资源环境科技发展态势分析平台

We propose a new index for quantifying the maximum rain intensity (IMAX) within a rainfall event. The goal of this paper is 2-fold: first, to analyse the characteristics of variability (trends, change points) for maximum values of this index calculated for spring (April-May), summer (June-August) and autumn (September-October) in comparison with other 2 precipitation indices (daily maximum and total amount). Secondly, to investigate the scaling of the IMAX high percentiles with temperature and compare the results to hourly precipitation extremes. The analysis was carried out at 6 Romanian stations over the period 1966-2007, extended over 1902-2007 for one station. Our results revealed a statistically significant increase of IMAX over the 20th century (intensified over the second half), in contrast with no significant trends in the other 2 precipitation indices. On the other hand, an opposite phase between IMAX variability and its corresponding duration (e.g. shorter durations correspond to higher intensities) was noted. Regarding the scaling behaviour of the IMAX percentiles, there was a difference between summer and spring/ autumn. In summer, the 90th and 99th percentiles showed a Clausius-Clapeyron (CC) scaling for temperatures between 14 and 26(circle)C and then a decrease, while the 99.9th percentile showed a super-CC scaling for temperatures lower than 18(circle)C and then a decrease. The spring and autumn 90th and 99th percentiles showed an approximate 2CC scaling for temperatures ranging from 8-20(circle)C. The 99.9th percentile exhibited a scaling close to the 2CC relationship for a temperatures range of 12-18(circle)C (spring) and then a decrease, while the autumn 99.9th percentile exhibited a dependence close to 2CC scaling for temperatures < 18(circle)C and then a scaling close to CC. Comparing these results with those referring to hourly and daily precipitation extremes, we conclude that the magnitude of the CC scaling in Romania is mainly dependent on the temporal scale of the extreme precipitation event, storm intensity and season. The proposed index (IMAX) is more sensitive to temperature increases than the hourly and daily extremes, and therefore it is more appropriate to quantifying the climate signal related to intense precipitation events in a warmer climate.