Please consider donating to Behind the Black, by giving either a one-time contribution or a regular subscription, as outlined in the tip jar to the right. Your support will allow me to continue covering science and culture as I have for the past twenty years, independent and free from any outside influence.
A little over a month ago I reported here on Behind the Black some recent results from the LEND instrument on Lunar Reconnaissance Orbiter (LRO) that had found significantly less water in the permanently shadowed craters at the lunar poles than previously thought. To quote again from that paper’s abstract, which I will henceforth refer to as Sanin, et al:
This means that all [permanently shadowed regions], except those in Shoemaker, Cabeus and Rozhdestvensky U craters, do not contain any significant amount of hydrogen in comparison with sunlit areas around them at the same latitude.
And from the paper’s conclusion:
[E]ven now the data is enough for definite conclusion that [permanently shadowed regions] at both poles are not reservoirs of large deposits of water ice.
You neglect to mention yet another possibility — that this paper and its conclusions are seriously flawed in almost every respect. The veracity of the LRO collimated neutron data [produced by the LEND instrument] have been questioned on serious scientific grounds. Other data sets (spectral, radar) suggest significant amounts of water at both poles, billions of metric tons in total.
When I read Dr. Spudis’s comment I immediately emailed William Boynton of the Lunar and Planetary Laboratory at the University of Arizona, one of the authors of the Sanin et al paper, to get his reaction. Today he sent me the following detailed explanation, describing the basis of the controversy and why he believes the LEND data is valid.
Paul Spudis questions the validity of the LEND data citing a recent manuscript that claims that all but a few percent of the signal in the LEND collimators is due to hydrogen. This manuscript is wrong in its quantitative assessment of the amount of signal due to hydrogen, and Spudis is wrong to suggest that our “conclusions are seriously flawed in almost every respect.”
LEND achieves its high spatial resolution by virtue of surrounding four of its neutron detectors with tubular shields, called collimators. The collimator tubes are open at the ends pointing toward the lunar surface allowing neutrons from directly below the spacecraft (within ± 5.6 degrees, which is about a 10-km diameter circle at the 50 km altitude of LRO) to be detected by the sensors located at the opposite ends of the tubes. The collimator walls cannot be made completely opaque to neutrons, so some small fraction of the neutrons outside the field of view of the opening will also be recorded by the sensors. The concern in the quality of the LEND data revolves around a disagreement concerning the amount of signal that is due to neutrons coming through the shield vs. those coming from the field of view.
Without getting into all of the “inside baseball” arguments on the proper way to estimate the contribution of the background signal coming through the walls of the collimator, it is instructive to simply look at the results. In the figure on the right I show a trace of the count rate measured in the four collimated sensors relative to the counts recorded in a presumably low-hydrogen area at low latitudes. Also shown is the surface elevation of the Shoemaker crater as determined by the Lunar Orbiter Laser Altimeter (LOLA) on LRO. An increase in the amount of hydrogen is accompanied by a decrease in the count rate of neutrons as seen here at the location of the crater. The data are taken along a line of constant longitude (45° and 225°) passing through the center of the crater. It can be seen that a very significant decrease in the flux of neutrons is seen that coincides with the topography of the crater.
The statistical uncertainty of the data is also shown. This uncertainty is something like the margin of error that one often hears associated with surveys of people on a particular subject. When scientists make measurements, there is always some error associated with the measurement. For example one could measure the diameter of a rod with a ruler and get it accurate to about 1/32 of an inch. If one could measure it with a good pair of calipers, one could get an accuracy of about 1/1000 of an inch, but there is still some uncertainty. The uncertainty shown implies that the true count rate of lunar neutrons at the location of the crater is likely to be between those limits of the value actually plotted. In this case the measured value is 0.18 counts per second (cps) less than that of the low-hydrogen region, but with the uncertainties, the true value is likely to be anywhere between 0.165 and 0.195 counts per second below that of the low-hydrogen region. In this case, even with the statistical uncertainty, we can be absolutely certain that the depletion of neutrons seen at the crater is real, and it indicates an enhanced amount of hydrogen in the crater compared to its surroundings.
The way we know that this suppression is due entirely to signal in LEND’s field of view is because the background comes from a very large region that extends all the way from horizon to horizon. The area over which the background is generated is so large, that moving the spacecraft a small distance, say 50 km away from Shoemaker, will have a negligible difference on the amount of background that must be subtracted. I have calculated the difference in background and find that it changes by only 0.002 counts per second over a 50 km distance. The disagreement between the two groups on the amount of signal coming through the walls of the collimator ranges between factors of 1.2 to 2.2. Even if I multiply the difference in background by the factor of 2.2, the change in background over 50 km is still just 0.005 counts per second, which is still completely insignificant compared to the observed suppression of 0.18 counts per second.
It would be better if the two groups could agree on the extent of the background leakage through the walls of the collimator, but the bottom line is that the data can speak for themselves. LEND has shown here that it can detect neutron suppressions with very high spatial resolution.
So, is the Moon a desert or not? I suspect we really won’t know until we get there. However, if I had to guess, my skeptical nature would favor a drier environment. It is remarkable enough that there is any water ice on the Moon. To believe that — as some data has suggested — there is a lot there is very counter-intuitive. And now that at least one instrument on LRO has found a lack of water at most locations is further confirmation of this intuitive conclusion.
Moreover, if I was a mission planner aiming to build a base on the Moon, I would take the LEND data very seriously and aim for those few craters (Shoemaker, Cabeus and Rozhdestvensky U) that LEND says do have water. Other locations may have water ice, as suggested by other instruments, but if they don’t, as LEND’s data indicates, any lunar base at these locations would have very significant problems. Better to be safe and aim for the most likely places to find water.
One interesting side note: The LEND instrument was built in Russia, and was chosen over another instrument designed by American scientists, some of whom are apparently leading the criticism of the LEND instrument. This makes me wonder if a political turf war battle is contributing to this conflict, partly fueled by leftover Cold War resentments. Just a thought.