I was really surprised that the band was as good as it was! It seems that the current G3 storm has primarily created wild QSB while leaving many of the typical paths open to one degree or another, at least at lower latitudes. If one could make it in between fades, it was probably possible to find some success. I am fairly certain that high latitude stations were not so fortunate as high absorption probably dominated their session. Even so, two trans-Atlantic reports were registered, one on CW and another on WSPR. Trans-Pacific opening continue to show lots of life, including reports between VK and North American stations. As was reported yesterday, looking at the WSPR map for WH2XCR might suggest that band conditions were normal. Joe, NU6O / WI2XBQ, reported during the evening that the session seemed average with North / South opening dominating in northern California.
It was noisier than I would have liked as storms located in southeastern Oklahoma drifted South across the Red River into Texas. On a few occasions I expected that I would have to secure my stations but the storms fizzled and split before they arrived in my area. This morning QRN levels remained higher than normal for this time of year but were an improvement over last night.
Geomagnetic activity continued at G3 storm levels during the session. The Bz has been variable but is currently pointing to the South and solar wind velocities are down from the previous report, averaging 575 km/s at the time of this report’s compilation. DST values continues to suggest disturbed band conditions but both Kyoto and Australia are improved over the previous session’s reported values.
There was a significant amount of JT9 activity in Europe as Mal, G3KEV, was QRV along with Spiros, SV8CS, Marco, DD7PC (transcript), G3XIZ, SV3DVO, DK6XY, and IW4DXW, who was in QSO with SV8CS when the Earthquake hit Italy. He QRTed immediately after reporting, via JT9, that an earthquake had just occurred. That doesn’t happen every day!
VO1NA’s CW signal was reported at PA0RDT, which seems rather remarkable given the absorption on high latitude paths. Roelof reported on the RSBG “blacksheep” reflectors that “Propagation was much worse and a massive thunderstorm over Italy did not help either, However, you are doing better than NDB QY-263, Sydney, CAN NS!”
The other remarkable trans-Atlantic report was of DK7FC at WE2XGR/3. That report can be viewed here.
Phil, VK3ELV, received reports from JH3XCU and 7L1RLL4 in the previous session. Those reports can be viewed here.
Roger, VK4YB, reported, “Almost identical stats to last night although QRN was higher and forecast to continue tomorrow.” Roger’s session details not found elsewhere in this report can be viewed here. He provided the following statistics for the session:
“Rx 3*wg2xxm (-27) 41*wh2xnd (-15) 7*wh2xgp (-22) 41*wh2xxp (-13) 4*ve7bdq (-24) 46*wh2xcr (-14)
Tx 7* wh2xgp (-21) 2*ve7bdq (-26) 1*wi2xbq (-26) 6*we2xbq (-26) 28*wh2xcr (-7) 1*jh3xcu (-27) 1*jf1lks_3 (-28) 7* ja3tvf (-26)”
Ward, K7PO / WH2XXP, was decoded by 58 unique stations including VK2XGJ, VK3ELV, and VK4YB. Ward’s trans-Pacific details can be viewed here.
Ron, NI7J / WH2XND, was decoded by 54 unique stations including VK2XGJ and VK4YB. Ron’s trans-Pacific details can be viewed here.
Ken, K5DNL / WG2XXM, reports that he was decoded by 56 unique stations including VK4YB. Those trans-Pacific reports can be viewed here.
Neil, W0YSE/7 / WG2XSV, reports “…very similar to previous session. No eastern states heard nor hrd by… ve6jy, Don, rcvd my JT9 at a -25 at 1330z, 1134 km, Very reliable path apparently.” Neil also reported that VE7BDQ was decoded on JT9 at -24 dB S/N just prior to 0430z.
Dave, N4DB, reported “Only 8 uniques last night, worst ever.”
Doug, K4LY / WH2XZO, reported, “Continuation of auroral conditions: 8 decoded and decoded by 36 inc surprising 11 decodes by XCR.”
Rick, W7RNB / WI2XJQ, reports, “Conditions here are unchanged this morning, with a wall to the East. Only XXM made it through … Rain has been constant and the ground is getting pretty mushy.” Rick decoded nine WSPR stations and was decoded by twenty unique stations. His unique receive and transmission reports can be viewed here.
