Seasonal Predictability of Bottom Temperatures Along the North American West Coast

Abstract

Bottom Temperature anomalies (BTA) along the North American West Coast strongly influence benthic and demersal marine species. However, to date seasonal BTA forecast efforts have been limited and sources of BTA predictability largely undiagnosed. Here, an empirical model called a Linear Inverse Model (LIM), constructed from a high-resolution ocean reanalysis, is developed to predict North American West Coast BTAs and diagnose sources of predictive skill. The LIM is considerably more skillful than damped persistence, particularly in winter, with anomaly correlation (AC) skill values of 0.6 at 6-month lead. Analysis of the LIM's dynamics shows that elevated BTA forecast skill is linked to developing El Niño-Southern Oscillation (ENSO) events, driving predicted BTA responses whose peaks occur at longer leads with increasing latitude. Weaker ENSO-related signals in the northern coastal region still yield high BTA skill because noise there is also weaker. Likewise, the LIM's forecast signal-to-noise ratio is highest for bathymetry depths of ∼50–150 m, maximizing forecast skill there. Together, these predictive components lead to “forecasts of opportunity” when LIM anticipates especially high prediction skill. For the top 20% of events identified by the LIM as the forecasts of opportunity, 6-month lead BTA hindcasts have AC skill averaging 0.7, while the remaining 80% hindcasts have mean skill of only 0.4, suggesting that the LIM can leverage ENSO-related predictability of BTA to produce skillful forecasts.