A charity is undertaking efforts to build a historical database of watercolour paintings, saving their fate from being hidden away within the corridors of private homes or the archives of public institutions.

In the pre-photography era, governments would train soldiers on military expeditions in watercolour painting to document the war efforts. In fact, this practice continues today. But before war photography was even possible, sketching battlefields or other operations was one of the few ways to record history in real-time.

Eruption of the Soufrière on April 30 1812 (Burning Mountain) 1815 Print after Joseph Mallord William Turner
Eruption of the Soufrière on April 30 1812 (Burning Mountain) 1815 Print after Joseph Mallord William Turner

Watercolours, however, are prone to quicker degradation than other forms of painting, so this has meant that many of these paintings need to be shielded from natural light sources, meaning even when they are stored within public institutions, many end up being filed away.

'Watercolour World', which launched this month, is striving to document as many of these historically significant pictures as possible. A small team of archivists have travelled up and down Britain, armed with scanners supplied by Fujitsu subsidiary PFU, to create digital copies - which are then uploaded to the site's database, which currently sits at 80,000 files strong.

The charity got its start when former diplomat Fred Hohler, chair of the Public Catalogue Foundation, was trawling Britain to try to photograph and catalogue oil paintings. As part of this decade-spanning project, Hohler noticed that watercolours were often forgotten and hidden away.

"The history of watercolour is rather amazing in the sense that it all started back in the 1740s when the military were actually trained to paint," Andra Fitzherbert, CEO of Watercolour World, tells Computerworld UK. "Prior to photography there was no other way of recording what you saw, so people were trained to paint accurate depictions of places and people and events.

"This is documentary evidence that is eroding, disappearing, and very forgotten and lost and hidden. What we really want, the whole purpose of this project, is to bring this out and make those images accessible for everybody, for research purposes, for education purposes."

As part of their early research, Fitzherbert's team looked into the kinds of scanners that would make such an endeavour possible and found that PFU's technology fit the bill. Unlike photography, which requires adequate and consistent lighting conditions, scanners are able to take high-quality copies of paintings that are agnostic to their surroundings. Crucially, they are highly portable, and the ultraviolet elements that are used in the scanning process are not known to be detrimental to the objects themselves. They are also able to scan through glass, adjusting for the reflection.

"We're often up in attics and funny little eaves and places in these big, old historic houses," Fitzherbert says, "so having something that is small and portable that can easily be set up has been so beneficial."

The process

Watercolour World researchers contribute on a voluntary basis. They find themselves travelling to attics, galleries, and other places of public record, where they work with the object owners to probe the viability of their collections for the database. Armed with ScanSnap SV600 scanners, donated to the charity by PFU, the record for image capture so far is "something like 1,200 images in a day," says Fitzherbert.

Generally, the criteria is that the images must be realistic enough - "not too impressionistic" - and that the team is looking for aspects of an image that can provide detail and information. Although they won't necessarily veer away from impressionistic or fantasy paintings if they tend to fit in as part of a collection, generally the team is looking for pre-1900 paintings.

Once the field work is completed, the researchers travel back to the office where the images are catalogued and meta data such as categorising or geolocation is performed, and then are uploaded either individually or as a bulk batch. Finally, they are published on the site.

There has been an overwhelming willingness to work with the team on the project, and quite often the owners of the paintings aren't aware of their historic importance. But a little bit of digging can surface treasure troves.

"I love the warmth we get in terms of the give-back, they actually all feel even though these are personal watercolours to them, actually they're very happy for them to be scanned and shared because why wouldn't you?" Fitzherbert says. "They've still got the actual item themselves, why shouldn't other people benefit from something that may be of no information to one person but somebody else finds deeply interesting?"

It's a nice narrative to the public-private art ownership debate. By cataloguing and uploading these pictures, the owners keep them in their personal collections yet also contribute to a public database to be enjoyed by all.

Even when these images are displayed physically in museums or galleries, Fitzherbert adds, they are limited to their location: so by uploading them into a central database they can be accessed by people who might otherwise not be able to.

Learning from history

Already there have been incidences where the images have proved beneficial to researchers. The charity provided watercolours from the 17th and 18th century to a researcher examining coastal erosions, to chart out the history of these in British territories but also even to be able to predict their future.

And the aggregation of various images from both public and private collections might also help researchers. Fitzherbert points to the images available for Mt Vesevius, Italy's famous volcano that overlooks the ruined city of Pompei, which will allow historians to compare and contrast.

"We really encourage people both to contribute and get in touch with us so we can digitise their images or they can share images with us, and upload them onto the site," says Fitzherbert.

PFU's customer engagement manager Steven Chad adds that it's not much of a leap to imagine this wealth of historic information in databases playing a sizeable role in cultural studies, or even for costume research in television and film.

One example Fitzherbert notes of interest is a scan entitled 'The Launch of Blanchard's Balloon at The Hague in 1785'. The image depicts a painting of Noordeinde Palace in the Netherlands, and zooming in on a detail at the right of the picture shows a man using a hammer to get the steam pressure just right.

"There will be specialists out there, which is what we think will be so exciting and pique their interest, who will say, 'actually I know a bit about this' because there's so much knowledge out in this world that isn't necessarily in the traditional academic houses," says Fitzherbert.

"[When] the naval drawing school was started in the 1740s, and watercolours evolved as Whatman paper, the which was the new innovation which meant that they could paint more accurately," points out Fitzherbert, "what is rather amazing about this project is how watercolours, that you don't think of being very impactful perhaps at first sight, when you actually zoom into them and see the level of detail, they are extraordinary - how some of them can actually be seen at these incredible resolutions, and still hold remarkable outlines and detail."

The future

While there are clear benefits for historians about being able to quickly access a database of paintings, tied to their locations, there are also hopes that the project might be open to image analysis in ways that are not yet clear.

Fitzherbert refers to this as a "wonderful, open-ended area of possibility".

"To have that bank of images is interesting, and that's going to be my next chapter, to see where that takes us, because I think there's definitely some legs there given our digital environment and artificial intelligence," says Fitzherbert.