RealClimate logo

Glory (not to) be

Filed under: — gavin @ 4 March 2011

This morning one of the most important (and most delayed) satellite launches in ages took place. The mission was to launch the Glory satellite into a polar orbit, where three key instruments would have been looking at solar irradiance, aerosols and clouds. Unfortunately, one of the stages failed to separate and the satellite did not make orbit.

The irradiance measurements were to be an important continuation of the SORCE mission results, and are needed to stably continue the Total Solar Irradiance (TSI) timeseries. However the big new measurements were those associated with the Aerosol Polarimeter Sensor (APS). A similar instrument has flown in space twice before (the French-developed POLDER instrument), but unfortunately only for short periods. Its uniqueness lies in its ability to detect aerosols over bright surfaces (like land), and more importantly, to distinguish what kind of aerosols it is seeing. (Update: There is a third POLDER instrument, PARASOL, that is currently in orbit, see comments).

It may seem surprising, but despite many different attempts, almost all remote sensing of aerosols from space is only capable of detecting the total optical depth of all aerosols. MISR can provide some discrimination in special cases (picking out dust via a retrieval of non-spherical particles, or using the single scattering albedo to distinguish black carbon), but overall the estimates mix up sulphates, dust, black carbon, sea salt, nitrates and secondary organics. These originate from different processes, have different properties and different impacts on both radiation and clouds. Sea salt comes from sea spray over the oceans, dust from dry desert areas, black carbon from burning of forests and fossil fuels, sulphates derive from ocean plankton and burning coal, nitrates derive from fertiliser use, car exhausts and lightning, and secondary organics come from the stew of volatile organic compounds from industrial and natural sources alike. There are also pollen, and fat particles from outdoor cooking etc.

Because we can’t easily distinguish what’s what from space, we don’t have good global coverage of exactly how much of the aerosol is anthropogenic, and how much is natural. That uncertainty is a big player in the overall uncertainty in the human caused aerosol radiative forcing. Similarly, we have not been able to tell how much of the aerosol is capable of interacting with liquid or ice clouds (which depends on the different aerosols’ affinity for water), and that impacts our assessment of the aerosol indirect effect. These uncertainties are reflected in the model simulations of aerosol concentrations which all show similar total amounts, but have very different partitions among the different types.

The APS technology is a big step forward on these issues. It turns out that while the reflected SW from many different aerosols is similar, the polarisation of that reflected light depends quite strongly on what kind of aerosol it is. This varies depending on the angle at which the light is shining, So by scanning through the angles and measuring the polarisation, we can get a better constraint on the distribution of key aerosols. Scientists have already been working with aircraft mounted versions of the instrument, and this will continue.

The story of how this launch actually happened is very long and twisted, and needless to say, has taken far longer than anyone envisaged at the start (over a decade ago). With the failure to make orbit this morning, the wait will unfortunately go on.

This is of course a huge setback for the mission team (many of whom I know), and I can only imagine how frustrating this must be. The loss of OCO two years ago was due to a similar problem, though 3 launches since then have been successful (and the same system is being replicated as OCO-2). With the postponement of CLARREO in the proposed 2012 budget, there is a huge hole building in the US contribution to Earth and Sun observing systems.

Working from space is hard, expensive and risky. We cannot take it for granted, and yet we need that information more than ever.

70 Responses to “Glory (not to) be”

  1. 51
    Ray Ladbury says:

    Speculating on the causes of the failure–and particularly looking for nefarious conspiracies–is completely irresponsible at this point.

    I would say, however, that provided the FRB did its job on the OCO launch failure (which is likely), the fact that we have seen a second straight failure that looks similar could point to a systematic vulnerability in the Taurus system.

    Looking deeper, though, what we should be asking is why satellites that could give us information critical to our survival are being launched on vehicles of questionable reliability. If you look for the cause of that, you will find that the Earth Sciences at NASA are not just run on a shoestring, but on a shoestring that has broken and been tied in square notes many times. Not only are projects too few and those few underfunded, the Earth Sciences budget is a favorite of Red State raiding parties looking to send some more money down the black hole of manned space flight. And with NASA again on the chopping block and Gulf State delegations again looking to raid Earth Sciences, look for more cancellations, more failures and fewer eyes in the sky watching over our planet when we need them most.

