The use of historical sources and data to improve knowledge of past extreme events is no longer to be demonstrated. Old quantitative and qualitative data on water levels available in non-digitized tidal charts, tide gauge records, newspapers, city council registers, engineers’ reports, etc. can be used to reconstruct water levels and skew surges that occurred during a storm. Data from primary and/or secondary historical sources require a historical critical approach. This paper presents an interdisciplinary study that required a combined work between historians, geographers, geologists and engineers. In this study, nine different variables that can lead to uncertainties on the quantification of historical skew surges have been identified, such as tidal predictions, chart datum references or the reliability of the historical data. Starting from the historical critical method developed by historians over decades, we propose a three-step method to assess the quality of historical documents and define their level of reliability. The method is applied to two case studies: the storm of 16th November 1940 and the storm of 15–16th February 1941. Interesting results are provided and the study shows that by paying attention to the potential causes of uncertainties, by ensuring the reliability of historical sources and data through the use of the historical documents quality method, by deepening historical researches and by combining different scientific fields and methods, it is possible to reduce uncertainties and errors when quantifying historical skew surges.