Where, exactly, was the target? A Google Earth story

“[O]ur boys joined in the attack on the Marshalling Yards at LILLE” – 463 Squadron Operational Record Book, 10 May 1944

The concept of a marshalling yard in Lille has long formed part of my understanding of what happened to the man known in our family as ‘Uncle Jack’. It’s been part of Purcell family folklore, I suppose, for as long as I can remember: that the target of Jack’s Lancaster on that fateful night was a set of railway lines in northern France.

But which set of railway yards, exactly?

The thing about Lille, you see, is that it’s a pretty important railway junction. It’s now a key stop on the Eurostar cross-Channel tunnel route between Brussels and London, an hour from Paris on the TGV and about 40 minutes from Brussels. There’s also a significant local railway network. There are two key stations in the centre of the city; Lille-Flandres, which hosts local and regional trains and some high-speed TGVs, into which I arrived when I visited Lille in 2009, and Lille-Europe, serving the Eurostar cross-Channel trains and international TGV services, from which I departed three days later. While a lot of these train lines and services have been built since the Second World War, their slower fore-runners also ran through the city: Lille sat on the route from the ports of Calais to Berlin and on to Warsaw, for example, and one of the first railroads in France, the line between Lille and Paris, opened as early as 1846. Locomotive building and repair workshops were also located in the city. 

It’s pretty clear, then, why the city’s railways were targeted as part of Bomber Command’s pre-invasion Transportation Plan. But which part of them, specifically, was the target on 10 May 1944?

In the International Bomber Command Centre’s superb Digital Archive I thought I found the answer: a bombing photograph from the night in question, which shows a distinctive set of marshalling yards amongst the smoke – just above the white bomb burst in the photo:

[

Extremely helpfully, the IBCC’s volunteers have geolocated this photograph over a modern-day map. I fired up Google Earth, bearing in mind that this is a modern-day image and a lot of this infrastructure wouldn’t have been present in 1944, and immediately found the relevant spot. The facility in the bombing photo is pretty clearly the Hellemmes workshops, circled in blue:

And it looks like the 50 Squadron crew who obtained the bombing photograph just missed the target – the centre of the photo, the point where the bombs themselves would have theoretically fallen, being plotted a little way south of the marshalling yards. In this view, which I’ve geolocated onto the Google Earth screenshot, the red ‘X’ marks the crosshairs:

If I remove the bombing photo overlay, you can see the red ‘X’ just above that little white building – not very far away from the marshalling yards at all:

Or so I thought… until I saw the other side of the bombing photo, which has also been scanned in the IBCC’s Archive:

Hmm. “1300 yds 114°” – that looks to me to be a bearing and distance. I wonder if it’s a location referencing the actual aiming point, or in other words, how far away from it this crew’s bombs landed?

If 114° is the bearing of the photo from the aiming point, as I suspect, its reciprocal (114 + 180 = 294) is the bearing of the photo to the aiming point. So here’s a Google Earth ‘ruler’ showing where that point is. The bottom right of the yellow line is located on the position of the red ‘X’ in the earlier screenshot. If I’m right, the other end of it shows where the actual aiming point was:

Looking promising – that’s clearly one of the other marshalling yards. But can I find any other evidence to corroborate this theory?

I wouldn’t be writing about it if I couldn’t. In the Night Raid Report[1] from this night’s operation, there’s a description of damage as shown by later photo-reconnaissance:

A great concentration of bombs fell on and around the railways and sidings 200yds S.W. of the steel and engineering works of the Fives Lille company. 2 locomotive sheds and a repair shop were destroyed, together with numerous smaller buildings, and many hits were scored on lines and rolling stock. The Fives Lille factory and several other industries were damaged.  

Where is or was the Lille-Fives company? It’s that white-roofed industrial area to the right – east – of the marshalling yard in the following screenshot. You might just be able to see the label above it that says ‘Fives Cail’; this is a new redevelopment project that aims to turn the old factories of the ‘Lille-Fives Company’, which later became known as ‘Fives-Lille-Cail’ and, now, simply ‘Fives’, into an urban project with housing, public areas and creative industries. But it’s where the Lille-Fives company was located during the Second World War, and its edge is clearly 200 yards north-east of the same marshalling yard identified by my yellow line earlier. In other words, the marshalling yard is 200 miles south-west of the factory, as noted by the Night Raid Report:

So it looks to me like the marshalling yards near Fives were the actual target for the bombers on the night of 10 May 1944. While the bombing was mostly accurate, some landed closer to the railway yards and workshops south of Hellemmes – and all of the six bombers that crashed within two miles of the target area, including B for Baker, fell east of the aiming point. That, however, is a story for another day.

