In 2013, a large upper-ocean thermal anomaly formed in the Gulf of Alaska (GOA) with sea surface temperatures (SSTs) warmer than 4 degrees Celsius above the climatological norm. This warm anomaly persisted for the next three years and has been linked to downstream effects on North American climate and negative impacts to the North Pacific ocean ecosystem. Formation of the anomaly has been associated with a persistent atmospheric high pressure ridge in the GOA which produced a lull in storm activity starting in the boreal winter of 2013 (Bond et al. 2015; Hartmann 2015). Whether this was a normal mode of oscillation within the North Pacific basin is still being debated as this warm anomaly is not unprecedented in the contemporary data record (e.g., Liang, et al. 2017). This ridging was hypothesized to suppress winter ocean heat loss to the atmosphere, leading to an anomalous upper ocean warm pool, colloquially named the blob. While storm activity was anomalously low during this and subsequent winters, we present cases where storms nonetheless significantly shaped the evolution of the GOA sea surface temperature anomalies (SSTa). During the almost four years that the anomaly resided in the GOA, it is shown here that synoptic-scale cyclones significantly eroded its surface signature and manipulated its spatial structure on synoptic timescales. The primary goal of this research is to test the hypothesis that the interaction between the upper ocean thermal anomaly and individual storm events during fall was a significant mechanism for dissipating excess heat from the warm anomaly. Not all storms, however, affected the thermal anomaly equally. Case studies using satellite observations, reanalysis, and ARGO hydrographic observations show that early fall storms produced the largest surface heat fluxes and the greatest cooling of SST compared with storms during other periods. The second goal of this study was to determine how much of the warm anomaly thermal energy was transferred to the atmosphere during individual storm events. Vertically-integrated heat budgets were computed using ARGO upper-ocean temperature profiles and surface turbulent heat and radiative flux estimates from the ECMWF and NCEP reanalyses. Storm-induced surface heat flux anomalies can account for 40-60% of the warm anomaly cooling observed with ARGO profiles during large fall storms. This rapid heat loss occurred over the course of one ARGO profile cycle, which is approximately 11 days. The erosion of the warm SSTa occurred much quicker than predicted by models forced with stochastic atmospheric forcing (Frankignoul and Hasselmann,1977; Wills and Thompson, 2018). Analysis of the individual surface flux terms indicated that the latent heat flux was the dominant contributor to heat exchange with the atmosphere during the early season storms.
Given the relative importance of evaporative decay of the warm NEP-SSTa demonstrated by this analysis, this research also seeks to understand how surface latent heat flux during and in-between storm periods have been changing in the North Pacific over a 36-year time period. A monthly, spatially-varying extreme state threshold is used to separate latent heat into two separate states; intra- and inter-event periods. A power law model is used to describe the duration of these different states to determine how the duration of storms, intra-events, are changing as well as the time between storms, inter-events. East Pacific intra-event periods are found to be shorter in duration with larger flux values with respect to the mean latent heat flux during these events. This is in reference to the most recent time period of 2005-2016 relative to 1981-1992. Inter-event periods in the East Pacific were seen be getting longer with weaker fluxes. This type of pattern, if it persists, may create favorable conditions for subsurface thermal anomalies like the NEP-SSTa to be a recurring feature in the future.