To: plucinsk Subject: EPHIN Analysis/ACIS Thresholds/CME results cc: das, rac, ybutt, swolk Date: Wed, 20 Oct 1999 03:16:20 -0400 From: Shanil Virani Hi Paul, My scripts work and I am finally done my analysis! Well, partially; I still want to do my leakage current analysis tonight but I will get to that in a moment. Anyway, since the results and plots I am about to discuss cut across several different topics, I've included Dan, Yousaf, and Rob on this distribution. Once we've had a chance to talk, show to Mark and discuss further, we can combine the following text plus additional comments/corrections into one mammoth memo. First of all, what you need for this email to be meaningful is the many, many plots I have made. You can find them physically at /data/acis0/svirani/ephin/radhist or on the web at http://asc.harvard.edu/acis/radbelt/ Yes, I know this amounts to MANY, MANY plots and pages currently, but I think the fine resolution this affords is worth it at this point in time given the context of establishing ACIS threshold limits. For an official memo, I will change the x-scales such that the entire 80 day data set used in this analysis spans 2 pages at most. That is, instead of having a two day x-length per plot, I will make it 8 days per plot. At any rate, that is for a future memo and is open to discussion. After reading this email and looking over my plots, you can decide whether to forward this email to Martin/Harvey/Mark, etc. At any rate, if you go to the above path, the files you want to print are sce150-gm-*.ps sce300-gm-*.ps sce1300-gm-*.ps sce3000-gm-*.ps scp4-gm-*.ps scp8-gm-*.ps scp25-gm-*.ps scp41-gm-*.ps scinteg-gm-*.ps DATA REDUCTION The files listed correspond to all FOUR electron channels, all FOUR proton channels, and the INTEGRATED channel. Here is a handy table that gives the relevant energies for each scientific channel: EPHIN CHANNEL ENERGY RANGE (MeV) E150 0.25-0.70 E300 0.67-3.00 E1300 2.64-6.18 E3000 4.80-10.4 P4 5.0-8.3 P8 8.3-25.0 P25 25.0-41.0 P41 41.0-53.0 INT e > 8.7, p,h > 53 Dick Edgar maintains a FITS file which has all thirteen EPHIN scientific channels counting rates along with a time column (amongst others). His data starts at day 207 and ends at day 290. For this analysis and the plots produced, I've used data from 210-290. Note that we did not reach our final orbit until day ~215. Specifically, the 8 electron and proton channels used for this analysis are: SCE150, SCE300, SCE1300, SCE3000, SCP4GM, SCP8GM, SCP25GM, and SCP41GM. The "M" in the proton channels refer to the middle portion of the G detector. As such, they are *geometry independent*. That is, while EPHIN does autonomously switch between large and small geometric factors depending on the radiation environment, the proton channels DO NOT need to be corrected for geometry since the middle detector is always on. The electron channels, however, must be. During low radiation environment periods, all EPHIN detectors are on and hence the "large" geometry factor should be used. During high radiation environments (ie, during perigee passage for instance), EPHIN uses only the middle detectors. Hence, the "small" geometric factor must be used. Below is the geomtry factor table for electrons. (Geometry factor has units of cm^2 sr.) EPHIN LARGE SMALL E150 0.25 0.01 E300 1.78 0.14 E1300 2.01 0.12 E3000 1.58 0.09 Therefore, to correct for the geometric factor, once must divide by the appropriate factor. Within Dick's FITS file is a HKFAILMODEA column; this identifies periods of high or low radiation periods. When EPHIN experiences high radiation periods, it has a value of 65; for periods of low radiation, it has a value of 127. To account for this correction, I wrote a simple IDL program that checks each element of HKFAILMODEA and then divides each electron channel by the corresponding geometric factor. The plots cited above then are EPHIN electron, proton, and the integrated channel rates vs. time for the duration of the mission thus far. The counting rates are in units of counts/integration time/cm^2/sr; the integration rate for EPHIN is 65.6 seconds. NOTE THAT PROTON PLOTS ARE INCORRECTLY LABELED. THE UNITS SHOULD BE JUST COUNTS/INTEGRATION TIME. The last key data set used in this analysis are the tabulated list of perigee crossings thus far. At http://hea-www.harvard.edu/SOT/chron, one can find the start and end times of the electron 1,2 and proton 1,2 periods. I am uncertain at this point what each refers to. The above web site has taken those times starting from the first daily load. In particular, the file begins on Day 226. To get times prior to then (ie, from Day 215 to Day 226), I accessed the definitive orbit files located at /dsops/GOT/aux on the OPS LAN. However, as you will see in the plots, these time spans are highly suspect. One other point that I was not able to determine, is what the "official" start time and end time of the radiation belts are. That is, that span of time during which the SI have been safed and absolutely no science is taking place. Since the electron 1 time span is the longest of the 4, I will use this time span as a proxy for the radiation belt duration. Before beginning the analysis and results portion, let me just briefly state that I did look at the lowest energy helium channel (SCH4GM), however, there is virtually no data to report so I have not included any helium data in this report. ANALYSIS AND RESULTS The data and plots produced will attemp to address three outstanding issues: ACIS thresholds, radiation belt duration, and lastly, the possible effect/arrival of the series of CME and solar activity that occurred last week. The first point to note is the well-established comment pointed out by Reinhold Mueller-Mellin (EPHIN IPI). That is, the EPHIN scientific channels saturate during perigee crossing so that counting rates reported during this time span are nonsense. However, since perigee behaviour is not the focus of this analysis, this observation does not affect my results. For EPHIN behaviour during perigee crossings, please see my memo entitled "Using EPHIN's Detector Leakage Current To Understand Radiation Belt Transits". Secondly, in order to bring out the small-scale variations (when compared to variations during perigee crossing) of each channel, I have suppressed the y-scale such that these variations can be observed. By observing how these particles behave outside of the radiation belts, we can determine crude ACIS thresholds for each channel. However, this analysis will need further input from Mark (and please see Steve Odell's reply to my email yesterday) before final values arrived at -- hopefully, at this afternoon's meeting with Mark/Martin, etc. Lastly, superimposed on each EPHIN plot are four horizontal lines: top horizontal line spans the proton 2 period, the next line down spans the proton 1 period, the third line spans the electron 2 period, and the bottom line spans the electron 1 period. Note there were FIVE (5) perigee crossings where there was no PROTON 2 time span; I do not know why. Without further information, I will assume the electron 1 period corresponds approximately to the "radiation belt transit" period. That is, that span of time during which the instruments are powered down, properly protected, and absolutely no science is taking place. Consider first the SCE150 (electron) channel vs time plots. As I stated above, I have very little confidence that the start and stop times I retrieved from the ephemeris files for Day 215-Day 226 (inclusive) are correct. However, if they are, it suggests that perigee crossings on Day 215 and Day 218 were longer than the time span of electron 1 and electron 2. Similarly, perigee crossings on Day 220, 223, and 226 began substantially earlier than any of the electron/proton periods. One can repeat this process for each transit but since I would like to get home before sunrise tonight, I will only point those passages for which there are signifcant discrepancies. We can talk about this more later today and elaborate further in a memo. To first order, I've simply tabulated each perigee crossing as either "good" or "bad" where "good" is defined as adequate electron 1 or 2 span of the perigee crossing; "bad" of course is the inverse. So by that definition, perigee crossing on Day 223 would be "bad" and perigee crossing on Day 228 (the first daily load data) is "good". At any rate, please look at each perigee crossings cited in the table below DAY PERIGEE QUALITY ------------------------- 228 GOOD 231 BAD 233 marginal/bad 236 marginal 239 marginal/bad 241 bad 244 mariginal/bad 247 bad 250 good 252 marginal 255 marginal/bad 257 bad 260 bad 263 bad 265 good 268 good 270 bad 273 bad 276 good/marginal 278 bad 281 bad 284 bad 286 bad 289 marginal/bad NOTE: There APPEARS to be slight raise in E150 profile beginning about DAY 287.5 until 290; this may be the affect of solar activity/CME. Also, to get the full import of perigee quality, it would be useful if you had TABLE 1 from my perigee crossing memo to compare to -- recall that table 1 provides the relative strength of one perigee crossing to another. Repeating the above analysis for SCE300: DAY PERIGEE QUALITY ----------------------- 228 GOOD 231 BAD 233 good 236 good 239 marginal 241 good 244 mariginal/bad 247 bad 250 good 252 good 255 good 257 good 260 marginal 263 bad 265 good 268 good 270 bad 273 marginal/bad 276 good 278 good 281 good 284 marginal 286 bad 289 marginal NOTE: There APPEARS to be slight raise in E300 profile beginning about DAY 287.5 extending until 290; these may be the affect of solar activity/CME. SCE1300: DAY PERIGEE QUALITY ----------------------- 228 GOOD 231 BAD 233 good 236 good 239 good 241 good 244 marginal 247 marginal 250 good 252 good 255 good 257 marginal 260 good 263 bad 265 good 268 good 270 marginal 273 bad 276 good 278 good 281 good 284 good 286 bad 289 marginal NOTE: There APPEARS to be a very (?) slight raise in the E1300 profile beginning about DAY 287.5 until 290; these may be the affect of solar activity/CME. SCE3000: DAY PERIGEE QUALITY ----------------------- 228 GOOD 231 good/marginal 233 good 236 good 239 good 241 good 244 good 247 good 250 good 252 good 255 good 257 good 260 good 263 marginal/bad 265 good 268 good 270 bad 273 marginal 276 good 278 marginal 281 good 284 good 286 margina/bad 289 good NOTE: There DOES NOT appear to be any rise in the E3000 profile beginning about DAY 287.5 until 290. GENERAL ELECTRON RATE CONCLUSION: We appear to have the radiation belt transits timed well from the point of view of high energy electrons, but the low energy electron plots clearly demonstrate a (significant?) problem. Let's now repeat the entire analysis for the for proton channels and the integrated channel. SCP4GM: (Recall definitions of good and bad relate to ELECTRON 1 TIME SPANS. In hindsight, I think a better way of measuring radiation belt duration would be to use the earliest time from any electron/proton period and the latest time from any electron/proton period; I do not think my results would change drastically, however.) DAY PERIGEE QUALITY ----------------------- 228 GOOD 231 BAD 233 good 236 good 239 marginal 241 marginal 244 marginal 247 marginal 250 good 252 good 255 good 257 bad*************************** NOTE THE INTERESTING BEHAVIOUR 260 marginal/bad 263 bad 265 good 268 good 270 bad 273 bad 276 good 278 marginal/bad 281 good 284 marginal/bad 286 bad 289 marginal NOTE: There APPEARS to be slight rise in the P4 profile beginning about DAY 287.5 until 290; these may be the affect of solar activity/CME. Also note the particularly odd behaviour on Day 257. DOES THIS NOT CORRESPOND TO AN EARLIER SOLAR FLARE/ACTIVITY??????????? Also, the analysis of the P4 results is particularly alarming. Low energy protons (E ~ 100 keV) have been implicated as the cause of our ACIS cti degradation. The above analysis for protons with 5.0