Ukraine’s Unmanned Systems Forces said on 23 August they hit two combat aircraft, an Orion strike-reconnaissance drone, short-range air-defence vehicles and eight radars from Vityazevo in Krasnodar Krai through Rostov and occupied Crimea. Commander Robert Brovdi listed 14 targets. Kyiv’s General Staff later said the two fighters and the Orion at Vityazevo had been damaged. Open-source outlets circulated strike video. Militarnyi judged the jet labelled a Su-33 was more likely a two-seat Su-30SM. Russia has not confirmed the losses. The full list of radar and launcher kills has not been independently verified. Treat the scorecard as a claim. Treat the lesson as real.
The useful fact is geography. Vityazevo is around hundreds of kilometres from the front, on the mainland, after Crimean parking became too hot. Aircraft that move still have to stop. When they stop in the open they become the target. A runway can be cratered and patched. A fighter with a burned wing, a radar that no longer sees, or a Pantsir that is a wreck beside the taxiway takes the base out of the fight without closing the concrete. That is why the claimed mix matters more than any single airframe be that jets, the drone that supports them, the sensors that warn them, and the point-defence that was supposed to save them.
Base survivability is now a package, not a fence. Hardened or at least revetted parking beats a row of jets on an exposed apron. Dispersal beats one famous base everyone already has on a map. Decoy airframes and emitters force an attacker to spend drones on empty metal. Rapid runway-repair kits still matter, but so do spare radars, spare generators and a plan to generate sorties from a second strip if the first is under smoke. Counter-drone sensors have to be stationed outside the perimeter, not only at the tower. A radar that is itself undefended is an invitation. Turnaround time, how fast a surviving jet can be refuelled, rearmed and airborne after a raid, decides whether the next wave finds parked aircraft or empty stands.
Asian air forces already live with versions of this problem. India’s western and northern bases, Taiwan’s west-coast strips, Japan’s southwestern islands, South Korea’s clustered peninsula fields and Australia’s northern bare bases all mix high-value aircraft with limited dispersal and civilian towns nearby. A cheap uncrewed aircraft does not need to destroy a hangar to cancel a day’s flying. It needs to hit the tanker, the AWACS, the air-defence radar or the jet that was left outside because the shelter programme is unfinished. The countries that treat shelters, decoys, mobile radars and counter-drone coverage as one budget item will lose fewer aircraft to a video released at dusk.
Runways will still be attacked. They are no longer the only thing worth attacking. The air force that learns that from someone else’s unverified film will spend less than the air force that waits for its own.
P.S. Russia has not publicly confirmed the losses.