Talk:Alleged Chemical Attack Khan Sheikhoun 4 April 2017/Location

Alleged bomb location
Telegraph article Gives specific location for one bomb strike "Corniche St" in the north part of the town. We have already seen an alleged fresh crater near the grain silos/bakery. I assume Corniche St is nearby? I can't find any street map to confirm. One other snippt is the wind is listed as blowing to the west - that's consistent with the Hama weather data. It also makes the Bellingcat 'analysis' somewhat dubious. --Charles Wood (talk) 03:19, 6 April 2017 (UTC)


 * Christiaan Triebert tweets that Pentagon and Brown Noses located the bomb crater, by which Hadi Abdallah is seen in his report video, at the same place - in front of the silos, near (somewhat) to where the strange US media image had its yellowish puddle. Pretty exactly here it is on wikimapia. --CE (talk) 03:40, 8 April 2017 (UTC)

A SMART News reporter visits the site of the bomb crater. Video: Testimony of the survivors of the regime chemical attack on Khan Shaikhoun in idleb At 1:08 he enters one of the buildings on the main street. He seems to be saying that people in the cellar floor had died of the attack. Who would live here? No furniture. Only a dirty rug on the floor for sleeping with an even smaller mattress serving as a common pillow. Internally displaced people, maybe. Or prisoners and hostages. The five shoes left next to the mattress belong to women and children. -- Petri Krohn (talk) 02:19, 9 April 2017 (UTC)
 * That's not quite the spot the US marked - this is on the highway, just SW of the silos. At 1:02 we see the crater, facing north and nearer (south of the crater) chunks of pavement and dirt. Could these be tossed from that impact? If so, direction of projectile was south, fired from the north. US map shows one track only flying apparently a bit south of Khan Sheikhoun. But I supposed a fancy missile could swoop north, turn around, head south, and do this? --Caustic Logic (talk) 03:19, 9 April 2017 (UTC)




 * CNN reports "US military and intelligence analysts have now confirmed it was Syrian regime warplanes that struck a hospital in Idlib province with chemical weapons. The US picked up the radar intelligence that regime warplanes were in the air and got the infrared signature of bombs detonating at the hospital according to a US official with direct knowledge of the information." But the US map shows a spot about 150-200 meters SW of the silos area as impact point. If the hospital (labeled separately in the right spot east of town) is where people were gassed - what were they all doing there? (I guess this is just some kind of mix-up? Or is there another hospital in the allegedly hit area? Nothing that looks like one, but they usually don't) --Caustic Logic (talk) 03:09, 9 April 2017 (UTC)


 * Is US intelligence claiming the White Helmets field hospital was the site of the exposure to sarin? Quote: "regime warplanes that struck a hospital in Idlib province with chemical weapons." Actually I agree, the cave complex is he most likely place for exposure. The victims where then moved to Shajul Islam's hospital downtown. But what were the civilian-looking people doing in this al-Qaeda / al-CIAda military complex at 6 am in the morning? See "" below. And why were only civilians affected and none of the White Helmets? -- Petri Krohn (talk) 03:40, 9 April 2017 (UTC)


 * See the Avigdor Lieberman quote Resup added to the section today. Apparently the claim is that "Assad" hit first near the silos, then the victims were evacuated to Whora Hora, where "Assad" hit a second time. All in 9 minutes. 100%. Or something  --CE (talk) 03:42, 9 April 2017 (UTC)


 * It may be that weird. Russia and Syria say the place they bombed was a place al-Nusra had CWs stored - US seems to say they say that was the hospital (cum-hostel?). But really .. is it that weird? I suspect not. For one thing, they're 8 hours off on timing, so we don't have a clear everyone agrees scenario. Something's amiss, and it might be six things or more. Takes untangling. (and sorry I've been so lame so far) --Caustic Logic (talk) 04:03, 9 April 2017 (UTC)

Crater
What kind of explosion would create the crater seen? A 122 mm Grad rocket hardly pierces the asphalt. Also note, that the electricity distribution box right next to the crater has not been damaged by the blast. The damage to the surrounding warehouses is said to be 6 months old. -- Petri Krohn (talk) 06:22, 9 April 2017 (UTC)


 * The size of the crater depends a bit on the fuzing type and fuze. The KS crater could easily be made by a 122mm HE rocket if it delayed a few milliseconds to detonate.
 * I find it less believable an S5 missile did it as that only has 800 gram warhead.


 * A third alternative is that the crater is old and a relic from a previous battle. This is less likely as it would presumably be filled in. However, the undamaged structure beside it could have been placed later if it was an old crater. It it was present at the detonation it would surely(?) have some shrapnel damage and paint scouring. --Charles Wood (talk) 08:47, 9 April 2017 (UTC)

Petri brought this, all images from that day, starting around noon. --Caustic Logic (talk) 10:16, 11 April 2017 (UTC)


 * These images are from at least 3 different days, I made a video collection instead, I also added some of my observations there. Note that there is a video tsunami from April 6th, but I only managed to find a single video from the day before. --Q (talk) 12:18, 21 April 2017 (UTC)


 * Interesting - how did you decide it was three different days? (sorry to be brief and not dig for an answer at the link - pressed for time - as I guess some readers will be) --Caustic Logic (talk) 13:08, 21 April 2017 (UTC)


 * There are many clues and I didn't reach the conclusion this way, but this is probably the most straightforward explanation:
 * * There are 2 videos published on April 4th.
 * * SMART has a batch of photos from April 5th, and these photos show the same scenes that also appear in their video published on April 5th.
 * * The next group of videos have several similarities, the "press briefing" setup, their publishing dates on Youtube are April 6th, late afternoon-evening(local time), there are common characters and other visual clues. The video in this Facebook post fits perfectly to this group, and the description has the exact filming date, April 6th("Thursday").
 * --Q (talk) 15:14, 21 April 2017 (UTC)


 * Not sure if this is/should be in your collection. Hadi Alabdullah by the crater on 4 April before any substantial alterations but with chemical staining on road significantly reduced. Had On Site --Charles Wood (talk) 21:43, 21 April 2017 (UTC)


 * Thanks, Q. I don't have time to review that, but there's some explanation for anyone else, or me eventually. it wouldn't be surprising if they come from different days. --Caustic Logic (talk) 12:02, 25 April 2017 (UTC)


 * I noticed in the 'professional' SMART video of April 5 some dead birds in a cage. I assume Krazy Kaszeta Kemistry ™ has a simple explanation why 'non persistent' Sarin was still killing birds a day later. I suspect that explanation won't include the locals faking it. --Charles Wood (talk) 12:38, 21 April 2017 (UTC)


 * I just found this video from Faruq Shami, it shows the White Helmet sample collection team in action on April 4th. It's a long video (15 min), and there are several interesting scenes, so I recommend everyone to watch it, the crater part starts at 07:56. It's April 4th, at ~17, this is so far the first recorded sample collection attempt. The red signs that appear in later videos originate from this team. Mostly soil samples are collected, but I can see a dead bird and something that looks like a metal piece as well. The small metal piece in the crater was moved, but it wasn't collected for some reason. --Q (talk) 17:04, 24 April 2017 (UTC)


 * I updated the pastebin with two new videos, White Helmets collecting samples, and a great drone video that shows the surroundings of the crater and the nearest neighborhood. --Q (talk) 17:14, 24 April 2017 (UTC)

It might have been there already. It looks like a power pole was knocked down by this impact, they set up this electric box to help (?) and that probably hadn't all been done by noon. But maybe. Anyway, the north view video (see ) shows no plume rising from the crater area after the first round of alleged air strikes.

So was it hit earlier? Or later yet, and they still had that fresh electric box placed before the first images? Or did no violence happen there at all this day? It doesn't look obvioulsy old - as Charles notes, it's not filled-in, and the debris spread looks realistic (but maybe lacking in tossed-up dirt) --Caustic Logic (talk) 10:16, 11 April 2017 (UTC)

Furthermore, doesn't that crater suggest a blast, and maybe a powerful one? And wouldn't that vaporize any sarin it also released? --Caustic Logic (talk) 10:16, 11 April 2017 (UTC)

Direction of travel: the crater details to me suggest it came from N-NW, seeming to raise and crack the pavement to the S-SE. It actually lines up pretty well with the damaged building that way, but if they've been that way for months, I suppose that's a coincidence. --Caustic Logic (talk) 10:16, 11 April 2017 (UTC)

It is an ODAB-500
I am looking at the HRW video at 1m 58s. It shows how the pieces of asphalt are pushed inwards, not out. There seem to be two effects that have formed the "sarin crater":
 * 1) I small explosion that thrown some small pebbles of asphalt around radially. The bigger pieces of asphalt have have not moved out of the crater.
 * 2) Some great force from above that has pushed the large pieces of asphalt deeper into the ground.

