[Crater-team] Another six months of GCR LET spectrum monitoring

[Mark D Looper]

Hello—

                Man, I wish I’d done this yesterday…. I just added the next six-month period to the time series of GCR LET spectra from D5/6, and in the course of assembling the precursor files I realized that some of the processing steps I had done before were incomplete the last time I ran them.  That is, one file that I use to determine solar quiet periods had only been updated to day 330 of 2019, so the code I used to calculate the GCR LET spectrum from 2019/248 to 2020/065 silently used only half that period, through 2019/330 – and this was the time period that I compared, in the paper I _just_ finished resubmitting, with the time period at the start of the mission to discuss how elevated the spectrum was in this solar minimum compared to the last!  I didn’t realize this until the calculation for 2020/066 to 2020/247 crashed because there were no days from that period in the quiet-time file.  It doesn’t change the conclusions in the paper, but it would have let me push the statistical noise in the comparison down by a third or so.  Aaargh!

                Anyway, I attach a couple of plots that replicate the comparison in the paper, but that add the new time period.  These are GCR LET spectra multiplied by LET squared to reduce the dynamic range and bring out differences better.  The four curves are red, start of mission (solar minimum); green, deepest suppression of GCRs (solar maximum); blue, last period in the paper (new solar minimum, with all the days’ data this time); and black, new time period (turning over toward next solar cycle).  The file with suffix _mag is magnified in the H/He region, which is where the statistics actually support a comparison, and you can see that the new time period is a couple of percent higher than the blue one, which was 4% above the red from last solar cycle!  So, through a couple of weeks back anyway, solar modulation has continued to decline.

                Actually, I had been thinking about this particular subset of the data some more lately, especially after reading Wouter’s and Fatemeh’s papers.  The humps in the attached figures due to relativistic protons and alphas at about 0.35 and 1.4 keV/micron are of course less modulated than are the spans to the right of each of them, which are due to subrelativistic protons and helium.  This is visible in the relative differences between the red and blue curves at each LET, as well as in the colorscale plot of LET vs. time that’s in the paper I just resubmitted.  Those particles constitute the “swooshes” off to the right from the diagonal in a plot of, say, D4 vs. D6 energy deposits; unlike the albedo doubles, though, we have a third pulse height for these on which to impose a consistency check.  Thus, in spite of the scruff due to all the inert TEP lying around, the lack of active collimation, etc., we can probably isolate these particles pretty cleanly, and have a good measurement of flux at least in this fairly narrow energy range.  Moreover, we have similar three-detector swooshes that are pretty well defined for a variety of heavier GCR species.  So we can get clean samples of these GCR species; I can use the simulations to pin down the energy ranges based on the cuts that I’ll devise, and I can also correct for any losses due to nuclear interactions in the stack (we can’t just assume unit counting efficiency within the swoosh energy range).

The reason Wouter’s and Fatemeh’s papers got me thinking about this is that they use the dose to back out the modulation based on modeling of the sensor’s dose response to changing GCR spectra.  Their dose time series has good statistics since it collects (my estimate) about half of the GCR events, which is way more than is in the swooshes; however, a lot of that dose is due to the less-modulated relativistic ions.  If I can extract from these swooshes a time series of multiple GCR species, with a stronger response to modulation than the dose and with absolute flux calibrations tuned via the simulations, would this hand you a better, or at least a different, “lever” with which to extract the time series of modulation?  Of course I owe Phillip and Tim the analysis that I’m doing now of nuclear showers in shielding, and I owe Fahad and Larry the analysis of the code-comparison albedo simulations that I’ll do next, and I still have that paper to write on the simulations of lunar albedo in the presence of hydrogen that I’ll do eventually; but with the GCR LET paper finally (I devoutly hope!) off my desk, maybe I’ll take a little time to see what kind of statistics these swooshes will support for each element on how fine a time cadence.

Let me know what you think—
--Mark

Mark D. Looper
Space Sciences Department
The Aerospace Corporation
M/S M2-260
P.O. Box 92957
Los Angeles, CA 90009-2957
Mobile: 310-529-3406
Voicemail: 310-336-6302
Publications: https://www.loopers.org/curvitae.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.sr.unh.edu/pipermail/crater-team/attachments/20200924/b8a4e2ae/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: d56combine_21_l2_rev3_mag.png
Type: image/png
Size: 79767 bytes
Desc: d56combine_21_l2_rev3_mag.png
URL: <https://lists.sr.unh.edu/pipermail/crater-team/attachments/20200924/b8a4e2ae/d56combine_21_l2_rev3_mag-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: d56combine_21_l2_rev3.png
Type: image/png
Size: 67023 bytes
Desc: d56combine_21_l2_rev3.png
URL: <https://lists.sr.unh.edu/pipermail/crater-team/attachments/20200924/b8a4e2ae/d56combine_21_l2_rev3-0001.png>