Geospatial data gathering bears fruit

Share this on social media:

Matthew Dovey examines the rise in academic geo-apps and the value of crowd sourcing in gathering data for research

‘The next web will be about space and time,’ said Marc Davis, Yahoo's social media guru in 2008 and, just three years later, we're seeing his prediction take firm hold. From counting horse chestnut trees to crime maps of London, the potential of geospatial tools, combined with the possibilities of crowdsourced data, has started to become clear.

2011 will be remembered as the year that academic geo-apps caught the imagination of the wider public. But should 2012 be the year we marry the excitement of these new tools with a commitment to training our researchers in the skills they need to use the data produced by the tools?

Geospatial information is now ubiquitous – three billion people have access to high-resolution satellite images, there are 400 million Google Earth users and $40 billion is spent each year on collecting geospatial information. Smartphones with built-in spatial location, giving easy access to online maps and apps, are owned by half the UK population. Across Europe, the INSPIRE directive is establishing a ‘spatial information infrastructure in Europe’ - the stitching to bind all the geodata from each of the EU's national mapping agencies – making data even more accessible and interoperable.

Geospatial tools and services are evolving in tandem with this take-up and some of the most exciting and innovative geo-apps are coming out of the higher education sector. JISC in the UK has funded a series of projects to increase the use of geospatial tools, infrastructure (data and services) and information for learners, teachers and researchers. These include GEMMA, from University College, London, which starts at first principles by creating an app to simplify the process of collecting, mapping, preserving, sharing and visualising geospatial datasets. Meanwhile, the Walking Through Time team at University of Edinburgh is developing an app that allows architectural historians, conservationists and tourists to download historical maps when standing in a specific location and then to annotate them.

One of the key benefits to emerge from the rise of academic geo-apps this year has been the potential to engage the public in crowdsourcing data for researchers to then work on in the lab. Learners are the focus of the JISC-funded crowdsourced language app, Cloudbank, and GeoSciTeach. Meanwhile the hit of the year was the Leaf Watch project. It attracted mass media attention with its appealing mix of a fun app to photograph conker trees and the laudable aim of helping researchers to build up a picture of which areas of the UK are affected by a non-native moth threatening the nation's horse chestnut trees. 10,000 people downloaded the Leaf Watch app and thousands of records were submitted to the project.

In times of austerity, these kinds of initiatives can seem to offer golden opportunities to researchers. By using the power of the masses and technology, gathering data from a wide geographical range suddenly becomes possible.

According to Dave Kilbey, Leaf Watch's project manager, ‘it has stripped the barriers to science down and opened up a massive pool of people. Projects that just weren’t feasible before have become possible…as people look to do more with less, this sort of citizen science, crowdsourced project will be happening a lot more.’

But, while geospatial crowdsourcing tools are an appealing new way to accumulate large datasets, they also raise questions. Apps such as Leaf Watch are not particularly complex in terms of the geospatial technology underpinning them – they could be seen as the electronic equivalent of putting pins on a map. This makes them attractive to the lay user – they are simple and fun to play with – but it may also have an impact on the kind of data that is submitted. How useful are the results for the academic community? Do researchers have the skills to manage and work with this kind of crowdsourced data, which may vary widely in quality, provenance and accuracy? The question of how academic rigour can be combined with ease of use is a critical one.

‘We started in the 1970s seeing GIS (geographic information system) used mainly by developers but in 2011 we have general users engaging with these tools,’ said Patricia Carbajales, geospatial manager of Stanford University, USA, at a recent JISC geo programme meeting. ‘We are moving to a place where there are hundreds or thousands of geospatial data users who really don’t care that much about the quality of the data and we have to give them the basic technology to understand and use the tools and data. And we are also looking at how those results impact our environment and our society.’

GIS skills are particularly important when combining datasets from different sources that may use different spatial areas and granularity. Misleading and erroneous statistics can result if the datasets are not combined correctly.

At Stanford University, Carbajales and her library colleagues teach ‘basic mapping know-how’ and encourage “thinking with maps” – the bread and butter skills of working with geospatial data. Their goal is to establish ‘a geospatial foundation for our learners’. Geospatial tools and training are being fully embedded in the curriculum, via the library, while undergraduate students are engaged with the research process at an early stage through initiatives such as the Spatial History Project.

Similarly, in the UK, projects such as Locating London's Past demonstrate how geographical analysis is being used by researchers in the humanities and social sciences. It allows anyone to explore fully geo-referenced detail using Google maps technology to reveal the distribution of crimes, wealth and poverty, mortality, archaeological finds, voting records and much more.

But, for more researchers to make full and effective use of the potential of sites, services and tools like this, the geospatial skills gap needs to be bridged. JISC is planning work to help do just that.

Matthew Dovey is programme director for the digital infrastructure at JISC

  • Do you have experience of this type of project and the opportunities and challenges of crowdsourced data that you would like to tell other readers about? Contact Siân Harris