Optimal Precursors of U.S. West Coast Marine Warming

The US West Coast hosts a very rich marine ecosystem which supports activities of large economical and societal value. Warmer than average conditions in this area are associated with reduced upwelling and reduced productivity, negatively impacting marine ecosystem dynamics. From the Fall of 2013 to the Winter of 2015 exceptionally warm conditions were experienced in the northeast Pacific and US West Coast with catastrophic consequences on marine ecosystems across multiple trophic levels. Positive tropical Pacific sea surface temperature (SST) anomalies associate with El Niño conditions usually lead warm conditions along the US West Coast by a few months due to atmospheric and oceanic teleconnections. However, only weak positive SST anomalies occurred in the tropical Pacific from the end of 2013 to the beginning of 2015, suggesting that SST patterns associated with canonical El Niño events may not be the most influential on the US West Coast. In this study, we use multiple linear regressions and singular value decomposition to identify the tropical oceanic state – sensitivity pattern - that can optimally force US West Coast SST anomalies at some later time. As such, the sensitivity pattern provides the conditions that most likely lead to extreme anomalies along the US West Coast. In other words, an initial state that has a strong projection on the sensitivity pattern can be expected to have a large SST anomaly signal along the US West Coast a few months later. We characterize the tropical state in terms of SST and sea surface height (SSH). The latter is dynamically linked to thermocline depth, thus providing information about the ocean memory associated with subsurface dynamical processes. Our results show that the optimal SST/SSH patterns do not coincide with a mature El Niño pattern, but include anomalies in the western equatorial Pacific and in the extra-equatorial tropics resembling some of the El Niño precursors. Large anomalies along the US West Coast, including the 2013-2015 warming, are indeed characterized by tropical conditions with a large projections on these sensitivity patterns, which thus represent more skillful predictors of the US West Coast conditions than the traditional El Niño Southern Oscillation indices.