The role of climate datasets in understanding climate extremes

A rapidly growing body of literature routinely employs historical observations to study extremes in past and current climate. The inconsistencies in observations often lead to erroneous results and drawing incorrect inferences when undertaking analyses of climate extremes at regional or global scales. Understanding the potential inhomogeneity in climate datasets is therefore central to the study of climate extremes, especially when attributing any shifts in extremes to a changing climate. Despite the best efforts in assembling quality-controlled input data sources, inconsistencies in data are inherently embedded within long-term records of observations. Knowing the strengths and limitations of climate datasets can potentially facilitate better analyses of climate extremes. This chapter begins with an overview of climate extremes and Climate Extreme Indices (CEIs). The importance of quality control input meteorological variables, tools for assembling the CEIs, and a detailed list of recommended CEIs suitable for examining a broad array of temperature- and precipitation-based extremes are described next. Different sources of global and regional input climate data along with their strengths and limitations for assembling the CEIs form the crux of the next section. Existing datasets of CEIs and recommendations for future research conclude the chapter as the final two sections.