On Temperature Characteristics From1979to2012in Xiangjiang River Basin

Climatic factors were being driven by the environment of global warming, which not only led a whole series of temperature to change in spatial and temporal, but also promoted the occurrence of extreme temperatures and promoted changes of probability (frequency) for extreme temperature events. It was of realistic significance and far-reaching strategic significance to the climate issue, which not only in academic inquiry, but also in the ecological environment and socio-economic systems of each region. This paper, around this issue about temperatue, based on reanalysis daily data(daily average temperature, daily maximum temperature and the daily minimum temperature), which were downloaded in ECMWF. And44sites data were generated by the interpolation process. The basic situation of daily temperatures, extreme temperatures, the probability distribution characteristics and interpersonal differences years were analysised through the climatic trend, partial correlation coefficients, skewness, kurtosis, and the threshold method. The main results are as follows:1) For the frequency distribution of mean temperature, maximum temperature and minimum temperature, they were to the right in spring relative to the normal distribution curve, and skewness coefficients of them showed negative bias, the same as in summer and autumn. But skewness coefficient of them were opposite, which showed positive bias in winter, and kurtosis showed significant differences in each season. Temperature frequency showed "high-low-high" distribution patterns during whole year. Spatially, skewness coefficient of the spring, summer and autumn were dominated by negative value for these three temperature values in most areas of the basin, and they were positive in winter. The performance of slope and steepness of kurtosis were different in spring (it’s less than3), summer (it’s greater than3), autumn (it’s less than3), winter (southern basin was less than3, the rest of area the more than3). Each site in the whole basin did not meet the normal distribution. For decadal differences, they had little different when the temperature threshold was T>18℃for the three kinds of temperatures during these periods (1979-1990,1991-2000,2001-2012), and the number of days of high temperatur increaseded significantly in recent years. The temperature threshold after (T>24℃, T>30℃, T>35℃), it was significantly higher than the probability of the first two periods during2001-2012. In conclusion, the days of high temperature increased significantly in recent years.2) For time variation of frequency of extreme temperature, throughout the Xiangjiang River, time trends of frequency distribution were similar to extreme high temperature during the day and night, they showed a decreasing trend until the1990s, it began to increase in the late1990s. Frequency distribution of extreme low temperature during the day and night expressed as three periods:it increased in volatility during1979-1984, it stability declined from1985to1996, and from1979to early20th century, it had a significantly volatility rises. From the spatial distribution of frequency, extreme high temperature during the day and night showed that the north areas was more than the south. Extreme low temperature events of day and night manifested less in central basin and showed more in the western and eastern. For frequency spatial variation trend, we can see that extreme high temperatures events during the day and night showed positive trends, and extreme low temperature events expressed as negative growth trend on most stations. It had more significant changes to frequency of extreme temperature events during the day. For the interannual variability, the probability distribution of extreme high (low) temperature event during day and night were "high-low high" pattern in three periods (1979-1990,1991-2000,2001-2012). Seasonally, probability distributions of daytime extreme high (low) temperature events had the same performance, it also was similar to decadal probability distribution of nighttime extreme high (low) temperature events in different seasons.