Development of a Statistical Subseasonal Forecast Tool to Predict California Atmospheric Rivers and Precipitation Based on MJO and QBO Activity

Abstract

This paper examines the empirical relationship between the Madden–Julian oscillation (MJO), the quasi‐biennial oscillation (QBO), and atmospheric river (AR) activity and precipitation in California on subseasonal time scales. We introduce an experimental forecast tool that uses observed anomaly patterns during a 38 yr period to predict the probability of above‐ and below‐normal AR activity and precipitation at lead times of 1–6 weeks based on the phase and amplitude of the MJO and QBO. The hindcast prediction skill of probabilistic AR activity and precipitation forecasts is evaluated for Northern, Central, and Southern California, as well as two sets of smaller geographical domains. These smaller domains are more relevant for water resource management and allow us to investigate the sensitivity of prediction skill to domain size. Consistent with previous studies, our results demonstrate that subseasonal AR activity and precipitation in California are strongly modulated by the MJO and QBO. The anomaly patterns of AR activity and precipitation vary considerably throughout the cool season, with a tendency toward below‐normal AR activity and precipitation during easterly QBO and above‐normal AR activity and precipitation during westerly QBO in JFM. The opposite patterns are generally observed in OND, but the anomaly signals are weaker and less coherent for AR activity. Certain combinations of MJO phase, QBO phase, lag time, and season yield notably higher skill scores, reinforcing the notion of “windows of opportunity” for skillful subseasonal‐to‐seasonal predictions. In California, these forecasts of opportunity are predominantly associated with easterly QBO in JFM and FMA.

From California’s Extreme Drought to Major Flooding: Evaluating and Synthesizing Experimental Seasonal and Subseasonal Forecasts of Landfalling Atmospheric Rivers and Extreme Precipitation during Winter 2022/23

“…However, the persistence of the AR activity along the California coastline during this period was unusually intense and unlikely to be explained by the MJO transition alone. We note that based on the results from Castellano et al (2023), the likelihood of wet conditions during NDJ under westerly QBO conditions, which were observed during the period of interest last winter, generally decreases over California. During JFM, very few combinations of MJO phase and lag time result in significantly increased probabilities of wet conditions under westerly QBO conditions.…”

“…End users stand to benefit from improvements in both subseasonal (2-6 week lead time) and seasonal (1-6 month lead time) forecasts of these variables (Gershunov and Cayan 2003;Waliser et al 2006;Gottschalck et al 2010;NASEM 2010NASEM , 2016Vitart et al 2017;Merryfield et al 2020;Mariotti et al 2020;DeFlorio et al 2021;White et al 2022;Sengupta et al 2022). During this recent period of substantially increased investment in improving forecasts of hydroclimate variables beyond weather lead times, CW3E and collaborating institutions have investigated research topics and designed experimental forecast tools for ARs, circulation regimes, and total precipitation at both subseasonal (DeFlorio et al 2019a,b;Gibson et al 2020a,b;Robertson et al 2020;Castellano et al 2023;Wang et al 2023;Zhang et al 2023) and seasonal (Gershunov and Cayan 2003;Gibson et al 2021;Kirtman et al 2014;Switanek and Hamill 2022;Scheftic et al 2023) lead times. In response to this growing suite of institutions, methods, and experimental forecast products, CW3E has recently created several new experimental subseasonal and seasonal synthesis forecast products to help summarize key experimental forecast information to stakeholders and end users (https://cw3e.ucsd.edu/s2s_forecasts/).…”

Atmospheric Rivers are Responsible for Cyclicity in Sierra Nevada Precipitation

“…The Sierra Nevada, however, sits astride the ENSO teleconnection dipole and as a result cool-season precipitation totals in this transition zone are not well correlated with SSTs in the tropical Pacific or elsewhere (Dettinger et al 1998;Higgins et al 2000a;Dettinger et al 2011;Cook et al 2018;Williams et al 2021). Some of the poor correlation between Sierra Nevada cool-season precipitation and tropical Pacific SSTs may be due to nonlinearity (e.g., El Niños are wetter than La Niñas are dry), interactions between multiple large-scale teleconnection modes of oceanatmosphere variability (Guirguis et al 2018;Henderson and Maloney 2018;Castellano et al 2023), subseasonal variability in teleconnections (Castellano et al 2023), and sensitivity to the exact location of deep tropical convection (Chiodi and Harrison 2015;Seager et al 2015;Chen et al 2017;Williams and Patricola 2018). Whatever the cause, atmospheric model simulations forced by observed SSTs have particularly poor skill in terms of replicating observed hydroclimatic variations within the transition zone of the western U.S. ENSO dipole (Huang et al 2019).…”