The two forces cannot be one and the same explosion. It is reasonable to assume that the large force is the weight of the bomb borrowing in and the small force is the primary igniting in the crater. HRW claims that it is KhAB-250 without the proximity fuze. This is typical HRW dishonesty. They claim that the filling cap is from a bomb only seen in a photo from a museum, when there are pieces of ODAB-500s lying all around. HRW + Michael Kobs put together may have gotten it right. It is a liquid-filled bomb that failed to ignite in the air, causing the crater by its mass alone. HRW may have gotten it right. After all, they have informants on the ground. ...except that it was not sarin. and it was not KhAB-250.


 * My hypothesis for the day:

The Su-22 dropped six ODAB-500s in a straight line. One parachute failed, causing the bomb to borrow in the pavement.

The only problem is that one would expect to see remnants of the ODAB-500 at the crater site. -- Petri Krohn (talk) 12:12, 15 May 2017 (UTC)
 * 1) Maybe it happened earlier and someone had already cleared up the bomb remains.
 * 2) Maybe the White Helmets removed all fragments early on April 4th to prevent anyone (but Michael Kobs) from finding out it was a ODAB-500.
 * I don't think we got solid evidence that this crater is related to the video showing 3+1 plumes (this area is not on that video, right?) It will be pretty tricky to make this hole with an ODAB-500. If it exploded above ground (it may have a corded trigger hitting ground first), we will see more damaged area, not just right there. If it hit  the ground, it will be hard to balance between going much deeper followed by blast excavation (not there at all) and having more damage on the surface (outside of small crater itself, none). Also it will knock off transformer box seen in some photos nearby, etc. I'd go for an explosion set to release CW (as far as we know CW showed up on tests, if they know what are they doing--and they should--that would not happen after ODAB explosion; or it's a false positive somehow). It may be like Postol described, or that + old crater used (or something else unrelated to the bombing video, looks more like IED, perhaps a mortar) --Resup (talk) 12:45, 15 May 2017 (UTC)
 * ODAB which dispersed stuff but did not ignite it? We will see more bomb parts around. + A lot of CW released, appears to be a lot more than as alleged. Another malfunctioning non-rigged ODAB, quickly rigged by rebels to look like CW? Seems too exotic, can't be planned. --Resup (talk) 13:06, 15 May 2017 (UTC)

Environmental Samples
HRW mentions 3 groups doing tests, below. French one performed environmental samples analysis.

French report The analyses carried out by French experts on the environmental samples collected at one of the impact points (by whom?)  of the chemical attack at Khan Sheikhoun on 4 April 2017 reveal the presence of sarin, of a specific secondary product (diisopropyl methylphosphonate – DIMP) formed during synthesis of sarin from isopropanol and DF (methylphosphonyl difluoride), and hexamine. Analysis of biomedical samples also shows that a victim of the Khan Sheikhoun attack, a sample of whose blood was taken in Syria on the very day of the attack, was exposed to sarin.

OPCW: biological samples described, several victims

Turkish: biological samples described, one victim.


 * --Resup(talk) 13:42, 15 May 2017 (UTC)

Whora Hora

 * ''(I have decided to name this place Whora Hora after bin Laden's more famous cave hideout and hospital. -- Petri Krohn (talk) 20:36, 6 April 2017 (UTC) )

I found one possible location for the White Helmets cave complex. There is a plot for a warehouse carved into the hillside on the eastern side of the M5 motorway, 1.2 km north, northeast of the point where the missile struck next to the grain silos. -- Petri Krohn (talk) 00:47, 6 April 2017 (UTC)

...but Aldin on Twitter found the real one. No coordinates. :( -- 05:09, 6 April 2017 (UTC)
 * This is at 35.441414, 36.671107 just outside the eastern edge of the town. Between the time of the google maps photo and the more recent satellite photo there have been large-scale excavations, and ventilation shafts have appeared in the field above the quarry.


 * As there was a hospital in town under the direction of struck-off Dr Shajul Islam, there was no reason for casualties to have taken to this quarry outside town. It's interesting that one of the photos uploaded by Petri shows a child being carried out of one of the bunkers by the WH cameraman who was photographed from behind by the Reuters cameraman. A careful comparison of videos and stills for continuity errors may show that children are decontaminated after they come out of the bunker, with obvious implications. Pmr9 (talk) 16:29, 6 April 2017 (UTC)

On April 3 Shajul Islam tweets a two minute video with a look inside a 800 m2 cave complex that he says is under 30 meters of mountain. One future hospital room is shown and seems to be about 7 x 3.5 meters in size. Cinder blocks are waiting for inner walls to be built. Shajul Islam says the hospital will have "about 10 to 15 rooms this size". Who said Osama bin Laden could not receive kidney dialysis at Tora Bora? This hospital will have CT scanners an everything - all paid for by the UK government. The video connects Dr Shajul Islam to the White Helmets cave hospital. SI operated other hospitals in Syria, but he was in Khan Sheikhoun and the soft earth in the caves is similar to what we see in the White Helmets base. This put Shajul Islam in control of both of the hospitals in Khan Sheikhoun. -- Petri Krohn (talk) 18:34, 7 April 2017 (UTC)
 * Dr Shajul Islam‏ @DrShajulIslam - April 3:
 * ''"After 5 hospitals hit and closed last week, this is the future of hospitals in Syria. Help by donating MedicalAidSyria.com
 * April 4: ''This is not a SAMS project. One nation UK will be buying equipment for this hospital."


 * If the cave hospital outside Khan Sheikhoun was so unfinished that the chambers were just cinder blocks with no inner walls, this is strong evidence against the opposition version (channelled by Kareem Shaheen in The Guardian) that victims were taken there to be treated, and consistent with the alternative that these chambers were used to hold captives. Do we have any images of victims inside the cave hospital, or being carried out of it (apart from the one photo of the WH cameraman? Pmr9 (talk) 20:27, 7 April 2017 (UTC)


 * "Whora Hora" is a huge complex and is being enlarged all the time. In other videos we see the inside of the existing hospital. I believe there are two other entrance courtyards to the cave complex. The two others may be used for weapons storage and for chemical weapons. -- Petri Krohn (talk) 20:56, 7 April 2017 (UTC)


 * Here is one video, from Petri's playlist, that shows the inside of the existing cave hospital. --CE (talk) 21:27, 7 April 2017 (UTC)


 * Shajul Islam didn't arrive in Khan Sheikhoun area until late March. On March 19 he was still in Idlib. His first twitter mention of KS area is on March 23 when he says "As the battle to free #hama continues we are saving lives from a cave on the front lines. Dua requested". He had previously been in Aleppo until its liberation and then in Idlib. His presence at the site of the attack was certainly opportune for the jihadis. Steve McIntyre 20:40 EDT 7 April 2017


 * I suspect that the to-be-finished larger "hospital" (plus whatever) is actually the al-Khazanat base (see below). The large structures don't seem to be created yesterday. Found some more info about the hostage-killing event in February (last link most relevant) at the former army underground refueling base, and the place was the headquarter of Liwa al-Aqsa, a super hardcore takfiri group affiliated with ISIS. After heavy infighting with Nusra they agreed to a deal and left to Raqqa, not before murdering 80 of their hostages from an "FSA" gang Jaish al-Nasr at some point supported by the US. Search for the hostages then took place on the 5km² large areal. It is planned to be taken over as base by Nusra (now known as Hayyat Tahrir al-Sham), says the February 22 article.
 * There's also the chance that the two are connected anyway as they are less than two kilometers away from each other--CE (talk) 03:26, 8 April 2017 (UTC)

I have collected 16 videos that show original footage from the compound into YouTube playlist. The videos are ordered by the approximate timeline. There are three different narratives or acts. I am not sure they can all play out on the same day. -- Petri Krohn (talk) 03:52, 8 April 2017 (UTC)

Kareem Shaheen of The Guardian has posted a video from the hospital cave from April 6th, two days after it was bombed. The place has been abandoned. I see no attempt to clean up after the blast. I am starting to think that the bombing was staged. -- Petri Krohn (talk) 10:47, 6 May 2017 (UTC)

Is this the al-Lataminah cave compound?

