Influence of regional and seasonal rainfall patterns on the ratio between fixed and unrestricted measured intervals of rainfall amounts

Abstract

Historically, most precipitation data have been measured by collecting rainfall, usually at intervals of 24 h, with a fixed starting time. Nonetheless, it is known that the use of fixed time intervals to measure rainfall quantities could lead to an underestimation of the true maximum precipitation amounts for the considered duration, so a single multiplicative correction factor is commonly applied, generally without taking into account the rainfall pattern of the place, nor regional or seasonal considerations. In the present work, hourly measurements from 120 stations of Catalonia (northeast of the Iberian Peninsula) have been used to analyse how the ratio between rainfall amounts measured by fixed and unrestricted intervals, i.e. the correction factor, depends on the considered duration and on the specific starting time of the fixed interval (local 00:00, 08:00, 12:00 or 16:00), as well as the influence of geographical location and seasonality and actual rainfall duration. For fixed sampling intervals starting at 16:00, the mean correction factor has been found to be higher (1.137) than at the usual 08:00 starting time (1.129). Some geographical patterns of the correction factor over Catalonia arose which, moreover, depend on the season, with a mean value of 1.161 in spring and a value of 1.093 in summer. Also, the value of the correction has been found to increase with the actual duration of the maximum rainfall events used in the analysis. Some of these extreme events had actual mesoscale durations between 6 and 9 h, linked to highly convective mesoscale organisations acting mainly in summer and the beginning of autumn. Other maxima episodes, with more advective rainfall lasting more than 12 h registered in the northern area of the territory, presented the highest values of the correction factor, especially in spring.