Retrieval of unexploited historical snow data from hydrological yearbooks in the Italian Alps

Snow depth is a key climate variable that plays an important role in the hydrological cycle, surface energy balance through albedo control and the functioning of mountain ecosystems. In Italy, snow monitoring was historically carried out by the Italian National Hydrological and Mareographic Service, which managed hydro-meteorological observations from 1917 to 2002 through 14 regional compartments defined according to the main river catchments, now transferred to regional authorities. Despite its potential scientific importance, the snow dataset is available primarily in paper-based hydrological yearbooks, remaining mostly unexploited. However, the scanned images of the original hydrological yearbooks are available via the ISPRA portal (http://www.bio.isprambiente.it/annalipdf/), thanks to a dedicated digitisation project carried out between December 2003 and September 2012.

Significant efforts have been made to digitize historical snow depth observations from the hydrological yearbooks, particularly for areas of the Apennine region. In parallel, for the Alps, additional programs have focused on the analysis and harmonisation of long-term snow depth records that were already available in digital form, taken from a variety of regional and local organizations throughout the Alpine area and surrounding countries. However, there are still significant temporal and spatial gaps, especially before 1970.

The objective of this study is to further close these gaps by recovering additional snow depth measurements from historical hydrological yearbooks for the Italian Alps that were not included in previous compilations, specifically from Parma,Venice and Genova sections. We are applying an optical character recognition (OCR) technique based on an algorithm designed to extract tabular snow data from scanned archival documents. This method will allow the digitization of previously unavailable observations, improving the data coverage both in time and space. The resulting dataset will be an important contribution to long-term snow variability research, climate change assessments, and hydrological applications in Alpine regions.