TUESDAY, 30 AUGUST 2022
I stare at the column of numbers, trying to parse meaning from the words written more than 150 years ago. I pause at one row, briefly unsure. Having familiarised myself with the handwriting of this page, I am fairly confident the number written down is a 3 and not a 5. I note that on this day, 2nd February 1863 at 8am, the H.M.S. Eclipse experienced an air temperature of 63oF, secure in the knowledge that my contribution will aid scientists in modelling the climate.
I am on Zooniverse, ‘the world’s largest and most popular platform for people-powered research’, with 1.6 million registered participants spread out over 234 countries. Its goal is to ‘enable research that would not be possible, or practical, otherwise’, by utilising the power of volunteers accessing and analysing datasets uploaded by researchers. This format lends itself well to hosting projects that cover a range of subjects beyond more traditionally scientific fields, like history, language, and literature.
The Zooniverse platform grew from the project Galaxy Zoo launched in July 2007, which aimed to classify nearly one million images of galaxies from the Sloan Digital Sky Survey into morphological categories. Several hundred thousand volunteers participated, with each image categorised an average of 38 times, such that the general public could classify sets of galaxies with a similar accuracy to professional astronomers. Since then, more than 380 projects across a diverse range of disciplines have been launched. More than 600 million classifications have been submitted, and over 250 articles have been published using data from these citizen science projects.
Citizen science, where scientific research is conducted in collaboration between non-professional volunteers and professional scientists, is by no means a recent concept - astronomer Edmund Halley in 1714 requested curious citizens to note down observations of a total solar eclipse visible in England. He received many ‘good Accounts’, which were ‘sufficient to establish several of the Elements of the Calculus of Eclipses, so as for the future we may more securely rely on our Predictions.’ The advent of the Internet has facilitated mass access to citizen science; while Halley credits only 24 observers of the solar eclipse, now thousands of volunteers can contribute to a single project.
Due to the sheer number of volunteers analysing the data for Zooniverse projects, reliable results can be produced by combining these individual contributions in a ‘wisdom of crowds’ approach. Whilst programming a computer algorithm would require a lengthy and complex process, Zooniverse projects take advantage of humans’ inherent ability of pattern recognition. By completing a few simple tasks, regardless of their level of expertise, volunteers with minimal training can analyse data, saving researchers time and resources. In fact, projects have used volunteer-made classifications to train machine-learning algorithms or used algorithms to sort data before presenting it to volunteers. Machines and humans working in collaboration can create superior classifications over either one working alone.
Having so many contributors, Zooniverse projects are in the unique position of facilitating what are described as ‘Known Unknowns’ and ‘Unknown Unknowns’. An example of the former is from the Exoplanet Explorers project, where volunteers are tasked to identify planetary transits from time-series photometry data taken from NASA’s K2 mission. The first discovery made by citizen scientists was the five-planet system K2-138, announced at the end of a three-day program of Stargazing Live featuring the project.
Particularly interesting are the ‘Unknown Unknowns’ - serendipitous discoveries driven by the human ability to notice the unexpected and having a platform designed to facilitate such discoveries. Each project has a discussion board where volunteers are encouraged to discuss their observations with each other and with researchers directly. The ‘green peas’ of the Galaxy Zoo project exemplifies this - after citizen scientists repeatedly noticed compact green shapes within the data set and set up the ‘Give peas a chance’ discussion thread, researchers worked with the volunteers to identify them. These objects are now known to be low-mass galaxies in low density environments with high star formation rates, creating a new category within galaxy type taxonomy.
Researchers can derive potentially huge benefits from citizen scientist contributions - however, what benefits can citizen scientists receive in return? Citizen science in general engages the public by exposing them to genuine scientific experiences. Projects promote greater interest in not only the science behind them, but also how research is conducted, developing participants’ scientific literacy.
Citizen scientists may therefore derive greater benefits from more ‘hands-on’ projects. A case study of youth-focused community and citizen science programmes which monitored coastal sites and local pollution showed these programmes helped foster a deeper appreciation for the site’s ecological importance and the impact of human activity. The programmes additionally aided in the youths’ understanding of environmental science, and in using their gained expertise as a foundation for change. They concluded that having a longer programme enabling participants to explicitly contribute to authentic research while strengthening a connection with the site encourages the development of environmental science agency.
Zooniverse therefore, simply by the less personal nature of the platform, is more limited in the scope of its potential benefits. Nonetheless, we should not dismiss the platform. In a survey of 1,921 volunteers participating in five projects, the participants were asked to respond to science quizzes where they needed to identify objects presented in a series of images. The researchers found a significant and positive correlation between the level of active engagement with the project and project-specific scientific knowledge. The majority of volunteers also self-reported that they believed they were developing their scientific literacy by participating in Zooniverse projects. However, another study carried out with 1,476 Galaxy Zoo volunteers found that project participation was not associated with an increase in astrophysical content knowledge. The authors note that perhaps a better measure associated with content knowledge is the activity level on related social networking spaces such as discussion boards and blog posts, based on anecdotal evidence of self-study by some volunteers within these spaces. This is perhaps unsurprising as it corroborates with the above conclusions that greater engagement promotes a deeper appreciation for the science behind the project. Zooniverse actively enables this by cultivating a platform wherein users are encouraged to ask questions and discuss their results.
The ease of access of the projects may also facilitate participation from a more diverse range of people. An interesting finding from a study of 104 users aged 5-19 years old is that, in contrast to what the authors describe as the largely white, middle-aged males who predominate citizen science field-based programmes, over half the surveyed users were females. Though this may simply reflect existing research that females volunteer more, as the authors note. The study is the first of its kind focusing on young people in online citizen science projects and has a very small sample size, so overarching conclusions should not be drawn. It is nevertheless an encouraging thought. Regardless, Zooniverse is an easy-to-access, low-commitment stepping stone facilitating potentially further scientific discovery for both citizen scientists and professionals.
Besides, for me there is something both satisfying and humbling in that, while I can relax performing the low-effort task of transcribing weather observations, researchers will use my results alongside that of over 1,700 other volunteers to improve estimates of the pre-industrial baseline climate. This is the Weather Rescue At Sea project, where to fill in the lack of early industrial era observations in climate datasets, volunteers transcribe ship logbooks from ships sailing in the 1860s and 1870s. The logbooks are a rich resource, as they provide usually sub-daily observations of measures such as air temperature and wind speed, alongside the ship’s latitude and longitude from locations across the world.
Ultimately, maybe this is what the Zooniverse is: a small gateway open for anyone to get a glimpse of the rich scientific background of your universe.
To learn more about this platform and the specific projects, visit zooniverse.org
Galaxy Zoo - https://www.zooniverse.org/projects/zookeeper/galaxy-zoo
Exoplanet Explorers - https://www.zooniverse.org/projects/ianc2/exoplanet-explorers
Weather Rescue At Sea - https://www.zooniverse.org/projects/p-teleti/weather-rescue-at-sea
Julie Tang is a third-year undergraduate studying Natural Sciences at Newnham College. Artwork by Josh Langfield.