 * ''(No, it is not.)

Yes, the location of the White Helmets base must be the Al-Lataminah cave compound just south of Al-Lataminah. It was bombed by Russia on the first day of their campaign. -- Petri Krohn (talk) 17:25, 4 April 2017 (UTC)
 * Lataminah is about eight kilometers south-west of Khan Sheikhoun, I don't think the SAA would be confusing them. The thesis that it happened and was filmed earlier is interesting, though. FWIW, here's a marker on wikimapia in Khan Sheikhoun which is eight years old, on something that could be a small rocky formation, with the first comment reading autotranslated "Ohla Abu Yahya .. King of Khan Sheikun and neighboring areas, we forget the sweet days in the university city and the moments of land bread and the voice of the breads of the bread from the decks like Grad missiles atmosphere". --CE (talk) 19:02, 4 April 2017 (UTC)

...or did al-Lataminah move to Khan Sheikhoun
I just noticed Adam's blog post on the Alleged Chlorine Attack in Latamna. (I have been too busy following the spy drama at the White House to have time for Syria.) The photos could be from the Whora Hora base in Khan Sheikhoun and not from the al-Lataminah cave compound. The chlorine cylinder is the same as seen in this photo from the Khan Sheikhoun base. I thought I had geolocated the photo to the second south-facing entrance, just east of what we know is the White Helmets Field hospital. I suspect the photo with the fire truck could be at a west-facing entrance to the cave complex, at the old stone quarry to the west. (I suspected there would be another entrance, but did not have proof before.) -- Petri Krohn (talk) 00:30, 7 April 2017 (UTC)

...or is it Camp Khazanat?
Rescanning wikimapia following this morning's Konashenkov statement, on the eastern outskirts of the town there is a "Camp Khazanat Army fueling base" complete with "vehicle shelters" and "underground fueling storage" that looks like it could fit our "quarry" of interest. --CE (talk) 06:06, 5 April 2017 (UTC)


 * The facility in the videos shows the facility was surrounded by green fields. The Camp Khazanat Army fueling base is surrounded by olive groves
 * There are numerous limestone quarries around Khan Sheikhoun especially to the East and North. Some of them are surrounded by fields, not olive groves.
 * Ian Grant suggests this facility to east of Town is the site. He has a better resolution photo showing air-vents in the fields to the North of it. See His Tweet with image
 * The gas related attack occurred before dawn. The Russian attacks occurred 11:30 to 12:30 local time.
 * The SyAAF doesn't have a history of night operational sorties - with the exception of some L-39 flights that don't appear to be offensive in nature. --Charles Wood (talk) 06:47, 5 April 2017 (UTC)

No luck finding any visuals of the place so far, but I found this tweet from February 23, 2017 by @markito0171:
 * #Syria ~80 corpses found in fuel-tanks of Camp "Khazanat"/Khan Shaykhun -executed by "Liwa al Aqsa" before their retreat to #IslamicState

--CE (talk) 06:24, 5 April 2017 (UTC)
 * That does sound intriguing - and I was going to suggest the same spot before work, noting the underground storage notes. But ... --Caustic Logic (talk) 10:06, 5 April 2017 (UTC)
 * This may be video of the 80 bodies. Massacre tanks in Khan Shaikhoun (Feb 25, 2017) -- Petri Krohn (talk) 10:54, 5 April 2017 (UTC)

Having now looked, camp khazanat doesn't seem at all cliffy enough. And now I'm about 70% sure it's the Latamnah compound as Petri suggests. The images are afternoon sunlight. If that conflicts with anything, great. Otherwise, I wouldn't say people lived or were gassed here. It's supposed to have a hospital, and the victims seem trucked in. It's not so far from Khan Sheikhoun that doesn't make sense - this might be the best hospital around for CW victims (the mud bath by un-suited men isn't encouraging, but...) Graphic forthcoming. --Caustic Logic (talk) 10:06, 5 April 2017 (UTC)

Here is a photo of the Al-Lataminah cave compound. I do not think they match. Besides, the Khan Sheikhoun caves must be facing south. -- Petri Krohn (talk) 11:04, 5 April 2017 (UTC)


 * Nice video find, Petri. I see nothing that would say it is NOT the place. It's a rather large area on the map, and they are on top of something. At 0:11 you can see that they are partly surrounded by the sort of green knolls we see in the event videos, and at 4:01 the camera goes over what seems to be a road carved into the surface through the whole picture, with people standing in front and behind it, and a higher structure to the left. Inside of that could be what we are searching for. Nothing definitive, but nothing to dismiss it either.
 * CL, your attempted match doesn't really convince me. The greens should be on different sides of the "road", the curves seem wrong, the buildings don't seem to match. --CE (talk) 14:38, 5 April 2017 (UTC)
 * I just noticed that all but one comment on the video are from the last 24 hours. I ran them through translation and they are all from knuckleheads cursing Assad in flowery poems, not realizing the date of the video. Somebody sells it as from the current "chemical weapons" event. --CE (talk) 15:07, 5 April 2017 (UTC)
 * Another video of the body recovery. At 0:29 a White Helmet is interviewed in front of a cliff similar to the White Helmets base. They could have come back to base for the interview. -- Petri Krohn (talk) 00:04, 6 April 2017 (UTC)


 * Okay, it seemed a good fit from the little I could see, but I guess it's not even that. The sun would be too low for that azimuth, come to think of it. I won't be digging for where this happened. It might be useful, but ... definitely too much going on. I'm skipping some stuff. --Caustic Logic (talk) 09:18, 6 April 2017 (UTC)

Scene Overview
Rough, basic idea map More precise map (by Michael Kobs on Twitter)

Discussion (starting with rough map, first version): This is what video shows at around 6:30 to 7:00 am (possibly not of April 4, but...) I used Bellingcat's field-of-view study (see here), which seems accurate. Here, blue lines from the camera to each plume, pink circles for rough area of high-rising, high-explosives plumes, and red for the oddball white plume that might be some CW. Just where each is along the line of sight is a guess - it's set broadly (big ovals), but they could be closer or further out on their lines (the ovals are not angled right to follow the line, so image that.) The red circle may be more exact: it's a short cloud but seems to cast a shadow on the tel, so must be only a bit northeast of it, approximately at that (school?). Crater date I think is/will be explained. Hospital timeline explains that later blast. --Caustic Logic (talk) 09:49, 11 April 2017 (UTC)


 * Bomb cloud sizes vary a lot. That could indicate either different size bombs or bombs that don't completely detonate. The really small one could be just the primer going off and not the main body. Of course a CW bomb is also an option but the variance in size should be considered.


 * Ultimately, actual fragments from the anomalous bomb will help resolve the question. They haven't been forthcoming, so the balance of uncertainty starts to favour a damp squib rather than a devil's brew. --Charles Wood (talk) 10:02, 11 April 2017 (UTC)
 * Agreed on smaller bomb/primer possibility. With fragments, there will be an issue of possession, that it's not cooked up--Resup (talk) 10:39, 11 April 2017 (UTC)

Amnesty upload time search indicates that video was posted before what we were told was the attack time ....--Resup (talk) 10:25, 11 April 2017 (UTC) Video ID: MYOMEDK_uVs Upload Date (YYYY/MM/DD): 2017-04-04 Upload Time (UTC): 04:59:28 (convert to local time)
 * Converting to local time means add 3 (2 for time zones, 1 for DST) giving 7:59, which is almost slow, but reasonable. --Caustic Logic (talk) 14:03, 11 April 2017 (UTC)

