Precipitation intensity-duration-frequency (IDF) curves are widely used for stormwater management planning and infrastructure design. Currently constructed based on past data, IDF curves do not reflect the potential impacts of current and future changes in the hydrologic system, potentially leading to over- or under-design of structures. Different disaggregation approaches have been developed to obtain IDF curves for sub-daily durations from coarse daily climate model projections. Here, instead of starting from daily IDF curves to obtain sub-daily ones, we use thermodynamic global warming simulations, which directly provide hourly precipitation across the contiguous United States (CONUS) for low- and high-temperature sensitivity models and future greenhouse gas emission scenarios. Our results indicate an amplification of precipitation extremes at different accumulation periods from 1–24 h in all scenarios across CONUS, especially towards the end of the 21st century. This is particularly true for the U.S. Northwest, where we report a statistically significant difference in the amplification of 1–6 h precipitation extremes compared to 24 h extremes. In general, we find the precipitation magnitudes are projected to increase with decreasing precipitation frequency (i.e. higher return period/lower annual exceedance probability), for higher emission scenarios, towards the end of this century, and for global climate models that exhibit a high sensitivity to anthropogenic forcing. For instance, in the northwestern United States under high emission scenarios and based on high-sensitivity models, the 100 year hourly precipitation extremes are projected to increase by over 200%, while the 5 year 24 h precipitation extremes are projected to increase by less than 50% by the end of the 21st century. Our results suggest that applying daily amplification factors to sub-daily extremes could lead to significant underestimation of sub-daily extremes. Therefore, a careful examination of changes in sub-daily extremes, and their regional variability is required to mitigate challenges with the design and operation of current and future urban infrastructure.

Read original article

Read original article