Statistical Parameters of Hydrometeorological Variables: Standard Deviation, SNR, Skewness and Kurtosis

Abstract

Global spatial patterns of standard deviation, signal-to-noise ratio (SNR), skewness and kurtosis of eight hydrometeorological variables, i.e. potential evapotranspiration (PET), average temperature (Tavg), precipitation (Pre), maximum temperature (Tmax), minimum temperature (Tmin), wet day frequency (Wet), diurnal temperature range (DTR) and vapour pressure (Vap) are studied and discussed in this chapter. Global high resolution gridded hydrometeorological data provided by Climate Research Unit (CRU) is used. Not much variation is found in the annual values of Pre, DTR and Wet with respect to mean. High values of standard deviation in annual values of Tavg, Tmax and Tmin are found. It is found that higher variability in the dataset is generally associated with a lower mean value. Only a few regions of the world are found to be significantly skewed. About 3.5–9.5% regions of the world are found to be significantly leptokurtic for all variables except for PET for which almost all datasets are found significantly leptokurtic.