Plume 3 Area
Reuters claims that this photo shows a bombed house, but the surroundings don't look anything similar to the other 3 locations we already have. I could swear that I have seen the interior of this house in a video somewhere, but I can't find it. --Q (talk) 16:11, 13 May 2017 (UTC)
 * The more I look at this, the more similar it looks to the southern bombed house, the white star on the ceiling and the wooded things on the walls suggest that the two rooms are the same. The photo from NYT most likely shows a much later state, it would explain the lack of furniture and the differences in debris placement --Q (talk) 16:53, 13 May 2017 (UTC)
 * As I recall, Postol in one of his versions believed that munitions depo is where fattest-at-the-bottom-yet-mushroomed plume is. Also, if you have street name you may try translating to Arabic by robot and see about that; I quickly tried but was not getting smth immediately useful (+translating back was needed). Seems to me, on google map of the place few streets are signed in whatever language; but don't know a better map source --Resup (talk) 17:03, 13 May 2017 (UTC)
 * House-4-1 must be the house with the hole in the roof shown at 3:25 in the NYT video. If you look on the satellite picture, in the street you can see the trailer shown in the photo. --Andrew1 (talk) 17:32, 13 May 2017 (UTC)
 * Seems to me, Q's 2-1, 2-2 homes, and NY Times @3:25 are all (apparently) about the same location which got a smallish pickup with cabin in place. But in 4-1, there seems to be not enough space for full cabin, just quite massive trailer part, with less bodywork, that side at least. But there is an another location on NYT video around 3:14, what about that one?
 * Also, pretty sloppy for NY Times to claim confidently those places were homes; double use is employed by rebel forces all over the place, and often, for the obvious media benefit --Resup (talk) 18:23, 13 May 2017 (UTC)
 * Looking at it again I have got it wrong, you are right- the satellite photo has shadows pointing to where the trailer would have been if it were at 3:25. It matches the southern house at 3:14. --Andrew1 (talk) 19:34, 13 May 2017 (UTC)
 * Just looking at the scenes now. Indeed the photos show the same place, just not the same white star on the cieling (light fixture). There's one in each room. All the concrete damage is the same pattern at the edge of the slab in the stairwell area. Noting in the fuller photo the small building just north of this is also laying flat - must have been pretty weak - and the buildings across the street were hit with shrapnel, suffered minor damage. --Caustic Logic (talk) 04:30, 4 June 2017 (UTC)

So here's the damage visible on the video mapped out - interir behind the left wall not included, flattened building to the north, light damage across the street not included. There seems to be more blast power to the north, so perhaps this was a projectile fired from the south? Otherwise, it could be no direction, which means it's a bomb dropped on the stairwell, or planted inside of it. I can see any of these making sense. --Caustic Logic (talk) 05:07, 4 June 2017 (UTC)


 * Seen Postol's latest? I'll just put it here for you, I'm not enough into the topic to know where it belongs best. Deals with the damage among other things. --CE (talk) 11:10, 4 June 2017 (UTC)
 * CE, thanks for popping in. Indeed, and I just posted a rough post on that and analysis of the sites. I felt the need to cast a fairly harsh light on his obviously baseless conclusions and the confusion they're helping perpetuate. That's over here. Puttin' my foot down on this one. --Caustic Logic (talk) 13:18, 5 June 2017 (UTC)

Affected Homes
(allegedly anyway) So far we have three identified sites where families were said to live and be affected by the gas, and to largely die. So far these are all in a narrow north-south strip in the town's northeast. Counting north-to-south:

1) Commentator and now ACLOS member Andrew brought this up in comments at the Monitor blog (first comment here) - marked a bit incorrectly on the map above as 'basement.' That means the corner building the host of this video starts out standing at. But the door he takes us into is actually a bit east of that marker. The alleged sarin crater is about 115-120 meters E-NE of the correct spot. Better graphic and notes in a bit... --Caustic Logic (talk) 08:21, 1 May 2017 (UTC)


 * This is a bit rough, but they go in the door boxed in red. Sparse basement living area shown. The place looks pretty fancy, but the inside looks more like squatters, hostages, or just some bedding for show was here recently. --Caustic Logic (talk) 09:24, 1 May 2017 (UTC)

2) ACLOS member Q located this and made the nice graphic - the corner building above boxed here in red - the home in question is south of that, marked here in green. Account details pending translation... --Caustic Logic (talk) 08:21, 1 May 2017 (UTC)


 * note: this one is not yet on the map shown above. It would be just south of "basement."
 * Added to the map now. --Caustic Logic (talk) 02:10, 7 May 2017 (UTC)

3) The furthest south is what a "white Helmets" volunteer claims as his family home, where his wife, children, nieces and nephews, brother and sister-in-law all or mostly died. Approximate location centered on Wikimapia, and shown on the map above. The alleged sarin crater is 275 meters to the north-northeast. --Caustic Logic (talk) 08:21, 1 May 2017 (UTC)

See also below for three more identified homes a bit to the southwest of these. --Caustic Logic (talk) 02:10, 7 May 2017 (UTC)

A Huffington Post report The Alyousefs brought their dead to a family member's home that was outside the worst attack area. The courtyard was turned into a makeshift morgue where surviving relatives tried for hours to resuscitate loved ones already dead. This is probably where all the family members are shown inside, and someone else saw them in the courtyard there. Has anyone located that yet? There's a home where he's interviewed, fancy but with a gate that doesn't open well - somewhere on Twitter this was placed. Might it be the same? I'll look into it. Among other things, this would help set the alleged gas area by being outside of it, or the worst of it. --Caustic Logic (talk) 10:04, 20 June 2017 (UTC)

Weather data for April 4th
According to Intellicast, weather at Hama (close enough) had light winds from the East / East NE sector at 6 mph around dawn. --Charles Wood (talk) 20:49, 4 April 2017 (UTC) Deleted reference to incorrect intellicast wind data --Charles Wood (talk) 08:28, 1 May 2017 (UTC)


 * Postol assessment 041117.pdf States winds were blowing to South-East - away from inhabited areas. I have seen two different sources that indicate the winds had a Westerly component at 10m blowing towards inhabited areas. This is not withstanding the apparent easterly component of the first video bomb clouds which I put down to different wind direction above and below the inversion. --Charles Wood (talk) 18:41, 12 April 2017 (UTC)

I can find no historical weather data for Khan Sheikhoun or even Hama for April 4th. The nearest weather stations are at the airport in Latakia, in Qoubaiyat-Akkar in Lebanon, and in Cezayirlioglu Ciftligi in Turkey. All show north or southeasterly winds at 6:30 am on April 4th. The bomb blast video however shows the wind blowing from the west. Which makes on question if it was actually filmed at some earlier date. -- Petri Krohn (talk) 06:50, 14 April 2017 (UTC)

The nearest reliable data is METARS data from a number of Syrian airports OSLK/Latakia has winds blowing from the North East direction at 03:00 UTC. Other data is intermittent --Charles Wood (talk) 07:06, 14 April 2017 (UTC)


 * Other sources tend to be very unreliable and often are an interpolated value on a mesoscale grid. METARS are the gold standard. --Charles Wood (talk) 07:11, 14 April 2017 (UTC)


 * Also forgot to mention. Wind direction above and below an inversion can be wildly different. There would have been a strong inversion there assuming the light winds everyone seems to agree on. The general name for this phenomenon is backing and veering with altitude (certainly not an Eckman spiral as the armchair experts are wont to say). This is especially obvious when an inversion is present.


 * I noticed in the video a direction change on some bomb clouds with altitude. The 'kink' will be at the inversion level which would likely be under 100m and perhaps significantly lower. --Charles Wood (talk) 07:32, 14 April 2017 (UTC)


 * Visual assessment: in this video, the faintly yellowish gas plumes seem to me to be moving right to left, when facing south, which means from the west, roughly (about opposite of what the Hama records suggested). One apparent origin point seems to be a bit SW of town. Overall breeze might be a bit from the W-SW. Reviewing. --Caustic Logic (talk) 11:19, 15 April 2017 (UTC)




 * I looked at the video. I didn't see any movement in the cloud/mist. The only clue is tapering of the density. I'm not sure that is a sound approach.