Screenshots from Google Earth. IBCC material used under the CC BY-NC 4.0 Attribution-NonCommercial 4.0 International Creative Commons licence. Analysis, additional geolocation and text © 2021 Adam Purcell


[1] The National Archives of the UK (TNA): Public Record Office (PRO), AIR 14/3411, B.C. (O.R.S.) Final Reports on operations, Night Raids Nos. 416-620, September 1943 to May 1944, vol. 4: Night Raid Report No. 602

Google Earth

The internet is full of tools to make the life of a Bomber Command researcher easier. There are forums to connect like-minded folk from across the globe. There is the National Archives of Australia website to view digital copies of original personnel and casualty files (among many many thousands of others). And there is Google Earth, which has for me proven invaluable in giving some sort of an appreciation of the geographical setting of the events I’m investigating from the other side of the world.

I made heavy use of the program while I was researching and writing my 467 Postblog series over the last couple of years. I pulled the approximate routes flown by the bomber streams on various operations from either logbooks or the Night Raid Reports, then stuck those ubiquitous virtual yellow pins into the map. Then I could plot the locations of casualties, nightfighter attacks or other interesting events in relation to the nominal track flown by the bombers. I could even, in some cases, locate the exact aiming point and trace the unfolding bombing operation in close detail.

As an example, take the Tours raid of 10 April 1944 (on which the crew of B for Baker did not take part). I wanted to find potential reasons for why the second wave of the attack was not as accurate as the first had been. I started with the Night Raid Report[1], which says that “the first red spot fires fell near the roundhouse and the bottleneck between the two yards”.

Off to Google Earth I went, where it was easy to find railway yards at Tours with a bottleneck between them and (though this image is quite low-resolution) a roundhouse visible just under the ’o’ in ‘Approx’. Excellent, I thought, and inserted a fire icon.

tours

What next? “The Master Bomber therefore ordered crews to bomb 500 yards to the E”, continued the Night Raid Report. So I picked an appropriate icon (a bullseye), and placed it some 500 yards east of the spot fire icon. It’s pretty close to the bottleneck – smash that and the entire marshalling yard complex will struggle to function effectively. So things are beginning to make sense.

The bombers approached the target roughly along the line from bottom right to the target pin (which is probably slightly displaced from the actual aiming point because of limits in the resolution of the coordinate system used by the aircrews). This image has been rotated slightly towards the east to fit the text in comfortably so the line of approach should actually be only slightly north of west.

While most crews were carrying high explosives only, the 463 and 467 Squadron Operational Record Books reveal that a number of crews also dropped incendiaries, the fires from which caused a lot of smoke to rise over the target. The ground wind, according to Wing Commander Willie Tait’s report on the 467 Squadron ORB, was coming from the west. One look at the target area on Google Earth reveals that the smoke would have been blown back along the path of the second wave of approaching bombers. It’s easy to see why the second wave had trouble identifying the target and thus, one reason for why their attack was not as accurate as the first had been.

This analysis came from a careful study of the data found in the Night Raid Reports and the 463 and 467 Squadrons Operational Record Books, combined with a Google Earth satellite view of the target area as it was in 2010. Perhaps a map of the area may have afforded a similar conclusion but it’s unlikely that the detail of the bottleneck and the roundhouse, which stick out clearly on the satellite photo, would have been on a map. I used a similar technique throughout the Postblog series, including tracing Phil Smith’s evasion as he walked south through France – I could even pinpoint the farmhouse where he spent a couple of nights in Orchies two days after his aeroplane crashed.

I visited the UK in 2009 on what was essentially a pilgrimage around various sites associated with the crew of B for Baker and with Bomber Command in general. However much they have changed in the intervening decades, there is nothing quite like seeing first-hand places I had, to that point, only read about in logbooks and diaries. But, when trying to trace these events from the other side of the world, if you can’t be there in person Google Earth comes closest for giving an impression of how everything fitted together.

 

[1] No. 576 – see here for full citation