Science - October 13, 2011

Many hands make light work

More and more ‘ordinary people' are recording scientific observations: crowdsourcing is gaining ground as a source of valuable data. And Wageningen scientists are in on the action. ‘People are keeping a critical eye on the transparency of processes. Which stimulates participation.'

21-crowdsourcing-science.jpg
In 2007, Hanny van Arkel was the first to spot a new heavenly body. The ‘object', as she called it, is probably a thin cloud of gas that glows green. It was a sensational discovery, but that is only half the story. Because Hanny is not an astronomer but a maths teacher from Heerlen who identifies constellations in online photos in her spare time. Together with an army of volunteers, she is building up a database which scientists are grateful to make use of.
This kind of outsourcing of work to the anonymous general public is known in the latest webspeak as crowdsourcing. The process is aided by new media such as twitter, smartphones and wikis, and the most prominent exemplar is the crowd-written encyclopaedia Wikipedia.
The scientific community is not lagging behind, as researchers increasingly often get laymen to collect or interpret data. Not least in Wageningen. In fact, Wageningen scientists were enlisting the help of the general public long before the rise of the internet, and have made a smooth transition to the digital age.
Take Alterra researcher Leen Moraal, for example. Moraal studies the way animal pests spread. He's never heard the fashionable term crowdsourcing, but his group has been making use of observations recorded by amateurs since 1946. ‘I have around 450 volunteers,' says Moraal. ‘Most of them manage forests and urban green areas. They report cases of damage and animal pests they come across in the course of a day's work.' These data are an ideal means of recognizing trends. In 1991, for example, he was the first to record the presence of the oak processionary caterpillar. But the cards that had been used since just after the war have now given way to the internet, and he hopes soon to bring out an app for smartphones. ‘It would be marvellous to be able to study this at an even finer level of detail.'

Nature calendar
So the crowdsourcing strategy is not new, but it is a booming business in the research world. ‘It is growing and speeding up', indicates Arnold van Vliet, associate professor of Environmental Systems Analysis. No other Wageningen researcher makes such intensive use of crowdsourcing as he does. With the help of well over 8,000 volunteers and hundreds of school students over the past ten years, his Nature Calendar has been tracking the timing of natural phenomena such as the flowering of the horse chestnut. This provides interesting data for studies of the effects of climate change, to give but one example. Van Vliet has a rapidly growing number of observation websites. There is Allergieradar, which collects reports of hay fever symptoms, and there is Splashteller, where motorists record the number of insects that have ‘splashed' onto their number plates.
Other researchers voice their dreams of establishing such networks. Martin Herold, professor of Remote Sensing, hopes to make use of crowdsourcing in new research on land use. ‘In remote sensing you work with satellite data', explains Herold. ‘From space you can easily see when land use changes, but you cannot see how. For Wageningen, that's easy: I just go outside and look. So you have to be on the spot.' Herold would very much like a network of volunteers who could supplement his satellite data with local information, making it much more interesting. He is launching a pilot project this year to see if it works.
Herold is optimistic about the future potential of crowdsourcing. There is an increasing amount of cheap but powerful technology available to the general public. Thanks to smartphones, you can take photos everywhere you go and locate yourself using GPS. ‘New sensors are becoming more and more accessible too. Take microphones for noise pollution, laser distance meters and soil humidity meters', says Herold. Once you have your data you can send it instantly through an internet connection. These are all developments of the last ten or even five years. Herold feels that the zeitgeist works in favour of this brave new world too. ‘People keep a critical eye on the transparency of processes. They want open governments, open data and free information. That stimulates participation.'

Motivation
But what motivates people in the ‘crowd' to work without pay? Fame such as that gained by Hanny van Arkel is only for the happy few. Besides the zeitgeist, Herold thinks the type of research plays an important role. Research on innovations and change, for example, appeals to the public. ‘And people are also interested in their own locality. The first thing they look up on Google Earth is their own house.' And then there are a number of subjects which always raise a lot of enthusiasm, such as nature and birds.
But, says Arnold van Vliet, ‘it is not easy to motivate volunteers'. In spite of the appealing and highly visible subject, the success of the Nature Calendar did not come by itself, he says. ‘We work very hard to keep people motivated. Many scientists do not realize what goes into this, how creative you have to be.' In order to recruit observers, Van Vliet publicizes his results in many different media, which is a time-consuming task. Another of his secrets is that he takes his public very seriously. His research themes, such as hay fever or Lyme's disease, are also partially decided by the wider society. And to very good effect. The network provides Van Vliet with the data to continuously pose new scientific questions and get results.

Limelight
Asked how long the ‘crowd' can keep up its interest, Van Vliet expresses astonishment. ‘There is not the slightest sign of a loss of interest. In fact I even see great potential for growth.' And yet doubts about motivation are not just plucked from the air. Wikipedia announced a lull in interest earlier this year. But Herold, too, largely dismisses this point, saying there is not so much left to write about on Wikipedia. ‘Crowdsourcing is not going to disappear.'
It does have other limitations though, according to Herold. ‘Crowdsourcing is applicable to certain selected themes. Not everyone can do it.' Van Vliet is a good deal more optimistic on this point. In his view it is all a matter of clear explanation and cultivating motivation. ‘We even get volunteers to collect ticks every month. That is hard work and has to be done according to strict guidelines.' You can just picture it. Anonymous volunteers sweating their way through the woods, driven by their enthusiasm. In the hope that one day, like Hanny, they will find themselves in the limelight.

Unreliable amateurs

There is still a fair bit of scepticism around about crowdsourcing. Data collected by amateurs is thought by some to be less reliable than that of professionals. But the experts disagree, certainly when it comes to large-scale trends in the recorded observations.
Moraal indicates that he uses the data in all his publications and reports. Van Vliet is firm too: ‘The data really is very usable; it is just like your own data. A great deal has been published, based on it, right up to articles in Nature.' He himself wants to take it a step further. ‘Sometimes the knowledge of our volunteers is far more extensive than the field knowledge of biologists. At university they only get limited field training, whereas the amateurs train for years.'

Re:act