 * Re Hama weather data. It's one of several sources, none of them consistent. The only constants I have seen from the different sources is there is a component from the North and the wind speed is very low ~ 2 m/s


 * The good thing about your new video is the clouds appear to be below the inversion, so whatever they are doing it will be consistent across the field of view - unlike the bomb plumes which had a clear change in direction with altitude. --Charles Wood (talk) 12:37, 15 April 2017 (UTC)


 * Looking again, it's not as clear as it seemed, but I still see slight movement to the left, but it is roughly still. Definitely no movement the other way. Also, it may have drifted the other way earlier. Partly I'm thinking about origin areas. Above I noted the white cloud northeast of the tel. That is the same color as this. It might be what spread all to the left, and something spread a bit to the right along the tel's north face. Or maybe a similar cloud emerged at the crater area off the left side and drifted right, as the other's smoke drifted southwest behind the tel. There also seems to be a center of spread off to the southwest, right side distance, spreading west and northwest, split around a low hill, perhaps. Might be that impression making me see the consistent movement. --Caustic Logic (talk) 01:35, 17 April 2017 (UTC)

Color: I retract yellow-ish - I can't make out a color, even after enhancing it. This seems to be a widespread white fog. Is this sarin? It was said to smell bad, burn the eyes, and be yellow, but this seems white, though it might stink. I should think if this was all sarin, half the town would have died. So if not, could it be anything legit from the attack, or is it maybe the other, less toxic poison used? Or just some effects smoke to paint their picture of big sarin release? --Caustic Logic (talk) 01:35, 17 April 2017 (UTC)

Theodore Postol now says he had the weather data 180 degrees wrong. The wind was blowing from the east. See: -- Petri Krohn (talk) 15:27, 22 April 2017 (UTC)
 * With Error Fixed, Evidence Against ‘Sarin Attack’ Remains Convincing - ThuthDig, April 21, 2017


 * Well he is simply wrong. I explained elsewhere about this. Not only does the regional data disagree with him, the video evidence disagrees with him. In this video at 1:05 wind is blowing towards the rising sun i.e. from West to East (plus some not well seen North/South component). They very helpfully zoom in in the base of the cloud and so we know it's under the inversion and therefore highly relevant to ground level gas dispersion. --Charles Wood (talk) 20:42, 24 April 2017 (UTC)
 * OK, we've switched positions somewhat on the wind and disagreeing differently. That's the video I called unclear above. Better version. There's a definite shear to the left higher up (so east, towards the sun, but it seems past tense, now sitting still), but surface wind matters (as you've noted, and know more than I about). At the base, I'd say the direction is a bit to the left, so the east, and towards the camera. This seems to be looking a bit SE, so that I think makes it almost northerly wind (to the north). It doesn't seem strong. AND I'll add the main 3-plumes view shows about the same, if clearer - a bit left and towards the camera, looking almost due south. So that would seem well-founded, but...


 * here we face north-NW and the wind clearly moves to the left, away from the sun, so to the west or W-SW. And here (and odd view, possibly not connected) it's the same, facing NE, it rolls away and to the left. So seems like winds changed a lot, if these all show the same day - almost seems wind always move left, whether you're facing north or south. Is that even plausible (to have such variation in one short span)? But the predicted direction Postol used is pretty close to one of the seen directions. So ... I just switched my position again after noting the 3-plumes view - it seems like a tie, and both directions are right. --Caustic Logic (talk) 09:39, 25 April 2017 (UTC)


 * Your video https://www.youtube.com/watch?v=fHQA7hliJCY is obviously a different day. Wind much faster and heavy cloud in background. The clincher of course is it was uploaded on April 3
 * The second one also was uploaded in 3 April.


 * Regarding wind and elevation. It's entirely possible for the ground wind to be a significantly different direction to the winds higher above an inversion. There was an inversion in KS on the day. The winds were very light, so changes with direction with altitude are entirely possible, if not expected.


 * Getting onto Postol. He got the direction wrong by 180 degrees in his first go and managed to produce something looking authorative.
 * He then switched direction (as a result of my telling him of his error) and produced yet another authorative looking report.


 * I don't think either report is worth consideration. The data he uses is completely unreliable to start with and he makes conclusions he can't possibly make given the lack of data. At best it's a hypothetical / descriptive exercise but nothing of scientific or forensic merit for the facts on the day at Khan Sheikhoun. --Charles Wood (talk) 10:35, 25 April 2017 (UTC)


 * OK, sorry then - both were uploaded even Aug. 3 in Syria time - one about midnight (almost the 4th) and the other hours earlier. No wonder the winds were so different. Pretty consistent visual direction then, FWIW. And the basement home would be upwind from the crater. --Caustic Logic (talk) 12:37, 25 April 2017 (UTC)

Wind direction as seen on videos

 * ''Syria denies bombing Khan Sheikhoun on the morning of April 4th. The three videos have not been independently verified to be from that date.

Wind Direction decided: Charles and I discussed this some, and I decided the winds is southwest, and blows TO the northeast, 45-50 degrees or, as this slightly too-early graphic shows, 45-55 with a broader range of 20 to 70 degrees possible. See Idlib Chemical Massacre: The When and Where for this and other geo-temporal findings. The significance of that to the alleged sarin spread to those homes is pretty evidently a mismatch. In fact, those homes are in exactly the opposite direction of the wind. That could be a total coincidence, or could actually be someone deciding where to say people died based it on an accurate wind reading - read backwards, like Postol did. --Caustic Logic (talk) 02:04, 7 May 2017 (UTC)
 * Update, June 1: a while back I noted a wrong line of sight estimate (my bad) that I used too centrally to set the estimate. Doing the same now with a correct line of sight, I estimated a core range of 30-45 degrees, as the direction again it was blowing TO. And finally here's the updated graphic. --Caustic Logic (talk) 12:07, 1 June 2017 (UTC)



Good work, but I don't agree with this wind direction estimation. I can see some movement to the left in the video from the northern filming location, and there is also a strong movement to the right in the NYT video from the other filming location in the SW edge of the town. Any wind below 54° azimuth would cause the fume from the 2nd strike to move towards the left side in the NYT video, but we see the exact opposite. As I see it, the movement is more pronounced in the NYT video than in the other one, at least near the base of the plumes. Overall, the wind direction should be somewhere in the 68° - 144° range, likely closer to 144° than 68°. --Q (talk) 20:19, 9 May 2017 (UTC)
 * Was not following all details and getting a bit confused; do we know where cameras were, and how do we know that? Kobs placing one camera to the North--why? There are two different episodes in NYT video (below), which one we discuss? (first, 0:50, second, around 1:38). I propose to collect separately what is that we see in each case, and what else we know from elsewhere. I'd say, in both cases the Sun (=East, for simplicity) is to our left. That means, we are North from whatever we observe, in both episodes (in the second one, we are close to be looking at the Sun position, less so in the first episode). We see movement to the left (and apparently away) in the first episode, and to the right and definitely away in the second one. Do we observe the same thing? --Resup (talk) 23:37, 9 May 2017 (UTC)
 * Yeah, I did not specify that I was talking about the part starting at 01:33 in the NYT video. The footage seen in first segment(at 00:50) was originally uploaded by a man named 'Mohammad Saloum', Bellingcat geolocated it, the northern camera position in Kobs's and CL's map refers to that video. The second segment in the NYT video contains 3 clips, all of them were filmed from the same location. The camera is facing NE in the first(01:33) and the third(01:44) clip, in the second clip(01:38) it's facing ESE, Kobs placed this camera on the W edge of the town. So, to reiterate what I said in my previous post, I see the fumes moving to the right in the NYT video at 01:33, I can also some movement to the left in this video in the same area, and based on these observations I place the wind direction in the 68° - 144° azimuth range. --Q (talk) 14:49, 10 May 2017 (UTC)


 * Okay, there is a slope to the right - if we mean the same thing - in the lower part of the two flumes seen from SW at 1:33 in the NYT video. If that's a main point, it's not a clear wind clue, IMO. When the weapon detonates, there's an initial blast shape, from the bomb and what it hits, directing force a bit this way or that, and in the case of fired rockets, a shape from the trajectory moving the explosives forward as they detonate and expand. Result may be an angled fireball rising, cooling to ash/smoke, forming a mushroom cap, then... having the wind take over (at the bend) and shift the smoke in a direction, as all clues suggest, mostly away from this camera (invisible movement to the NE) and a bit to the left (so some degrees more to the north than this line of sight I now revise to around 50° on the compass (was erred, 56-57°). Hence core estimate range 30-45°) (copying your °). Note the movement - not bend but movement - seems very slight here. That's from most of the movement being in an invisible (or hard to gauge) direction. The north view clarified it's away from, not towards the camera, and it's pretty evident.


 * If there wasn't another point, that should maybe do it. Agreed, Q? --Caustic Logic (talk) 13:44, 19 May 2017 (UTC)


 * Curious what weather report gives (perhaps not observation based but anyway): 2 mph SE at 6 am (here seems consistent),  changing to stronger E/ENE wind past 9 am --Resup (talk) 16:07, 10 May 2017 (UTC)


 * It's not a weather report. It's computer generated data using a proprietary computer model operated by WWO See WWO FAQ. It turns out the model for KS is mostly based on data from Basel al-Assad airport about 60k West of KS. The airport is 350 m lower in elevation so a lot warmer, and the wind there was a 'land breeze' blowing straight out to sea due to temperature difference between sea and land (the opposite of a sea breeze).


