Big hole around Gulf of Thailand

[NRDcubed]

I’ve spent the last couple weeks in Thailand, particularly Pattaya area And there seems to be a huge hole in detection over the Gulf of Thailand. Sitting on the beach looking west almost nightly, you can see cloud to cloud lightning Which could be very far because there is no sound But on the map, even 500 miles away there are no hits. Furthermore, almost daily there is cloud to ground strikes with no hit.

[GeezerD]

As I understand it (I have gotten a detector just this year) Cloud to Cloud lightning (IC, Intra-Cloud) has a couple of characteristics that make it harder for the Blitzortung network to detect or locate.

First is the polarization of the transmission: the antennas used are best at vertically polarized signals, and CG (Cloud-Ground) strike are what transmit most efficiently vertically polarized. These CG strikes also have the most efficient ground-wave propagation.

I'm unfamiliar with the algorithm that is used to identify particular strikes at multiple stations, but I suspect lightning transmission signatures are more directionally invariant with CG strikes than IC strikes.

South America and Africa have really low location rates as well, considering how well represented they are with other lightning detection networks. Blitzortung seems very biased towards northern and and western hemisphere detections.

D

[cutty]

I’ve spent the last couple weeks in Thailand, particularly Pattaya area And there seems to be a huge hole in detection over the Gulf of Thailand. Sitting on the beach looking west almost nightly, you can see cloud to cloud lightning Which could be very far because there is no sound But on the map, even 500 miles away there are no hits. Furthermore, almost daily there is cloud to ground strikes with no hit.

As I understand it (I have gotten a detector just this year) Cloud to Cloud lightning (IC, Intra-Cloud) has a couple of characteristics that make it harder for the Blitzortung network to detect or locate.

First is the polarization of the transmission: the antennas used are best at vertically polarized signals, and CG (Cloud-Ground) strike are what transmit most efficiently vertically polarized. These CG strikes also have the most efficient ground-wave propagation.

I'm unfamiliar with the algorithm that is used to identify particular strikes at multiple stations, but I suspect lightning transmission signatures are more directionally invariant with CG strikes than IC strikes.

South America and Africa have really low location rates as well, considering how well represented they are with other lightning detection networks. Blitzortung seems very biased towards northern and and western hemisphere detections.

D

The ANSWER is quite simple, actually:
   

...and about a dozen DOA...

[oppedahl]

I’ve spent the last couple weeks in Thailand, particularly Pattaya area. And there seems to be a huge hole in detection over the Gulf of Thailand. Sitting on the beach looking west almost nightly, you can see cloud to cloud lightning Which could be very far because there is no sound But on the map, even 500 miles away there are no hits. Furthermore, almost daily there is cloud to ground strikes with no hit.

Yes you are right.  I looked at https://map.blitzortung.org/#5/9.21/104.38 to see where there are active Blitzortung stations.  All of the stations that immediately surround the Gulf are presently "offline".  The way you can tell this is to look at the curved brackets that surround the colored dot of the station.  The eleven or so stations that immediately surround the Gulf all have curved brackets that are gray. 

• There have been five stations in Thailand and all are presently offline.
  
• There have been two stations in Cambodia and all are presently offline.
  
• There has been one station in southern Vietnam and it is presently offline.
  
• There has been one station in southern Myanmar and it is presently offline.
  
• There have been three stations in nearby parts of Malaysia and all are presently offline.
  

It would of course be most welcome if some of those stations could be returned to online status.  This could help with the hole in detection that you describe.

Every now and then, a lightning strike in the Gulf of Thailand will get detected using signals from other Blitzortung stations that are much further away.  For example just a few hours ago this strike was detected:  2025-09-16 20:39:17.239059456 .

With the help of forum member Robo I was able to learn more about how the Blitzortung system detected and localized that particular strike.  One Blitzortung station that helped to find that strike was the sole active station in Vietnam (2243).  Another that helped was in the Philippines (2710).   Eighteen stations in Japan helped to locate the strike.  And a station in Omsk, Russia (2769) helped -- that station was about 1700 km away from the strike.

This strike that I happened to notice on the map was, I suspect, a rarity for the location.  I bet there are lots of strikes in the Gulf of Thailand that don't get detected.

