Thesis advisor: Hilary I. Palevsky / The biological carbon pump in the North Atlantic Ocean is powered by the annual spring phytoplankton bloom. These primary producers use inorganic carbon in the surface oceans and convert it into organic carbon, a fraction of which is exported out of the surface mixed layer and sequestered at depth. Determining the rate of carbon flux below the maximum winter mixed layer depth, driving sequestration on annual or longer timescales, is critical to understanding the North Atlantic carbon cycle.To constrain daily-to-annual scale changes in carbon export in the subpolar North Atlantic, I analyzed seven years of daily optical backscatter depth profiles (200-2600 m) collected from the subsurface profiler mooring at the Ocean Observatories Initiative (OOI)’s Global Irminger Sea Array from September 2014 to May 2021. This is the longest-running time series of daily, year-round optical backscatter profiles that has been collected in this region, providing novel opportunities to assess seasonal and interannual variations in particulate organic carbon (POC) flux to depth. This analysis, focused on large particles and aggregates identified from optical backscatter spikes, shows annual pulses of sinking particles initiating in May to June during each year of our seven-year time series, consistent with these export pulses being driven by organic matter production during the spring phytoplankton bloom. These pulses of particles sink through the water column at rates ranging from 10 and 30 meters per day, and though particle concentration attenuates through the water column due to remineralization, coherent large particle pulses generally extend deeper than 1500 m, the deepest maximum annual mixed layer depth over this period. Although deep winter mixing in this region requires sinking particles to penetrate much deeper than in other parts of the ocean to be sequestered long-term, pulses of large particles consistently penetrate to below even the deepest annual mixed layer depths in the region, highlighting the importance of these large particle pulses to carbon sequestration at depth in the subpolar North Atlantic. / Thesis (MS) — Boston College, 2024. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Earth and Environmental Sciences.

A combination of in situ sampling and real-time ocean observations was used to investigate the processes responsible for the formation and the areal extent of Texas coastal hypoxia from 2002 to 2011. In situ sampling, real-time mooring and buoy observations, and multivariate statistical modeling were used to investigate the physical processes driving hypoxia formation. Geostatistical interpolation (ordinary kriging) models were tested to compare the differences in annual hypoxia area on the Texas shelf. Results from these two sections were integrated into recommendations for improving federal hypoxia monitoring and mitigation strategies in the northwestern Gulf of Mexico. Winds, currents, temperature, salinity, and dissolved oxygen records revealed the annual, seasonal, and daily variability of hypoxia formation on the Texas coast from 2009 to 2011. Hypoxic events occurred from late May to late October lasting from hours to weeks. Hypoxia formation was either the result of salinity stratification, associated with the freshening of surface waters by the advection of Mississippi-Atchafalaya River freshwater westward or the wind- and current-driven upcoast or downcoast flow of Brazos River discharge. Records from 2010 and 2011 showed the variability and frequency of stratification development differs on the north and south Texas shelf. Multivariate linear model results showed contributing factors on the north Texas shelf vary annually and that primary factors for hypoxia development are near-surface current speeds and salinity-driven stratification. Interpolation models resulted in three size categories for hypoxia area: small (100 – 1,000 km^2), moderate (1,001 – 3,000 km^2), and large (3,001+ km^2). Moderate years include 2002, 2004, and 2007 and a large year was 2008. There was no increase in hypoxic area from years 2002 to 2011, but years 2007 and 2008 resulted in a hypoxic area over 5,000 km^2, which is the federally mandated hypoxia reduction target for the northwestern Gulf of Mexico. Geostatistical interpolators represent and predict the structure and spatial extent of the hypoxic area on the Texas shelf by accounting for the anisotropy of physical processes on the Texas shelf. Geostatistical interpolation models are preferred to deterministic models for developing and improving federal hypoxia monitoring and mitigation strategies on the northwestern Gulf of Mexico shelf.

Globally sustained observations of the marine ecosystem and biodiversity are crucial to understand changes in the ocean environment, manage ocean resources and assess progress towards internationally agreed targets such as the SDGs. Efforts in the observing community are growing to close the current gap in the collection of ecosystem and biological data through harmonizing and coordinating monitoring activities around the Biological and Ecosystem Essential Ocean Variables (BioEco EOVs). This thesis looks into the challenges the implementation of a sustainable and coordinated BioEco ocean observation system could meet through conducting a systematic literature review and key informant interviews. It identified a number of key areas of challenges for BioEco ocean observation and corresponding challenges within these areas. The discussion and analysis of the results led to the identification of 11 priority recommendations to implement successful and sustained observations of life in the sea globally: (1) Clear communication of the BioEco EOV concept to the global observing community and international agreement on standards and best practices for data collection; (2) Create an overview of the various scattered databases for BioEco ocean data to realize a “world wide web of oceanographic data”; (3) Promotion and widespread adoption of the FAIR data principles coupled with the development of strong and adaptive data infrastructures and architectures to enable data and database interoperability; (4) Increased capacity in marine science to increase the understanding of large-scale ecological processes and interactions and thereby the quality of data analysis, which can enable better data products catering to the needs of society and decision-makers; (5) Implement mechanisms for better coordination, communication and collaboration across disciplines, institutes, monitoring programs and geographical scales to promote knowledge exchange, resource and capacity sharing; (6) Unification of the fragmented ocean governance framework, implementation of clear governance structures for glocal BioEco ocean observation and harmonization of ocean data integration into policies and decision-making; (7) Unification of the scattered ocean observation efforts under one transparent system that is adaptive to its user needs and has strong links between its components; (8) Implementation of standards and best practices within the system while still encouraging innovation; (9) Implementation of sustainable long-term funding mechanisms at all scales while making the observing system more cost-effective and -efficient; (10) Implementation of continuous capacity development activities for all system components; (11) Improved participation of developing nations through targeted capacity development and strong collaboration processes including capacity and resource sharing as well as knowledge and technology transfer.