Geolocating an Oil Refinery in Ukraine using NASA Fire Map

(Click here to watch and listen to the video version of this blog entry)


A few days ago I came across a video on twitter that at first seemed fairly easy to geolocate. The tweet mentioned the name of the city where the incident was happening and the type of infrastructure affected. The footage had a decent view of the surrounding area so everything was pointing towards a less-than-5-minute geolocation. However this was one of those instances in which my assumption was wrong. It has happened before when thinking that some photo or video might take too long (or even be impossible for me to geolocate) and then turns out I solved it in a few minutes, such as the one I did on the pro-Russian Chechens in Mariupol last month.
This time it was the other way around. I immediately assumed, since I was given so much data to start with, that I would just hop in Google Maps and immediately find the area. I did find the correct coordinates eventually but for that I had to use an extra piece of information I don’t often get to use: NASA Fire Information for Resource Management System, or FIRMS for short.
So let’s see how we can use this useful (and free!) tool to our own advantage.

The Video

The video we are about to geolocate was found on Twitter and depicts a “fuel depot on fire today in Kremenchuk after Russian strike“.
So there’s already a lot of information we can gather from this. We know that this will be a big fire, probably visible from quite a distance since fuel is extremely flammable. We also know that this is located in a place called Kremenchuk in Ukraine and that it was happening on the 2nd of April 2022, the day the tweet was published.

Feel free to watch the video below.
If Twitter goes down, here is a screenshot the archived tweet and a link to the video.

Fig. Footage of a fire at an oil refinery in Kremunchuk, Poltava Oblast, Ukraine. (Archive)

After watching the video we can also notice a few landmarks and features that will make it easier (ha!) to identify the exact location. The footage starts by showing some sort of white structure that starts above our head and then goes down, to our left, ending on a much lower height. The person filming is walking on a paved pathway and there are two rows of trees on the left separating them and the silos on fire. At the 0:15 minute mark we can see on the left some small set of stairs leading up to the silo area. Then towards the end we see a connecting path on the right, with a sign on the grass. There are also more silos ahead at the end of the path.


At this point we already have enough data to start looking at google maps and navigate around Kremenchuk in order to find this industrial oil facility.

If you just search for “Кременчук” (Kremenchuk in Ukrainian) this is what you’ll see.

Fig. Satellite view of Kremenchuk, Poltava Oblast.

It’s a fairly big industrial city in central Ukraine but it doesn’t take much to spot the big fuel depot at the North of town. Some of the silos have such an enormous size that they can be seen from quite higher up.

Fig. Highlighted area of Kremenchuk where the oil refinery is located.

To make it easier to see without having to turn my head a lot, I switched to Google Earth Pro which enabled me to rotate the camera in any direction.
I aligned the oil refinery map in front of me and these are all the silos that we can spot at first glance. There’s a lot.

Fig. Satellite image of the oil refinery in Kremenchuk with all the silos’ locations highlighted.

This is what I meant when I said that at first it looks like it will be super easy and fast. We went straight to Kremenchuk and immediately spotted the industrial area with the silos. But when you start looking into it, there’s so many places it could be that checking them all one by one in search of our specific landmarks and features would take a very long time. And we don’t have that amount of time as data from Ukraine keeps coming at an alarming rate. Geolocations like these need to be done swiftly and precisely so we can move on to the next one.


This is where NASA’s Fire Information for Resource Management System tool comes in handy. This free OSINT tool is perfect to locate fires and hotspots around the globe. It provides both data in real time, which is perfect for ongoing fires, and historical information for when we need to verify what happened in a certain area within a specific time range.
NASA FIRMS, is also a fairly simple to use tool. When entering the website you will see on the left the location tool. After selecting it we will choose the “find location” option and write “Kremenchuk”. The list will provide a few options and we can simply select the first one as (after a quick google check) we can confirm that Kremenchuk is indeed located in the Poltava Oblast region, in Ukraine.

Fig. Image showing the location tool at the NASA FIRMS website.

The image will, at first, be quite zoomed out and might not even look like the correct place but if you keep on zooming in you’ll soon recognise the area.

Fig. Image showing Kremenchuk on the map at the NASA FIRMS website.

Here it is, our oil depot in Kremenchuk.

