EVIDENCE FOR THE INHERENT STOCHASTIC NATURE OF FUNCTIONAL GROUP DISTRIBUTION IN DISSOLVED ORGANIC MATTER ACROSS VARYING WATER MASSES

Dissolved organic matter (DOM) is one of the most complex mixtures known, with only 5% of marine DOM characterized at the molecular level. Ultrahigh-resolution mass spectrometry (UHR-MS) is the primary method for analyzing DOM, but it typically provides limited structural insights. We converted carboxyl groups of DOM recovered from three different water masses (fluvial as well as surface, and deep Pacific Ocean) to their corresponding methyl esters using isotopically labeled diazomethane and characterized the reaction products by UHR-MS and nuclear magnetic resonance spectroscopy. The conversion of reacted compounds closely followed a log-normal distribution. In accordance with the central limit theorem, this indicates a stochastic reactivity of carboxyl groups across all DOM compounds. Additionally, carboxyl groups were stochastically distributed across the molecular space, with their abundances only dependent on the elemental composition of the respective molecular formula and water type. Computational isomer enumerations supported the stochastic distribution. We established a model based on the stochastic nature of carboxyl groups in DOM to assess their abundance in more than 1,000 samples across major oceanic water masses on the globe. By combining this approach with further molecular analyses, we assessed structural differences between DOM from land, the surface and deep ocean. Photochemical processes and the presence of vascular plant-derived phenols are likely the key drivers behind these differences.

Presentation Preference: Oral

Primary Presenter: Nico Mitschke, Carl von Ossietzky Universität Oldenburg (nico.mitschke@uni-oldenburg.de)

Authors:

Aaron von Seggern, Carl von Ossietzky Universität Oldenburg  (aaron.von.seggern@uni-oldenburg.de)

Thorsten Dittmar, Carl von Ossietzky Universität Oldenburg  (thorsten.dittmar@uni-oldenburg.de)

Nico Mitschke, Carl von Ossietzky Universität Oldenburg  (nico.mitschke@uni-oldenburg.de)