An algorithm for daily temperature comparison: Co.Temp - comparing series of temperature

Having a high-quality database is a necessary condition to study the climate variations in a location, in a nation or in a continent. The World Meteorological Organization, WMO, is aware of the importance of identifying the non-climate factors that can modify the recorded values of the meteorological variables. According to this, WMO wrote a set of guidelines recognizing the need for National Meteorological Services to improve databases managing this kind of information (WMO 2007). One of these guidelines points out that this work can be done using parallel observations. Comparing series of Temperature, CoTemp, is a free and open source software written in R language that could help the identification of discontinuity caused by non-climate factors, change in instrumentation, change in positioning or change in the network. It works on parallel series with an overlapping period of at least 1 year. CoTemp starts with a statistical analysis, classifies the events (in cold, mean and heat events), and then shows their differences. CoTemp is a cross-platform software, easily adaptable to different needs, that takes in input a single text file with daily information of two temperature series and outputs tables (in CSV format) and plots (as PDF images) that help in the interpretation of the data. CoTemp code has been tested and used on different temperature series in the Piedmont region (northwestern Italy) and in the Gaspè Peninsula (Canada).