In spite of QRN, last night was fun for me. I opted to start WSPR late and call CQ on CW through the sunset period and into the early evening. Coinciding with dinner time here, I had to press the wireless headphones into service. I typically use these in the morning when I am wandering the house trying to get the day started and they worked ‘OK’ but lack the fidelity and isolation that my wired, 25-year old Kenwood HS-5’s offer as they squeeze my head to the point that I get a headache (Its amazing to me that they are still so tight after 25 years!). Without the wireless headphones, however, I would not be able to operate as much as I have recently so there is value, lack of fidelity and isolation or not. I heard WG2XJM calling me in the noise from the kitchen while working on dinner so they are not too poor! Eric and I engaged in two QSO’s, really. The first was on a peak during extremely unstable band conditions where Eric reported that I went from RST 549 to nothing in a two-minute cycle. Eric was fine copy but experience says that the bottom was due to drop out, which it did just after signing. We both lamented in the ON4KST chat the fact that we had hoped to make a ‘ragchew’ out of the contact but knew it would likely only lead to frustration so early in the evening. I CQ’ed a bit more and a few minutes later Eric called me again on a peak while running 900-watts TPO. He was and easy RST 599 and I was back to RST 549. We chatted for about 6-minutes and the path was solid with no hint of QSB or loss of understanding of what was being said. At one point Eric turned the amp off, using 100-watts, and was 100% copyable, only being taken out by the lightning crashes from the storms to the North of me. On any quiet night, 100-watts would have been more than sufficient but I appreciate his 900-watts TPO under this session’s conditions. We signed and I continued to call CQ for another 20-minutes as I nervously watched the storms to the North move to the South. I should note that the “phantom ditter” made numerous appearances during this CQ session. As Eric noted in the ON4KST chat, “They are chomping at the bit” referring to stations that can’t wait to get on the band.
Andy, KU4XR, recorded me calling CQ shortly before I transitioned back to WSPR and this is what I sounded like in Tennessee:
Note that there are two calling sequences and the intent here is to show the stability of the band through each. It was like someone had flipped a switch from when Eric and I made our first QSO to our second QSO and Andy’s recording. Note that this recording has been edited to remove the one-minute listening period between both CQ sequences.
This morning’s CW session was uneventful. I did not have a sked this morning so I called CQ until just before 1200z when I needed to get the day into “high gear”. Below is a screen capture from my 472-grabber showing my CW calls in addition to several WSPR signals, all of us in QRN:
WSPR reports were relatively typical, with numerous CW levels indicated although the QSB was apparent from one transmit cycle to another. Receiving was also good after QRN subsided some after local midnight. My WSPR transmission reports can be found here and my WSPR reception reports can be found here.
106 MF WSPR stations were observed at 0310z on the WSPRnet activity page which has had over 100 stations reported at some point in the session for over a week if I recall correctly. KK4WJF and NM5SW appear to be a new receiving stations in North America. Bill, PY2GN, was listening for the first night on 474.2 kHz WSPR after Doug, K4LY / WH2XZO, made contact, helping him get on the correct frequency in the 600-meter low allocation. Bill indicated in an email that he would be using an 80-meter Windom for receive. Welcome aboard!
Regional and continental WSPR activity follow:
There were no stations reporting from Africa.
Eden, ZF1EJ, experienced a good night of reporting, decoding WH2XGP and WG2XKA, both near the auroral zone, in addition to several stations in the southern US:
Laurence, KL7L / WE2XPQ, was only receiving during this session. Even under the circumstances, Laurence managed to provide reports for VK4YB and WH2XCR. Those reports can be viewed here.
Merv, K9FD/KH6 / WH2XCR, seems largely unaffected by the current G3 storm if one focuses on the map distribution of stations. Reports into the eastern areas of the US have returned, notably at WG2XJM and WH2XZO. There were reports at JF1LKS_3 and JH3XCU and two-way reports at VK4YB and VK3ELV. Living at lower-mid latitudes appears to have some advantages. Merv’s DX reports can be viewed here.
Jim, W5EST, presents, “PART 2: 160/630M WSPR 2-BAND COMPARISON TEST, K4LY/WH2XZO into SWL/K9”:
“In today’s first illustration, at left, seventy-one 630m (red) WSPR transmissions from Doug WH2XZO as decoded by Ken SWL/K9 execute a “random walk”* of SNRs up and down like a drunken sailor. Time flows from right to left. Likewise, seventy-one 160m (blue) WSPR SNRs–decoded by Ken from Doug’s ham station K4LY in the very same time slots–also stagger about during the night, but do it considerably less and with a somewhat differently shaped curve.
At right in that first illustration, sliding window averaging on a spreadsheet provides a “low pass filter” (LPF) to show the general SNR behavior through the night. I had supposed this SNR behavior indicated some large scale propagation-enhancing effect was at work, but I’ve come to seriously doubt it. Here’s why.
