Thread by @bellingcat

Over late October and early November, military cargo flights in and out of Kaliningrad were at an uncharacteristic high. Bellingcat went to investigate what these flights could be moving using SAR imagery… bellingcat.com/news/2023/11/1…

Sentinel-1 Synthetic Aperture Radar (SAR) imagery is freely available on a variety of platforms such as Sentinelhub’s EO browser. SAR satellites emit signals onto the ground and create an image based on what is reflected back. They don’t rely on light sources.

In late October, internet users noticed an uptick in Russian military cargo flights. An X user highlighted the anomaly with a graph showing fluctuations in weekly military flights dating back to 2021.

https://twitter.com/auonsson/status/1721282205874327667

Konrad Muzyka, an independent defence analyst who studies Russia, said it is likely that the increased activity was in connection with the movement of military equipment, but stopped short of saying what equipment it may be.

https://twitter.com/konrad_muzyka/status/1719810483165806727

Bellingcat has found a marked change at certain air defence sites on SAR imagery since the reported cargo flights, indicating that at least some of Russia’s S-400 batteries have indeed been moved from the Kaliningrad Region.

Using a map file of Russian military installations created by X user @Archer83Able, we studied ten air defence radar and missile sites for change in November on Sentinel-1 (S1) Synthetic Aperture Radar (SAR) imagery.

From these manual checks, four locations of interest were identified at air defence sites where returns were lower in November 2023 than in the previous month.

In order to limit the possibility of false negatives, Bellingcat created aggregates from two to four separate days each of before and after imagery to be used for comparison, from this two sites of particular interest became clear.

At one site near Gvardeysk, referred to as location A, points which had previously shown high returns in October appeared indistinguishable from regular noise in November. These points aligned with two radars and a row of launchers of an S-400 air defense system.

At another location near the Primorskaya bay, referred to as location B, returns from November appeared significantly lower when compared with October at points associated with launchers and radars in reference imagery.

Umbra imagery of the sites were consistent with Sentinel-1 imagery. For location A, points that should've had highly SAR reflective objects were dark.

At location B, the results of the imagery were much less conclusive, due to factors including the direction of the satellite and tree coverage. But points where values had dropped in November S1 imagery also appeared dark in Umbra’s imagery.

Read the analysis and see the full dataset via the article here: bellingcat.com/news/2023/11/1…

Bellingcat is a non-profit and the ability to carry out our work is dependent on the kind support of individual donors. If you would like to support our work, you can do so here bellingcat.com/donate/