    This failure is the fruits of the “Starve the Beast” approach to shrinking government writ small.

  2. 52
    Adam says:

    Maybe you know who Severn Suzuki is. Her speech is famous from 1992 UN Conference….
    I’ve found a remix of her speech, for fun :DD (its actually good)

  3. 53
    Isotopious says:

    On the contrarian, Gavin.

    “Since 1981 the following volcanic eruptions are known to have placed material directly into the stratosphere: El Chichon (March 1982), Nevado del Ruiz (Nov. 1985), Mt. Augustine (March 1986), and Mt. Pinatubo (June 1991). After each of these eruptions we have noted the presence of (generally) submicrometer-sized ash particles and aerosol droplets on collectors, although we cannot always be certain of the identity of the volcano responsible for the material. For example, collectors from March 1981 contain abundant silicic volcanic ash although no volcanic eruption was known to have directly penetrated the stratosphere (Zolensky and Mackinnon, J. Geophysical Research 90, No. D3, pp 5801-5808, 1985). In addition, we have noted the presence of coarse-grained (less than or equal to 25 microns) volcanic ash on collectors with samples from August 1989 to April 1990, which cannot have been derived from any of the aforementioned eruptions. ”

    [Response: No idea what point you are making. How does the presence on occassional of volcanic sourced material from an unidentified eruption (local? tropospheric?) imply that material from pinatubo last forever? – gavin]

  4. 54
    Isotopious says:

    It doesn’t. But if a 25 micron particle can maybe last for 4 to 5 years, how long would a sub micron particle last?

    I would expect that after an eruption cloud has dispersed, that all you would find is the occassional particle, even if all of the material was still up there. The volume of the atmosphere is enormous compared with a tiny volcano.

    Sure the volcanic cloud will have an effect, because it’s concentrated near the tropics, but spread out it does nothing.

    So I’m a layman wondering how you relate the distribution of dust to a climate forcing, thats all.

  5. 55
    Hank Roberts says:

    Iso’s arguing for uncertainty.

    He starts by ‘thinking for himself’ instead of looking up facts. He imagines something must be true forever — itty bitty things float in the air due to Brownian motion. He didn’t look at what’s known about aging over time of particulates of all sorts. He could find this by starting with one of the early surveys, e.g.
    Atmospheric effects of the Mt Pinatubo eruption
    McCormick, LW Thomason… – Nature, 1995

    and looking at the hundreds of papers citing it:,5&hl=en

    Clue: they react chemically; they adsorb moisture; they act as nuclei for condensation. They fall out. After a while what’s left makes no difference large enough to detect in ‘noise’ of natural variation.

    Look at the work documenting radioactive traces from bomb tests. Stuff will show up for a while. Fallout happens.

    What if there were particles this small?

    “if you take a deep breath right now, at least one of the molecules entering your lungs literally came from Caesar’s last breath. That’s what they say…. you will find professors who say we take in three of Caesar’s molecules per breath, or eight, or 10. It all depends on your assumptions ”

    Does this molecule or eight or ten make a difference to you? Nope.

  6. 56
    tamino says:

    The chance of two successive mission failures can be pretty substantial:

    So I think we shouldn’t entertain conspiracy theories.

  7. 57
    chris colose says:


    You’re touching on a number of different issues here, and you’re mostly off-base with all of them.

    For one thing, in the troposphere particles less than 0.1 μm (i.e “Aitken particles”), have a short residence time since they are removed quickly by diffusion. Giant particles are also removed relatively quickly (say, over a micron) by gravitational settling. There’s a range in between (called the Greenfield Gap) where it’s tougher to remove aerosols, but it’s still done. If water vapor condenses on aerosol particles and then precipitates you can remove that way too. In the stratosphere, particles last longer but can still be removed eventually by gravitational sedimentation or troposphere-stratosphere exchange (maybe upper-level fronts?). The stratospheric post-volcano removal has an e-folding time of about ~1 year. I haven’t studied aerosol removal mechanisms in any detail though I am slightly familiar with a few, but a brief look through google scholar suggests there’s a large literature on the theory of aerosol removal and on the settling after volcanic eruptions. For confirmation, there are also optical depth time series available derived from solar extinction measurements (McCormick et al 1993). See also the series from Ammann et al 2003 or this paper based on Mauna Loa Lidar measurements showing the decay to pre-eruption conditions. You can search this stuff out a bit better.