The differences of consecutive WSPR SNRs** on each of 160m and 630m turned out to be totally random during the nine hours of that Oct. 21 two-band test. (Such differences are collectively called “time-differenced SNRs” here.) See 2nd illustration for these SNR differences on each band. These SNR difference graphs have zero average and essentially steady variance on each band: 160m (blue) and 630m (red) for the 832 km Oct. 21 nighttime path from South Carolina to northern Indiana.
I believe this means that the SNRs themselves acted as if over time the path accumulated random positive and negative dB contributions to SNR, especially any part of the path traversing any ionospheric region(s). Time flows from right to left, and the time scale spans nighttime hours with two-minute WSPR time slot granularity. Even though the SNR increased slowly and decreased sometimes as if it had some slow major geophysical force gradually determining it, this random walk result indicates that no slow major physical forcing was pushing SNR first higher and then lower over hours during the night.
If instead there were deterministic trends on both 160m and 630m bands, they would leave a non-zero “DC level” in the time-differenced SNRs on different quiet nights. The 2nd illustration lacks this deterministic feature on both bands, as indicated by the brown dashed lines. As an example of what I mean by “deterministic trend,” think of sunrise leading into daytime when a dramatic decrease of SNR usually occurs on 630m due to the sun’s ionizing radiation on the D-layer.
My seemingly undramatic nighttime summary of the one-hop XZO-swl/k9 test actually describes a quite important result—one that I didn’t expect and even now do hesitate to recognize as a general rule in the quiet nighttime. A random walk by 630m or 160m SNR nighttime SNR means that one can’t predict how SNR will behave the rest of the night, or indeed any night, after the post-sunset regime has faded into evening nighttime.
As shown in the 2nd illustration, the time-differenced SNRs on 160m and those on 630m turned out to be uncorrelated (-0.06) random sequences*** with zero trend and continually constant variance—the “signature” of statistical drift. Because of the uncorrelated randomness in the time differenced SNRs, then on other quiet nights the 160m and 630m SNR random walks may be correlated more, not as much, or even opposite to those that Ken received at SWL/K9 on Oct. 21.
However, if random walk behavior is not infrequent on 630m, any departure from a random walk type of SNR sequence can be sensed rather soon by detecting a departure from zero by the average of SNR differences. Such a departure-sensing method could be a very convenient and important way for us to distinguish fundamentally different types of 630m nighttime SNR behaviors when we find them.
Oct. 21 was a quiet night in the sense that Dst was near-zero, and no geomagnetic storm was in progress. Moreover, terrestrial storm activity was only small-to-moderate and was localized partway along the path from South Carolina to Indiana.
Based on Doug and Ken’s Oct. 21 test, I won’t be surprised if SNR random walks are found to happen on other quiet nights on 160m and 630m. Statistical drift need not lead to the same moderate correlation of Oct. 21 SNRs on other quiet nights, like some steadily driven physical trend or force would produce. TU and GL!”
*“Random walk” for SNR is statistical talk for a process typified by ΔSNRi = θεi. Note 2**.
http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&ved=0ahUKEwj5h-fwi_LPAhXDNiYKHZzIB5cQFggxMAM&url=http%3A%2F%2Fcmapskm.ihmc.us%2Frid%3D1052458884462_996058812_7176%2Frandomwalk.pdf&usg=AFQjCNFk9F-OGf7svpWL9fqE3lj2dcFecA . For what it’s worth, correlation of SNRs in 71 simultaneous time slots was a modest 0.34, about the same as for the superset of 145 time slots (0.35). The smoothed SNR curves (Illustration 1, at right) for 63 LPF points had a higher correlation, 0.77.
**Time-differenced SNR ΔSNRi is shorthand for SNRi+1 – SNRi. Index i means the time slot index. Epsilon ε signifies a random variable over time that has unity standard deviation and zero average, so theta θ signifies whatever value of standard deviation an actual randomness has. Epsilon ε is randomness in the differenced signal S part of MF SNR. Epsilon ε does not mean noise. (Noise instead is the “N” in SNR.) For some general background on various time-series processes, see https://en.wikipedia.org/wiki/Autoregressive_integrated_moving_average .
***SIQ is the range of SNR dB that span the middle 50% of decodes overnight. SIQ was 8dB on 630m and only 4dB on 160m for a ratio of 2.0. By a somewhat smaller ratio, the variabilities represented by standard deviations of the time-differenced SNRs of Illustration 2 for 630m and 160m were 4.8 dB and 3.5 dB for ratio of 1.4. The square of 1.4 is about 2, but I don’t know if this indicates some general statistical principle.
Soon this site will transition to a new web server to handle more capacity. It should be a seamless transition but note that you may have to reload the site after it happens in order to refresh the browser cache with new information.
Additions, corrections, clarifications, etc? Send me a message on the Contact page or directly to KB5NJD gmail dot (com).