Abstract: Based on the historical disaster data of Hubei Province, we constructed a regression model between disaster loss rate, rainstorm intensity and Per Capita GDP. The daily precipitation data under RCP4.5 and RCP8.5 scenarios and population and GDP data under corresponding SSPs paths were selected to analyze temporal and spatial distribution characteristics and changes of extreme precipitation events, population and population exposure during different periods in Hubei Province. The results showed that: the disaster loss rate is significantly (p < 0.01) correlated with rainfall intensity and per capita GDP. By constructing a disaster loss rate model, it can evaluate population exposure more accurately. Extreme precipitation events in Hubei province were found to decrease spatially from southeast to northwest, and the overall intensity increased with time. The total population of Hubei Province is dense in the east and sparse to the west, and total number under SSP3 path was found to be always higher than SSP2 path. The per capita GDP presents a pattern of decreasing distribution from Wuhan City to the surroundings, and grows rapidly over time, the value under the SSP2 path is always higher than SSP3. The high value center of population exposure to extreme precipitation in Hubei Province was found to shift south and eastward, and due to the rapid growth of per capita GDP, the greatly increase of fortification level, the decreased total population, the total exposure population has been continuously reduced, the extreme precipitation population exposure and total exposed population under the scenario of RCP8.5 are always higher than it under RCP4.5, and the gap will continue to increase in the medium future.