Potential predictability of seasonal precipitation over the United States from canonical ensemble correlation predictions

[1] Potential predictability of seasonal precipitation over the US is explored using a new canonical ensemble correlation (CEC) prediction model, which optimally utilizes intrinsic sea surface temperature (SST) variability in major ocean basins. Results show that CEC yields a remarkable (10-20%) increase in baseline prediction skills for seasonal precipitation over the US for all seasons, compared to traditional statistical predictions using global SST. While the tropical Pacific, i.e., El Nino, contributes to the largest share of potential predictability in the southern tier States during boreal winter, the North Pacific and the North Atlantic are responsible for enhanced predictability in the northern Great Plains, Midwest and the southwest US during boreal summer. Overall, CEC significantly reduces the spring-summer predictability barrier over the conterminous US, thereby raising the skill bar for seasonal precipitation predictions.