Friday, January 27, 2012

For Whom the Bell Tolls

It's a bad time to be a nuclear physicist.

Never mind that the battle cry is "better education equals better jobs," that a PhD is supposed to be one of the highest achievements one can attain, that our own government continues to tell us that science is important. This patina just serves to hide a rusting inner truth. Nuclear physics in the US is on its deathbed.
Now, don't get me wrong. FRIB will be a great boon to the nuclear physics community. But the cost of FRIB - not only financially - is too much burden for us to bear. In order to "get" FRIB, we had to trade HRIBF. We had to give up Yale. And that's just recently. For many years, smaller university labs have been closing, unable to fund themselves or else seemingly at the end of their useful lives. Slowly, the community has become, instead of vibrant individual groups doing competitive research at a number of labs, an array of satellite groups, each competing against another - not for results - but for experimental time and resources.
To use a (limited) analogy, the heyday of nuclear physics was like having dozens of Mom-and-Pop stores, each competing, and specializing slightly. But slowly, individual shops are being lost, and are being replaced by Walmart - a gigantic monopoly over the whole scientific field. This is what FRIB will be. And unlike our corporate example, FRIB will not provide more of its commodity. We will have more isotopes available for research, but the same amount of time to study them. FRIB can't run three-years' worth of experiments in one year. It's still just a year. So now more of the community is fighting for the same amount of experimental time and resource - which can only mean that fewer in the community are successful. Smaller university groups will either disappear, or be forced to scientifically amalgamate themselves with larger, more successful, groups.
Particle physics has already gone this route, if not for any other reason than money. It simply isn't financially viable to run two dozen versions of CERN. But now that the Tevatron at Fermilab has closed, we only have CERN. The Department of Energy has, of late, threatened to close even more facilities.
One of the great things about science is the objective way it views the universe. We pride ourselves on having others in our field try to prove our answers wrong. But they have to be independent in order for the system to really work. And if the US government keeps closing labs, we'll soon only have one place to do nuclear physics - and thus no truly "independent" researchers.
This collapsing of a field into only one facility (from many) has other problems as well. Fairly or unfairly, our merit as researchers is partially determined by the number of publications we have in a given year. With more people fighting for fewer resources, the total number of publications will decrease, as will the number of publications any one researcher has to his or her name. Because outside researchers are forced, in this scenario, to collaborate with larger and larger groups, only the lucky few who have a job at the one remaining facility will have plenty of publications. Everyone else has to "go through" these few staff members in order to do an experiment. This is already happening in nuclear physics - while on paper things such as a Program Advisory Committee or Experimental Evaluation Committee still exist, experimental proposals submitted to these governing bodies are less successful if a member of the staff at the facility isn't involved. It's simply a matter of familiarity. Proposing a successful experiment at Oak Ridge is difficult if you're not intimately involved with the day-to-day goings-on in the lab. You can gain that expertise by including someone from Oak Ridge in your experimental planning and proposal. So if, in ten years' time, the only place we can do nuclear physics experiments is FRIB, then all experiment proposals to FRIB will involve, in some way, FRIB staff. This creates a potentially unfair advantage when applying for jobs - those from small university groups will have far fewer publications than those who are laboratory staff, even if otherwise either person is equally intelligent, capable, and qualified. A loss of experimental facilities also leads to difficulties in performing different types of experiments. Just because you've spent $500 million to build a lab that can create something special and specific doesn't mean a $50,000 experiment can't also inform what you're trying to figure out. Sure, I'd love to do experiments with radioactive aluminum, but I learned a little something about it recently using stable silicon (the same stuff computer chips are made from). Both approaches are complementary - and necessary. With fewer experiments and fewer ways to do them, the potential for systematic errors increases. And with fewer experiments (as there is only one place to do them), more people will be on each experiment, so the contribution of each is lessened. Particle physics papers have hundreds, if not thousands, of coauthors (and it's always suspect as to how much each coauthor actually contributed to the work). If a PhD candidate had to design, set up, run, debug, analyze and publish an experiment with minimal help, he/she would have substantially more knowledge than the PhD student who spent the same amount of time simply making sure all of the cables on one side of one detector were plugged in for one experiment. But in the end, both get the PhD. It has become a bit of a running gag in the physics community that taking a two-year postdoc position with a particle physics collaboration is roughly equivalent to taking a paid vacation. Nuclear physics is headed this way. Potential PhD students may see this as a boon - less work for equal reward - but it spells death for the field. What good are a thousand coauthors on a paper if not one of them knows how to do more than one-thousandth of anything?
I know I'm picking on particle physics a bit, but it's because particle physics stands already as the model for what happens when you only have one place for your entire scientific field to do experiments. It weakens your science, it weakens your scientists, and it leaves you unprotected from the whims of your funding bodies (one Mom-and-Pop store closing is sad; Walmart going bankrupt is catastrophic). Nuclear physics research may not recover from this; we may already be too far down this road to turn around. And nuclear physics in the guise of nuclear energy and nuclear engineering won't save us, either. There will not be a nuclear renaissance, not in the near future. The public is too sensitive to incidents like what happened at Fukushima, rare (and not as dangerous) as they may be. And even if we did start building and running more nuclear plants, the research community is still at a loss. This is yet one more symptom of a dying field; there is such a degree of specialization that, as a nuclear physics researcher, I can't get a job in nuclear engineering research.

