Written by tim


A couple weeks back, NASA’s Orbiting Carbon Observatory (OCO) failed to achieve orbit and crashed into the ocean, having measured…absolutely nothing.

The nose cone somehow failed to disengage from the satellite, and the whole thing was too heavy to actually enter orbit. Ugh. I’d have snarkier things to say if I weren’t so disappointed by the news.

The satellite system was intended to measure CO2 sources and sinks, and to fill an egregious gap in the satellite missions put on by the U.S. Although there is a relatively large variety of earth-observing satellites already in orbit, the majority of these are well past their designed lifespans — some of them decades beyond their planned fail dates. When one of these sensors breaks down (as they routinely do), scientists back on earth are left with holes in the data that seriously undermine attempts at understanding the earth’s carbon and nutrient cycle.

All is not lost: thankfully Japan launched a nearly identical satellite (but with a lower resolution) just a week earlier. And the stimulus package included $400 million for space-based Earth sensors. (The failed OCO satellite cost $278 million, so at the going rate, we’ve now got enough for 1.39 new satellites.)

If we’re going to get serious about climate change, if we’re going to try and understand the earth’s many feedback cycles and interactions between carbon sources and sinks, we need to have a global, high resolution, and continuous data set from orbiting satellites. That’s a hard feat to accomplish, and right now we’re flying pretty blind.

You May Also Like…


  1. Ross

    …$400 million for space-based Earth sensors. The failed OCO satellite cost $278 million, so at the going rate, we

  2. Steven

    are these launches not insured against this kind of failures to recover the cost?

  3. Tim

    I’d be horrified if there weren’t some sort of insurance, but there is no replacement for this satellite. Generally, most of these satellites are produced in pairs for just such an (unlikely) event. In the face of budget cuts under the previous administration (satellites take 7-10 years to design, build, and launch, and earth observing satellite systems were heavily cut under the previous administration), no backup was built for this system. It’s not hard to see why; NASA has a pretty good launch record, after all, but I lament the fact that countless DoD satellites have backups but a backup for a carbon observatory? Apparently we can’t afford that. More information about the OCO can be found here:

  4. GetCaughtDead

    I suppose it could be different for a government customer, but almost all of them do have insurance.
    No, there wasn’t a backup per se, but as far as a replacement, NASA is in negotiations for a replacement satellite identical to the one that fell in the drink. Keep in mind that once a satellite like this has been built, the design and development aspect has already been completed so it certainly costs far less time and money to build another one.
    NASA’s launch record is not a factor here since they did not actually design, build or launch the spacecraft. But it’s important to realize that no company in the world has a perfect record. Not Boeing, not Lockheed, not NASA. The only way to guarantee 100% success is to never launch at all. Fortunately, the company that did build and [tried to] launch the satellite does have a great record as a low cost, high reliability space systems provider. Let’s just hope NASA’s next launch on a Taurus XL this winter goes a little more smoothly! 🙂