144MHz eTEP Propagation

[Tim Fern]

Hello all, thank you for accepting me into the HamSci community.

HamSci was recommended to me by G0KYA, Chair of the RSGB Propagation Studies Committee of which I am an associate member.

Recently, I have been investigating 144MHz trans-equatorial propagation (eTEP) and have some questions that I hope someone here may be able to help with:

I particularly need to understand the mechanism that creates Equatorial Plasma Bubbles (EPB’s), their shape (curve) when created, and how their dimensions and shape changes (if at all) during their ascent up through the ionosphere.

No doubt there will be other questions too, there are many unknown-unknowns here :-)

If there is someone with this knowledge please reply!

Best 73

Tim G4LOH

[Kornyanat Hozumi]

Hi Tim!

I'm a postdoc with HamSCI currently working on Equatorial Plasma Bubbles (EPBs)! So, I’d be happy to share what I know and hopefully help clarify a few things.

EPBs are formed through what’s known as the generalized Rayleigh–Taylor instability, which is an extension of the classical instability from fluid dynamics. In the classical case, the instability occurs when a heavier fluid sits above a lighter one in a gravitational field. A common example is dense water sitting on top of oil. The system becomes unstable, and the heavier fluid sinks into the lighter one, creating finger-like structures.

In the ionosphere, while the basic instability criterion is similar, the dynamics are more complex. The ionosphere consists of a magnetized plasma rather than neutral fluids. The instability is aligned with Earth's magnetic field and driven not only by gravity, but also by electric fields, E × B drift, plasma density gradients, and ion–neutral coupling. Although not identical to the classical case, the same principle applies: a configuration where denser plasma lies above a depleted region becomes unstable. Under the right conditions, these low-density regions can grow and rise into the higher-density plasma, forming plasma bubbles along magnetic field lines.

Here is a brief timeline. After local sunset, photoionization stops, and the bottomside F-region begins to recombine quickly, forming a steep vertical plasma density gradient. About 30 to 60 minutes later, the ionosphere experiences a pre-reversal enhancement (PRE)—a temporary increase in the eastward electric field. This happens because the E-region (roughly 90–150 km altitude)—famous among hams as the magical layer where 10 m signals suddenly skip across the continent—loses ionization quickly after sunset, leading to a sharp drop in conductivity. At the same time, the F-region above remains more ionized. This contrast in vertical conductivity, combined with the influence of neutral winds, breaks the daytime dynamo balance and generates a polarization electric field that enhances the eastward electric field near the equator. This eastward field lifts the F-layer upward via E × B drift, further steepening the density gradient and creating favorable conditions for instability. Although it lifts plasma upward, it does so along nearly horizontal magnetic field lines, helping redistribute plasma off-equator and enabling new current paths in the F-region, which compensates for the loss of E-region conductivity.

EPBs typically span tens to a few hundred kilometers in zonal width and can extend from about 300 km up to over 1000 km in altitude, with their field‐aligned structures reaching the Equatorial Ionization Anomaly (EIA) peaks at roughly ±15° magnetic latitude. Large-scale wave structures (LSWS) are thought to seed the growth of EPBs. Gravity waves, which originate in the lower atmosphere and propagate upward, can also perturb ionospheric density and help initiate EPB development. These waves modulate neutral winds and thereby induce localized electric fields, altering E × B drift patterns and enhancing the likelihood of localized depletions that grow under unstable conditions.

As a bubble rises along magnetic field lines, its shape evolves due to wind dynamics. In satellite and airglow observations, EPBs commonly appear in C-shaped, reverse-C-shaped, and even X-shaped forms (see plain-language summary at https://www.eurekalert.org/news-releases/1049771). Observations often show a reverse-C structure when EPBs are viewed from above in geographic coordinates. This shape occurs when the lower-altitude, equatorward part of the bubble is displaced farther eastward than the upper, off-equatorial parts. While zonal winds are typically westward after sunset, perhaps eastward winds can occur at lower altitudes during tidal wind reversals or specific seasonal and longitudinal conditions. These coincident tidal wind reversals and the peak of the PRE electric field perhaps combine to push the bubble base more than its top, producing the reverse-C appearance seen in satellite and airglow observations.

Alright, that’s a lot to digest. I hope it helps without overwhelming you!