The final question remains. Why should you care?
Why should the public care if one scientific field, an aging one at that, finally dies? You should care because it affects you, too.
You may not realize it, but without nuclear physicists, you'd have an entirely different world. Nuclear physicists may have given the world The Bomb, but nuclear physicists are also the only ones who know how to safely dismantle and dispose the bombs we have. How else would you know whether the new TSA scanners are safe? How would you know how old ancient artifacts are, how would you keep astronauts safe in space, and how would you treat cancer?
And to take a broader perspective, it isn't just nuclear physics that we're considering. Death of one scientific field is a partial death for all science. Any time we stop pursuing objective truth, we lose a piece of ourselves. It tolls for thee.

"No man is an island, entire of itself; every man is a piece of the continent, a part of the main. If a clod be washed away by the sea, Europe is the less, as well as if a promontory were, as well as if a manor of thy friend's or of thine own were: any man's death diminishes me, because I am involved in mankind, and therefore never send to know for whom the bell tolls; it tolls for thee." - John Donne, Devotions upon Emergent Occasions, Meditation XVII (1623)

7 comments:

  1. Hi, Kelly-

    I feel your pain. It is sad to lose livelihood and community. I understand the damage wrought by concentration of available facilities as well.

    But allow me to advance a few critical points as well, from a lay perspective.

    One problem is that sciences such as particle physics have by their success worked themselves into this position where their scientific frontier is also at the outer reaches of the technical (er financial) frontier. Any bigger experiment would require some rather intrusive use of the moon, or other miracle. From there it becomes an observational science insofar as something can be gleaned from the cosmos, or frankly theoretical and fully speculative.

    This all boils down to great success. All the smaller questions that are humanly-empirically accessible have been answered, and that is a great thing. (Or were they the biggest questions? Hard to say, perhaps.) No one said that every question has an answer let alone an accessible one- only religious people insist upon it. And funding should follow the currently productive quests, not fund every form of inquiry that has ever been productive in the past. (Though the issue of complete loss of relevant expertise is a huge cultural issue- we have to find better ways to record such knowledge, but we really can't .. it is a very bloggable topic.)

    A related point is the baby boom in sciences. I don't know what it is like in nuclear physics, but in biology every professor has students and postdocs, all lowly paid, and then they are cast off.. a vast excess of students trained so that the field can pick the cream for faculty postions. What happens to the rest? Who cares? There is no law that every biologist must have a biology job, and the same is perhaps true in nuclear physics.

    Lastly, I could pick a few bones with the justificatory examples provided. I accept your rationales for archaeology/geology, for nuclear bomb care / disposal, and for scanner analysis (though is microwave radiation imaging such a close application of nuclear physics?) ... I am sure that they hire as many nuclear scientists as required, however short that is of providing places for all who need them. There is surely interesting science to be done and applications to be found elsewhere as well, perhaps in materials science generally.

    But astronauts in space? The whole concept is a bit passe, really. Robots are doing a great job. Despite the fun / excitement of putting people in space, it is neither efficient, safe, climate-friendly, nor needed. And for cancer, radiation of any kind is an extremely blunt instrument, indeed, frankly, barbaric. Molecular understanding, while very slow in coming, is starting to provide genetically-specific treatments that will far outstrip the ability of blunderbusses like any of the various chemotherapeutic cell-killing toxins that have been used to date. Not that they will be cures necessarily .. that remains speculative. But we can and will do better than radiation.

