Guiding the creation of a comprehensive surface temperature resource for 21st century climate science

Surface temperature data – observed primarily for weather-related purposes – are disparate, originating from ever evolving instrument types and observational practices. Although several global databases are in use internationally, no comprehensive global repository exists and many data are undigitized or restricted. Scientists have painstakingly obtained vast quantities of data, carefully removed random errors and accounted for systematic biases. The 21st Century demands go further - requiring highly detailed (spatially and temporally), globally complete, long-term products, with quantified uncertainties, and created from freely available, fully traceable data. Many decisions of substantial socio-economic importance rely on the accuracy of such products. An international meeting was held to plan how best to facilitate such efforts. A central repository is to be created, where data are traceable from their origins to final product. Strategies are outlined to rescue non-digitized data and move towards entirely freely available data. Creation of multiple methodologically independent products is recommended for quantifying uncertainty. Methods of benchmarking and assessing multiple products to aid inter-comparison and end-user product selection are described. Data-products would be obtained and visualized using in-house tools from the planned data-portal. Structure and governance include engagement with bodies such as WMO and, importantly, with experts other than climatologists.