[robo]

The ANSWER is quite simple, actually:

I bet some of the station operators whose stations are offline haven't even noticed this yet. We should ask Egon to finally implement an email notification when a station is offline for more than an hour.

[MDuperrier]

South America and Africa have really low location rates as well, considering how well represented they are with other lightning detection networks. Blitzortung seems very biased towards northern and and western hemisphere detections.

Indeed 

My attempt to plug a hole as big as a continent is not yet very effective. Some more detectors (mini+ferrite) are ready to be deployed but I still need to find some collaborative host with electricity and internet. The last constraint is even harder than I though in the region I'm trying to cover.

I feel like the MCG (Max Circular Gap) defined at 270° is the main limiting factor, more than the pure distance. Not very precise math, but with 2 detectors, increasing the MCG to 300° would increase the max distance by ~73%. This would make a big difference. The system will likely be less precise in other scenario 

No plan for Thailand, but in central Africa, I'll keep pushing and deploying some more (and tuning, I know, I know, ...)

   Marc

[oppedahl]

South America and Africa have really low location rates as well, considering how well represented they are with other lightning detection networks. Blitzortung seems very biased towards northern and and western hemisphere detections.

Indeed 

My attempt to plug a hole as big as a continent is not yet very effective. Some more detectors (mini+ferrite) are ready to be deployed but I still need to find some collaborative host with electricity and internet. The last constraint is even harder than I though in the region I'm trying to cover.

I feel like the MCG (Max Circular Gap) defined at 270° is the main limiting factor, more than the pure distance. Not very precise math, but with 2 detectors, increasing the MCG to 300° would increase the max distance by ~73%. This would make a bid difference. The system will likely be less precise in other scenario 

No plan for Thailand, but in central Africa, I'll keep pushing and deploying some more (and tuning, I know, I know, ...)

   Marc

"I still need to find some collaborative host with electricity and internet."  Well and not only that, as we all know.  The would-be station operator would ideally be able to draw upon some experience with RF engineering, so as to maximize antenna success and minimize RFID issues.  I suspect an active ham radio operator might be a good candidate for this kind of thing.   Or an electrical engineer or person with a physics degree.

We have all seen the occasional forum posting from a person who is very well intentioned but is not able to get a station functioning well, due in part to limited experience in these areas.

[MDuperrier]

"I still need to find some collaborative host with electricity and internet."  Well and not only that, as we all know.  The would-be station operator would ideally be able to draw upon some experience with RF engineering, so as to maximize antenna success and minimize RFID issues.  I suspect an active ham radio operator might be a good candidate for this kind of thing.   Or an electrical engineer or person with a physics degree.

We have all seen the occasional forum posting from a person who is very well intentioned but is not able to get a station functioning well, due in part to limited experience in these areas.

You have a very valid point, but we can't have everything 

I have tested and practiced in Switzerland, now packed the antennas and detectors with good quality power supply and shielded cables and I had some success with remotely guiding in order to find a good antenna placement (e.g. not on top of the air conditioning units). Not ideal for sure, and it require collaboration and patience from the host for a few days. When they run, #3111, #3113, #3155 & #3157 provide some useful signals. How do I know ? I ran query for all strikes in Africa below 20N (50 million km2) and the listed stations are involved in 99% of the strikes.

I would love that my host where electrical engineer, but they are not.

Another issue (more an anecdote), when any storm is too strong and too close, the power usually cut 

My very rough estimates is that the detection was nearly null and it's now about 1% This is a challenging hobby.

Regards,
    Marc

[oppedahl]

"I still need to find some collaborative host with electricity and internet."  Well and not only that, as we all know.  The would-be station operator would ideally be able to draw upon some experience with RF engineering, so as to maximize antenna success and minimize RFID issues.  I suspect an active ham radio operator might be a good candidate for this kind of thing.   Or an electrical engineer or person with a physics degree.

We have all seen the occasional forum posting from a person who is very well intentioned but is not able to get a station functioning well, due in part to limited experience in these areas.

You have a very valid point, but we can't have everything 

I have tested and practiced in Switzerland, now packed the antennas and detectors with good quality power supply and shielded cables and I had some success with remotely guiding in order to find a good antenna placement (e.g. not on top of the air conditioning units). Not ideal for sure, and it require collaboration and patience from the host for a few days. When they run, #3111, #3113, #3155 & #3157 provide some useful signals. How do I know ? I ran query for all strikes in Africa below 20N (50 million km2) and the listed stations are involved in 99% of the strikes.

