Advisor

Andrew L. Rice

Date of Award

8-8-2019

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.) in Applied Physics

Department

Physics

Physical Description

1 online resource (xi, 160 pages)

DOI

10.15760/etd.7003

Abstract

Nitrous oxide (N2O) is the third most important greenhouse gas (GHG) behind carbon dioxide (CO2) and methane (CH4). Sulfur hexafluoride (SF6) does not add significantly to climate forcing by itself due to the low concentration in the atmosphere; however, it is one of the most powerful GHG known. Measurements of atmospheric N2O made prior to mid-1990 have larger uncertainties than later periods due to advancements made in gas chromatography (GC) methods. Few atmospheric SF6 measurements pre-1990 exist, especially in the northern hemisphere. Archived samples may be analyzed using updated measurement techniques to reduce uncertainty in past periods. Additionally, measurements of the isotopic composition of N2O (15N/14N and 18O/16O) can address questions regarding how specific sources contribute to the observed atmospheric composition and changes in time. This information also has been used to identify changing contributions of nitrification versus denitrification processes to the global N2O budget, determined by the rate of change in the site preference (SP, defined as δ15Nα - δ15Nß of 15N. Here, we present the findings of 159 measurements of N2O and SF6 mixing ratio and N2O isotopic composition from the OHSU-PSU air archive, containing samples collected from Cape Meares, Oregon between 1978 and 1997.

Based on these analyses, N2O mole fraction at Cape Meares in 1980 is found to be 301.5 ppb and rises to 313.5 ppb in 1996. The average growth rate over this period is 0.78 ± 0.03 ppb yr-1 (95% CI). Seasonal amplitude maximum and minimum is 0.34 ppb near April and -0.42 ppb near November respectively and are statistically different from one another (p2O were found to match well with previously reported values for Cape Meares from AGAGE and other comparable locations, suggesting that the N2O in archived samples has stored well.

For SF6, the mole fraction in 1980 was found to be 1.06 ± 0.05 ppt and increased to 3.91 ± 0.07 ppt in 1996. The average growth rate over this period is 0.17 ± 0.01 ppt yr-1 (95% CI). Seasonality shows peak amplitude of 0.04 ppb near January and minimum amplitude of -0.03 ppt near July. There are no previous reported measurements of SF6 from Cape Meares to compare against directly; however, comparisons against archive measurements from Cape Grim, Tasmania suggest these results are accurate.

Isotopic composition measurements of N2O are made by cryogenically concentrating N2O before analysis using an isotope ratio mass spectrometer operated in continuous flow (CF-IRMS). From replicate analysis of the working standard, precision in measurement of the measured isotopologues is 0.05‰, 0.10‰, and 0.28‰ for δ45, δ46, and δ31, respectively. When calculating the desired isotopic composition, these translate to 0.05‰, 0.10‰, 0.37‰ and 0.39‰ for δ15Nbulk, δ18O, δ15Nα, and δ15Nß, respectively.

For archived samples from Cape Meares, no distinguishable seasonality is found in δ15N or δ18O while δ15Nα, δ15Nß, and SP show statistically significant amplitudes. δ15Nα and δ15Nß show nearly opposite phases to one another, with SP matching the phase of δ15Nα. These results suggest processes that contribute air enriched in N2O mole fraction at Cape Meares in the spring also contribute enriched δ15Nα and depleted δ15Nß, causing a positive SP. During the fall, processes that contribute air depleted in N2O mole fraction also contribute depleted δ15Nα and enriched δ15Nß, causing a negative SP.

Secular trends (except δ15Nß) calculated by applying a linear fit to the deseasonalized data show negative trends statistically significant at high levels of confidence. Secular trends for δ15N and δ18O match well with previously reported values while secular trends for δ15Nα and δ15Nß for Cape Meares are significantly different than those reported by other groups, appearing to be nearly inverted for δ15Nα and δ15Nß. To address this inversion, the sensitivity of the numerically calculated δ15Nα and δ15Nß on the scrambling coefficient was investigated and ruled out. We also investigated the possibility of an error in the numerical algorithm used to convert measured 45R, 46R, and 31R values into δ15N, δ18O, δ15Nα, δ15Nß and found this too was not responsible for the inverted results.

A 2-box model of the atmosphere was used to investigate changes in measured atmospheric composition to characterize source isotopic composition. From the results of the box model, the magnitude of the pre-industrial natural source match well with previous literature. The isotopic compositions of the natural source match well with previously reported values, as do the modeled anthropogenic δ15N, δ18O, and δ15Nα. However, our modeled δ15Nß is significantly enriched compared with previously reported values. Additionally, the modeled anthropogenic SP is significantly depleted than previously reported values. Assuming the laboratory measurements of intramolecular SP are globally relevant, these results suggest there have not been significant changes to the balance between contributions from nitrification and denitrification to the observed isotopic composition of N2O during this period (1978 -- 1996). However, a more sophisticated model maybe needed to investigate this hypothesis.

Persistent Identifier

https://archives.pdx.edu/ds/psu/29602

Available for download on Sunday, August 08, 2021

Included in

Physics Commons

Share

COinS