSL's cost overruns (http://www.space.com/spacenews/spacenews_summary.html):

"During a March 12 presentation at the Lunar and Planetary Science Conference held in Houston, Stern said the MSL's total cost has grown to about $2 billion.

"MSL tried to put too much in the bag from the beginning," Stern said, recalling to the audience that it begn as a $750 million mission. "We were kidding ourselves as a community and as an agency that you could do that."

"We need to do this" mission, Stern said, "so we need to come up with the money."

Stern said no single aspect of the MSL undertaking is to blame for the overrun. "It's rife through the whole project," he said, adding that NASA and its contractors underestimated the scope of the task.

What launch vehicle is MSL going to use?

Atlas-V 541

Alas, the cited article is nearly content-free about the actual problems MSL had. Your "best defense" above; I don't disagree, but how do you pay for "predesigning" a mission? You're just moving the cost from one column to another, you may not save anything.

The best defense against overruns is to relax the requirements/descope as the cost ceiling is approached. If you don't do this, you better make sure you can cover the costs, because you can't have it both ways.

I disagree. Five years -- a typical time from new start to launch -- is a very short period for doing a design, developing technology, and resolving problems. It is essentially a crunch time in which your options are to descope or throw money.

I worked in the high tech industry for years. My company spent extensively on technology development so that when a product (our equivalent of a mission) wanted to use the new technology it had already gone through its problems. The most successful programs were those where the new technologies and combinations of new technologies were extensively prototyped before hand. You always want to find out where the problems will be as early as possible so that you can address them in as a normal part of design and not as a crisis.

It was always obvious which programs would run into trouble. They were the ones that either (1) incorporated new techniques without either gobs of resources (2-3X normal staffing) or extensive predesign and prototyping, and/or (2) were approved on optimistic schedule and/or budget projections (we always knew which those were, but management, it seems, too often believes in Santa Claus). It never failed.

All serious design challenges have to resolve lots of problems. You just want to uncover them as early as possible. If you want to fly to Europa -- very challenging mission because of the radiation -- begin your technology development and design work 5 years before the new start. Resolving problems before its a crisis is almost always cheaper, and it leads to fewer problems discovered when your deep into design, cutting metal, and testing and there's a launch date just a couple of years away. I think NASA needs to put serious funding now into design and technology development for both MSR and a Titan mission. The Europa mission has had a decade plus of technology development and mission design, and it showed in the maturity of the Flagship proposal and the medium risk rating it received (compared to the Titan mission's high risk rating). You are right, you may not save money in the long run with all the up front work. You can budget that upfront work, though, and not have to rob other programs because it has become a crisis.

mepag.jpl.nasa.gov/meeting/mar-09/index.html

There is a wealth of interesting presentations from the MEPAG meeting early this month at the above website. Some of the highlights I found:

--MELOS, an ambitious Japanese orbiter/lander Mars mission, set for 2016-2020
--still talking about a possible U.S. science orbiter for 2016, descoped for the tight budget situation, but with increased interest because of detailed methane studies that could be conducted
--an ExoMars presentation with some cool graphics
--current thinking seems to be for 2016 ExoMars to be European-led with significant U.S. contributions, and a 2018 rover that is U.S.-led with significant European contributions
--Mars Sample Return is out of the question in the current budget climate

Why is Sojourner not considered a successful mission?

I have a hard time defining MPF as a "rover mission." It was basically a technology demonstrator for the airbag landing technique, with a rover technology demonstrator.

Sojourner's science package was, to be honest, next to useless. Its cameras returned poor images with poor resolution, and its APXS took some great measurements of the dust on the outsides of some rocks (without a RAT or any other way of removing dust from the rocks, it never got a really decent look at the composition of the rocks themselves). The best science of the mission, IMHO, was done by the Imager for Mars Pathfinder (IMP) camera on the lander.

Not that MPF was a failure -- it just wasn't really a mission designed to land a rover and then do science with the rover. It was a mission to validate the airbag technique and to pioneer wheeled-vehicle operation at a Mars-to-Earth remove. It was an engineering mission, with a few science instruments added on to take advantage of anticipated engineering successes.

In purpose, it was rather like Surveyor I, which had all scientific instruments stripped off so it could serve as an engineering test. The only "science" payload on Surveyor I was its TV camera, which was used as much for engineering analysis of the Surveyor itself as it was for scientific analysis of the lunar surface. And looking at it, the only real science payload on MPF consisted of the IMP camera, the meteorology boom, and the APXS on the rover. The (rather deficient) wide-angle navigation cameras on the rover were designed solely for engineering/operational purposes; rather as with the MERs, these hazcam-like imagers weren't designed to do science, they were designed to help the operators drive the rover.

So, it's not like MPF and Sojourner were failures, it's just that it wasn't what I would call a "rover mission." The Mars 2001 lander wouldn't really have been a "rover mission," either, IMHO, even though it was scheduled to carry Sojourner's sister-toaster, er, rover, Marie Curie. It would have done more of its science from the lander than from the mini-rover, and once again would have been more like an engineering demonstration than a full-fledged rover ops mission.

-the other Doug

Sojourner, while limited, certainly wasn't useless. The APXS results gave a clear indication of water in the past, as evidenced primarily by excess oxygen in the rocks from water bound up in them. The problem is that due to calibration issues and the problems caused by the lack of a brush to clean the rocks off, these results did not come out until late 2003, so they were quickly eclipsed by Oppy at Eagle Crater. Had there been a longer gap between missions, this result would have gotten more fanfare. Still, it did serve more like a really long robotic arm than a rover.

http://adsabs.harvard.edu/abs/2003JGRE..108.8096F
http://adsabs.harvard.edu/abs/2003AGUFM.P12C..06F

Thanks, Ted -- that's exactly what I was struggling to say, that Sojourner acted a lot more like a really long robotic arm than like a true rover.

The only true rovers that have been actually deployed, IMHO, are the Lunakhods and the MERs. All of which worked (or are working) pretty well, all things considered.

BTW, I had not heard that there had been significant information returned about the rocks under the dust on Sojourner's APXS. Well, OK -- I had heard that the only "clear" patches they had tested indicated andesitic basalts, but there was a lot of discussion at that point as to whether this reflected composition of the rocks or the dust. Especially since andesites weren't exactly expected.

-the other Doug

OK, that makes sense.

What are the big problems of sample return from an engineering standpoint - why couldn't we do it now [if we had funding]? Carrying enough fuel to be able to lift out of the gravity well of Mars?

Well, from a VERY big-picture view, it's probably the most complex engineering problem ever proposed in a number of ways. Consider the sheer number of interdependencies & critical events that have to happen just right for it to work beyond the already-formidable problem of getting the orbiter/lander package into Mars orbit & then executing a successful EDL. MSR literally enters unexplored territory shortly thereafter.

A top-level functional decomposition/method decision tree for post-landing operations probably looks something like this (grossly oversimplified), and bear in mind the hundreds of not thousands of sub-functions required as well as the overarching paradigm that they all must be executed autonomously:

-Acquire samples.
--Deliver samples of max scientific value to Earth-return module with minimal contamination.
---Fly a sample-acquisition rover?
---Land near MSL & design MSR to accept a sample from it?
-Depart Mars with samples.
--DTE return? (Hard.)
--Mars orbit rendezvous? (Also hard, and complex.)
-Deliver samples to Earth.
--ISS rendezvous? (REAL hard.)
--Earth orbit injection with retrieval via dedicated manned flight? (Also hard, and expensive.)
--Direct re-entry? (Easier, but possibly more risky?)

...and that's all ridiculously incomplete, with far from all options & risks explored.

There's also the problem that the ascent vehicle and Earth return craft both must carry enough fuel from Earth to (1) achieve Martian orbit and (2) enter a low Martian orbit (aerobraking will help) and two break Martian orbit to return to Earth.

If I remember correctly, MSL is already pushing the weight and size envelopes. Anyone here know for sure?

There are some proposals to manufacture the ascent vehicle's oxidizer from the Martian air, but that's another piece of technology.

I suspect that our current suite of launch vehicles means that the technology push to minimize weight makes this even harder. If we could use Ares V, things would be easier. That though, is a development project in its own right.

@vjkane: That was what I was asking about, the take-off fuel problem.

The idea of picking up samples from MSL is intriguing, but sounds really difficult (due to the precision landing necessary). Wouldn't it be easier just to drop in some very interesting region with abundant loose rock and dirt, extend an arm Phoenix-like to scoop stuff up, and then blast back to Earth? (Maybe a paired mission hitting 2 different spots, like Spirit & Oppy)?

What are the problems with DTE return? The others sound really complex ... why not just drop a sealed capsule in some un-populated zone, like Stardust did? I never realized that that wasn't the obvious option.

We already have random samples of Mars. What is wanted next is carefully selected samples that address specific questions. There was once a proposal for a no rover sample return. That died when Opportunity landed. Just imagine if we had a sample return mission and couldn't collect samples from that beautiful exposed bedrock a few meters away...