    ReplyDelete
  2. I'l add that another future prospect is a revitalized nuclear power industry. I have been very interested in the pyroprocessing cycle.. is that a realistic path for a sustainable nuclear fuel cycle? Sustainability in general requires that everything become electrified.

    ReplyDelete
  3. Hi, Burk.
    One problem is that sciences such as particle physics have by their success worked themselves into this position where their scientific frontier is also at the outer reaches of the technical (er financial) frontier. Any bigger experiment would require some rather intrusive use of the moon, or other miracle.
    You're totally right there. But more and more, I find myself inexplicably drawn to string theory...

    And you're right, too, in that funding need not continue to prop up a dying subject. But I'd argue that the fact the DOE agreed to build FRIB in the first place (our replacement) was because we're not a dying subject, and there is still much we can learn. Even here, at this facility. We haven't exhausted our capabilities yet, not by a long shot.

    As for astronauts, sure, I'm a bit of an idealist and I think there's something inherently romantic about manned spaceflight. But the argument still stands with robots. Nuclear physicists work at places like Ball Aerospace and Lockheed to design radiation-resistant chips. And with cancer, one rather gaping hole in our nuclear knowledge is what long-term low-level radiation exposure does to the body. Perhaps such exposure actually helps to prevent cancers from forming... we don't know. We'd like to know. So I'm sure that molecular targeting of cancer treatments, etc, will be the future, but that's not to say that radiation is no longer of interest in biological systems.

    As for next-gen nuclear power, I'd love to see it happen! We've had feasible designs for reprocessing plants since the Cold War. Politics, not science or engineering, prevented it from becoming reality.

    ReplyDelete
  4. Actually, things are not too bad in Nuclear Physics. CEBAF is finishing up its 12 GeV upgrade and FRIB is at the onset of construction - these are two major state-of-the-art research tools that will can serve the Nuclear Physics Community for two decades more. There are several other facilities in the US for Nuclear physics such as Argonne. The Holifield was built in the 80's (from accelerators constucted in the 50's and 70's) and the Yale tandem was built in the 60's. It is unreasonable to expect equipmemt to be relevant forever. Yes, it would be nice to keep these facilities going but to advance the field in times of constant or declining funding hard choices need to be made. The ~$30 M that went to these two labs are needed by FRIB.
    Cheer up. These things tend to go in cycles.

    ReplyDelete
  5. While I appreciate that there is still money in nuclear physics - and trust me, I do appreciate it, I was there in the UK when STFC slashed 52% of the NP budget - the overall funding situation wasn't my main concern in writing this.

    It is still uncertain whether FRIB will be constructed; Steven Chu has said as much. And everyone in the nuclear physics community knows that the $10 million that DOE saves by closing the HRIBF (which was actually dedicated as a RIB facility in 1996 - it's not as decrepit as you make it out to be) doesn't do much to change the financial status of FRIB. Also, Argonne as yet can't produce any of the radioactive beams that HRIBF can, so it's hardly a valid replacement. The fact of the matter is that the HRIBF is unique in the US and only one of three similar facilities in the entire world. So when it comes to the argument that the HRIBF may close because it has become old and irrelevant, one can see it simply doesn't hold water. HRIBF will become irrelevant as soon as someone else in the US can produce half a million particles per second of radioactive tin-132.

    But like I mentioned before, the issue isn't simply one of funding. The particle physics community is flush with funding, but I do not feel like it is a good time to be a particle physicist (at least, not a graduate student in particle physics). There is much more to a field than simply whether it has enough money.

    ReplyDelete
    Replies
    1. Particle physics isn't flush - they get $800M vs $700M for Nuclear Physics. Nuclear Physics is spending $50M on CEBAF upgrades this year that will be winding down - that (plus $30M from shutdowns) is where FRIB funding will come from. The big issue is ITER - the US has a high priority in funding this and will likely take it from the other programs.

      Delete
    2. High energy gets $800M to nuclear physics' $600 from DOE. NSF provides roughly $166M to nuclear physics (most of that being the current NSCL funding, which will disappear when NSCL transitions to FRIB) and $133M to elementary particle physics ($240M if you include particle astrophysics).
      So that's a total of $760M to nuclear (really, it's more like $660M, since otherwise we're double-counting NSCL/FRIB) and $1.2 billion to particle physics. That's nearly a factor of two different.
      But as I said, it's not just about funding.

      Delete

Think carefully before you post. I reserve the right to moderate any comments posted to my blog.