    Concerning sensitivity, I respectfully disagree with Ray that “volcanoes favor a sensitivity of 3 C” although they do help rule out very low sensitivities. They do virtually nothing to eliminate very high sensitivities. I still can’t follow your logic though in what you’re trying to say with this.

  8. 58
    Isotopious says:


    Thanks chris, the Lidar measurements are robust enough. I was just trying to find a hole in the method of estimating climate sensitivity from eruptions…..but crash landed in the south pacific.


  9. 59
    Vendicar Decarian says:

    But some Republicans, who hold a majority in the House of Representatives, want to see NASA give up climate science so it can focus on returning astronauts to space once the 30-year-old shuttle program ends later this year.

    “NASA’s primary purpose is human space exploration and directing NASA funds to study global warming undermines our ability to maintain our competitive edge in human space flight,” said Republican Congressman Bill Posey last month.

  10. 60
    Chris O says:

    @51: Ray, I completely agree that the Glory failure mode does seem to point to a systemic problem with the Taurus. Meaning, i see it as unlikely that the OCO and GLory launch failures were unrelated and due to different problems. But of course there will certainly be a new investigation to try to determine exactly that.

    @56: Tamino – well, OCO and Glory did use the same launch vehicle before. let’s push your math just a bit further. Prior to OCO, the Taurus had had k=1 failures in n=7 launches. applying your (k+1)/(n+2) formula, yields a failure probably of 22%. Applying it for your two-in-a-row formula yields a probability of ~7%. Seems reasonable. But, GIVEN the failure of OCO, we get a probably of glory failure at 30% (only slightly higher than the failure probably for OCO, which seems counterintuitive to me). A failure probability of 30% seems unacceptably high for a $424M mission. But maybe that’s just me ;).

    Now, let’s carry this through to a discussion that really matters for the future – namely for OCO-2, which is the next, and last, scheduled launch on a Taurus. Given the history (3 failures in 9 launches) yields a whopping 36% chance of failure. And if you consider that 3 of the last 4 launches have failed (!), your failure chances seem higher – 67%. This is just nuts to me, and therefore I hope and pray OCO-2 will *not* launch on the last Taurus as currently planned.

  11. 61

    Isotopius (42) wrote: “As Lindzen has pointed out, with regards to aerosols, even the sign is in question.”

    The source that Lindzen uses for this claim (Ramanathan) doesn’t actually claim what Lindzen claims he claims.

    Lindzen’s argument is implicitly built on a high level of confidence that aerosol net forcing is around zero, whereas according to the avialable evidence it is extremely likely negative while the exact value if highly uncertain. So Lindzen implicitly assumes high certainty where it doesn’t exist.

    Not to mention the peculiar fact that Lindzen recently signed a skeptic letter in which the NIPCC report was approvingly mentioned. The NIPCC report makes the *opposite* claim as Lindzen does, namely that “The IPCC dramatically underestimates the total cooling effect of aerosols.”

    Guess Lindzen tries to have it both ways.

    See for more details and refs:

  12. 62
    gavin says:

    From Bruce Wielicki (via Scott Mandia) responding to why the loss of this satellite was a huge setback for climate science:

    “Your questions are good ones and I’ll try to put them in context. First: who am I? I have been a co-investigator or lead investigator on NASA Earth Science missions since 1980: ERBE, CERES, CALIPSO, CloudSat, CLARREO. So I am very familiar with space missions in general and climate missions (all of the above) in particular. If you need more details they are in the attached resume.