 * Regional winds based on the European isobar map indicate very light air flow over all of Syria coming from the North. However, the wind speeds are so low local variations in just about any direction are possible. --Charles Wood (talk) 23:01, 10 May 2017 (UTC)


 * I think Charles and I agree the best method of predicting the wind was doing is - if possible - is to read a reliable locally-recorded measure, or to look at video recordings from different angles of what it actually was doing. That's possible here, shows it blowing from the SW, to around 30-45 degrees northeast, and as a side-effect, it helps show the limitations of the various predictive models used around. --Caustic Logic (talk) 13:59, 19 May 2017 (UTC)

Also, it might be useful information that the bomb from the second strike most likely exploded in closed space, see image. video from inside, video from outside(at 01:18). The fumes from this explosion could only leave the room through the door/window in the northern side, or through the hole in the ceiling. I'm not sure how these circumstances affected the shape of the second plume, but I think the fumes raised with more power initially because all the trapped hot air in the room wanted to escape through one exit point. --Q (talk) 20:42, 9 May 2017 (UTC)
 * Indeed, that's one of the things that can give a flume its initial shape and direction. Another is if they were, say, powerful FAE rockets fired at a mid-distance from the north. I think this is possible, and seems to fit some evidence, but I need to do a review... --Caustic Logic (talk) 13:44, 19 May 2017 (UTC)

Question on BC geolocation (did anybody checked it? ). Camera is North from what we see Sun is on the left, but it can be anywhere. Lines he draws are essentially parallel (we are essentially at infinity) and may go anywhere you want. It needs to be fit of distortions of visible distances between objects, or comparing visible heights with actual heights (which we do not know). Was any of that done, or else how it was placed wherever it was placed? Don't have time at the moment for doing it. It may not tell very much, but if we are to draw our wind directions ourselves, some of that sort needs doing. --Resup (talk) 15:32, 12 May 2017 (UTC)
 * I never did verify it myself, but I ind their work in this area is usually trustworthy and technical. Here, all features seem to keep lining up along those lines of sight like they would only from that point or one really close (at least, barring an extra-smooth trick that might not even be possible... not sure, really). In the right area are three small buildings; I wouldn't say which one it is, but I think it has to be one of them. not open field. The lines of sight seem exactly the same between the videos, but the later one (showing the fog spread) seems filmed from a bit higher looking down, for different vertical line-ups. So he went up to the roof for the second video. --Caustic Logic (talk) 13:44, 19 May 2017 (UTC)

Revision: 30-45 degrees NNE Sorry I totally flaked on this discussion, guys. I'll see about responding to parts anyway. First, I set out a fuller explanation of how I decided on this wind direction, here. Along the way, I noticed I had set up an important field of view wrong, and Michael Kobs got it right (or, not 100% verified, but not stupid like mine). That had set the direction most directly. But the view is about 7 deg. further to the north, so the direction is about the same, and I expanded the core estimate from 10 deg. to 15. 40-50 deg. becomes 30-45. All visual clues to me are pretty clear in this, with the same direction at ground-level, maybe 6-8mph, and faster at the higher levels. The only things suggesting another direction are other features, like the ininital (lower) bend of rising smoke plumes (some kind of blast dynamics mostly set that) and the expanding white cloud cited by Bellingcat's Timmi Allen here on Twitter. (side-effects of that: Allen confirms general left (east) direction, and shows there's some reason suddenly to discuss the wind direction is this case and argue against what the video seems to show...) --Caustic Logic (talk) 13:26, 19 May 2017 (UTC)

Bubslug Challenge
My analysis shows the wind direction being more towards the east. I used the apparent drift angle of bomb clouds #1 and #2, viewed from the north in the "Saloum" video, and from the southwest in the NYTs video, using the accepted video geolocated camera positions, and view azimuths to target clouds, indicates the bomb clouds were drifting between azimuths of about 80° - 120° which narrows the range somewhat from Q's estimate. Here's the graphic and table of results.--Bubslug (talk) 21:58, 13 June 2017 (UTC) As a further comment this method allows wind speed calculation using vertical velocity of the cloud from videos in seconds per metre, using a known length/height reference for cloud height at X seconds, and the actual true drift angle.--Bubslug (talk) 22:17, 13 June 2017 (UTC)
 * Huh. That's way different, even south of east instead of north. I'll have to look at this later on. Did you use the new position for camera 3? Yes, looks so. You see a drift to the right? Ok... that's where we disagree. I didn't use the lower slant - that, I think, is from the initial blast. Look as high up as possible for pure wind. It's been said there's almost no movement visible here, but I see it. It counteracts that right slant and starts pushing to the left. That will be the wind. Later, it shifts to even more north (plume dispersion, as seen on cam 3) --Caustic Logic (talk) 23:35, 13 June 2017 (UTC)
 * Looking again, I visually think I'm seeing a slight left drift. But comparing an early frame and a late one - 1:33 and 1:36 - with lines at various spots, it seems more like no movement, or just a hair to the right, but that includes the expansion - the left side doesn't move right, it just hold as the right side shows the plume is just growing a bit. There's expansion here as well as a lot of rising, and note this is some moments earlier than the view from the north, so the initial blast angles and dynamics will apply a bit more, and the wind a bit less. I think we're seeing where the wind is just halting the right movement - individual particles and sub-plumes seem basically stationary over 3 seconds. From there - it might appear still (meaning movement almost entirely away from the camera, making the wind TO about 50 degrees), or a bit left-moving, which I think I see the start of here - nearest the very top, especially on plume 2, red line area.

--Caustic Logic (talk) 12:32, 14 June 2017 (UTC)


 * After looking at the evolution of the bomb clouds, I think you are right that close to the ground there is little wind, but the higher the cloud rises the more it deflects, although the effect is subtle. But clearly the angle of the "stem" to the "mushroom" represents the effect of wind. The mushroom is not quite as clear as it is expanding in all directions, including into the wind direction, making the upwind side of the cloud look somewhat stationary as it rises. Here's a history of the bomb clouds #1 and #2, shown at seconds 0,4,8,12, and 16 seconds into the Saloum video, so you can judge for yourself whether the angle of the "stem" represents wind direction, or in the case of any video view, apparent wind direction. Looking at this I might shallow my angles out a bit (the blue and yellow lines in my apparent drift angle interpretations) to represent better the wind vector nearer ground level.--Bubslug (talk) 20:04, 14 June 2017 (UTC)



I largely agree, what we see has to do with late stage slow rise of hotter less dense air due to buoyancy (with fingering giving it stem and mushroom shape), while that rise is drifted by the wind. So if we project that mushroom down onto Earth surface, it will be in the wind direction. This is not that much about the blast where stuff moves at near sound speed (ie much faster than we see here, and by he time mushroom forms that is already gone). Blast itself, on this sort of scale, will appear to us as pretty symmetric fireball, will be hard to detect any slant, especially if this is fuel-air type, and moreover parachuted. But even Grads or shells initial blasts from large distance appear to us as fireballs, followed by dusty stuff rising (while looking at the crater, or watching from nearby will tell more about which direction it came from) --Resup (talk) 22:33, 14 June 2017 (UTC)


 * Okay - of course I'm not really any expert, but here's how I see this. "clearly the angle of the "stem" to the "mushroom" represents the effect of wind. " If there's just one direction, it could be no initial trajectory and just the wind (as with a dropped bomb), or maybe trajectory and wind appear the same from that angle. But if we see two different rise angles, one low and one high, that's two different forces, and we're probably seeing the initial trajectory and then the wind. The initial fireball is gone by then, burned out an risen, but I think it creates sort of a chimney in the air - as hot particles rises, it creates a vacuum, which pulls in more from below. It would continue as long as the area is hot enough, and that tower of smoke would maintain the basic shape created by the fireball, with the wind affecting the direction a bit at the bottom and a few bits higher up. Some smoke that's too far away will drift out side windows, etc. But the majority will be pulled up by the suction of the prior stuff into that same angled column. At least, that's my reasoning. (I'll see if I can find an example...)
 * Here, it first angles right seen from the SW, then changes direction higher up - first vertical, then at the very top, very slight movement to the left. I think I did overrate the left movement seen; it's basically zero. But just going more to vertical means the wind is countering the trajectory, so it must be (partly) to the left. As for the north view, we just see one prevailing angle, with some wiggling. So in that case, we're seeing little or nothing besides wind drift. Of course there is a bend, we just can't see it from here because it bends roughly along the line of sight.
 * If this is wrong, I'm open to learning that. Just trying to make sure we all get what I'm talking about in the first place. It is sort of complicated. --Caustic Logic (talk) 11:01, 15 June 2017 (UTC)
 * Example video: A combination of trajectory and blast dynamics has the red portion of the fireball roll right and towardsa the camera. It starts in slow-motion and resumes normal speed at 0:33, so I start counting from there. For about 16 seconds, the plume continues to rise out the top of this angled chimney (orange box). It grows a bit longer, but keeps the angle, until the heat below diminishes and around 0:50. Here the chimney disappears, and the plume visibly separates from the source. Just what the wind is doing is not as important, but it seems to be blowing left, halting the right expansion and, at the video's end at 1:03, starting to blow the plume a bit to the left. I think this is pretty much what we're seeing here.

