We investigated sinking velocity (w_s) of suspended particulate matter (SPM) in the German Bight. By inferring w_s indirectly from an extensive turbidity data set and hydrodynamic model results, we found enhanced w_s in a coastal transition zone. Combined with known residual circulation patterns, this led to a new conceptual understanding of the retention of fine minerals and nutrients in shallow coastal areas. The retention is likely modulated by algal excretions enhancing flocculation of SPM.

The ice-core record shows a stabilisation of atmospheric CO₂ in the 1940s, despite continued emissions from fossil fuel burning and land-use change (LUC). We use up-to-date reconstructions of the CO₂ sources and sinks over the 20th century to evaluate whether these capture the CO₂ plateau and to test the previously proposed hypothesis. Both strong terrestrial sink, possibly due to LUC not fully accounted for in the records, and enhanced oceanic uptake are necessary to explain this stall.

The Arctic contains a large pool of carbon that is vulnerable to warming and can be released by rivers and coastal erosion. We study microbial lipids (BHPs) in permafrost and shelf sediments to trace the source, transport and fate of this carbon. BHPs in permafrost deposits are released to the shelf by rivers and coastal erosion, in contrast to other microbial lipids (GDGTs) that are transported by rivers. Several further analyses are needed to understand the complex East Siberian Shelf system.

We explored the relationship between organic carbon and mud (i.e. silt and clay) contents in seagrass ecosystems to address whether mud can be used to predict soil C content, thereby enabling robust scaling up exercises at a low cost as part of blue carbon stock assessments. We show that mud is not a universal proxy for blue carbon content in seagrass ecosystems, but it can be used to estimate soil Corg content when low biomass seagrass species (i.e. Zostera, Halodule and Halophila) are present.

We address the imbalance between the supply and loss of organic carbon to the upper layer of the ocean by measuring a previously poorly quantified term: particle-associated microbial respiration of in situ particles. We find rates that are too low to account for the missing sink of carbon and suggest instead that zooplankton drive the transformation of large fast-sinking particles into slow-sinking and suspended particles. This apparent loss may help explain imbalances in the carbon budget.

We identified biogeochemical plant–soil feedback processes that occur when oxidation, drying and modification by plants alter sediment conditions. Wetland construction in Markermeer (a lake in the Netherlands) is used as a case study. Natural processes will be utilized during and after construction to accelerate ecosystem development. We conducted a 6-month greenhouse experiment to identify the key biogeochemical processes in the mud when Phragmites australis is used as an eco-engineer.

Phytoplankton spring bloom phenology was derived from a 15-year time series (2000–2014) of ship-of-opportunity chlorophyll a fluorescence observations in the Baltic Sea. Bloom peak concentrations have declined over the study period, while bloom duration has increased. It is concluded that nutrient reduction efforts led to decreasing bloom intensity, while changes in Baltic Sea environmental conditions associated with global change corresponded to a lengthening spring bloom period.

This study demonstrates that intensively managed agricultural grasslands can sequester significant amounts of carbon in soils, especially under long-term applications of cattle slurries. There is also evidence that soil carbon sequestration can contribute to offset between 9 % and 25 % of total greenhouse gas emissions associated with the management of permanent grassland. However, to make livestock-based grassland systems more sustainable, further carbon-gain efficiencies are required.

Previously identified positive feedbacks within the nitrogen cycle in the eastern tropical South Pacific (ETSP) have challenged our understanding of the observed dynamics and stability of the nitrogen inventory. We present a box model analysis of the biological and biogeochemical relations in the ETSP among nitrogen deposition, benthic denitrification and phosphate regeneration. Our results suggest dominant stabilizing feedbacks tending to keep a balanced nitrogen inventory in the ETSP.

Thawing permafrost may release large amounts of terrestrial organic carbon (TerrOC) to the Arctic Ocean. We assessed its fate in the marine environment with a suite of biomarkers. Across the Laptev Sea their concentrations in surface sediments decreased significantly and showed a trend to qualitatively more degraded TerrOC with increasing water depth. We infer that the degree of degradation of TerrOC is a function of the time spent under oxic conditions during protracted cross-shelf transport.

We extend the Community Land Model coupled with carbon and nitrogen (CLM-CN) decomposition cascade to include simple organic substrate turnover, fermentation, Fe(III) reduction, and methanogenesis reactions, and assess the efficacy of various temperature and pH response functions. Incorporating the Windermere Humic Aqueous Model (WHAM) describes the observed pH evolution. Fe reduction can increase pH toward neutral pH to facilitate methanogenesis.

Wetlands are the largest global natural methane source. Peat-rich bogs and fens lying between 50°N and 70°N contribute 10–30% to this source. The predictive capability of the seasonal methane cycle can directly affect the estimation of global methane budget. We present multiscale methane seasonal emission by observations and modeling and find that the uncertainties in predicting the seasonal methane emissions are from the wetland extent, cold-season CH₄ production and CH₄ transport processes.

Soil organic carbon can be used for climate reconstruction and carbon cycle study. This work assessed the effect of temperature on surface soil δ¹³C by an investigation from 27 sites across a temperature gradient along an isohyet. This study shows that soil type and vegetation type had influences on soil δ¹³C, while temperature was not related to soil δ¹³C. This suggests that organic carbon isotopes in sediments cannot be used for the paleotemperature reconstruction.

Ocean carbonate observations from surface buoys reveal that marine life is currently exposed to conditions outside preindustrial bounds at 12 study locations around the world. Seasonal conditions in the California Current Ecosystem and Gulf of Maine also exceed thresholds that may impact shellfish larvae. High-resolution observations place long-term change in the context of large natural variability: a necessary step to understand ocean acidification impacts under real-world conditions.

Australian vegetation phenology is highly variable due to the diversity of ecosystems on the continent. We explore continental-scale variability using satellite remote sensing by broadly classifying areas as seasonal, non-seasonal, or irregularly seasonal. We also examine ecosystem-scale phenology using phenocams and show that some broadly non-seasonal ecosystems do display phenological variability. Overall, phenocams are useful for understanding ecosystem-scale Australian vegetation phenology.

In the context of the VAHINE mesocosm experiment in the Nouméa lagoon (New Caledonia), a 1-D vertical biogeochemical mechanistic model was used together with the in situ experiment to complement our comprehension of the planktonic ecosystem dynamics and the main biogeochemical carbon, nitrogen and phosphate fluxes. The model also showed the fate of fixed N₂ by providing, over time, the proportion of diazotroph-derived nitrogen (DDN) in each compartment (mineral and organic) of the model.

The increasing seasonality of atmospheric CO₂ is strongly linked with enhanced land vegetation activities in the last 5 decades, for which the importance of increasing CO₂, climate and land use/cover change was evaluated in single model studies (Zeng et al., 2014; Forkel et al., 2016). Here we examine the relative importance of these factors in multiple models. Our results highlight models can show similar results in some benchmarks with different underlying regional dynamics.