Alerivan here.
This article will be about how I used NASA FIRMS, Sentinel EO browser and geolocation skills to determine the time, location and perpetrators of the razing to the ground of an area covering roughly 41,028 square meters in the Masnaa District of El-Fasher, Sudan (see page 13 of this report). But first, a small primer on the Sudan civil war, for those unfamiliar with it.
The Sudan Civil War
The fighting began on 15 April 2023 in Khartoum. What began as a power struggle between the leaders of two armed factions that deposed the civilian led government in 2021 escalated into a full blown armed conflict. These leaders are General Abdel Fattah al-Burhan, the head of the Sudanese Armed Forces (SAF) and General Mohamed Hamden Dagalo (also known as Hemedti) as the head of the Rapid Support Forces.
Fighting is taking place all over the country and a breakdown of regions of control and frontlines is beyond the scope of this article. However, here is a map from the Sudan War Monitor group to give a brief overview. It is dated February, but should suffice to get a general idea of what’s going on and where.
Just as important as the what and where, however, is the who. Due to the fact that both sides have allied themselves with multiple militias, who bear no official uniform, it can be difficult to ascertain who to attribute a particular incident to. Fortunately, it is not uncommon to see one or two uniformed members of either the SAF or RSF amongst a larger group of civilian clothes wearing militia. So to aid attribution, here is a picture of an RSF soldier in uniform. Note the tricolour desert pattern, RSF Emblem and Sudanese flag emblem:
And a soldier of the SAF in uniform. Note their camo pattern is generally more green and brown, but also with a Sudanese flag emblem:
Now, onto the incident.
El-Fashir, 28/04/24
On 29th April, I began to see various posts detailing the advancement of RSF forces into SAF controlled El-Fashir and the subsequent fighting that began the day before. Unfortunately, a consistent feature of such skirmishes throughout the war has been the deliberate burning of civilian dwellings; the Centre for Information Resilience has been mapping these fires since the conflict began and you can view it here.
Following their approach, I sought to gather media relating to the fighting to see if any such destruction would take place in El-Fasher. Sadly, after searching Twitter with a search string of ‘El-Fasher’ in Arabic (الفاشر (filter:media), I found footage showing RSF fighters celebrating next to the burning of what is clearly civilian dwellings here and here, uploaded on the 29th April.
I quickly realised they were both of the same fire, as both videos contained some of the same people, but were filmed a few hundred metres apart and a short time apart. The man on the left is easily identifiable as the same person; the man on the right, with the poor screenshot quality, I had to identify by his rifle with matching stock and white tape on the barrel, more than his face.
As for the assertion that they were RSF, here’s a screenshot of one of the soldiers in the tri coloured uniform with RSF patches, at 0.03 of this video:
The next step was to chronolocate the time of the fire with the footage, to ensure that the footage was not of another place or another time.
Chronolocating the Fire
Reverse Image Search
First off, I reverse image searched a screenshot of videos as they appeared on the tweets, to check if they have been uploaded before, potentially in a completely different context; if they have, then this would be a case of misinformation.
Also if you’re lucky, this sometimes works to show other posts containing the video that may contain valuable context for your investigation.
This does not always work as well as it does with images, so returning no results is not conclusive proof that the video has not been uploaded before, but it’s still worth doing. In any case, I did not find historical postings of the footage nor did I find other current posts of it with added context. It therefore seemed likely to have been filmed the same day it was uploaded; the 29th.
NASA FIRMS
The next step in chronolocating the fire was to see if NASA FIRMS picked anything up in the area at the time the footage started surfacing. It did, and it confirmed the suspected timeframe: the night of 29th April.
Both the time and the area were correct. I now had to wait for the Sentinel Satellite to pass over to get a more precise image of the burned area to assist in geolocating it, which it did on 30/04/24.
(Side note: I used FIRMS here following to confirm footage I found on social media, but an investigation may also begin in the reverse order; that is, you may find a new signature in FIRMS, then search for user generated content regarding the area, find footage of a fire and proceed to geolocate it.)
Sentinel Imaging
This is a timelapse of all Sentinel passes from 31/03/24 to 30/04/24. As you can see, several distinct fires have taken place over that time, but the one I was looking for is this one circled blue, in the same rough area as the FIRMS signature:
I then measured the area of the burn, which showed as roughly 0.5 square kilometres:
Then took the coordinates and transferred them to Google Earth Pro to begin the geolocation.
The Geolocation
Drawing a circle of 0.5 square kilometres centred on the coordinates from Sentinel, and considering how close to the fire the footage was, I figured I was looking for somewhere in this area:
Turns out I was correct; this is how I did it.
Analysing the Footage
Given the absence of landmarks, the three things I mainly relied on were trees, telephone poles and the width and colour of the road. With just these three features, I was able to geolocate the footage to this area and confirm that the fire was the same as that registered by Sentinel. Here’s the geo for this video:
And the geo for this video:
Final Thoughts
The war shows no sign of abating, fire will keep being used as a weapon of war, and the need to geolocate such incidents in pursuit of accountability will continue. Going forward, while I do what I can to contribute to the OSINT efforts towards accountability in Ukraine, I intend to also devote time to the OSINT efforts in Sudan, and I hope that this article has shown how you can do the same.
Thanks for reading.