I would love that my host where electrical engineer, but they are not.

Another issue (more an anecdote), when any storm is too strong and too close, the power usually cut 

My very rough estimates is that the detection was nearly null and it's now about 1% This is a challenging hobby.

Regards,
    Marc

I offer you my thanks on behalf of the community for your efforts to try to get some stations working in unserved areas.

[oppedahl]

[quote pid="27446" dateline="1758212092"]
You have a very valid point, but we can't have everything 

I have tested and practiced in Switzerland, now packed the antennas and detectors with good quality power supply and shielded cables and I had some success with remotely guiding in order to find a good antenna placement (e.g. not on top of the air conditioning units). Not ideal for sure, and it require collaboration and patience from the host for a few days. When they run, #3111, #3113, #3155 & #3157 provide some useful signals. How do I know ? I ran query for all strikes in Africa below 20N (50 million km2) and the listed stations are involved in 99% of the strikes.

[/quote]
I looked at the station reports for those four stations.  The green line is very low in the graphs, but the other lines look very encouraging.  I think you have accomplished quite a lot.  Thank you.

I see that for each of these four stations, you chose ferrite-rod antennas.  I think that is smart.  For the host with what might be limited experience, any other kind of H antenna has more ways it might go wrong.

Do you add shielding to the FRAs?

Do you mount the FRAs and preamp to a board, so reduce the risk the host will damage the fine wires from the FRAs?
   

[MDuperrier]

I see that for each of these four stations, you chose ferrite-rod antennas.  I think that is smart.  For the host with what might be limited experience, any other kind of H antenna has more ways it might go wrong.

Do you add shielding to the FRAs?

Do you mount the FRAs and preamp to a board, so reduce the risk the host will damage the fine wires from the FRAs?

Hello,
I started with loops but the ferrite are definitely easier to package and 'tune'. With a 15m cable, I only instruct the host to place it anywhere away from any obvious source of noise and watch with the "FFT html page" provided by Mike. Sometime on the ground below the bed, sometime on top of a wardrobe, etc... Not optimal for sure but again, my host are not engineer and I try not to bother them more than a few days 

I have not shielded the ferrite rod, only using shielded cable between the detector and pre-amp. My antenna packaging was discussed at: Thread 3983. Since then I have read post that indicate 'L' is better than 'X', but too late (and it works).

Thanks and regards,
   Marc

[oppedahl]

I see that for each of these four stations, you chose ferrite-rod antennas.  I think that is smart.  For the host with what might be limited experience, any other kind of H antenna has more ways it might go wrong.

Do you add shielding to the FRAs?

Do you mount the FRAs and preamp to a board, so reduce the risk the host will damage the fine wires from the FRAs?

Hello,
I started with loops but the ferrite are definitely easier to package and 'tune'. With a 15m cable, I only instruct the host to place it anywhere away from any obvious source of noise and watch with the "FFT html page" provided by Mike. Sometime on the ground below the bed, sometime on top of a wardrobe, etc... Not optimal for sure but again, my host are not engineer and I try not to bother them more than a few days 

I have not shielded the ferrite rod, only using shielded cable between the detector and pre-amp. My antenna packaging was discussed at: Thread 3983. Since then I have read post that indicate 'L' is better than 'X', but too late (and it works).

Thanks and regards,
   Marc

Thank you for your posting, and again, thank you for your work that helps the goals of the project by providing a chance to get some stations working in a part of the world that previously did not have much coverage.

I see that you labeled the "ANT" port on the processor box.  That was smart.  It reduced the risk that your host might inadvertently plug the wrong cable into the wrong port.

I admire your antenna packaging.  It looks like you found some black plastic box that was very lucky in its size and shape, that you were able to repurpose as a way to secure the two ferrites and the preamp.  You end up with a very nice looking box that is easy to ship and you do not need to worry about the host possibly damaging the thin wires from the ferrites.  It looks like the box has two keyhole slots on the back for wall mounting.  But I cannot guess where the plastic box came from.  Can you share how you happened upon this box with its ideal size and shape?

Again thank you for this service to the community.

Carl
DE AA2KW