Marine dissolved organic matter (DOM) is one of the largest reservoirs of organic carbon and nitrogen in the world. Yet, despite its global importance, most DOM remains molecularly uncharacterized. Solid-state nuclear magnetic resonance (NMR) spectroscopy of isolated DOM fractions represents one of the most powerful techniques to understand overall structural composition. However, it is well known that standard cross polarization magic angle spinning (CP/MAS) NMR, the technique used for almost all past solid-state NMR studies of DOM, is at best “semi-quantitative,” and underestimates fully substituted NMR active nuclei. Additionally, almost all past solid-state NMR work analyzed high molecular weight (HMW) material isolated by ultrafiltration, which is now understood to represent mostly 14C-young, “semi-labile” compounds. In contrast, there is far less information regarding the composition of older, low molecular weight (LMW) DOM, which represents the vast majority of the ocean’s accumulated refractory DOM pool.Here, we applied 13C and 15N solid-state multiCP/MAS NMR, improved NMR methods optimized to more quantitatively resolve fully substituted NMR nuclei, to both HMW and LMW DOM isolated from the surface and deep North Pacific Subtropical Gyre. These methods confirm past work indicating most nitrogen containing HMW DOM as amide compounds, but also demonstrate a modest heterocyclic N component not previously identified. In contrast, we found that LMW DON is almost entirely aromatic heterocyclic N, consistent with the hypothesis that heterocyclic N structures may be largely responsible for the accumulation of the ocean’s refractory DON pool. Surprisingly, however, we find DOC aromatic functionalities still represent only a very minor portion of either the HMW or the LMW refractory carbon pools, in marked contrast to refractory DON composition. Together, these more quantitative solid-state NMR techniques likely represent the most accurate picture of DON and DOC functional and compound-class makeup to date, and so have broad implications for our understanding of marine DOM structure and cycling. Specifically, our new data suggests that while chemical composition likely acts as a key control on DOM lability, the most refractory components of DOC and DON have very different compositions, sources, and cycling, supporting the idea that DOC and DON cycling in the ocean may be largely decoupled.