    The loss of the Glory satellite is a tragedy for climate science. The Glory satellite included two critical instruments: one to monitor the total energy reaching the Earth from the Sun, and a second to unravel some of the key remaining mysteries about tiny particles called aerosols: especially about the aerosols that humans emit when we burn fossil fuels in cars, power plants, or our homes. Aerosols remain one of the key uncertainties in how fast our fossil fuel burning is pushing the climate system to warmer levels. So the Glory mission was a key part of understanding how both natural (Sun) and human (aerosols) forcings are acting to change our current and future climate.

    If this loss is so serious, why was there no back up strategy? Why was this allowed to be a single point of failure? The answer is that space missions are expensive by nature, risky by nature, and our nation has decided not to spend the kind of resources it would take for a more robust set of climate research observations. Such an observation system might easily cost 4 to 5 times the current NASA Earth Science budget. Would it be worth it to make more intelligent future decisions about climate change? Without a doubt. But is there a national will to do it? Evidently not. One way to look at this is that we have a football team with only one player at most positions, and none at a few positions. When one of the players we do have gets hurt: there are no replacements. You play without him and wait until he heals. The time to heal a lost space mission is typically 3 to 7 years depending on budgets and how many spare parts remain from the last instrument builds.

    What is NASA’s role in climate science? NASA Earth Science missions are a critical part of climate science. Space is the only way to get truly global observations of the Earth and its climate system: from equator to pole, from the U.S. to China. Those observations include everything from the atmosphere to the oceans to the ice sheets to polar sea ice to land cover including vegetation and snow. They include the energy we receive from the sun as well as the solar energy we reflect back to space and the thermal energy we emit to space to shed the solar heat that we absorb. Climate is an interlinked global system including all of these key parts. Looking at just one or even a few of them typically leads to large uncertainties and low scientific confidence. NASA has led the world in global climate science since the advent of the Earth Observing System that started in 1990: the first attempt at a global Earth observing system. Ironically it was the deficit federal budgets of the mid 1990s that reduced the effort to about 1/3 of its original plan. What we have now are pieces of that system that have lived well beyond their design life. For example, the Aqua spacecraft was launched in 2002, designed for a 5yr mission life, and was originally supposed to have 3 copies launched on 5 year intervals to achieve a continuous climate record over at least 15 years. Only 1 spacecraft was ever built and launched, and has now been operating successfully for about 9 years on orbit. A follow on mission called NPP is finally planned for launch the end of this year. But there is no climate observing system in the same sense that there is a weather observing system. NASA is doing the best it can with the limited resources it has. There are no backups.

    Should NASA be doing climate science? The National Aeronautics and Space Act established the agency in 1958. In the Space Act, the first objective of the agency was listed as “the expansion of human knowledge of the earth and of phenomena in the atmosphere and space.” Earth science has been a key part of NASA’s mission throughout its history. The need for that mission today is more critical than it was in 1958. When the Space Act was written, we had little idea of potential climate change issues. The Keeling record of carbon dioxide in the atmosphere was just starting in Hawaii. The Keelilng record was not the first carbon dioxide observation: but it was the first with the high accuracy over a long time period needed for climate change research. Many people confuse weather with climate. Why can’t weather satellites be used for climate data? In general they lack the high accuracy needed for climate change. Weather accuracy is 1 or 2 degrees in temperature, while climate accuracy is a tenth of a degree: a factor of ten more difficult. In the end, climate observations have requirements that are typically ten times more accurate than weather, and require 10 times as many variables to be observed. In the U.S. we have a dozen agencies that contribute to climate science and are coordinated using the U.S. Global Change Research Program (USGCRP). NASA resources are the largest contributor to the USGCRP of all the agencies, but none of the agencies has climate as its highest priority. This results in a “curse of the commons” situation where none of the agencies can really lead the development of a climate observing system. Each does the best it can within its limited scope and resources.