Fig. Image showing the oil refinery in Kremenchuk at the NASA FIRMS website.

What we need to do now is to select a few options on the right in order to get some visual information.
As we are now looking at the incident in the past, we’ll need to choose the “historical” tab and input the correct date. We are quite lucky that we know exactly when this fire happened as it says in the tweet so we can just select the 2nd of April. The results are immediately visible on the map.

Fig. Image showing the hot spots on the 2nd of April at the NASA FIRMS website.

If you are a bit curious you can select different settings to see what they do. Most options will have the “i” symbol meaning you can click on it and read more information about each type of imagery provided and their provenience.

As an example, if you go to the “Advance Mode” you’ll see that by default all 4 types of satellite images are available; two VIIRS which stand for Visible Infrared Imaging Radiometer Suite, located in both the satellites Suomi NPP and NOAA-20, and the MODIS which stands for Moderate Resolution Imaging Spectroradiometer, located aboard the Terra and Aqua satellites. All these instruments are used to collect data on active fires and thermal anomalies.

In this case, we can see how the satellite that picked up the data of the fires in Kremenchuk on April 2nd, was Terra. Thank you Terra.

Fig. Image showing the hot spots on the 2nd of April gathered by the MODIS satellite at the NASA FIRMS website.

Now that we have a few specific hotspots we can compare it with the location of the silos in a previous map and narrow down our search (by a lot!). Below are the only 2 areas which contain silos and have also been marked as having big thermal changes on the 2nd of April.

Fig. Image showing the areas of interest in the oil refinery in Kremenchuk.

Now we just need to zoom in and look around. From looking at the image above there was a 50% chance of picking the correct section out of the two highlighted ones and I was lucky enough to go straight for the winning spot (it’s the bottom one).

Now that we are here we need to find the exact fuel silos that were on fire in the footage. The more precise geolocation you can do, the better.
At a quick glance you might already be able to see the winning spot.

Fig. Satellite image of the Southern area of the oil refinery in Kremenchuk.


As always, identifying the location is not enough. You need to prove it with clear evidence. So let’s get it.

For this part I switched to Google Earth Pro (again) as it has the great feature of rotating the camera around. It’s much easier to compare it with the footage if both images are facing the same way.

Below you can see the interesting features that got my attention. On the number 1 spot we can observe what looks like that strange white metal frame that comes from above where the person starts filming. On 2 you can see the long row(s) of trees and above it the stairs leading up to the silos. On 3 there’s a connecting path on our right and on 4 we can see another set of silos in the distance, straight ahead.

Fig. Image showing the interesting areas at the oil refinery, highlighting the landmarks.

We can now attempt to match them all with the frames from the footage.
Starting from the top, in dark pink, we can see the silos at the end of the path. On the right, in purple, we can see where a new pathway connects towards the right hand of the screen. Below that one, in orange, you can spot the area where the stairs leading up to the silos can be seen. At the bottom we can observe that white metal structure that goes above and across the pathway, and finally, in bright green we can see the long line of trees between the silos and the pathway.

Fig. Image showing all the details that match the footage with the satellite view of this section of the refinery.

From looking at the footage we know that the affected silos were the first two on our left so we can confidently pinpoint our coordinates to that area.

Fig. coordinates of where the fire was located at the Kremenchuk oil refinery in Poltava Oblast.

And here is the end result at the Centre for Information Resilience’s Russia-Ukraine monitor at MapHub with the final coordinates: 49.147932, 33.464159
The map has been moved and updated in December 2022 and here is its new location:

Fig. Image showing the geolocated video with the correct coordinates at the Centre for Information Resilience’s monitor map in MapHub.


Although for most of the time you don’t need anything else other than Google Maps to geolocate photos or videos coming from Ukraine, occasionally specific open source tools such as the NASA FIRMS are actually quite useful to speed up the process.

Would we be able to get to the correct spot without it? Absolutely.
Would it have taken way longer? Definitely.

In situations where there’s so much data that you want (need) to go through it at a fast pace, learning what tools are available is very important. Even if you don’t get to use them on a regular basis, just knowing that they exist and how to operate them is enough to cut down the time spent geolocating or analysing something.
Thank you for reading and hopefully you learned something new and useful today.
Take care!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Website Powered by

Up ↑

%d bloggers like this: