Investigating spatio-temporal patterns in subsurface chlorophyll using biogeochemical-Argo floats and novel statistical methods

Abstract

Chlorophyll concentration is a widely used proxy for phytoplankton biomass, and its continued monitoring is essential for understanding phytoplankton’s role in the global carbon cycle and as the foundation of marine ecosystems. The work presented in this thesis explores the large-scale spatio-temporal variability of vertical chlorophyll structure, particularly its relationships with environmental conditions, using data collected by Biogeochemical-Argo floats through a suite of statistical approaches. First, a spatio-temporal modelling framework was employed to identify the drivers of subsurface chlorophyll maxima (SCMs) on a global scale. This method pooled observations across space and time and revealed that the euphotic depth (zeu) was the main driver of SCM depth and intensity. However, these insights were limited by the need to extract SCM properties prior to modelling. Consequently, functional regression models were used to examine how environmental conditions influenced the profile shapes of chlorophyll and particle backscatter (bbp), which enabled the study of SCMs within the context of entire profiles. Results showed that SCM depth was primarily governed by the zeu, while peak bbp was linked to the nitracline depth. Additionally, photoacclimation, the physiological response of phytoplankton to low light conditions, emerged as a key driver of SCMs throughout the low latitudes. Finally, a novel measure of variance for oceanographic profiles was applied to chlorophyll and temperature data, and their correlation was assessed. Spatio-temporal autocorrelation of both variables was examined in Eulerian and semi-Lagrangian perspectives. This analysis revealed that the similarity in spatio-temporal length scales of chlorophyll and temperature varies by region and spatial extent of the dataset. In summary, this work highlighted the importance of light in determining the vertical distribution of phytoplankton and how contemporary statistical tools improve ecological insights into subsurface biogeochemical observations from autonomous platforms.

More information

Identifiers

ORCID for Sujit Sahu: orcid.org/0000-0003-2315-3598

Catalogue record

Export record