Propagation enhancement at 7.85 and 10.0 MHz during the 2024 solar eclipse
Dave Lewis
Propagation enhancement of 7.85 MHz (WWV) and 10.00 MHz (CHU) transmissions received in Florida during the April 8, 2024, solar eclipse
Belated greetings from Miami Florida (grid locator EL95us). Being unable to get to the eclipse path this time around, and with little equipment here, I figured that the best ways I could contribute to the HamSCI mission during the recent eclipse were to generate WSPR signals on 20M from Florida and to record carrier strengths from beacons transmitting either from locations close to the path of totality or from locations the same distance from the eclipse path but on the opposite (northwest) side. This is a report of the beacon monitoring effort.
Study protocol. I recorded the signal strengths (S-meter readings) of nine steady time/frequency standards: six from WWV at Fort Collins, Colorado (about 2825 km away) at 2.5, 5.0, 10.0, 15.0, 20.0 and 25.0 MHz, and three from CHU just southwest of Ottawa, Ontario (about 2200 km away), at 3.33, 7.85 and 14.67 MHz. I also listened for N8GA's beacon from Ohio at 50.08 MHz. I used a Yaesu FT-710 transceiver with an inverted Vee 20/40 M trap dipole antenna orientated N/S, about 20M above the ground. I entered the ten frequencies into the memory and set the transceiver to receive narrow CW mode, which generates a fixed 700 Hz side-tone for easy recognition and verification that I was in fact receiving the intended stations.
I sampled the ten frequencies in five-minute blocks, 30-seconds for each station. On each frequency I recorded the maximum sustained S-meter value over the 30-sec, ignoring noise bursts and short-term QSB. Besides the standard S-meter readings 1 through 9, five db over S9 was recorded as a "10," 10 db over S9 as "11," and 15 db over 9 as "12." For weak signals below S1, "clearly audible" was recorded as 0.5 and "barely audible" as 0.2. Total absence of any detectable signal was zero. Quirky, yes, but quite reproducible.
On April 8, I recorded S-meter readings beginning at 8:30 am EDT (1230 UTC) and extending to 9:20 pm EDT (0120 UTC on 4/9). This was repeated most every half hour, except that during the eclipse period 2:00 to 4:20 pm EDT (1800 to 2020 UTC) I cycled through the ten frequencies every five minutes, to catch brief periods of signal intensification or degradation (see 04-08-24 data). For comparison, this sampling process was repeated over the same time period on April 10 (see 04-10-24 data); the only procedural changes were that readings were made just at half hour intervals, and no data were recorded between the 1500 and 1700 UTC due to an unplanned interruption. On a few other days before April 8, I also listened on those ten frequencies through the mid-afternoon to gauge what was normal propagation under non-eclipse conditions and how it might vary, and to develop a listening pattern that I could sustain over a 13-hour period on eclipse day.
Results. Comparing the data from April 8 with data from April 10, two columns jump out as very different: WWV @ 10.00 MHz and CHU @ 7.85 MHz. A time plot of April 8 and April 10 data for those two frequencies is attached, showing strong correlation between time of signal strength increase and time of the eclipse. The peak of the CHU signal lagged that of the WWV signal, although the time delay is not quite as long as the time difference between maximum sunlight blocking at WWV and CHU. Perhaps that is because the radio waves had to penetrate both the D and E layers above the different points of origin (Colorado or Ontario) and the common point of arrival (South Florida, where about 50% of the sun was covered). On both frequencies, the data also appear to show first a small intensification lasting only a few minutes, followed shortly by a stronger intensification lasting 15 - 25 minutes.
In comparison, the variations in strengths of the other eight beacons' signals on April 8 appear less remarkable. The higher-frequency WWV stations (15.0, 20.0 and 25.0 MHz) and CHU (14.67 MHz) were heard regularly throughout daylight hours on both April 8 and all other days sampled. It is possible that all four were slightly stronger and more variable during the eclipse --and note especially the sharp decline in 20 MHz WWV at 1920 UTC; but there were also abrupt decreases in some signal strengths on the reference day -- see WWV (15 and 25 MHz, but not 20 MHZ) at 1800. I didn't hear WWV (2.5 MHz), CHU (3.33 MHz) or N8GA (50.08 MHz) at all, and WWV (5.0 MHz) only in the evenings.
Although low-tech, I hope these measurements and observations will be useful. Comments will be appreciated.
David K Lewis, W2HMT, with assistance from Henry J. Lewis
Postscript. My first radio-encounter with a solar eclipse was in 1963, on July 20. Someone at work had tacked a brief newspaper article on a bulletin board announcing an upcoming eclipse passing across southern Canada and Maine, so I drove north from my home near Boston to take in the spectacle. I took along a couple of portable HF radios and a roll of antenna wire to see if I could hear anything unusual. There was no media build-up, no above-normal traffic, and no one else around the open field in very-rural Granby Quebec where I set up my viewing and listening station. The officer at the Canadian checkpoint in a little town on the border hadn't even heard they were about to have a solar eclipse. This was soon after the Cuban Missile Crisis, so I thought it best not to tell him I had radio equipment in the trunk; he didn't check. My how things have changed!
Besides some interesting changes in HF band conditions (especially a daytime band opening on 75 M) which I later described in an article in the January 1979 issue of QST, the view was spectacular. I was lucky to be one of the only viewers who was not completely clouded out. This year, my son and his family went to almost the same spot and also got a clear view of totality.