Dissolved organic matter (DOM) is a highly diverse mixture of compounds, accounting for one of the world's largest active carbon pools. The surprising recalcitrance of some DOM compounds to bacterial degradation has recently been associated with its diversity. However, little is known about large-scale patterns of marine DOM diversity and its change through degradation, in particular considering the functional diversity of DOM. Here, we analyze the development of marine DOM diversity during degradation in two data sets comprising DOM of very different ages: a three-year mesocosm experiment and highly-resolved field samples from the Atlantic and Southern Ocean. The DOM molecular composition was determined using ultra-high resolution mass spectrometry. We quantify DOM diversity using three conceptually different diversity measures, namely richness of molecular formulas, abundance-based diversity, and functional molecular diversity. Using these measures we find stable molecular richness of DOM with age >1 year, systematic changes in the molecules' abundance distribution with degradation state, and increasing homogeneity with respect to chemical properties for more degraded DOM. Coinciding with differences in sea water density, the spatial field data separated clearly into regions of high and low diversity. The joint application of different diversity measures yields a comprehensive overview on temporal and spatial patterns of molecular diversity, valuable for general conclusions on drivers and consequences of marine DOM diversity.

Dissolved organic carbon (DOC) is the main energy source for marine heterotrophic microorganisms, but a small fraction of DOC resists microbial degradation and accumulates in the ocean. The reason behind this recalcitrance is unknown. We test whether the long-term stability of DOC requires the existence of structurally refractory molecules, using a mechanistic model comprising a diverse network of microbe-substrate interactions. Model experiments reproduce three salient observations, even when all DOC compounds are equally degradable: (i) >15% of an initial DOC pulse resists degradation, but is consumed by microbes if concentrated, (ii) the modelled deep-sea DOC reaches stable concentrations of 30–40 mmolC/m3, and (iii) the mean age of deep-sea DOC is several times the age of deep water with a wide range from <100 to >10,000 years. We conclude that while structurally-recalcitrant molecules exist, they are not required in the model to explain either the amount or longevity of DOC.

Marine dissolved organic matter (DOM) is a complex mixture of chemical compounds. At 750 Pg C, it is one of the biggest pools of reduced carbon on Earth. It has been proposed that the diversity of DOM is responsible for its recalcitrance. We hypothesize that the chemodiversity of marine DOM is a reflection of the chemodiversity of bacterial exometabolomes. To test this, we incubated two model strains of the Roseobacter group; Phaeobacter inhibens and Dinoroseobacter shibae in pure culture using three different simple organic compounds as sole carbon sources (glutamate, glucose, and acetate and succinate for P. inhibens and D. shibae , respectively). The exometabolome of the model organisms was characterized using Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (FT-ICR-MS) and ecological diversity measures. We detected thousands of molecular masses in the exometabolomes of P. inhibens and D. shibae (21,105 and 9,386, respectively), reflecting the capability of single bacterial strains to diversify simple organic compounds. The chemical composition of the exometabolomes changed with growth phase and also differed according to the strain incubated and the utilized substrate. We mimicked a higher diversity of substrates, bacterial species and heterogeneous growth (different growth phases) to approach the complexity of natural environments, by computationally creating combinations of detected exometabolomes. We compared the chemodiversity of these combinations, indicative for chemodiversity of freshly produced microbial DOM to that of refractory DOM from one of the oldest oceanic water masses (North Equatorial Pacific Intermediate Water). Some combinations of exometabolomes showed higher richness than the deep ocean refractory DOM, and all the combinations showed higher functional diversity. About 15% of the 13,509 molecular formulae detected in exometabolomes and refractory oceanic DOM were shared, i.e., occurred in Roseobacter exometabolomes and in deep water samples. This overlap provides further support for our hypothesis that marine bacteria from the Roseobacter group contribute to the sustainability of marine DOM chemodiversity and stability.

Ecological stability refers to a family of concepts used to describe how systems of interacting species vary through time and respond to disturbances. Because observed ecological stability depends on sampling scales and environmental context, it is notoriously difficult to compare measurements across sites and systems. Here, we apply stochastic dynamical systems theory to derive general statistical scaling relationships across time, space, and ecological level of organisation for three fundamental stability aspects: resilience, resistance, and invariance. These relationships can be calibrated using random or representative samples measured at individual scales, and projected to predict average stability at other scales across a wide range of contexts. Moreover deviations between observed vs. extrapolated scaling relationships can reveal information about unobserved heterogeneity across time, space, or species. We anticipate that these methods will be useful for cross‐study synthesis of stability data, extrapolating measurements to unobserved scales, and identifying underlying causes and consequences of heterogeneity.