Linking spiciness and oxygen variability on isopycnals off Vancouver Island
Date
2026
Authors
Maier, Michaela
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Subsurface oxygen variability in coastal upwelling systems is often difficult to quantify due to sparse observations, strong seasonal dynamics, and the combined influence of physical and biological processes. Along the slope and shelf off of the west coast of Vancouver Island in the northern California Current System, subsurface oxygen concentrations are strongly influenced by the relative contributions of oxygen- rich subarctic and oxygen-poor southern source waters. This thesis investigates the link between physical properties and subsurface oxygen variability in the study area and evaluates whether numerical models can be used to estimate oxygen changes where observations are limited. Using over three decades of temperature and salinity observations and more than two decades of oxygen measurements, subsurface variability is examined on key isopycnals (density surfaces) that characterize the seasonal upwelling–downwelling cycle.
This thesis identifies the σθ26.6 isopycnal as the deepest regularly upwelled isopycnal (shoaling over 100 m seasonally), highlights important seasonal cycling in both spiciness (the relative temperature and salinity of a given isopycnal, Chapter 2) and oxygen (Chapter 3), reveals notable differences in interannual variability, and uncovers significant trends.
Chapter 3 also shows that spiciness effectively traces source water mixing and explains much of the observed oxygen variability on the slope, where this physical driver dominates. A robust linear relationship between spiciness and oxygen is identified, allowing oxygen to be estimated from spiciness where biological influences are weak.
In Chapter 4, output from a global physical reanalysis model and a regional coupled physical–biogeochemical model is evaluated against observations. In regions and seasons where spiciness variability is realistically represented in the model data, oxygen derived from spiciness consistently outperforms directly modelled oxygen, accurately reproducing the observed variability and trends in oxygen. These results demonstrate that a physically grounded proxy-approach based on the link between spiciness and oxygen can provide meaningful estimates of subsurface oxygen variability and trends where observations are sparse. However, the reliability of these estimates depends critically on the accurate representation of the physical water-mass structure in the model, highlighting the continued necessity of sustained local observations to evaluate model performance and interpret model-derived oxygen estimates.
Description
Keywords
oceanography, spiciness, oxygen, Vancouver Island, California Current System