It's really a weight problem. For the same reasons that we use multistage launcher on Earth, you want a multistage launcher for Mars. If you carry the extra stage to the surface, you have to size the lander for it, carry the fuel to land the extra fuel.... The Mars orbiter essentially acts as that second stage.

Not trying to start a debate (& the mods will quickly stop one if it begins!), and definitely not expressing an opinion one way or another, but a capsule return al a Stardust or Genesis will have to also deal with the potential back-contamination issue. It's just another system element in this very, very complex construct.

In the case of Opportunity, it took a 250-ton rocket to send a 1-ton payload to Mars, which landed a 185 kg rover. So at 3300 tons, an Ares V ought to be able to land a 2-ton vehicle on Mars, all other things being equal.

Now it's easier to send something from Mars back to Earth than it is to send it to Mars, but even so, with a 2-ton rocket, you're not going to get a whole lot back -- even if you can get the money for an Ares V in the first place.

Of course there are lots of clever ideas to improve on that -- but that means "untested" and "we're not sure how to do it yet" and, more importantly, "we can't do it with what we've got today."

--Greg

If we want a really random mars dust sample, a cheap way to do it would be the old SCIM proposal http://www.lpi.usra.edu/meetings/lpsc2002/pdf/1721.pdf

If we do an all-out sample return mission, being selective is important.

Dvandorn, going back to what you said about Sojourner, from a scientific standpoint, you hit the nail on the head (from an engineering demonstration standpoint, it was certainly more than a robotic arm, but that isn't what we are talking about here).

True, plus a DTE burn from the surface direct to interplanetary earth trajectory would probably only be possible from certain latitudes on Mars and certainly only at very specific dates & times, the best trajectory would almost certainly be to burn to Mars orbit first and then do a TEI burn from Mars orbit. If you have to get into orbit first, then it makes a lot more sense to leave all fuel, engines, and complexities of the return trip in Mars orbit and dock the ascent stage with it before burning out of orbit back to earth. This is exactly the same problem which long ago let to the Apollo LOR method, it saves you a lot of weight if you use only a small lander which docks with an orbiting returnship for the homeward journey.

Personally I still think the best method is using a lot of small lander+rover units, who each shoot a small container with samples into mars orbit, then later have one or two dedicated orbiter missions collect all of those orbiting sample containers and bring them back to earth. This allows for simpler and lighter lander units as they won't need all the complexities of the return-to-earth part, and it allows you to sample a lot of different sites without having to fly a dedicated (& complex) return-craft for each of these missions. Only restriction will be that all those sample-containers will need to end up in the same orbital plane (unless you like to spend lots and lots of fuel from the orbiters for plane-changes), so this will restrict your landing-points somewhat but most of the surface will still be in reach. I don't think it is useful to design a whole mission just to sample one specific site. Mars is such a complex planet and there are so many different landscapes that you would always need samples from a lot of different (carefully selected) sites.

I'm with you. I think that a number of rover missions should be flown that characterize locations in depth. From ExoMars on, I believe they should be able to cache samples. However, even if they don't cache, we will know which site(s) are most worthy of sampling. Imagine spending $6B, flying to a new site, and getting skunked like we were at Gusev Crater.

As I armchair assess the risk, it seems to me that the high risk elements are first the Mars ascent vehicle and then the landing. For this reason, I would fly at least two lander/rover/ascent missions (probably in sequential Mars opportunities) and then fly the Mars orbiter/Earth return vehicle.

The science community will also have to look hard at the cost of Mars sample return. For the likely cost of that mission (~$6B, I hear), both a Titan/Enceladus and a Venus flagship mission could be flown. If a series of rovers find no sites with organics, then which is more valuable: carefully selected geological samples or a Titan orbiter+Titan balloon+Titan lander AND a Venus orbiter (radar imaging at 5m)+two balloons+two landers (~24 hour life)?

If any rover finds organics, then I'm all for Mars sample return. Otherwise...

I don't like to turn this discussion to organics for obvious reasons, but I'm very worried about all the talk of 'finding organics'. I remember Viking all too well, all the PR about how it would 'prove' everything, and when it didn't Mars missions went belly-up for a very long time... You wont 'find organics', or at least not in the way the public expects, you get experimental results which 'might indicate', and then you get a debate which lasts for years and which should be done quietly within the science community and not on worldwide media!