Best regards,
Kornyanat Hozumi (Kukkai)

> --
> Please follow the HamSCI Community Participation Guidelines at http://hamsci.org/hamsci-community-participation-guidelines.
> ---
> You received this message because you are subscribed to the Google Groups "HamSCI" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to hamsci+un...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/hamsci/2b6ed3d5-ed68-4222-8f6e-f045fcb7f02bn%40googlegroups.com.

[Tim Fern]

Wow, thank you :-)

(I love this place already)

I'll take the weekend to digest and figure out what it means for the propagation we observed around the Spring Equinox at 144MHz, (on the 6,000km path between St Helena and South-West Europe )

[Tim Fern]

Many thanks for such a detailed response, I think I understand most of what you wrote .. :-)

Whilst “TEP” propagation has been seen for decades at VHF, there does not appear to be much recent analysis of it. 

The path from South Africa to Southern EU was seen in the 1980’s, long before digital modes were available to us.  I can’t find any analysis of the communication channel EPB’s create so it is unclear that the digital modes we have been trying are optimal, or even whether they are more effective than simple CW…

I think it might be helpful to describe what we have observed:

In 2024 a new station appeared on 2m band from St Helena Island (IH74db) and has been making many contacts into South West Europe. Garry, ZD7GWM runs just 50W, into a vertical co-linear antenna with ~9dBi.

There have also been many 2m contacts between a station in Namibia (V51WW) and Italy/Greece/Malta/Slovenia, and contacts between Northern Australia and Japan.

• The contacts are clustered around the equinoxes +/- 2months either side
• Contacts happen after local sunset, for a few hours.
• High Solar Flux Index (SFI) seems significant.
• Stable geomagnetic conditions seem to be needed (There is an obvious negative correlation with GeoMagnetic storms/Aurora)
• There is a complete loss of polarisation in the channel (V51WW transmits horizontal polarisation, stations with H and V antennas available confirm equal power is received on both polarisations)
• Signal paths are normal to the GME within ~7 degrees
• Whilst being normal to the GME, stations up to 1,000km North from the expected symmetrical path have heard signals.
• The channel causes signal distortion such that FT8 (with 6Hz tone spacing) does not decode on the St Helena path.
• The degree of distortion observed varies, and can change within a few minutes, and always gets worse (As the EPB ascends?)
• A 2m path is also observed between northern Australia and Japan, here stations always use FT8 (perhaps because the path is shorter (~4,800km?)

My own station is ~1,000km North from the footprint of stations that have made contact so far… however, this distance frequently experiences intense tropospheric ducting that might couple a signal into the eTEP channel. Previous analysis of using tropo ducting to extend ionospheric modes is that signals must arrive from the ionosphere at no more than ~3 degrees from the tangent….. So I’m particularly interested to understand signal coupling in/out of the EPB duct and the angles involved.

From your description of the EPB mechanism, I have a mental image as to the shape of the bubbles, their formation, ascent and dissipation in the high ionosphere….but what is the take-off angle to get a signal into the EPB duct? How does this angle vary as the EPB ascends?

Are the bubbles curved such that the entry angle aligns with the magnetic dip angle? If so, coupling into tropo ducts seem unlikely.

Whilst the observed eTEP contacts happen around the equinoxes, might this simply be due to the geography of the stations attempting it?

Do EPB’s form all year round? but perhaps they are only symmetrical across the GME around the equinoxes, thus providing the longest paths?

Are there any remote sensing systems like NASA G.O.L.D, or NOAA GLOTEC  etc…? That might be predictive for VHF eTEP propagation?

Might atmospheric tides be a factor in the occurrence of eTEP openings?

Best 73

Tim G4LOH

Images from EI7GL blog

[Ethan Miller K8GU]

Tim,

As-promised, a response...

You're asking a lot of the right questions.  As to the seasonality:

The first-order seasonality of EPBs is driven by the orientation of the solar terminator (sunset) with respect to the geomagnetic equator.  The EPB growth conditions are most favorable when the E-region sunset in both the northern and southern hemispheres is approximately coincident.  There are other factors which may dominate on a given day, but this dominates the seasonality.  You can observe this effect where the declination is large and the seasonal maximum departs from the equinox.  Likewise, high solar flux conditions "turn up the gain" on the instability due to the increase in available electron density principally.  In low solar activity years, EPB activity may not even start until near local midnight and the season may change.

