Investigating the effects of extreme rainfall trends on the flow capacity of streams over the Northeast United States

While various research efforts investigate the direct effects of climate change on hydrometeorological variables, the incidental consequences of extreme rainfall trends on the flow capacity of open channels remains an open question. Hydrological modeling for the assessment of flood events and the organization of protection strategies usually include precipitation fields transformed by climate change factors. The latter, however, simply account for the relation (frequently through a ratio) between past and future Intensity-Duration-Frequency (IDF) values. Along these lines, epistemic uncertainties introduced by the choice of the IDF estimation techniques and/or the extensive incorporation of climate model simulations are accounted for through the application of safety factors on the yielded results. Yet, this practice may lead to a misestimation of flood risk, accompanied by costly, yet ineffective, protective measures. Moreover, the employment of high-resolution distributed hydrological models over extensive areas can be computationally cumbersome, while introducing an additional layer of uncertainty. In this study, we attempt to link the occurrence of channel overflowing to the evolution of the magnitude and frequency of extreme rainfall over the Northeast United States. More precisely, we: a) use measured streamflow data offered by the United States Geological Survey (USGS) during the 41-year period from 1979 to 2019, to assess the rate of occurrence of flood events over gauge locations across the study domain, and b) link the observed evolution of the aforementioned overflow rates to that of extreme rainfall for different return periods and durations of temporal averaging. In this context, we attempt to develop a conceptual basis for studying the effects of climate change on the linkage between rare precipitation events and the reliability of existing channels.