SSX360
Book a call

Security Services X-Ray / Field Report 002

Free to read

Reconstructing the Road to Khaled bin Walid Camp

From “Mocha intersection” to the camp gates

The source footage

Reuters published a video said to show Houthi forces capturing the Khaled bin Walid camp in Yemen, noting that it could not independently verify where or when the footage was filmed.

The original footage is available via Haystack News. As redistribution rights have not been confirmed, the full video is linked rather than reproduced here:

Watch the full source video on Haystack News ↗

Source excerpt · Mocha intersection. Supplied with the investigation. The full source remains linked on Haystack News.

For difficult geolocations, I usually start by reconstructing the spatial structure of the scene: roads, terrain, buildings, sightlines, and the position of objects relative to one another. I then look for a place that can explain the whole structure, rather than a single similar-looking object.

The first useful clue came from the Reuters description of the footage, which refers to a reporter in the video identifying the location as “Mocha intersection.”

Reuters description naming Mocha intersection.
Figure 01. Reuters description naming Mocha intersection. Reuters description reproduced in the supplied investigation. Select the image to inspect the full-size figure.

Searching that phrase directly was not very helpful. But “intersection” or “junction” can correspond to the Arabic مفرق, mafraq, while Mocha is المخا, al-Mukha.

That led to the local name مفرق المخا, Mafraq al Mukha, giving me a much more precise area to search.

A cut slope beside the road

Once I had a search area around Mafraq al Mukha, I went back to the video and examined the first scene more closely.

The strongest visual feature was a steep exposed cut slope immediately beside the road. Its face runs right up to the road, creating a very distinctive terrain-road geometry.

Source excerpt · Cut slope. Supplied with the investigation. The full source remains linked on Haystack News.
The exposed cut slope beside the road.
Figure 02. The exposed cut slope beside the road. Video still and annotations supplied by GeoSudoku. Select the image to inspect the full-size figure.

I then searched the satellite imagery for a place where the relief meets the road in the same way.

That was the first point where the scene started to lock into place.

Comparing the cut slope with satellite relief.
Figure 03. Comparing the cut slope with satellite relief. Google Earth imagery and GeoSudoku annotations. Select the image to inspect the full-size figure.

Reading the 2023 imagery

The match was not perfect. Some buildings visible in the video were absent from the latest satellite imagery available to me, dated 2023.

I considered the possibility that these differences reflected changes to the area after the 2023 satellite image was taken, rather than a mismatch in location. On the satellite image, those video-only structures are shown as dashed outlines.

Video-only structures marked against the 2023 satellite image.
Figure 04. Video-only structures marked against the 2023 satellite image. Dashed outlines show the investigator’s proposed positions, not buildings visible in the satellite image. Select the image to inspect the full-size figure.

So I focused on more stable features, the cut slope, road geometry, surviving buildings, and their relative positions. Together, they still made this a strong candidate.

Testing the tower hypothesis

Around five seconds into the first scene, a tower-like structure becomes visible on a hill to the right of the road.

After the cut at about 12 seconds, the camera is much closer to the camp gates, and a very similar tower-like structure is still visible on the right.

My first hypothesis was simple: both scenes might be showing the same tower.

The tower-like structure at about five and twelve seconds.
Figure 05. The tower-like structure at about five and twelve seconds. The same-tower hypothesis is tested in the following sightline comparisons. Select the image to inspect the full-size figure.

The pattern of low dark clouds changes between the two scenes, although this is only a secondary clue and does not establish the camera's orientation. The suspected same tower remains on the right in both views. If it is indeed the same structure, a bend in the road could explain the different perspectives. My working hypothesis was a significant right-hand bend before reaching the camp gates.

GeoSudoku’s working sketch.
Figure 06. GeoSudoku’s working sketch. An investigator’s rough spatial sketch, not a measured survey. Select the image to inspect the full-size figure.

This is a rough working sketch I made for myself while analyzing the footage, not a polished diagram.

Following the sightline

The strongest test came from the first scene. I traced a sightline from the road beside the cut slope, through the parking area and the one-story building, toward the tower on the horizon.

When I transferred that same line to the map, it pointed directly toward the tower-like structure beside the camp gates.

Tracing a sightline through the first scene.
Figure 07. Tracing a sightline through the first scene. Video tracing supplied by GeoSudoku. Select the image to inspect the full-size figure.
Transferring the sightline to the map.
Figure 08. Transferring the sightline to the map. Map overlay supplied by GeoSudoku. Select the image to inspect the full-size figure.

The map then resolved both working hypotheses at once.

The road does make the expected right-hand bend before the camp. And the tower seen in the first scene aligns with the same tower beside the gates in the second.

The road bend and tower alignment.
Figure 09. The road bend and tower alignment. Comparison supplied by GeoSudoku. Select the image to inspect the full-size figure.

What had started as two separate assumptions, the same tower and a change in road direction, now converged in one real spatial layout.

The observations in one spatial layout.
Figure 10. The observations in one spatial layout. Spatial comparison supplied by GeoSudoku. Select the image to inspect the full-size figure.
The camp entrance and tower-like structures.
Figure 11. The camp entrance and tower-like structures. Annotated video frame supplied by GeoSudoku. Select the image to inspect the full-size figure.

What the match establishes

The result: the footage can be independently geolocated to the road leading to the Khaled bin Walid camp in Yemen.

The match is based not on a single object, but on the combined geometry of the terrain, road, buildings, sightlines, road bend, and the two video scenes.

This investigation verifies the location. It does not independently establish the filming date or who controlled the camp when the footage was recorded.

Reported coordinates

Coordinates supplied by the investigator. The decimal places reproduce the case record and do not state survey accuracy.

Sources

  1. Reuters Connect: Iran-aligned Houthis said they captured Mocha. The source record states that Reuters could not independently verify the filming location or date.
  2. MarketScreener: Reuters transcript and description.
  3. Haystack News: full Reuters video.
  4. Google Earth satellite imagery. The latest image used in this investigation is dated 2023.

Credits and citation

George Sudo (GeoSudoku)

Investigation, geolocation, working sketch, and figure annotations by George Sudo (GeoSudoku). SSX360 editorial preparation by Ryan James York.

George Sudo (GeoSudoku). “Reconstructing the Road to Khaled bin Walid Camp.” SSX360: Security Services X-Ray, Field Report 002, October 26, 2026.

ssx360.com/publication/002/

Final web edition · Free to read · Published October 26, 2026

Questions or corrections: mission@ssx360.com. See our editorial and corrections policy.