TEP propagation mechanism:

TEP signals are guided by refraction via sharp plasma gradients - in the afternoon, these gradients are spatially quite smooth and one may think of it a bit like the Secant Law (fMUF = f0 sec theta) that we normally think about for computing the MUF at lower frequencies.  As the zenith angle (theta) of the ray approaches 90 degrees, the secant becomes large and the MUF trends toward infinity.  Structuring and the rise of the equatorial ionosphere in the late afternoon enables such a scenario--the zenith angle of the ray is normally limited to something less than 90 by the curvature of the earth.  Before any professional radio scientists jump on me--this is not the whole story, but it should give you some intuition for how to think about the problem.

After sunset, the plasma instability processes retain some of the propagation mechanism described above, but for TEP paths also include scattering from the now-rough gradients.  It's likely that the scattering processes dominate for the first few hours.  Note that this is not the same process to results in "coherent backscatter" observed by VHF radars perpendicular to the geomagnetic field, which is a Bragg-scattering process that samples the irregularities in a particular geometry that do not directly affect TEP.

This should give you a sense as to why the polarization is "scrambled" as well.

Duration and Doppler characteristics:

The cascade of plasma instabilities that create the depletion grow very quickly in the first 10s of minutes from initiation.  However, as they stop growing vertically, other parts (e.g., the western wall) become unstable as well.  Although this continues for perhaps an hour or two, as the conditions for the instability are no longer satisfied, diffusion dominates and begins to filter the shorter-scale irregularities.  Eventually, only the large 10-100-km (horizontal) scale irregularities remain through the night. 

During the fast-growing initial phase of depletion development, the envelope of the depleted region is rising (and therefore extending in latitude, due to efficient mapping of the generated/imposed polarization electric field along the geomagnetic field lines) at supersonic velocities (e.g., ~600 m/s) for 10 minutes or so.  This means that the imparted Doppler velocities and spreads on the channel are enormous--there are many multipath components that are emerging and disappearing and varying quickly over that period of time.  This is a spectrogram of WWVH (Hawaii) 15 MHz (multiply the Doppler frequencies by 10 to approximate 2 meters effects) recorded near the geomagnetic conjugate point in American Samoa.  The vertical striations between 0600 and 0800 UT are the explosive growth of bubbles.  The discrete negative Doppler traces prior to 0600 are multipath from the 1F2, 2F2, 3F2, etc, and probably the chordal hop.  After 0800, the bubbles are more or less "frozen into" the eastward plasma flow...more on this in the next paragraph.

As the depletion stabilizes over the next few hours, the also becomes more stable; however, there is some residual Doppler spreading due to the instabilities as they decay.  Furthermore, there is a Doppler shift that remains as the depletion structure moves from West to East at a velocity of perhaps 50 to 150 m/s depending on solar conditions and the particular day.  This example, from the previous day to the one above, contains no explosive growth signatures, but if you look carefully, you can see many discrete paths in the 0700-1200 UTC time period.  The spread of Doppler shifts in this case is due to the westernmost "acceptance" of the 15-MHz wave into the EPB TEP duct on the positive Doppler side and easternmost extent on the negative Doppler side.  I believe American Samoa is west of the conjugate point, which explains the asymmetry.

Onto some other questions:

Acceptance criteria into the TEP / EPB duct:  An individual EPB is approximately 100-km in longitude extent.  However, at the high-latitude footprints, it may have bifurcated (one bifurcation is almost guaranteed) several times as it grew vertically so the area covered is much larger.  With some simple geometry, one can work out that a 100-km wide duct on a 2500-km half-path is about 2.3 degrees...there are other details in play, such as the "tilt" of the EPBs to the west as they grow vertically, as well as the bifurcation...but, it gives you a sense that single-digit degrees is not uncommon.

Why FT8 might not work in one longitude sector versus another:  I know there has been some work done on this in the literature but I don't know how comprehensive it is:  EPBs may form as solitons spaced 300-400 km or more apart or they may form in groups of 2-3 closer together, with the groups spaced farther apart.  I do not know if there are longitudinal trends, but this could be one explanation.  It also might be explained by the types of antennas and discipline/practices employed by the operators.  Narrower beamwidth antennas (versus the vertical) might be better for FT8 because they reject more spurious paths.  A thought or two...

Takeoff angle:  I would think low elevation is generally a good idea.  Unsure if you could couple in/out of a tropo duct...the big thing is whether you could get out the right point, I suspect.

