资源环境科技发展态势分析平台

In Australia and globally, changes in the entire distribution of daily precipitation are poorly understood. An analysis of spatiotemporal changes in the intensity and frequency of wet and all day Australian rainfall between 1958-1985 and 1986-2013 is presented using three different gridded data sets of daily precipitation based on varying underlying station data and interpolation methods. Our analysis method provides a complete picture of changes in the entire distribution as well as a coherent picture of the spatial changes across Australia. We find that the spatial pattern of changes in total rainfall is similar to the pattern of change in wet day frequency but not to change in wet day intensity. Furthermore, more grids across Australia show statistically significant increases than decreases in rainfall intensity throughout the wet day distribution. This means that when it rains (i.e., limited to the wet day distribution), whether it is light, moderate, or extreme rainfall, it rains more.

Plain Language Summary Rainfall has substantial impacts on our society, agriculture, urban environment, and industry and is strongly affected by climate change. However, understanding changes can be complex due to the variable nature of rainfall. For example, a region can receive fewer rainy days (rainfall frequency) even though the amount of rainfall when it rains (rainfall intensity) could be higher. Our study conducts a comprehensive analysis of rainfall changes by analyzing changes in the frequency and intensity of light, moderate, and heavy rainfall using daily gridded observational data sets for Australia between two 28-year periods, 1958 to 1985 and 1986 to 2013. We found that when it rained, it rained more intensely in the second period compared to the first irrespective of whether it was light, moderate, or heavy rainfall. However, changes in the frequency of rainfall varied regionally. Specifically, there are now more rainy days in northern Australia compared to the past, while in the east and south of Australia there are fewer rainy days. Finally, we also show that some aspects of rainfall are sensitive to the underlying station network, while others are sensitive to the gridding methodology of the data sets used.