Skillfully Predicting Atmospheric Rivers and Their Impacts in Weeks 2-5 Based on the State of the MJO and QBO

With the advent of Week 3-4 forecasts beginning in September 2015, the NOAA Climate Prediction Center (CPC) now provides a continuous suite of precipitation outlooks targeting mean conditions spanning from the extended range (i.e. Days 6-10 and 8-14) through approximately a year. While the majority of these outlooks are skillful, Week 3-4 precipitation outlooks only exhibit skill similar to that expected from random chance. Of interest to this proposal is improving CPC precipitation forecasts for Week 3-4 and beyond by improving forecasts of relatively short duration (~day), high-impact, extreme precipitation events associated with atmospheric rivers (ARs) along the west coast of North America.

Skillful forecasts of ARs on subseasonal-to-seasonal (S2S) timescales would support many aspects of society, e.g. emergency management, water managers, shipping route designation, and agricultural practices. While current dynamical model forecast skill of AR events decays rapidly beyond Week 2, recent work by the PIs (published in Mundhenk et al., 2017) demonstrates that AR activity can be skillfully predicted into Week 5 with an empirical forecast model based on two prominent modes of tropical variability: the Madden-Julian oscillation (MJO) and the quasi-biennial oscillation (QBO). This empirical model is based on the joint MJO and QBO probability distributions and largely follows the recently operational method of Johnson et al. (2014) to forecast 2-meter temperature anomalies.

Given CPC’s existing lack of AR-related guidance, the methodology of Mundhenk et al. (2017) to skillfully predict AR frequencies and precipitation at S2S timescales will be transitioned into CPC operations. Refinement and extension of this methodology will be explored and transitioned into operations, pending demonstrable skill. The goals of the proposed activities are threefold: (1) successfully transition the AR frequency forecast tool to operations, (2) refine and extend the methodology of Mundhenk et al. (2017) to maximize skill of AR activity and AR-related variables, and (3) leverage additional predictors, including dynamical model MJO forecasts, to extend the skillful forecasts beyond Week 5. The tasks will be performed by scientists at the Climate Prediction Center (CPC) and the Department of Atmospheric Science at Colorado State University (CSU), who have extensive experience working together and transitioning similar empirical tools into CPC operations.

Relevance and Suitability for NOAA
The proposed project addresses the CTB goal of testing, demonstration, and implementation of tools for S2S predictions relating to atmospheric rivers. The competition Information Sheet explicitly encourages the prediction of heavy precipitation, and the proposed work will directly address these extremes by improving the prediction of atmospheric rivers and their impacts. This project leverages the PIs past experiences in weather and climate extremes, empirical guidance development, and transition of guidance into forecast operations. The proposed work initially meets NOAA Readiness Level 5, which is the final stage of development where methodologies and products that have been successfully evaluated in experimental settings (i.e. Mundhenk et al., 2017) are tested and prototyped in their relevant environment.