Predictions:  This is one of the enduring problems in Space Physics for at least 50 years.  While I would be delighted to see someone from HamSCI make a contribution to it, it's a hard problem and has consumed the brainpower of quite a few bright minds over those 50+ years.  Yes, I would say the atmospheric tides probably play a role.  Yes, one might learn something from GOLD, ICON, and other datasets.  I would offer the following:  the growth rate equations for the Rayleigh-Taylor instability that we find in the literature are both linearized (incomplete) and contain multiple terms.  There are least 2-3 possible mechanisms to initiate an EPB and some of them are very localized--even state-of-the-art nonlinear numerical models can only predict the occurrence of EPBs that were generated over the site where the drivers were measured.  They do score well on forecast skill scores until other depletions drift in from the west.  But, do feel free to do some analysis of your own!

Citations:

Hopefully that was somewhat helpful.  This is a pretty good paper about the climatology of EPBs that will also key you into some references:

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2005RS003340

I have to cite one of my own papers here that talks about the out-of-season EPBs. I believe now that one of the conclusions of this paper (about the locations of the field-aligned Bragg instabilities and the altitude of the airglow) is incorrect.  I should probably write and submit a correction; but, the climatological conclusions should still be solid.

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2009JA014946

Here's a paper on the numerical forecast model using ICON data--see references by the same author for more details/variations:

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023SW003427

This is a seminal paper on the topic of transequatorial propagation, which is regrettably behind a paywall:

https://www.sciencedirect.com/science/article/abs/pii/0021916973900160

The American Samoa spectrograms are unpublished work from my AFOSR Young Investigator grant.

Whew...that was a lot...

73,

--Ethan, K8GU.

To view this discussion visit https://groups.google.com/d/msgid/hamsci/b700969c-558b-4f89-b897-07c71fd3b930n%40googlegroups.com.

--

http://www.k8gu.com/
Repair.  Re-use.  Re-purpose.  Recycle.   

[Tim Fern]

I’ve now been able to observe many more 2m eTEP contacts between Europe and Africa and have a few more questions I’d like to ask the community here

1) Polarisation

It appears that the polarisation of 2m signals is completely lost through an eTEP channel. One station with an EME array switchable V-H, has repeatedly shown reception of equal signal strengths on *both* polarisations. 

Is this as expected? 

There have been times when the paths have appeared to be one-way, this might be explained by a polarisation rotation.

Is it conceivable from theory, that the signal polarisation might sometimes be rotated rather than lost completely?

2) The signal frequency distortion (spreading) is seen to change quite quickly (within a minute) such that changes to the modulation needs to be made quite frequently sometimes.

Using Q65, based on the appearance of the distortion in the waterfall, stations switch between modes 30B, 30C

The 13Hz tone separation of 30C ensures decodes when distortion is high with some penalty in C/N performance.

Could you conjecture as to what is happeningin the ionosphere that is causing this sudden change in distortion? 

Is this a single bubble right across the F-layer path? Might polarisation distortion also be smaller when the frequency spreading is lower?

3) Whilst the contacts noticeably peak at the equinoxes, there were still some 2m contacts being made just one day after the summer solstice. I think this shows the benefit of digital modulation that this has not been observed before.

As many stations in Europe try to widen the footprint of contacts made (trying to contact the tiny number of African stations) I have noticed the paths be North - South within ~8 degrees of Normal to the GME.

It occurred to me that the solar terminator is only Normal to the GME close to the equinoxes and at the Solstice, it will be 23 degrees from Normal.

As we move away from the equinoxes might we observe paths more skewed from Normal?

So, the paths might be skewed slightly NNW-SSE near the summer solstice and skewed NNE- SSW when the we move towards the winter solstice?

4) My mental image of the EPB’s rising through the F-layer leads me to imagine that the footprint of the possible radio channel is likely to be longest when the bubble is highest?

So, if a station 20 degrees South of the GME was in contact with a station 20 degrees North of the GME….. 

Would that path likely become 21 degrees South to 21 degrees North, as the EPB rises a few minutes later?

5) As for predicting EPB’s, I should have asked a slightly different question relating to the remote sensing satellite systems currently operating:

Similarly to how we watch magnetometer data to foresee the start of radio Aurora paths, are there remote sensing services of EPB’s that we might watch in real-time, to see the EPB’s (location/intensity/altitude?) and anticipate eTEP openings on VHF/UHF?

Best regards

Tim Fern G4LOH