Julian Borger and Eliot Higgins
February 18, 2015
Russia shelled Ukrainians from within its own territory, says study
When Ukrainian forces came under withering attack in the east of the country last summer, soldiers were surprised as much as scared by the ferocity of the attack. The separatists they were up against had proven fierce and organised. But this was something else.
Now a group of British investigative journalists using digital detection techniques, satellite imagery and social media has provided near conclusive proof that the shelling came from across the border in Russia.
The work by the Bellingcat investigative journalism group highlights a murky aspect of the war in Ukraine, which continues to sputter despite last week’s attempt in Minsk to draw up a ceasefire, with reports of heavy fighting around the railway hub of Debaltseve on Tuesday.
Russia has long been accused of funnelling soldiers, munitions and military vehicles into eastern Ukraine to help separatists take on the Ukrainian army. But until now, little has been written about Russian military units shelling across the border into Ukraine.
The Bellingcat team analysed crater patterns from satellite photos of three battlefields where the Ukrainian army came under particularly savage attack last summer and traced the estimated trajectories back to likely firing positions, where it identified scorch marks and tyre tracks on satellite images consistent with Russian rocket-launchers.
With a single exception, the identified firing positions were on Russian soil. Furthermore, the tracks to and from the firing positions led further inside Russia, further evidence that they were Russian units, not separatist fighters who had strayed across the border. Images of the same terrain just before the attacks show no track marks or scorched earth.
An independent military forensics expert warned that the accuracy of crater analysis in determining direction of fire on the basis of satellite photography was scientifically unproven, but said that the images of firing positions on the Russian side of the border were compelling and raised questions of what they were doing there.
The incidents happened last summer, during an intensive period of fighting in which the Ukrainian army began to gain the upper hand against separatists and Russia began to supply more overt aid to ensure the rebels were not defeated. In mid-August the Guardian saw a column of Russian armour cross the border, and Kiev claims that thousands of regular Russian troops effectively invaded. A ceasefire was signed in Minsk in September, though it broke down almost immediately.
In early February, the governor of Luhansk, Gennady Moskal, made new allegations that Russian forces were shelling Ukrainian territory from inside Russia, as part of the battles that raged before the signing of the new Minsk accords last week.
Russia has repeatedly denied the involvement of its troops in eastern Ukraine, insisting the war against government forces there is being fought by local insurgents. Vladimir Putin has described them as “volunteers”. “We’re not attacking anyone; we’re not warmongers,” he declared in December.
The families of Russian soldiers killed in Ukraine have been put under pressure not to talk publicly about it, and have reportedly been threatened with a withdrawal of state death benefits if they do, but some have begun to speak out.
In an earlier investigation, Bellingcat – a group of investigative journalists specialising in image analysis – tracked the movement of the Buk, which was photographed by members of the public near the crash site on the day MH17 was shot down, and the photographs were posted online. Bellingcat showed that the same launcher was part of a Russian unit.
It has also published research on the use of munitions, including chemical weapons, in the Syrian civil war.
In its new investigation of artillery use in Ukraine, Bellingcat focused on three battles in July as pro-Moscow separatists pushed back a Ukrainian government offensive that had regained a large section of the Russian-Ukrainian border. The separatist counter-offensive was supported by heavy artillery, which proved decisive in driving the Ukrainian army out of the Donetsk and Luhansk regions.
Bellingcat used publicly available satellite imagery of the battlegrounds and adapted established procedures for analysing craters on the ground for determining the trajectory of artillery fire, applying them to the photographs. It looked at two main types of crater: low angle, which produce a distinctive diagonal spray of soil from the central crater, and high angle, which make a triangular shaped crater. Both can show the direction of fire.
The first battle studied was near the town of Amvrosiivka, where there was a crater field showing 330 separate impacts from an artillery attack on Ukrainian army positions on 14 July. From the shape of the craters, an average trajectory was worked out. The Bellingcat team traced that line back through satellite photographs of the area until they found a potential firing position, identified by burn marks on the agricultural land, of the sort caused by multiple rocket launcher systems (MLRS). This was nine miles away, across the border and near the Russian village of Seleznev. Scrutiny of the imagery showed a pattern of tyre tracks at the suspected firing position, suggesting a number of vehicles parked in a line at the site.
The report said: “The visible tracks that lead to the site come from further inside Russian territory. This leads us to believe that there was no crossborder (Ukraine to Russia) movement of military equipment for this particular location.”
A second site analysed in the report was in the region of Chervonopartyzansk, where Ukrainian units came under heavy artillery barrages between 14 July and 8 August, forcing a Ukrainian retreat. Looking at a field of 813 craters, it appeared there were six separate attacks from five different directions. Using the same methods, the Bellingcat team found five separate firing positions, four of which were in Russia. In each of those cases, “all the observable tracks near the firing sites were exclusively within the territory of Russia”.
One of the identified firing positions was near the Russian town of Gukovo. Six videos uploaded by local residents to YouTube, and another social media video site, VK, showed MLRS (multiple launch rocket system) salvos being fired on 16 July, throwing up large plumes of smoke. By working out the direction from which the videos were shot using visible geographical features, the team estimated two firing positions near Gukovo – where satellite photos showed telltale burn marks and tyre tracks.
The video footage taken by members of the public in Gukovo showed rockets leaving the launchers, so the investigative team could measure the angle of elevation at which they were fired. In each case, that was found to be 20 degrees. Using a firing table for a 122mm rocket fired from a BM-21 Grad launcher, the most likely system used, that suggested a range of between 15 and 16 km. The actual distance between the estimated firing positions and the crater fields over the border in Ukraine was 9.5 miles.
The investigation made similar findings on a third artillery barrage, on 25 July, south of Sverdlovsk, where Ukrainian forces came under heavy fire – an attack Russian media attributed to the separatist Luhansk People’s Militia. Trajectories calculated from a crater field, however, led to two firing positions in Russia, one at a military base near Pavlovka, just across the border.
Stephen Johnson, a weapons expert at the Cranfield Forensic Institute, part of the Defence Academy of the United Kingdom, said that the application of crater analysis techniques to satellite imagery was “highly experimental and prone to inaccuracy”.
“This does not mean there is no value to the method, but that any results must be considered with caution and require corroboration,” Johnson said in an email after reviewing the Bellingcat report. He added that “the most significant part of the report” was the discovery of the apparent firing positions on the border.
The ground markings do not seem to be consistent with agricultural machinery, Johnson said. “They indicate an orientation of vehicles that would not be unusual for artillery vehicles, and there does appear to be some ‘scorch’ damage that is not a wheel or track.”
No comments:
Post a Comment