In this week’s issue of The Space Review, Taylor Dinerman examines the NPOESS weather satellite system, a program whose problems have been discussed here recently. Dinerman makes a good point in that much of NPOESS’ woes can be traced to the program’s technologies, which are far ahead of previous systems. As he puts it, “Why do so many US government technology development efforts aim at revolutionary improvements in capability, instead of settling for incremental progress?” It can explain not only the problems facing other large space programs, but also the concerns raised by Congress about future systems, like TSAT and Space Based Radar.
Meanwhile, this week’s issue of Aviation Week reports that the DOD is facing a “perfect storm” caused by the convergence of “operational, budgetary, manpower and transformation crises” that could threaten any number of procurement efforts. The article primarily focuses on aviation, not space, programs, but it’s clear that space is weighing on the minds of many planners at the Pentagon. In particular, the article notes how those working on the Quadrennial Defense Review (QDR) have seen their effort shifted to “a lightweight budget drill” seeking minor savings that end up being swamped by cost overruns. “All the people that take QDR seriously as a policy exercise spend 3-4 months scraping together a couple of hundred million dollars in savings from here and there in order to buy the new policy initiatives,” one source told the magazine. “Then, in comes a bill for a $1.3-billion fix on a satellite program.”
This disaster in slow motion suggests two things to me:
(1) For whatever reason, researchers are unable to deliver technologies the Pentagon wants
(2) For whatever reason, somebody is allowing these programs to proceed from tech development to full-blown engineering development without the underlying technology being ready.
I know what I think causes these things; I know what Dfens thinks causes these things. I even suspect that Ron Sega knows precisely how to fix these things.
The question is, does he have the authority to do so? If not, why not? Does it really take a disaster unfolding in a “three act play” of political theater with genuine tragedy at the end to empower individuals to make the necessary changes within the Pentagon?
NPOESS has many more screwups than this article goes into. As I’ve complained about before, on top of the original 13 instruments, they also added the Landsat OLI instrument to it in 2004 much against the wishes of GSFC landsat program office. It looks like as part of this shakeout, they are finally getting rid of that godaweful decision.
Another big problem with this system is that instead of just marrying up the two different types of weather satellites, they are also combining the instruments of three other NASA EOS missions, Aqua, Terra and Aura, to the program making it even more complex. This added the following 4 instruments to the program: VIIRS, CrIS, ATMS and OMPS. The overall philosophy of one satellite fits all swiss army knife approach pervades this program. It is no surprise to anyone following just how badly overbudget this program is.
Looking back at the schedule and for Tiros and other previous weather satellites, one can see why they are shorter and cheaper in that their missions were more direct and focused than NPOESS. The schedule and cost overruns are as much a factor of trying to make swiss army satellite than what Taylor Dinerman’s pointed out to being just indicative of today’s working environment. The government really needs to look at the advantage of huge satellites with huge development costs and delays verses multiple smaller satellites with fewer instruments and quicker schedules.
I strongly suspect the problems is, to quote a computer friend, the American propensity to “high tech everything to death.”
We have perfectly good comsats that are known to work. We even have military comsats that are known to work. Why aren’t we buying gobs of commercial XTar spacecraft, or military Wideband Gapfiller spacecraft, et cetera, while developing TSat at a more leasurely pace?
Likewise, why do the new weather satellites feature all new instruments? What’s wrong with the old instruments, while slowly evolving improvements.
We Americans are supposed to have invented mass production, why aren’t we using it?
— Donald
Let me try to add a little information to this, although I cannot guarantee that I have any real _insight._
For information on the early Tiros program, you might want to look at my recent Spaceflight article. That is the first of what will probably be a 3-4 part series about the development of military metsats. Although it was a relatively quick development program by today’s standards, the early Tiros benefitted from a long technology development program. The vidicon instrument (essentially a television camera) was actually in the works since the late 1940s, sponsored under USAF seed money to RCA.
Taylor Dinerman asks the rhetorical question of why the US military does not do incremental upgrades. Although I am sure he is aware of this, I would point out that they _do_ incrementally upgrade their satellites. Tiros and DMSP, both flying now, have been updated with new instruments for decades. I’m not sure when the last DMSP or Tiros block change occurred, but it was at least in the mid-1980s.
If someone ever writes the history of NPOESS (which is unlikely–note that no government agency even bothered to write a history of Tiros when it had its 40 year anniversary), it would be interesting to delve into the question of why they decided to add so many instruments onto the spacecraft. Why not a simple merger of DMSP and Tiros into a single platform with growth potential?