--Caustic Logic (talk) 12:06, 15 June 2017 (UTC)
 * Yeah, it can happen if explosion happens to throw stuff in a directed way; what I am saying, typically it does not, looks like a pretty round fireball, but there exceptions --like yours, directed big hit on a hilltop, or a building wall, etc. But just falling at an angle is not enough, like eg slow motion here, many slow motions here and on many grad videos. Steming is unstable so fancy stuff can occur --Resup (talk) 19:13, 15 June 2017 (UTC)

And let me put it this way, as I should have before - there must be two forces at work to explain the curve that we actually see, isntead of a simple slant angle. One force clearly is the wind, more like the one that prevails higher up. The other - could be a different wind at the other level. This can happen. But many independent signs suggest the same basic direction applied at all levels and over the area for the relevant time - only speed increases higher up, that we can see. So the lower angle might just be kind of like the above example. It only needs to hold the shape for ... however long it's actually sen for. Maybe 20 seconds or so? This angled "chimney" is more dramatic from its less explosive nature, no big competing plume sucking material in other directions, etc. Its shape is more like a clean and simple mushroom cloud you'll get from a mid-power FAE explosion. I think the damage suggests just that, and fired from the north (see Monitor analysis, and sorry all that I haven't brought more of this over here). --Caustic Logic (talk) 13:42, 16 June 2017 (UTC)

Couple of points: this apparent change in direction appears to be on all 3 tall plumes, at about the same height, as far as I can tell (that favors non-random cause  as random things are not likely to be biased in the same direction, and can be wind change with height). Another thing, there is somewhat hilly terrain (+ city-'terrain') which will cause wind to go around and over top (some vortices could be formed too). Terrain and buildings can also direct initial blast prevailing direction sending plume there initially (quickly), picked up by buoyancy and wind once that is slowed moving through air, by air resistance/drag. (No need for extra 'second force', if it is sent flying there by blast it will keep flying for awhile; force is needed to change velocity, not to keep it --air +wind drag + buoyancy forces here. Either that or wind change with quite undirected blasts--and those are common) --Resup (talk) 08:55, 17 June 2017 (UTC)
 * From the north, with plumes 1+2 anyway, there's no clear bend, so presumably that angle is all from the wind. What it does at altitude is increase, but keep pushing left. For the west view, I don't see buildings and topography being the second force to make both plumes angle to the south. Topography: good point. The big hill seems to affect the 'white cloud's' expansion, situated just NE of the tel, it would be partly in its wind shadow - this would block the wind on its right side, and we see it billowing freely to the right, while wind is hitting its northeast corner (left side as seen) and smearing it to the northeast. The white cloud location is dispuited, but I think it's there, and it halps confirm the wind direction.

And I haven't shown this here yet, though it's mentioned above. Here's the same video some minutes later, after any right-leaning smoke chimneys would have faded. This should all be wind, apparently faster the higher up we look, but all clearly to the left. This might be a faster left movement than present before, but it seems to be the same direction. --Caustic Logic (talk) 12:10, 17 June 2017 (UTC)

Bubslug challenge, re-considered much later. All that above about blast direction - maybe, to some to degree. Otherwise, nevermind. I re-considered right vs. left wind in this view and what it means. Some new graphics in this tweet explain how we're likely seeing two different northeast-pushing winds, one a bit more east at low levels, one more north at upper levels. That's not deduction - both directions of movement are actually seen in the NYT video of smoke dispersal, besides being suggested by the shape (if we take it as plumes 1+2). I didn't notice that before. I now say this wind is the main or only cause for that lower slant - as Michael Kobs suggested, there's a left wind at upper levels, right at lower. I was being dense to deny that, thinking it challenged my otherwise solid NE wind, But it doesn't. The key is in what I think Bubslug missed. You can't take left and right of straight ahead from the camera as the actual directions, not even if you average two of those together. Both views also have towards-away movement to consider, if hard to quantify. The known minimum speed (at least as fact as the l-r movement seen from the north, + unnoticed towards-away movement), compared to very little l-r movement seen in this NE view, means most movement must be away (because that movement is clearly partly to the east, so it can't be towards this camera off to the southwest). Therefore, as the scene suggests, it's all blowing NE, just a bit differently at each level, at depth pushing the zones of smoke further and further apart, left and right of this guy's line of sight, which is northeast. The bulk of lower smoke seeming to still be in his FOV means it's not blowing much differently, like due east or southeast, or it would be further off frame to the right. Knowing I've been wrong before, I can't say this is a final answer, but it's a best try anyway. I hope the floundering is over on this point anyway... --Caustic Logic (talk) 03:46, 12 March 2018 (UTC)
 * Adding: I suppose the comparative angles graphic is correct as for the angle of the lower smoke column, so scratch what I said about right angles. Two of them weighted for apparent angle should give the correct not-right angle for each view. I think that means these do angle to the E-SE as you show. But this isn't necessarily the wind direction - the SW fog spread at least suggests north and east and no south - it could be the points I make above play in like a rocket flying south or s-se has its fireball formation on that line blunted a bit by the largely opposite wind, and then slowly shifted. In fact such a hot event as an FAE blast (suspected weapon type) might alter the wind for a moment right in that area (?) Either way, the later drift doesn't seem southeast to leave this much of the smoke in our view to the NE. Re-considering now what the lower direction might be, could be a bit ESE, but I suspect ENE is still more likely, and east might be the best word in the end. Hard to say how far the lower smoke continues off frame to the right, but it seems further than the upper stuff has shifted left, so less northeast than I was thinking. --Caustic Logic (talk) 10:36, 13 March 2018 (UTC)

Date of videos?

 * ''Moved to Talk:Alleged Chemical Attack Khan Sheikhoun 4 April 2017/Timeline

Fog?
Crude (and possibly nonsensical) estimate: estimate drop size by the condition that wind drag on spherical droplet is roughly equal gravity force ; obtain radius r = (3/4) D (v^/(2g)) (rho_a/rho_sar) ~ 0.03 mm (too small ? or what? ). Then take 'fog' of visible size L to mean that on distance L you are sure to hit q ~ 1  droplet. That means, L Pi r^2 (N/V) ~ q, with N = V_sar/(4/3 pi r^3) and V ~ L^2 h, h --fog height. That gives, L~V_sar/(4/3 r h q); taking V_sar of 0.25 m^3, h = 15 m, q=1, gives L < = 400 m = visible fog size of ODAB 500 converted to CW. (Any better ideas about such sort of estimate?). I think we see some 1-1.5 km of fog on the vids (Alternative to that is to see attenuation length of water at size L, but that is some 20 meters of water to be seen, and will not work at all, unless sar is much more absorbent than water. So, decided that hitting few scatters means opaque enough. Unsure what proffi do here)
 * If only 1/2 of 250 or so liters goes up the air, and 2 scatters needed for opaque, get L~ 100 m, etc... --Resup (talk) 17:36, 10 May 2017 (UTC)
 * (In reality, more complicated; this estimates size of droplets which will drop down, really, smaller ones may still float, this is saying those are comparable, Will look more into this when will get some time; if there are better ideas, appreciated. Have seen sizes 50 micrometers for initial ice particles sizes in clouds, this is comparable; but sarin will stay liquid. For water, somehow it takes -40 C for droplets to uniformly freeze, otherwise phases will coexist)--Resup (talk) 09:30, 11 May 2017 (UTC)

Temperature and Humidity
I used the dawn METARS from Damascus Airport (OSDI) to calculate temperature and humidity at Khan Sheikhoun (KS). The reason being there is no actual data from KS and the published data purported to be from KS is in fact inexpertly interpolated/extrapolated from other locations.

