Spatiotemporal characteristics of extreme precipitation and its response to temperature change in Xinjiang, China

Abstract: Extreme events happen frequently in the context of global warming, particularly in the arid and semi-arid regions, such as Xinjiang. The global surface temperature has increased to allow more water vapor entering the atmosphere, which intensifying the global water cycle. Meanwhile, extreme precipitation events can be more frequent and intense both regionally and globally in the future. Global warming has led to variation in the precipitation types, resulting in more precipitation in the form of rainfall in some regions. In previous studies, extreme precipitation has been often studied as a whole, whether the extreme rainfall or snowfall has been discussed separately in some cases. But, only a few studies have been reported on the differences between extreme rainfall and extreme snowfall. The Clausius-Clapeyron equation hypothesis (hereafter referred to as the C-C equation) can be considered as the physical mechanism under extreme precipitation (rain and snow) events. At the same time, the atmospheric water holding capacity can be assumed to increase by approximately 7% with the increase of every 1°C in temperature under a constant relative humidity. However, there are some differences in some regions, such as most parts of Europe, North America, and the Tibetan Plateau, where the increase of extreme precipitation intensity can be much less than 7% with the increase of every 1°C in temperature, (so-called sub-C-C rate). Even some regions experience extreme precipitation intensities with the increase of temperature up to 14% (the "super-C-C" rate), such as Hong Kong and the De Beers region of the Netherlands. And even more, there is a negative relationship between extreme precipitation intensity and temperature in tropical regions. Therefore, it is necessary to discuss the relationship among extreme rainfall, extreme snowfall intensity, and temperature at various stations. In this study, the rainfall and snowfall events were separated from the wet-bulb temperature method using the data from 75 observing meteorological stations from 1961 to 2020, which has an overall accuracy of 95%. The percentile approach (95th percentile) was also used to define the extreme precipitation (rain and snow), further to determine the overall characteristics of extreme precipitation and frequency changes in this region, while the spatial variation differences between the northern and southern part in Xinjiang, as well as the response mechanisms of the intensity of extreme rainfall and extreme snowfall to temperature changes. The results show that there was an increasing trend in the amount and frequency of average annual precipitation (rain and snow) and the average annual extreme precipitation (rain and snow). The increase in the amount and frequency of rainfall, extreme rainfall, snowfall and extreme snowfall was greater in the northern region than that in the southern region (except for the frequency of rainfall). The mean contribution of extreme precipitation (rain and snow) to precipitation (rain and snow) was greater than 26%, both of which showed an increasing trend. The response of extreme rainfall and snowfall to temperature changes in both the northern and southern parts was less than the expectation of the C-C equation, i.e. the "sub C-C rate". Once a certain temperature was exceeded, there was an inhibiting effect on the intensity of extreme rainfall and snowfall. The study will contribute to the scientific understanding of the changing characteristics of extreme precipitation events in the Xinjiang and is of great significance to the rational regulation of regional hydrological and water resources.