Date of Award

Fall 2021

Project Type

Dissertation

Program or Major

Natural Resources and Environmental Studies

Degree Name

Doctor of Philosophy

First Advisor

Joseph Salisbury

Second Advisor

Robert Byrne

Third Advisor

Peter Raymond

Abstract

As anthropogenic climate change continues to elevate the amount of carbon dioxide (CO2) in the Earth’s atmosphere, the absorption of a large portion of this CO2 by Earth’s oceans has resulted in a steady decrease in pH. The consequent phenomenon of ocean acidification (OA) is a result of shifts in the carbonate chemistry system of the ocean- a system which can be analytically described by several factors, including total alkalinity (TA). TA in the oceans has been measured for over a century, but analytical and operational constraints have limited these measurements in time and space. Additionally, recent work has highlighted gaps in our knowledge of the species which collectively comprise TA. This dissertation describes efforts to examine TA through several novel applications: by deploying an automated TA analyzer aboard a survey vessel to map East Coast USA TA distributions, using the same analyzer in a long-term fixed coastal location to build a timeseries and examine seasonal biogeochemical dynamics, and measuring the concentrations and properties of the poorly understood organic component of TA in two Gulf of Maine estuaries. East Coast regional distributions of salinity (S) and TA generally agreed with prior findings, but linear TA:S regressions varied markedly over time and deviated from previously developed models. This variability is likely due to a combination of biological, seasonal, and episodic influences and indicates that substantial errors of ±10-20 μmol kg−1 in TA estimation from S can be expected due to these factors. This finding has likely implications for numerical ecosystem modeling and inorganic carbon system calculations. New results presented in Chapter 1 provide refined surface TA:S relationships, present more data in space and time, and improve TA modeling uncertainty. Coastal timeseries observations were collected hourly over 28 months representing all seasons between May 2016 and December 2019. Results presented in Chapter 2 indicated that endmember mixing explained most of the observed variability in TA and dissolved inorganic carbon (DIC), concentrations of which varied strongly with season. For much of the year, mixing dictated the relative proportions of salinity-normalized TA and DIC as well, but a fall season shift in these proportions indicated that aerobic respiration was observed, which would decrease buffering (β-H) by decreasing TA and increasing DIC. However, fall was also the season of weakest statistical correspondence between salinity and both TA and DIC, as well as the overall highest salinity, TA and β-H. Potential biogeochemically-driven β-H decreases were overshadowed by increased buffering capacity supplied by coastal ocean water. A simple modeling exercise showed that mixing processes controlled most monthly change in TA and DIC, obscuring impacts from air-sea exchange or metabolic processes. Advective mixing contributions, more than biogeochemically-driven changes, are critical to observe when evaluating local estuarine and coastal ocean acidification. Chapter 3 describes the first comparison study of both organic alkalinity (OrgAlk) distributions and acid-base properties in contrasting Gulf of Maine estuary-plume systems: the Pleasant (Maine USA) and St. John (New Brunswick CA). Four surveys of each estuary were conducted between May 2018 and October 2019. Substantial amounts of OrgAlk were measured in each estuary, whose distributions were sometimes not conservative with salinity. Two measures of OrgAlk produced consistently differing results, indicating acid-base characteristics that may be inconsistent with the definition of TA. OrgAlk and dissolved organic carbon (DOC) concentrations varied seasonally in the St. John Estuary, but not in the St. John. The fraction of TA represented by OrgAlk ranged from a maximum of 78% at low salinity in the St. John Estuary to less than 0.4% at the coastal ocean endmember. While the range of St. John OrgAlk concentrations was comparable to other studies, the St. John Estuary demonstrated a broader distribution. The acid dissociation constant (pKa) of the estuary samples was modeled according to a combined speciation and mixing approach, while the organic carbon acid dissociation constant (pKDOC) was estimated using a separate method. Results showed general agreement, but with some notable exceptions in the St. John estuary. OrgAlk modeling results from the Pleasant Estuary were more consistent than the St. John, despite St. John OrgAlk, DOC and pH results exhibiting much less seasonal variability. The mean OrgAlk pKa was higher in the Pleasant than in the St. John, while the mean Pleasant pKDOC was higher or lower than that in the St. John depending on which OrgAlk analysis approach was employed. Application of a bulk pKa or pKDOC to model OrgAlk from more common measurements such as pH, salinity, or DOC may offer promise (as in the Pleasant), but should be undertaken with caution as variability can pose challenges (as in the St. John). Future work should blend the analyses described in the chapters of this dissertation. For example, by collecting discrete samples aboard the survey vessel or at the coastal laboratory organic alkalinity contributions could be used to refine carbonate system calculations. Regional shifts in TA:S could be used to differentiate local and remote coastal endmember TA shifts. While this work utilized novel TA and OrgAlk analyses in three specific applications, the applicability of these analyses is broad and offers the potential to greatly enhance monitoring efforts and ecosystem biogeochemical studies.

Download

Share

COinS