The two nearest weather-stations with observations are Latakia / Basel al-Assad airport aka the Russian Khmeymin airbase aka OSLK, and Damascus airport OSDI. OSLK is 70 km due west of KS and OSDI is around 200km south.

OSLK is in a different micro-climate zone to KS, being on the coast and significantly lower, so temperature is higher and the wind is strongly influenced by the adjacent Mediterranean sea. This is the station that contributed most to the Postol weather data but without any correction for geography. The wind was recorded as blowing due West (due to a land-breeze) and the temperature was a lot warmer. I reject data from that location as inapplicable.

OSDI is about the same distance inland as KS but at ~ 260m higher elevation. It has a similar microclimate to KS but is around 2.6 degrees C cooler. The difference in elevation between OSDI and KS affects the temperature and humidity, however there are simple formulas to correct for that difference.

Overall, Syria was in a very slow moving wind from the north under the influence of a region of slightly higher pressure centered on Cypress. Inland areas thus had very similar weather conditions.

Net result using OSDI METARS is that KS temperature around dawn was 7.6C dry, 3.5C dew point and 75% humidity - based on 5C dry and 3C dew-point at OSDI and correcting for pressure differences using appropriate adiabatic corrections (NB, OSDI data reported to nearest degree C). --Charles Wood (talk) 13:09, 16 May 2017 (UTC)
 * Using air density around 1.27 (for those temperatures) and adiabatic constant about 1.40 (7/5, mostly just dry air anyway at those temperatures), I am getting recomputed temperature of 7.5  (adiabatic correction to 260 meters down). Doing adiabate again for vapor partial pressure, with adiabatic constant 1.33, getting in the end relative humidity of 75%. So, same as you say, and agree about dew points (using Magnus formula, which is in wikipedia). But having an issue, that I guess this says, if no fog higher up, no fog a bit down. Perhaps. But fog is like over-saturation, saturated vapor pressure + water droplets. So I guess, if there is fog higher up, could be some a bit down too. Which brings an issue, can it be fog at 86 % air humidity. I guess it can be, in non-equilibrium, with water droplets getting gradually evaporated --Resup (talk) 19:59, 16 May 2017 (UTC)


 * Fog at 86% (1.2m AGL observation) is impossible unless there is a source of moisture such as a warm pond and then the fog will only be around the moisture source. Fog by definition requires 100% humidity --Charles Wood (talk) 02:12, 17 May 2017 (UTC)
 * I do not mean equilibrium, I mean dynamic/transient effect. Lower temperature so that vapor saturation pressure is below actual vapor pressure, so that droplets of water form and stay as a mist. Then raise temperature fairly quickly so that humidity becomes a bit below 100%. What will happen is that water droplets in the mist will start evaporate . Eventually they will, and the fog is gone, but for a little while it will still be there, takes time for water droplets to evaporate.
 * To get static fog at 1 m, indeed in needs to be very close to 100 % at the ground 0 m . With 86 % humidity and that temperature, stable fog and 100 % humidity is at about 250 m further up, not 250 m down --Resup (talk) 02:49, 17 May 2017 (UTC)

Sarin moist volume
Similar to the above logic also gives an estimate for volume of sarin mist which can stay for a while (quasi-static fog volume). Vapor pressure of sarin at around 5 C is some 100 Pascal. Condition is that we expand 200 kg of sarin gas to such a volume that at temperature of 5 C its partial pressure is under 100 Pascal. Even higher volume will cause droplets starting evaporating. The volume in question can be computed as usual and is some 33 thousand cubic meters, 46 m^2 if height is 15 m (its below minaret height). Ref, p12 here--Resup (talk) 03:07, 17 May 2017 (UTC)

New York Times video
The New York Times has done good work matching photographs of bomb damage to fresh satellite images. Their key claim is fraudulent. The witness who made this video says the Syrian jet dropped four bombs. NYT agrees. They match three blasts to photos from the ground. They then claim: "More video from that day show a fourth strike nearby. We don't see the strike in the first video, but we know the crater is new because it was not there in earlier satellite photos." The "video from that day" clearly show the fourth strike in question happened nowhere near the crater. Yet they use the witness testimony of four bombs and video of a fourth blast as proof that a Syrian bomb created the crater. This is fraud! -- Petri Krohn (talk) 05:06, 3 May 2017 (UTC)
 * How Syria and Russia Spun a Chemical Strike - The New York Times, April 26, 2017 (video)


 * The NYT video rubbishes the Russian statement of a bomb run between 11:30 and 12:30. What the NYT doesn't do is reference the two different videos from the cave complex about that time showing bombs being dropped - confirming the Russian statement of a bombing raid on teh Eastern outskirts.


 * Separately, I have read but can't find the reference that there were two bomb runs in the early morning with a total of six bombs dropped. I'm unsure if that was two aircraft, or the same aircraft doing two bomb runs. --Charles Wood (talk) 05:22, 3 May 2017 (UTC)

BBC link victims to houses

 * ''Moved from Talk:Alleged Chemical Attack Khan Sheikhoun 4 April 2017/Victims

Short BBC clip showing victims and walk-through by one victim 'Abo Rabeeea' showing the relationship of his house to other named victims Syria 'gas attack' victim: 'I can't sleep' --Charles Wood (talk) 05:07, 4 May 2017 (UTC)


 * I have a good idea where the old man's house is, but the video is cut in a very weird way, I'll work on it a bit later. According to him, the dead kids were around the same place as one of airstrikes. The name "Amira Saleh" sounds suspiciously similar to the name of the mother in this family(Amira/Ameerah = Abeerah?). If they are the same person, it would be interesting to see how the kids ended up at that place, common sense tells me that they should have ran away from the fume, not towards the middle of it. --Q (talk) 15:52, 4 May 2017 (UTC)


 * The BBC clip was indeed cut in a retarded way(seriously, why?), so I made an image to make its understanding easier. The old man seems to have some medical condition that causes amnesia/loss of orientation. I still find it interesting that there were unconscious bodies ~320m SW of the claimed chemical attack crater, and apparently the old man was also in that area when he started to feel dizzy. --Q (talk) 23:05, 4 May 2017 (UTC)
 * BBC clip starts from showing him 'returning to the scene' (possibly not his house), so I guess this is what you geolocated, 'the scene'? May be good to add a link with location on interactive map like google/yandex so geolocation can be checked by somebody having doubts but not the time to do it oneself. 320 meters SW from crater may be not much affected. At last in one version, wind is from S-SW making this place upwind. Downwind strongly affected area was estimated at 50-200 meters (reliability of that is uncertain, but being upwind and far amplify each other in having little effect)
 * Wonder do we know where is his house relative to crater ? (BBC shows him entering his house after a cut, so probably not. He stayed in the house for a long time after feeling dizzy elsewhere; so house must be in unaffected area).  --Resup (talk) 00:00, 5 May 2017 (UTC)
 * Yeah, I totally forgot to add the locations to the page, thank you for mentioning that. The clip has some erratic storytelling with no clear beginning and end, I'm not even sure whether the path they took (marked with the white line in the map) is the same as Abo Rabeeea's route on that day, or it's just an arbitrary one. In any case it sounds like the old man first went to the place of the second "airstrike", probably just to check out what happened. There he saw the kids (dead? unconscious? struggling?), but soon he started to feel dizzy so he headed home, and just as he arrived, he lost consciousness. His home is actually pretty close to the closest (allegedly) affected home we know of (~50m), but his exposure to the chemicals happened (started, at least) at the other place, also ~320-340m from the crater. If he took the BBC route, then he stayed more than 300m away from the crater for the entire duration of his journey. --Q (talk) 10:06, 5 May 2017 (UTC)

Who is Abu al-Baraa? Kareem Shaheen of The Guardian claims he spoke to him. Note, that Amira / Ameerah = al-Meerah ~ Abeerah ~ al-Baraa. The "L" is not pronounced when it precedes a consonant. Instead it adapts the same sound. -- 10:10, 6 May 2017 (UTC)