    What do the successive failures of the OCO and Glory missions mean for NASA climate science? They will have a serious impact and will delay advances in understanding carbon dioxide sources and sinks (OCO), natural and man-made aerosols (Glory), and solar climate forcing (Glory). They will also force NASA to evaluate the best balance between use of small less reliable but less expensive rockets, versus larger more reliable but more expensive rockets. Unfortunately there are no easy answers. It is likely that NASA will continue to find the best option is a range of small to large missions, with a range of small to large costs, and low to high reliability. The resources are not there to design and implement a global climate observing system with a 90% chance of success. Maybe someday that will change.

    Sorry this is so long, but these are not easy questions. It is a sobering time to consider them.”

  13. 63
    Marlowe Johnson says:

    are there insurance policies for these sorts of thing? Can we expect NASA to try again (third time a charm), or is it too soon to say?

  14. 64
    jfb says:


    It really doesn’t matter what House Republicans are saying; they’re not working on the rocket.

    I’ll ask the same questions I did over at Tamino’s – *who* would have performed the sabotage (it would have to have been be either an OSC or NASA engineer), *when* would they have performed it (during payload integration, during closeout inspection, or after closeout), and *how* would they have performed the sabotage with so many eyeballs checking and double-checking everything?

    It’s not like just any idiot can walk up to a rocket and monkey with it.

    And as so many other people have pointed out, we don’t *need* to invoke sabotage; plain bad luck accounts for all of it.

  15. 65
    Aaron Lewis says:

    Sabotage? Too complicated! It is like a small town car rental shop. The guys with more money (and friends in congress) get the better rides, and poor folk get the cars that were bought lowest bid. No climate science program has as much money or as many friends as any of the national security programs.

    If DHS decided that knowing what was going on with the climate was essential to national security, then DHS would put the right gear into orbit even if they had to use the very pricy launch vehicles that they use for intelligence satellites. And, they would order backups, so that there would be redundant measurements.

    If DHS was smart, they would be launching more “climate” satellites and fewer “intelligence” satellites. Then, if they were really smart, they would share the data so that it could be sliced and diced by guys that are too smart to work for DHS.

  16. 66
    AIC says:

    Thank you to Bruce Wielicki for the extended explanation.

    Could an outside outfit like The Aerospace Corporation come in and help in the investigation? We cannot afford to lose the second OCO satellite, just as we could not afford to lose the Glory satellite.

  17. 67
    Ray Ladbury says:

    In light of Bruce’s lament above, I thought I’d post a link to Tom Bodett’s wonderful essay, “Home Planet” again:

    It’s the only explanation that makes sense.

  18. 68

    Very very sad for science, but then the people of the world have plenty of information & science since at least 20 years ago to have been mitigating GW to the hilt (we don’t need high confidence of disaster to strive against it), and absolutely no excuses not to.

  19. 69
    John Dixon says:

    Why do you need that information? You told us the science was settled.

    [Response: Did I really? Perhaps you could point that out. – gavin]

  20. 70
    Susan Anderson says:

    69 JD:

    I suspect you think you’ve scored a point with your smug drive-by comment, but just in case you would like to learn, I’d suggest you start by reading the article and the many excellent comments above, which are informative and build knowledge (though I’d skip the conspiracy stuff, which is neither useful nor likely). In particular, the post from Bruce Wielecki which addresses factually and in detail the problems with politics and finance that clog the arteries of this necessary work. This work is partly needed because of people like you, who will never acknowledge the truth of measurement and data, or that uncertainty is a given in life and we still put one foot in front of another.

    I particularly enjoy the trenchant commentary of Ray Ladbury. The nice thing here is that the total adds up to more than the sum of its parts due to the careful avoidance of arguments that are in part intended to prevent these guys, who have work to do, from getting on with it. The denialosphere calls this censorship, but they, who *do* practice censorship, of things they disagree with, are focused solely on eliminating dissent, while here legitimate arguments are carefully answered until they reveal obdurate positions.

    As to “settled science” that mostly comes from fake skeptics. A little study will demonstrate that the words of people like Dr. Schmidt are rarely left unchanged in the reports from the political deception campaign. Those commenting on it rarely look at the original, they take the word of their “friends” as to what was said. DeepClimate has a particularly good section on this issue – the claims about “settled science” being attributed backwards and misquotation proliferation: