[600MRG] TA/TP wspr correlating to phases of the moon?

[James Hollander]

Comments from the 630m community are invited, especially anysolar/geomagnetic correlation with TA/TP receptions in the last month.  
73,  Jim Hollander  W5EST
 
630M TRANSATLANTIC/TRANS-PACIFIC (TA/TP) WSPR AND LUNARPHASES
    If TA/TP occurrencesand SNR levels are correlating with lunar quarters (first, full, third, new),this is not accidental.  The lunarquarters pertain to the four points in the 28 day lunar period of revolutionaround the Earth at which the moon is essentially in-line with or perpendicularto the line between the Earth and the sun. 
    The TA/TP opportunitytime window is defined by sunset and sunrise at the respective station locations involved,with the time window for those two stations starting about one hour or later after the relatively-westwardstation’s local sunset (Zulu) and ending about a half-hour or earlier beforethe eastward station’s local sunrise (Zulu). Consequently, the midpoint (Zulu) of the TA/TP opportunity time window, forwhatever TA/TP signal path, is roughly midnight locally over the signal pathmidpoint.  This means the line betweenthe Earth and the sun passes through the meridian over the signal path midpointat that time.  For the moon to be at the longitude of the signal path midpoint duringthe TA/TP window therefore implies fullmoon or new moon.  The other twoquarters are the neap tide positions of the moon relative to the sun.   Occurrence of the neap tide low-tide times during the TA/TPopportunity time window implies first or third quarter.
 
TENTATIVE PREDICTION ABOUT TA/TP IN FALL 2014
    I venture that a weekly lunar pattern in occurrence of TA/TP existsand will persist for perhaps another month, but will soon be overtaken by moreTAs that are more evenly spread through each week.  The weekly lunar pattern--period one week--is supported by databelow. The upcoming more-even spread in TA will be due to a noisyrising/falling TA SNR level having a one-week period. Only the TA SNR peaks arerising above wspr decode threshold now in late northern hemisphere summer, but more parts of the week-long SNR "wave" are likely to rise abovedecode threshold as fall commences and progresses.  
   Regarding TP, a spotty lunar pattern may persist in N.America/VK-ZL trans-equatorialTP for the next six weeks, whereupon VK-ZL TP will fade out and then resume in the February-through-April timeframe centered around spring equinox.  Northern Pacific TP such aswith Japan will be more likely in the November, December, January period whensufficiently-long midwinter nights can open opportunity time windows not onlyfor the higher latitudes in western N. America but also with lower latitudes incentral and eastern N. America.  
  Trans-equatorial TP is different from TA because the TP time windowinterval does not change much in duration. Also the equatorial anomaly in the ionosphere is near the TP signalpath midpoint.  By contrast, the TA time windowto UK/EU lengthens significantly as we go into fall, so eastern USA/CanadaTA improves and TA begins for central and western areas.  Moreover, TA isentirely in the northern hemisphere.  The mid-winter TA path from USA to UK/EU inherently lacks summer effects. By contrast, summer inVK-ZL degrades a mid-winter TP path to N. America. Until a South Africa station or otherequatorial/south African 630m wspr receiving or transmitting stationbecomes available, TA will not involve summer effects in the northernhemisphere’s mid-winter.
  The 630m community’s alertness to solar and geomagnetic indices iswell-placed regardless of whether a weekly lunar effect is present or not.  Multiple contributing factors to thevariations in 630m propagation are of course to be expected.
 
COMPETING LUNAR TIDAL HYPOTHESES. HYP. #3 IS BEST SUPPORTED.
     HYPOTHESIS #1:Full/new moons and lunar/solar eclipses tidally lift ions at signal mid-pathinto high F-layer altitudes and increase DX rather than first/last quarter moonwhich would not lift ions this way.  http://en.wikipedia.org/wiki/Atmospheric_tide; http://en.wikipedia.org/wiki/Ionospheric_dynamo_region (The mid-path is a common point ofionospheric reflection of signal for all the odd-numbered multihop propagationrays.)   Evidence for Hypothesis #1: 
        NEW MOON
8/25/14 WG2XIQ into -30dB at VK2DDI
8/26/14 WG2XJM 3 TAs -21dB G3XKR, -19dB G3HUH, M0LMH (? onmap not database)            
            WE2XGR/3-27dB G3XKR.
            Storms inVK no reception.
8/27/14 WD2XSH/20 one undecode by VK2DDI 
     FULL MOON
9/07/14 WD2XSH/17 TA DK2FC
9/08/14 DK7FC 1w into 6sp pk -24dB WD2XSH/17.  
             G0PKT .2winto -26dB W4HBK.
 
    HYPOTHESIS #2:Moon in first/last quarter has neap tides displaced away from signalmid-path.  The quieter ionosphereconstitutes a less-disturbed "superior mirage" reflector and shouldincrease DX, rather than full/new moon.  Evidence: 
       3rd QUARTER
8/15/14 DL5RBD rx VE3PLE TA.
8/17/14 WG2XKA TA GM4SLV
       1st QUARTER
9/01/14 WG2XIQ -28dB into ZL2AFP
             WH2XGP-27dB into ZL2AFP
        WD2XSH/20-25dB into VK2XGJ
 
    HYPOTHESIS #3: Allfour quarters (first, full, third, new) produce weekly TA or TP opportunitiesbecause of more-nearly-pure tidal or non-tidal effects.  Also, interactions of tides with equatorialanomaly may turbulently disturb propagation because mid-path USA/VK-ZL liesnear the equator.   Evidence:  
     Bunching of TA/TP around all fourquarters as above.  
     No TA/TP whateveroccurred in last month except around all four lunar quarters.  
 
    HYPOTHESIS #4:Moon/sun tides in ionosphere exist but amount to insignificant effects onionospheric DX propagation compared to solar and geomagnetic events and randomarrangements of ionospheric clouds and reflecting surfaces within the E/Flayers.  Evidence: <please fill in, JHnot monitoring solar/geomagnetic indices>.  Eloquent verbiage can make any hypothesis sound plausible. Hypotheses ##1-3lack credibility because no conclusions are justified on only one month of datain late summer in only one year.  Thesun’s synodic rotation period 26¼ days makes sunspots on the Sun rotate to thesame apparent position seen from Earth, similar to the moon’s 28 day period. (Butin favor of Hyp.#3, if rotation of the sun or solar events were wholly responsiblefor TA/TP SNR, wouldn’t they produce a less-periodically occurring, more erratic set of TA/TPevents over the days?)
 
STATISTICAL CONFIDENCE LEVEL FOR LUNAR HYPOTHESIS #3
    A conservativeestimate of the probability of Hypothesis 3 occurring by random chance assumesthat every “1/3 week” has an unknown but equal probability P of one or both ofTA and TP occurring at least one-way between USA and EU or USA and VK-ZL. Thenconsider a probability p of the4-week pattern with TA/TP actually observed during one “1/3 week” interval andnot observed during the next two “1/3 week” intervals in every week of the fourweeks.  That probability p of the 4-week pattern wouldapproximate the product of exponentiated quantities
   p = P^1x4(1-P)^2x4  = P^4(1-P)^8  
Since TA/TP probability P is unknown, assume the worst caseP value at which the one-month pattern probability p is maximum.  Thisestimates a confidence level p of thepattern.  The analysis does not purport to be an estimate of theTA/TP probability P except as a worst case value for confidence p estimation purposes only.  Using calculus to determine P at the themaximum for confidence p,  
  P = 1/3
  p=  .5 x .001 = (1/3)^4(1- 1/3)^8  = 2^8/3^12 =256/531,441 
This confidence level pis favorably significant for earth science purposes. In statistics-speak, confidence level p< .001 or triple-star (***) on results tables. 
 
Because the moon is the only nearby celestial object havingweekly quarters, the conclusion follows that this confidence level p should be associated with alunar-phase-related pattern in the ionosphere. 
 
This confidence level pcalculation assumed that the TA/TP probability P was constant—that TA/TP occurs completely randomly.  However, solar and geomagnetic influences areat work on the ionosphere so TA/TP is not random.  Accordingly, further work would be needed toevaluate the relative importance of lunar and solar/geomagneticinfluences.  Moreover, the equatorialanomaly is not included in the probability estimates. 
--end--