The Role of the Atmosphere in Marine Heatwaves

Thesis (Ph.D.)--University of Washington, 2020Marine heatwaves (MHWs) are events of abnormally warm sea surface temperatures (SSTs) that last for an extended period of time. MHWs have devastating impacts on marine ecosystems and coastal economies, and thus there is motivation to better understand these extreme events and forecast their evolution in order to improve the adaptive capacity of communities experiencing these impacts. Although MHWs are extreme oceanic events, both the atmosphere and the ocean affect the buildup, maintenance, and decay of MHWs. This dissertation focuses on the role of the atmosphere during MHWs. While it is well documented that the atmosphere can trigger MHWs through a stalled ridge of high pressure and/or a decrease in winds, not much is known about the role of the atmosphere after SST anomalies emerge. This dissertation documents atmospheric behavior during MHWs. Chapter 2 surveys the data needed for MHW analysis. I outline the variety of atmospheric and oceanic data products that are available for studying the physics of MHWs and provide an evaluation of which products are best suited for certain research questions. For individual MHW events where regionally well-validated reanalysis products are available, reanalysis data provide a large suite of atmospheric and oceanic variables over a longer time period than the newer generation of satellite observations. However, reanalysis products are not recommended for global MHW analyses, as most reanalysis products are not well-validated over the entire globe and errors are regionally variable. For global MHW analyses, satellite data are preferred, as they provide the best available global estimates of SSTs, radiative fluxes, and clouds. Chapter 3 expands on the survey of data products by providing an in-depth evaluation of reanalysis products compared to satellite observations over the Northeast Pacific Ocean, with the goal of finding the best reanalysis dataset for examining the 2013-2016 Northeast Pacific MHW. There is large variability in performance between reanalyses, including how well they capture variables within the datasets and sub-regional variability within the Northeast Pacific. However, for radiative fluxes and cloud fractions, the Climate Forecast System Reanalysis (CFSR) product generally has the smallest errors compared to NASA’s Clouds and the Earth’s Radiant Energy System (CERES) satellite observations, and thus CFSR is selected as the best dataset to analyze MHWs within the Northeast Pacific region. Chapter 4 analyzes the role of clouds and radiative fluxes during the unprecedented 2013-2016 Northeast Pacific MHW, known as the Blob. The warm waters observed during the Blob altered the surface energy balance and disrupted ocean–atmosphere interactions in the region. In principle, ocean–atmosphere interactions following the formation of the MHW could have perpetuated warm SSTs through a positive SST‐cloud feedback. The actual situation was more complicated. While CFSR reanalysis data show a decrease in boundary layer cloud fraction and an increase in downward shortwave radiative flux at the surface coincident with warm SSTs, this was accompanied by an increase in longwave radiative fluxes at the surface, as well as an increase in sensible and latent heat fluxes out of the ocean mixed layer. The result is a small negative net heat flux anomaly (compared to the anomalies of the individual terms contributing to the net heat flux). This provides new information about the midlatitude ocean–atmosphere system while it was in a perturbed state. More specifically, a mixed layer heat budget reveals that anomalies in both the atmospheric and oceanic processes offset each other such that the anomalously warm SSTs persisted for multiple years. The results show how the atmosphere–ocean system in the Northeast Pacific is able to maintain itself in an anomalous state for an extended period of time. Chapter 5 zooms out and takes a broader perspective on the role of the atmosphere during MHWs all across the globe. Here I use satellite data from 2001-2019 to identify MHWs and anomalous atmospheric variables, including radiative heat fluxes, turbulent heat fluxes, and cloud cover, associated with these events. CERES satellite data are used instead of reanalysis data, despite the shorter time series, because satellite data are well-validated worldwide. We find robust patterns in SST-cloud and SST-heat flux relationships that show important geographical differences in atmosphere-ocean interactions during MHWs. Because of these regional differences, we don’t expect MHWs to evolve the same way in all regions. We also find that the cloud response observed during MHWs globally corresponds well with the cloud response to future warming, as identified in the Cloud Feedback Model Intercomparison Project (CFMIP) ensemble of global climate models. This suggests that MHWs can provide valuable insight to anomalous atmosphere-ocean interactions under future warming. Chapter 6 employs a surface heat flux feedback framework in order to quantify the response of surface heat fluxes to underlying SST anomalies during MHWs. Physically, the net surface heat flux feedback is expected to be strongly negative over the world’s oceans (the atmosphere strongly damps underlying SST anomalies) due primarily to enhanced upward turbulent and longwave radiative heat fluxes over warm SST anomalies. However, the atmospheric response can modulate the negative feedback. It is useful to understand regional and seasonal variability in climatological net heat flux feedbacks, as this sheds light on the nature of regional ocean-atmosphere interactions. Climatologically, there is large spatial and seasonal variability in net heat flux feedbacks. This is driven primarily by variability in the shortwave and latent heat flux feedbacks. Although computed feedbacks show that the global net surface heat flux is largely negative as expected, certain regions- including the Northeast Pacific, central and eastern subtropical and tropical Pacific, Northwest Atlantic, and west tropical Atlantic- have positive feedbacks during certain seasons. A statistical analysis shows that net heat flux feedback parameters and MHW length are negatively correlated. This is an important finding, as it indicates that regions with near zero or positive feedbacks are more prone to persistent MHWs. This dissertation lays out multiple lines of evidence showing that the atmosphere plays an important role during the evolution of MHWs. After warm SST anomalies form during MHWs, anomalies in clouds, radiative heat fluxes, and turbulent heat fluxes are observed. These atmospheric anomalies feed back onto SSTs and affect the progression of MHWs. There is large spatial and seasonal variability in the atmospheric patterns during MHWs, therefore, we do not expect MHWs to evolve the same in all regions and all seasons. Furthermore, some areas are more prone to persistent MHWs due to near zero or positive climatological net surface heat flux feedbacks in that region. These new insights into the role of the atmosphere during MHWs are key for helping develop our understanding and get closer to properly modelling and forecasting these extreme events. Using results from the dissertation, we know that coupled atmosphere-ocean models will be needed to capture MHWs. Furthermore, models will need to adequately represent the spatial variability in atmosphere-ocean interactions in order to capture the heterogeneity in the evolution of MHWs around the globe