My suspicion is that they viewed NPOESS not simply as a unification of two separate systems, but as an opportunity to achieve revolutionary change. They also probably fell into some of the traditional traps of not spending enough money on the early technology development. During the congressional hearing on NPOESS the program managers claimed that they did have a technology development program underway, but that it had proven inadequate to the task.
Before the NPOESS agreement in 1994, there were apparently no less than _eight_ previous attempts to merge DMSP and Tiros into a single platform. I have at least one and possibly two of those study reports (recently declassified). A merger was always rejected as impractical for reasons that were apparently too sensitive to acknowledge. It is worth noting that back in the early 1960s the original intent was to merge the programs by 1963 (DMSP was really only supposed to be an interim system until the upgraded Tiros, called Nimbus, arrived). What happened was that NASA became more interested in scientific data collection instead of operational weather data collection. Ultimately the Weather Bureau (predecessor to NOAA) defected and built a separate system, and NASA pursued Nimbus as an environmental scientific satellite. There are some interesting lessons in that experience about how NASA has done a poor job of satisfying customers outside of the agency. They later started down that road with EOS.
This might lead into some interesting questions as to how the EOS scientific instruments ended up on NPOESS. It seems that some major management mistakes were made with NPOESS–too many missions combined into a single platform, and too many advanced technologies adopted at once.
Finally, I think an interesting question to ask is how come the replacement for Tiros and DMSP will only fly two decades after the decision was made to merge them. How much of this was simply long development time versus reluctance on the part of the institutions to give up their existing successful platforms?
I suspect that Mr. Dinerman’s analysis has more than a little relevance to another milspace debacle, the optical component of FIA. As we recall, the contract has reportedly been (or is in the process of being) taken away from Boeing and given to LockMart, the loser of the original competition.
Which seems odd, because on the face of it designing an electro-optical spysat is something that many people have done over the past few decades and building another one shouldn’t be impossibly difficult. So what wasn’t Boeing able to do? Apparently its winning design in 1999 was thought to be more “innovative” than LockMart’s, but what does that mean?
I’ve wondered whether it might refer to some new mirror (or even lens) technology that would allow lighter satellites to deploy KH-11 size or larger apertures once on orbit, but that’s really just a WAG.
Any suggestions for other things that might have brought FIA to its current unhappy pass?
While we’re on to conspiracy theories, I can’t help wondering if this has something to do with the size ultimately chosen for the NGST, and if the problems might not be one and the same.
— Donald
“I can’t help wondering if this has something to do with the size ultimately chosen for the NGST, and if the problems might not be one and the same.”
Unlikely. The JWST (no longer NGST) is not experiencing technical problems. It is experiencing cost overruns, which are not the same thing. The contractors seem to be making effective progress, but the initial cost estimates were bad.
Donald Robertson – I agree wholehearted only the XTar sentiment, but in full disclosure the company I work for owns 56% of that bird. The problem we are seeing on XTar right now which is upsetting is that prior to launch, the DoD was very keen on purchasing capacity on this spacecraft but is now mired within its own red tape and unable to commit to any purchases yet. However, the State Department has signed up as have several foreign governments, so there is hope.
Allen – I think global security said it best (http://www.globalsecurity.org/intell/systems/fia.htm) in that Boeing overpromised and under delivered. Boeing had no experience doing electro-optical and found out it is alot harder than they thought.
Dwayne – Thanx for the history on the mil side, I look forward to reading the rest when you publish it. I fear that I agree with the questions you have raised and I have a feeling this program will continue to be plagued with problems and I won’t be surprised if it does not launch before 2015.
Synchronistically enough, there’s an AP story about FIA today that touches on where the Boeing -> LockMart transfer stands:
http://www.startribune.com/stories/535/5756889.html
Ryan, in the full disclosure department, I once owned a significant amount of stock in your company until I trusted the CEO’s repeated communications to stockholders, didn’t sell, and had it stolen (not too strong a word) from me during your reorganization. (I understood the company’s difficult position and would have accepted a fraction of a penny on the dollar, but I wanted something. I don’t generally believe in punitive class actions, but in this case, had I bought at the correct time, I would have joined the shareholder class action against your company in an instant.)
Nonetheless, I still stand by my statement that the government should buy a bunch of XTars, or the equivalent, rather than developing T-Sat at this time.
— Donald