No abstract

The focus of this study is on the climatology of atmospheric rivers (ARs) over the central United States using six atmospheric reanalysis products. This climatology is used to understand the long‐term impacts of ARs on annual precipitation, precipitation extremes, and flooding over the central United States. The relationship between the frequency of ARs and three prominent large‐scale atmospheric modes [Pacific‐North American (PNA) teleconnection, Artic Oscillation (AO), and North Atlantic Oscillation (NAO)] is investigated, and the results are used to statistically model the frequency of ARs at the seasonal scale. AR characteristics (e.g., frequency, duration) are generally robust across the different reanalysis products. ARs exhibit a marked seasonality, with the largest activity in winter (more than 10 ARs per season on average), and the lowest in summer (less than two ARs per season on average). Overall, the duration of most ARs is less than 3 days, but exceptionally persistent ARs (more than 6 days) are also observed. The year‐to‐year variations in the total annual precipitation over the central United States are largely explained by the variations in AR‐related precipitation. Moreover, 40% of the top 1% daily precipitation extremes are associated with ARs, and more than 70% of the annual instantaneous peak discharges and peaks‐overthreshold floods are associated with these storms, in particular during winter and spring. The seasonal frequency of ARs can be described in terms of large‐scale atmospheric modes, with PNA playing a major role in particular in winter and spring.

The Bulletin 17B framework assumes that the annual peak flow data included in a flood frequency analysis are from a homogeneous population. However, flood frequency analysis over the western United States is complicated by annual peak flow records that frequently contain annual flows generated from distinctly different flood generating mechanisms. These flood series contain multiple zero flows and/or potentially influential low floods (PILFs) that substantially deviate from the overall pattern in the data. Moreover, they often also contain extreme flood events representing different hydrometeorologic agents. Among the different flood generating mechanisms, atmospheric rivers (ARs) are responsible for large, regional‐scale floods. The spatial and fractional contribution of ARs in annual peak flow data is examined based on 1375 long‐term U.S. Geological Survey (USGS) streamgage sites with at least 30 years of data. Six main areas in which flooding is impacted by ARs at varying degrees were found throughout the western United States. The Pacific Northwest and the northern California coast have the highest fraction of AR‐generated peaks (∼80–100%), while eastern Montana, Wyoming, Utah, Colorado, and New Mexico have nearly no impacts from ARs. The individual regions of the central Columbia River Basin in the Pacific Northwest, the Sierra Nevada, the central and southern California coast, and central Arizona all show a mixture of 30–70% AR‐generated flood peaks. Analyses related to the largest flood peaks on record and to the estimated annual exceedance probabilities highlight the strong impact of ARs on flood hydrology in this region, together with marked regional differences.

Flooding over the central United States is responsible for large socioeconomic losses. Atmospheric rivers (ARs), narrow regions of intense moisture transport within the warm conveyor belt of extratropical cyclones, can give rise to high rainfall amounts leading to flooding. Short-term forecasting of AR activity can provide basic information toward improving preparedness for these events. This study focuses on the verification of the skill of five numerical weather prediction models in forecasting AR activity over the central United States. We find that these models generally forecast AR occurrences well at short lead times, with location errors increasing from one to three decimal degrees as the lead time increases to about 1 week. The skill (both in terms of occurrence and location errors) decreases with increasing lead time. Overall, these models are not skillful in forecasting AR activity over the central United States beyond a lead time of about 7 days.

Short‐term weather forecasts have the potential to improve reservoir operations for both flood control and water supply objectives, especially in regions currently relying on fixed seasonal flood pools to mitigate risk. The successful development of forecast‐based policies should integrate uncertainty from modern forecast products to create unambiguous rules that can be tested on out‐of‐sample periods. This study investigates the potential for such operating policies to improve water supply efficiency while maintaining flood protection, combining state‐of‐the‐art weather hindcasts with downstream conjunctive use to transfer surplus flood releases to groundwater storage. Because available weather hindcasts are relatively short (10–20 years), we propose a novel statistical framework to develop synthetic forecasts over longer periods of the historical record. Operating rules are trained with a recently developed policy search framework in which decision rules are structured as binary trees. Policies are developed for a range of scenarios with varying forecast skill and conjunctive use capacity, using Folsom Reservoir, California, as a case study. Results suggest that the combination of conjunctive use and short‐term weather forecasts can substantially improve both water supply and flood control objectives by allowing storage to remain high until forecasts trigger a release. Further, increased conjunctive use capacity reduces the importance of forecast skill, since surface storage can be moved to groundwater during the flood season without losing water supply. This analysis serves the development of forecast‐based operating policies for large, multipurpose reservoirs in California and other regions with similar flood hydroclimatology.