Australasian Plant Conservation
Originally published in Australasian Plant Conservation 20(4) March - May 2012, p 6-7
Could crowdsourcing be used to detect and monitor invasives?
Lynne Sealie and Bryan Kalms
Atlas of Living Australia, Canberra ACT. Email: firstname.lastname@example.org
How the relative level of scientific and community involvement offers a simple model for understanding the relative contribution of each group to crowdsourced research projects.
Can members of the community use new technology to help Australia track pests and diseases? After all, imported red fire ants, a significant pest, were first detected in Australia by domestic gardeners who noticed their aggressive stinging and swarming behaviour. We all know of amateur naturalists and photographers, farmers, landholders and others who spend time observing, photographing and recording animals and plants. Could this valuable information, gathered by vast untapped human resources across Australia, be utilised in a coordinated way to monitor invasive species?
The Atlas of Living Australia
Enter the Atlas of Living Australia (the Atlas). The Atlas (www.ala.org.au) has been funded by the Australian Government as a national repository for information on all species found in Australia: plants, animals, fungi and bacteria.
The data in the Atlas comes from a huge variety of sources - museums, herbaria, botanic gardens, national specimen collections, state government databases, NGOs (eg BirdLife Australia), consultants, DNA barcode libraries, community groups and individual citizens.
Crowdsourcing refers to outsourcing problem solving to a network of people, or the crowd. Allowing people to provide data to an organisation like the Atlas does not of itself constitute crowdsourcing. Crowdsourcing is as much about engaging with a network of people as it is about the issue or the tools used to engage: the crowd engages with the issue and uses the tool to record that engagement. But a good tool will make engagement easier and more interesting, as is shown by the rise of social networking sites.
All over the world, researchers in fields as diverse as astronomy, cancer research, biodiversity and the search for extraterrestrials have discovered the power of crowdsourcing to engage with motivated individuals and groups to collect and/or analyse data for their research projects.
And there’s an interesting twist starting to emerge: citizen driven research. Through crowdsourcing, not necessarily facilitated by computers, community groups are approaching researchers, offering crowdsourced data and asking the researchers to study the phenomenon represented by those data. This is especially so when researchers find it difficult to obtain funding and/or are not interested in the issues that the community wants researched, like some diseases that only affect a very small percentage of the population. Maybe the scientific paradigm is changing, perhaps becoming more democratic in some ways.
The Atlas and crowdsourcing
The geographic spread of Australian biodiversity provides a challenge in collecting data. Australia is a huge country and information about biodiversity in many areas has gaps; gaps that local people can help to fill. So, we see a valuable role for crowdsourcing; for encouraging landholders, NGOs, community groups, consultants and other interested individuals to provide data on the state and distribution of Australia’s biodiversity, including invasive species.
To facilitate crowdsourcing, the Atlas offers data collectors two main mechanisms to capture, aggregate, analyse, model and map their data, free of charge. Firstly non-scientists and scientists alike have been able to enter their sightings and photos of species directly into the Atlas for some time now. All data provided to the Atlas, irrespective of its source, undergoes a series of data quality checks, eg to identify outliers and correct misspellings of species names. Results are annotated on each sighting record. This allows Atlas users to easily check the provenance, and quality, of all records in the Atlas to assess their ‘fitness for use’. For example, a researcher may decide to include records contributed by an annual BirdLife Australia survey, but not include records contributed by a primary school student.
Secondly, the Atlas has a software suite—FieldData—that provides a dedicated web portal as a focal point for community groups and researchers for recording observations of plants, animals and other living things. These sightings can be ad hoc, entered as and when desired by users, or based on organised surveys. Data entered into FieldData is instantly displayed on a map so users can compare their sightings with those of other users of the portal. As with the Atlas site itself, photos can also be uploaded, a capability that is especially useful for species identification; where the photo can be viewed by more experienced community members or experts to help with identification.
The Atlas and invasives
So how could the Atlas be used to detect and monitor invasives? Imagine this scenario…
A keen ecologist and Bushcare member — let’s call her Liz — notices an unfamiliar weed starting to appear in her neighbourhood. Liz mentions this to her Bushcare group and does some research—she thinks the weed may be known but newly arrived in her area.
For her own interest, Liz looks up the weed species in the Atlas, checks the distribution records for it and generates a map of its recorded distribution. There are no records of the weed in her neighbourhood. So Liz starts adding her sightings of the weed to the Atlas and encourages her colleagues and Bushcare members to do the same, which they do. They all then contact similar groups and interested individuals in the surrounding area to encourage them to register sightings of the weed on the Atlas site.
Slowly, the ‘crowd’ engages with the spread of the weed and, using the Atlas as an enabling tool, the number of sightings of the weed increases reflecting the increasing and known distribution of the weed. Within a year the weed sightings are widespread but with an inconsistent distribution. Liz decides to investigate further. Back in the Atlas, Liz pulls up all the sightings of the weeds and maps their distribution. Liz works her way through the environmental layers available in the Atlas’s spatial portal looking for some correlations. Finally she discovers that the weed is only found in areas with particular rainfall, temperature and soil characteristics.
She reports these findings to the relevant state government agency that can now use this information to monitor and assess the spread of the weed, and decide whether it needs to monitor and/or control the weed and its spread.
Crowdsourcing is expected to continue to reinforce its role in providing valuable, credible and quality data for research. The Atlas will continue to support crowdsourcing activities intended to inform our understanding and management of Australian biodiversity.