They should put more emphasizes on HiRISE and the MER images, don't try to 'explain' everything but just show the public the beauty of those landscapes! I don't care whether there are organics or not, Mars is a fascinating place, every time I wander through HiRISE I find new wonders and we are still nowhere near to global coverage, this place is amazing! People spend thousands for a holiday to the Rocky Mountains or other similar places, they should see Mars just 'as it is'. You don't give people a guided tour through the Rocky Mountains while talking about difficult science questions, you show them the 'views', that's what they want to see!

Getting back to MSR, I think there is a moment when you just get diminishing returns from ('single trip') robotic missions, we haven't reached that moment yet but all the budget-overruns on MSL are a first indication. At a certain moment it just gets too expensive and too complicated (and too risky) to get all the equipment you would need down on the surface and then, in the end, it will be cheaper and better to get the samples back on earth. When we reach that moment, and only then, MSR should be ready.

Getting in some 'race' to get the samples would be the worst possible thing for Mars-science I guess. You can probably fly MSR today in the same manner as the Soviets flew their Y8E sample return missions from the moon (and as the Chinese will be repeating soon), those missions were scientifically close to worthless, they were extremely restricted in their landing locations (they could/can only land close to 60 deg longitude) and they could only randomly grab the nearest handful of sand and pebbles and take off. It is a 'quick and dirty' method of getting samples but scientifically I think you could save yourself the trouble...

I love to discuss MSR and the best method to fly it, but personally I don't think we are anywhere near to needing to fly it (yet). At the very least we would need global HiRISE and CRISM coverage first, and more rovers down on the surface. And sure, I'm all in favor of more flagship missions to Venus and to Jupiter/Saturn and their moons, there is beauty out there as well!

Once again, I guess the worst possible outcome would be some 'race to get the samples', that would be disastrous, you'll get a handful of close to worthless samples and the media will expect you to immediately solve all big questions...

I think the argument from much of the scientific community is that we are already at that point. There are a number of questions about Mars that can only be answered with samples, and the mission has been endorsed by numerous blue ribbon communities.

I don't doubt the scientific value of the mission, but approach the question of when to fly from a different perspective. Whenever MSR flies, it will suck up much of the world's (since this will be an international venture) budget for planetary exploration for a decade. It will fly once in the professional lifetime of any but the youngest scientists. Therefore, I ask two questions. First, do we have excellent confidence that we know where to gather the sample from? I would argue not -- we have simply explored too little of the surface. We may explore a couple of the sites with clays, for example, and find that they are thin verniers. The other question I ask is whether MSR is more valuable than the missions that could otherwise be funded. None of the blue ribbon panels that I am aware of have stated which missions we should forgo to other targets to enable MSR.

My personal call would be to land a mid-range rover every opportunity for a decade (5 rovers plus Spirit and Opportunity, MSL, and ExoMars). If one of them shows that the site is absolutely compelling ("organics"), then move post haste to MSR. Otherwise, get the experience on the ground, and then decide which of the 9 sites explored is the one to go back to for samples.

I fully agree with you on that!



Agree, but don't forget the orbiters! In a way one could argue that even Spirit and Oppy launched too early (or MRO too late), if we would have had those HiRISE images and CRISM data we have now they might have selected other landing sites. As of yet I don't think we have explored all the available options to get data from orbiting instruments and there are a lot of 'grey spaces' on the map.

With regards to rovers, they are wonderful marvels of engineering, but looking through the orbital images, there are so many sites which look extremely interesting but are simply out of reach for rovers. Maybe we need some 'climbers' or 'crawlers' or some kind of airship or helicopter-analog to reach some of the potential sample-points...

Unfortunately, it will be the orbiters that will suffer the biggest impact of the MSL cost overrun. The Mars Science Orbiter (MSO) was to have a HiRISE-class camera, which meant that it would have also the precision pointing and high data rate that might also have enabled other precision imaging spectrometers. Even expanding the HiRISE spectral bands to another 4-6 to pick up key minerals would have greatly expanded our exploration of Mars. Alas, the camera (and more importantly for cost cutting, the precision pointing and high data rate) has been dropped from MSO plans.

I hope that NASA is funding the MRO extended mission sufficiently to get every last bit of science possible because the capabilities it has isn't likely to be repeated for another 20 years or more. Typically for extended missions, NASA reduces funding, which also reduces science activities per unit time. Does anyone know whether or not MRO is working as hard now as it did in its prime mission?

As said, "don't forget the orbiters". Organics have been found on Mars -- the methane. But we need a specialized orbiter to look for methane hot spots. Otherwise looking for organics with rovers is shooting in the dark.