This dissertation reports new measurements of stratospheric ¹⁷ O-CO₂ and ¹⁸O-CO₂, doubling the existing dataset and providing new knowledge regarding the variation in ¹⁷O-CO₂ and ¹⁸O-CO₂ throughout the stratosphere. The measurements reveal an unexpectedly large slope of 2.3 for the ¹⁷O/¹⁶O:¹⁸O/¹⁶ O relationship, compared with values of 1.2 to 1.7 from previous measurements and values of 1.0 to 1.7 in models, differences that are robust with respect to regression statistics and analytical precisions of 0.2‰ for &dgr; ¹⁷O and 0.1‰ for &dgr;¹⁸O. Hypotheses explaining these discrepancies include 2-endmember mixing between less isotopically-enriched, lower stratospheric air with more enriched, upper stratospheric or mesospheric air, differences in the mesosphere-stratosphere mixing predicted by models, and/or latitudinal variations in the anomalous or mass-dependent isotope effects controlling the ¹⁷O anomaly in CO₂. As such, the new measurements provide important new constraints on the underlying isotope effects and how they are modified by atmospheric transport. These results also indicate that the optimal locations for future measurements are the upper and the tropical stratosphere.

Modeling performed for this dissertation shows that the anomalous ¹⁷O enrichments in stratospheric CO₂ can also provide a new constraint on gross carbon fluxes between the atmosphere and biosphere. A box model was used to investigate the balance between stratospheric production of the ¹⁷O anomaly and destruction and dilution by the biosphere. Tropospheric Δ¹⁷O-CO₂ is predicted to be 0.1‰ and sensitive to changes in variations in gross primary productivity. Therefore, measurements of Δ¹⁷O-CO₂ will provide an important additional constraint on gross carbon fluxes, similar to &dgr;¹⁸O-CO ₂ but dependent on fewer complex hydrological variables, such as &dgr; ¹⁸O-H₂O.

This dissertation also reports net stratosphere-troposphere fluxes for &dgr; ¹⁸O-CO₂, &dgr;¹³C-CH₄, &dgr;D-CH ₄, and &dgr;D-H₂, determined empirically from stratospheric observations. The magnitudes of these fluxes indicate that the transport of isotopically-enriched gases from the stratosphere likely influences the meridional distribution or seasonality of the isotopic compositions measured at the surface. These isotope fluxes can now be included empirically for the first time in 3D models used to determine the latitude distribution of the various sources and sinks.