Abstract
Upon landfall, atmospheric rivers (ARs)—plumes of intense water vapor transport—often trigger weather and hydrologic extremes. Presently, no guidance is available to alert decision makers to anomalous AR activity within the subseasonal time scale (~2–5 weeks). Here, we construct and evaluate an empirical prediction scheme for anomalous AR activity based solely on the initial state of two prominent modes of tropical variability: the Madden–Julian oscillation (MJO) and the quasi-biennial oscillation (QBO). The MJO—the dominant mode of intraseasonal variability in the tropical troposphere—modulates landfalling AR activity along the west coast of North America by exciting large-scale circulation anomalies over the North Pacific. In light of emerging science regarding the modulation of the MJO by the QBO—the dominant mode of interannual variability in the tropical stratosphere—we demonstrate that the MJO–AR relationship is further influenced by the QBO. Evaluating the prediction scheme over 36 boreal winter seasons, we find skillful subseasonal “forecasts of opportunity” when knowledge of the MJO and the QBO can be leveraged to predict periods of increased or decreased AR activity. Certain MJO and QBO phase combinations provide empirical subseasonal predictive skill for anomalous AR activity that exceeds that of a state-of-the-art numerical weather prediction model. Given the wide-ranging impacts associated with landfalling ARs, even modest gains in the subseasonal prediction of anomalous AR activity may support decision making and benefit numerous sectors of society.
Highlights
A comparative gap in forecast guidance exists between mediumrange weather forecasts and seasonal outlooks (3 + months).[1,2,3] opportunities abound to add far-reaching value to society with skillful predictions of extreme, and frequently hazardous, weather events that occur within this so-called subseasonal-to-seasonal gap.[3]
We focus our analysis on the December through March (DJFM) period, as shaded in Fig. 1b, when atmospheric rivers (ARs) frequently occur near both the British Columbia and California landfall boundaries, when teleconnection patterns are expected to be the most robust over the North Pacific, and when the aforementioned Madden–Julian oscillation (MJO)–quasi-biennial oscillation (QBO) link has been observed
A worthwhile question is how does the level of skill from this empirical prediction scheme compare to the skill available from numerical weather prediction models? To answer this question, we evaluate a suite of 46-day European Centre for Medium-Range Weather Forecasts (ECMWF) retrospective forecasts initialized from 1995 to 2016.2 As with the empirical method, we target 5day average anomalous AR activity for each landfall boundary
Summary
A comparative gap in forecast guidance exists between mediumrange weather forecasts (up to 2 weeks) and seasonal outlooks (3 + months).[1,2,3] opportunities abound to add far-reaching value to society with skillful predictions of extreme, and frequently hazardous, weather events that occur within this so-called subseasonal-to-seasonal gap.[3] Sectors such as agriculture, energy production, resource management, and insurance stand to benefit from advance notice of weather extremes in order to prepare for such events. A recently developed empirical model for predicting North American 2-m temperatures based on the MJO, the El Niño-Southern Oscillation (ENSO) cycle, and linear trends produces skill and provides valuable guidance beyond a basic climatological forecast.[13]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
