[Crater-team] Status of follow-up on last week's discussion [was: Re: CRaTER mtg today]

Wed Jul 17 16:32:48 EDT 2019

Hello--

	Many thanks -- I see at least one big area where we were talking past each other last week.  I thought you were saying that proton _albedo_ resulted only or at least mostly from He and greater GCRs; what you actually said was that the proton albedo _increase_ with a wet surface layer only resulted from these.  This renders most of my comments in (3) moot; good, since that was the longest section!  I remain concerned about how the mostly-downward-aimed high-energy neutron fragments would scatter back upward in such a way as to produce the strongly collimated and energetic vertical beam of albedo protons that is evident in the difference between "Wet_Regolith_Albedo_Protons.pdf" and "Dry_Regolith_Albedo_Protons.pdf."  I also note the colorscale label, counts per (source particle) per MeV per degree; if the "per MeV" and "per degree" refer to the energy and angle of the albedo particles, as seems to be the case, then scaling the angular bins to be "per sr" would enhance near-normal colorscale values relative to near-horizon values.  This would probably flatten out the contours for the "Dry_Regolith" case so that they are nearly horizontal, i.e., isotropic, as in my "all processes" plot, "all_dry_3_1.png", but it would also even more strongly enhance the vertical collimation of the albedo proton beam in your "Wet_Regolith" case.  If you multiply out the "per (source particle)" part using your assumed GCR input fluxes and convert the albedo to physical flux units per (m^2 sr sec MeV), accounting for the cosine factors as on slide 3 of "angular_normalization.pptx" dated October 3, 2018, then we will have a basis for quantitative comparison of the different sets of simulations with the observations.

	I'll keep working on (1) and (2).

Many thanks--
--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

On 7/17/19, 9:38 AM, "de Wet, Wouter" <Wouter.deWet at unh.edu> wrote:

    Hey Mark,

    Thanks for the thorough and well thought out response! This is very helpful for me to interpret what you were saying last Wednesday. 

    Do the two modeling efforts disagree?

    	I don’t think there is substantial disagreement in the modeling results. It seems like we are getting similar results; although, perhaps we are interpreting them differently. The attached plot of 'Dry_Regolith_Albedo_Protons.pdf' shows the total albedo spectrum as a function of energy and angle for dry regolith. If I understand your plots correctly (and please correct me if I'm not) this plot should be the same as the 'all_dry_3_1.png' you shared. At first glance, they seem to be telling the same story: The behavior in energy and angle (apart from very slightly surface glancing angles) are roughly the same. Also, both plots exhibit roughly 4 orders of magnitude variability in flux between glancing and normal emission angles. When we map what the population of protons that are observable by CRaTER via D6D4 coincidence in Nadir viewing mode (see energies above ~60 MeV in 'Dry_Regolith_Albedo_Protons_D6D4_Nadir.pdf') it confirms we are not likely to experience any protons above 1 GeV/n. 

    	Wet regolith results:

    	However, things are different if we consider the (very) wet regolith results. Keep in mind that we used a whopping 1 cm, 10% H2O by mass, regolith layer on the surface. The presence of a standing population of hydrogen near the surface changes the Nadir albedo spectrum somewhat dramatically (see 'Wet_Regolith_Albedo_Protons.pdf'). Note that the wet simulations had a global lower energy cutoff placed at 60 MeV/n to speed things up. The albedo proton spectrum for the wet case is nearly identical to the dry case for angles greater than 40 degrees off-normal. The primary difference between the wet and dry cases lies in angles less than 20 degrees off-normal. The flux of protons coming straight up from the surface (~0 degrees off-normal) at energies around 1 GeV is about 4 orders of magnitude greater than the dry case. When this is mapped to show the population of protons that are observable by CRaTER via D6D4 coincidence in Nadir viewing mode (see 'Wet_Regolith_Albedo_Protons_D6D4_Nadir.pdf') this shows that there is a not-quite-insignificant population of 1 GeV protons to be seen by CRaTER. I'm showing D6D4 coincidence mappings only because we haven’t made the triple coincident correlations yet. I expect they will tell a similar story in energy, and will just be more narrow in the emitted angles that are observable.

    Where are the albedo protons coming from?

    	"I did not understand why GCR protons should not scatter up as well as any fragments of heavier nuclei.  Moreover, my own simulations produce albedo protons from GCR H, He, and heavies roughly in proportion to their primary intensities, i.e., 100:10:1; this and the fairly isotropic upward-going distribution of protons (and other albedo species) suggests that the source of most albedo particles in my simulations is isotropic disintegration of regolith nuclei after being struck by a GCR primary, regardless of species."

    	I think I see where the miscommunication/confusion is coming from, but I am not sure I can communicate it well. When a nuclear collision (at these energies) occurs, there are generally two populations of secondaries: the projectile fragments, and the target fragments. The projectile fragments will be very similar to the initial projectile in direction and energy. The projectile fragments are actually emitted isotropically in the rest frame of the original projectile nucleus, but their emission energies are on the order of a few MeV/n, so it does very little to change the momentum in the stationary observer rest frame. However, after these secondaries undergo enough collisions of their own in the regolith, their direction has been slightly changed so many times they will become isotropic. Likewise, the energy and direction of the target fragments, as you correctly stated, generally have very little to do with projectile nucleus. The target nucleus, once struck, will de-excite by breaking into smaller fragments via the same de-excitation mechanisms as the projectile nucleus. The only difference is that the rest frame of their parent nucleus (the target) is the same as the stationary observer, so they are emitted isotropically with kinetic energies of singles of MeV/n. The target fragments typically do not travel very far because of their low energies. An exception to this is nucleon-nucleon collisions (such as neutron striking a stationary H nucleus). In this case, the target can inherit up to 100% of the projectiles kinetic energy, although on average it only inherits 50%.

    	When we run the wet regolith simulation with just GCR protons as a source term, it looks nearly identical to the dry case. When we include He and heavier ions in the source term, suddenly this population of high energy protons coming from near-normal angles appears. In order for this new population of albedo protons to appear you need both the standing layer of H near the surface and GCR source particles that can produce high energy neutron fragments. The heavy projectiles are fragmenting and their high energy neutrons are bouncing around inside of the regolith. When these neutrons strike the standing population of H near the surface, the H nucleus inherits much of the neutron's kinetic energy, allowing them to escape the regolith.

    Does that help solve the miscommunication, or have I just made things even more confusing?

    -Wouter

    On 7/17/19, 4:00 AM, "Crater-team on behalf of Mark D Looper" <crater-team-bounces at lists.sr.unh.edu on behalf of mark.d.looper at aero.org> wrote:

        Hello--

        	Last week I kind of hogged the telecon time based on having had about 15 minutes to look over the slides that were attached to the email below.  I figured if I was going to say anything else, I should put in more time than that thinking them over; however, I have not finished what I had hoped to send to the group within the week, as demands from other projects intervened.  Below are the calculations I am in the middle of; I attach two figures, to which I will refer briefly in (3).

        1) I am in the middle of stripping the mission-long dataset into a D2/D4/D6 cube of PHA channels, so that I can slice and dice it both to replicate the plots on the attached slides and to see if this gives any additional insight into the origin of the particle-event populations under discussion.

        2) You will recall that I expressed concern as to how directional the response identified as albedo MIPs would actually be.  That is, if we see a big D6 pulse height and a D4/D2 pair consistent with an upward-going proton penetrating both of them, can we be sure that the primary particle apparently striking D6 came from below?  To address this, I am preparing a simulation of a simplified lower-half of CRaTER, with D6, the TEP, and the surrounding aluminum (approximated as cylinders), so that I can illuminate this from all directions with GeV protons and see how collimated are the subset of these that result in modest-energy protons coming out the other end of the TEP within the D4/D2 acceptance cone.

        3) You will also recall that we had some discussion as to how many albedo MIP protons would actually be present, given that the albedo spectrum is falling at energies above the D4/D6 doubles energy range and we already don't see any distinct population of D2/D4/D6 albedo triples at the lowest possible energies for those.  If I understood last Wednesday's discussion correctly, the assertion was that the source would be energetic fragments from He and heavier GCR nuclei that scatter in the regolith and come back out.  Also if I understood correctly, the simulations being used in this discussion do not produce many albedo protons from GCR protons, but only from GCR He and heavier, which is why the source is identified as being from such fragments.  I did not understand why GCR protons should not scatter up as well as any fragments of heavier nuclei.  Moreover, my own simulations produce albedo protons from GCR H, He, and heavies roughly in proportion to their primary intensities, i.e., 100:10:1; this and the fairly isotropic upward-going distribution of protons (and other albedo species) suggests that the source of most albedo particles in my simulations is isotropic disintegration of regolith nuclei after being struck by a GCR primary, regardless of species.  With regard to scattering, I attach energy-angle plots of all simulated albedo protons, and of the subset of these that are due to primary GCR protons that scatter back out.  Comparing the two, my simulations assert that scattering is very much focused toward shallow angles; when one looks leftward in the scattered-primaries plot, toward upward-going rather than surface-grazing albedo trajectories, the intensity and the energy fall off very rapidly.  Either I need to be shown what I am misunderstanding in the simulations being discussed in last week's slides, or we should try to understand why our results are so different.  I thought that some time ago Larry Townsend, after dissecting the Geant4 "Shielding" physics list and throwing the appropriate swi
tches in his codes to use the same models, was able to get results similar to mine; if our results really differ as much as last week's discussion led me to believe that they do, we should figure out which set of assumptions/models is more representative of the actual processes, by comparing predicted albedo spectra with observations.  I did a preliminary (no GCRs heavier than He) version of this in the 2013 Space Weather paper, and if I ever get around to finishing the two papers that I'm trying to get out the door, I should include an update; Joe Mazur already did a comparison with whole-moon gamma and neutron observations from other spacecraft, getting pretty good agreement with the simulations.

        4) Sorry, that was pretty darn long.  If (1) and (2) do not persuasively identify the particle population in question, I can slice and dice the results of my simulations of the sensor-head response to see what kind of particles (species, energy, incidence direction) are predicted to produce a signature like that being discussed in last week's slides.  I'll send out another PPTX when I get (1) and (2) done, and if necessary (4).

        Let me know if we have a telecon tomorrow/today--
        --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

        On 7/10/19, 7:29 AM, "Crater-team on behalf of Andrew Jordan" <crater-team-bounces at lists.sr.unh.edu on behalf of ajordan at guero.sr.unh.edu> wrote:

            Hi folks,

            Attached are some slides I can talk about. They're about how to get 
            directional information from some minimum-ionizing protons in CRaTER.

            Andrew

            On 7/10/19 11:33 AM, Sonya wrote:
            > Team,  sorry about the late notice, but if anyone is available we will go ahead and have a call at 2pm today.
            >
            > Usually call in #.
            >
            > Sonya
            >
            > _______________________________________________
            > Crater-team mailing list
            > Crater-team at lists.sr.unh.edu
            > https://lists.sr.unh.edu/mailman/listinfo/crater-team
            >