Abstract
Cli-DaRe@School is a Citizen Science project (https://aisam.eu/progetti/cli-dare-at-school/) aiming at digitising unexploited Italian meteorological observations still available only on paper or as scanned images. It was launched in 2022 as an initiative of the Italian Association of Atmospheric Sciences and Meteorology (AISAM) involving researchers, high-school students, and teachers.
Until now, the project focused on digitizing four monographs published by the Italian Hydrographic Service on monthly temperature and precipitation data for the Italian territory. The temperature data covers 1926-1955, while the precipitation data refers to the years before 1950. During two academic years, a team of more than 500 students from more than 10 high schools was engaged in supporting digitization in the framework of a national work trainee program for students, making about 7931 station records available. Each school was provided with pdf files containing the scanned pages to be digitized, along with spreadsheet templates for data entry, and related tutorials. Students also had the opportunity to participate in a training program offered by the project, which consisted of seminars and activities specifically designed for them. Their goal was to allow the students to delve into various aspects of climate change and to make them aware of the potential of the recovered data. Cli-DaRe@School, on the one hand, demonstrates the potential of high school students to make an enormous contribution to rescuing past meteorological data; on the other hand, it has a great educational value, offering young students an experience with climate data.
At the end of each year, we sent a questionnaire to both students and teachers to gauge their satisfaction with the project activities and to gather suggestions for the following years. The questionnaire revealed a good level of satisfaction among the teachers and students, with the most critical point being the number of hours devoted to digitization.
The quality control of the digitized data, partly automatic using an R code and partly manual, is finished for precipitation over the period 1921-1950, whereas it is still in progress for the previous years and for temperature. The station metadata proved to be a frequent source of errors, while the digitized precipitation and temperature records have a very low error rate. Project results have been published in the Bulletin of the American Meteorological Society (https://doi.org/10.1175/BAMS-D-24-0078.1). The newly rescued precipitation records for the period 1921-1950 are already freely available (https://doi.org/10.5281/zenodo.15084062) while, for the other periods, data publication is still in progress. In parallel, the application of Optical Character Recognition techniques to extend the data rescue also to the daily records is currently under evaluation.