The Global Historical Climatology Network (GHCN-M), version 3 and 4: Improving the global surface temperature record

Since the early 1990s the Global Historical Climatology Network-Monthly (GHCN-M) dataset has been an internationally recognized source of data for the study of observed variability and change in land surface temperature. It provides monthly mean temperature data for 7280 stations from 226 countries and territories, ongoing monthly updates of more than 2000 stations to support monitoring of current and evolving climate conditions, and homogeneity adjustments to remove non-climatic influences that can bias the observed temperature record. The release of version 3 monthly mean temperature data marks the first major revision to this dataset in over ten years. It introduces a number of improvements and changes that include consolidating “duplicate” series, updating records from recent decades, and the use of new approaches to homogenization and quality assurance. Although the underlying structure of the dataset is significantly different than version 2, conclusions regarding the rate of warming in global land surface temperature are largely unchanged. The version 3 dataset has already provided a foundation for the more ambitious version 4. An experimental version 4 dataset is already under development, with a nearly quadrupling of the monthly mean temperature stations (28,000+). In addition version 4 will continue to refine and enhance the existing version 3 quality control algorithms and procedures. It is expected that GHCN-M version 4 will become the official repository for all monthly element data publicly distributed by the National Climatic Data Center (NCDC).