The scientist behind citizen science

Any number of definitions can be found for ‘citizen science’ but generally they are thought of as volunteers collecting scientific data which will be then used for research, conservation or management of species or groups. But while a number of projects refer to their volunteers as ‘citizen scientists’, volunteers are often unaware of the limitations of the data they collect and in some cases filtering of data removes some of the citizens from the citizen science. In order to advance ‘citizen science’, volunteers must be treated like scientists and scientists (at least like to imagine) that they never blindly follow instructions without understanding the underlying reasons. Except chemists. So what follows is a brief introduction to the concept of citizen science recording and some of the issues that arise.

In an ideal world, species recording and population monitoring would involve counting all of the individuals of a given species in a given area on a regular basis. Where scientific teams would consist of hundreds of specially trained individuals able to identify all species, have unlimited budgets, be free of all time constraints and it would only rain between 3-7am except when conducting bat surveys (and your birthday because no world can be truly ideal). Meanwhile, on this planet, scientists are constrained by both time and money making the collection of large quantities of data difficult. Typically, this results in either periodic national surveys (e.g. National Survey of Breeding Hen Harriers or National Otter Survey 2010/12) providing a snapshot of population change over a large area or detailed projects showing species interactions or population interactions over a much smaller spatial scales (typically research projects that require intensive surveying in an area, repeat measurements or specialist licences). For example, Sheehy & Lawton’s work detailing the negative correlation between pine martens and grey squirrel and the positive effects on red squirrel populations was carried out across 3 counties (Wicklow, Offaly and Laois). However, the level of effort required would have made a nationwide study impossible from a practical point of view. Therefore, in order to bridge this gap, between costly large scale studies and specialist spatially explicit projects, ‘citizen science’ is heralded as the solution to all ills as it allows large quantities of data to be collected across as large an area and time frame as interest from volunteers is maintained.

Citizen science data was traditionally used to produce atlases, or presence only maps over a large area, and these were then used to detect coarse changes in biodiversity and species distributions. However, presence only maps are chiefly useful when done over a long time scale (traditionally ten years), with easily identifiable species, and in order to detect trends in species abundance filtering of data must occur (the removing of citizens from the citizen science). Most issues with citizen science are based around the concept of ‘variation in recorder effort’ which must be resolved to make all the data collected available for analysis. Variation in recorder effort is a catch all term that can include: different number of visits to the same sites each year; sites visited by different recorders; differences in identification skills and training of recorders; and differences in recording effort per site visit amongst others (Isaacs et al., 2014 cover this topic in considerably more detail). This variation in recorder effort makes it difficult to identify areas of differing species structure and abundance or to track population density and distribution changes for individual species.

Variation in recording effort and detectability can be best exemplified by a terrestrial example. If you record 1 mistle thrush in your garden on the 1st of January (while the record in of itself is useful), there is a no way of quantifying recording effort. Did you see 1 mistle thrush on your way to the car in the morning, rushing for work with a struggling child under each arm and your car keys in your mouth? Or did you quietly sit at your kitchen window all day and only one solitary mistle thrush visited your garden all day? Do you record every mistle thrush you see in the garden or just this one? Do you record every day? Or even the dreaded, was it actually a song thrush?

Without information on recording effort, unfiltered data gives you distribution maps similar to the one shown. While some analysis of the data may be made, by and large, the distribution pattern seems to follow human population density in Ireland. Is this simply a matter of more individuals, more recorders, more records (an artefact of recording effort) or urban foxes are performing better than their country cousins? Obviously filtering of the data or the addition of further information would be needed to answer those questions. Similarly, absence of records from an area on an atlas may be due to a lack of recording effort in an area, difficulties in identifying or observing species. Finally, small scale projects collecting the same data may result in a number of patchy datasets across an area rather than few large datasets and is why repositories of data (e.g. National Biodiversity Data Centre and Global Biodiversity Information Facility) are so important.

Figure 1 Distribution records of fox (Vulpes vulpes) from Ireland, data from gbif.org, showing records of fox density per 10km2 divided into areas of high (red), medium (orange) and low (beige) fox density.
Figure 1 Distribution records of fox (Vulpes vulpes) from Ireland, data from gbif.org, showing records of fox density per 10km2 divided into areas of high (red), medium (orange) and low (beige) fox density.

The easiest way to avoid issues with recording effort is to use a standardised methodology (though obviously this does not eliminate them entirely) which can include same number of site visits by the same recorders at set intervals. For example, Bat Conservation Ireland’s All Ireland Daubenton’s survey, the Irish Whale & Dolphin Group’s constant effort sightings scheme, the NBDC’s Butterfly and Bumblebee monitoring schemes, i-Webs, and many, many more, all use a standardised methodology and provide training to volunteers undertaking surveys. However, once you begin to standardise methodology and introduce some level of training, we return full circle to the issue of the volunteers, no longer being volunteers, and becoming the ‘citizen scientists’ of citizen science. So congratulations, if you contribute to any of the schemes mentioned, you can consider your honorary B.Sc. well earned.

Rory O'Callaghan
About Rory O'Callaghan 5 Articles
Rory O'Callaghan is currently the National Coordinator for Seasearch's citizen science site monitoring scheme. He originally completed a B.Sc. in Applied Freshwater and Marine Biology in GMIT before undertaking a M.Sc. in Animal Behaviour in Anglia Ruskin University Cambridge. His primary research interests are on the use of citizen science for species monitoring, competition theory, invasive species and anything shiny that crosses his path.
Contact: Website

Be the first to comment

Leave a Reply

Your email address will not be published.


*


*