Spatial Correlations of Storm, Monthly and Seasonal Precipitation

Abstract

Abstract One approach to defining sampling requirements for precipitation measurement networks is through statistical correlation methods. Data from three dense raingage networks in Illinois were used with this method on rainfall measurements ranging from 1-min rates to total storm, monthly and seasonal amounts. Effects of rain type, synoptic storm type, and other factors on spatial correlations were studied. Correlation decay with distance, used to indicate sampling requirements, was greatest in thunderstorms, rainshowers and air mass storms. Conversely, minimum decay occurred with steady rain and the passage of low pressure centers. Seasonally, the decay rate is much greater in May–September storms than in cold season precipitation. Sampling requirements are extreme in measuring rainfall rates; thus, assuming a minimum acceptance of 75% explained variance between sampling points, a gage spacing of 0.3 mi is needed for 1-min rain rates compared with 7.5 mi for total storm rainfall in summer storms.