Caves are natural archives for reconstructing past climates and environments. A 480 cm sediment profile was dated at 74.7 ± 12.3 to 56 ± 8 ka, with some sections influenced by Last Glacial Maximum flooding. Bacterial diversity shifts with depth, from soil bacteria in upper layers to thermophilic sulfur bacteria in deeper sediments, possibly linked to historic hot springs or sapropels. This study highlights the potential of bacteria as proxies for reconstructing paleoenvironmental conditions.

The open ocean helps mitigate climate change by storing CO₂ via the biological carbon pump (BCP), which involves processes like organic carbon production at the surface and transferring it to the deep ocean via various pathways. By deploying an autonomous platform, we found significant marine snow accumulation from the surface to the mesopelagic zone in frontal regions between eddies. We suggest that the coupling of hydrodynamics at eddy edges and biological activity may enhance this process.

Research on the influences of dynamic pH on the marine ecosystem is still in its infancy. We manipulated the culturing pH to simulate pH fluctuation and found lower pH could increase eicosapentaenoic acid and docosahexaenoic acid production with unaltered growth and photosynthesis in two marine diatoms. It is important to consider pH variation for more accurate predictions regarding the consequences of acidification in coastal waters.

We introduce argon saturation as a prognostic variable in a suite of coupled general ocean circulation biogeochemical models off Mauritania. Our results indicate that the effect of increasing the spatial horizontal model resolutions from 12 km to 1.5 km leads to changes comparable to other infamous spurious effects of state-of-the-art numerical advection numerics.

This study examines the effects of geochemistry and microbial communities on crustacean assemblages in a volcanic aquifer. It uncovers that major ions, trace elements, and microbial traits dictate the composition and functionality of crustacean assemblages, revealing variations across adjacent groundwater bodies. The study highlights the importance of approaches that integrate geochemical, microbial, and biological indicators for understanding the dynamics of groundwater ecosystems. 

Using comparative metagenomics and metatranscriptomics, we examined the functioning of airborne microorganisms in clouds and a clear atmosphere. Clouds are atmospheric masses where multiple microbial processes are promoted compared with a clear atmosphere. Overrepresented microbial functions of interest include the processing of chemical compounds, biomass production, and regulation of oxidants. This has implications for biogeochemical cycles and microbial ecology.

The emissions of greenhouse gases (GHGs) carbon dioxide (CO₂), methane (CH₄) and nitrous oxide (N₂O) were measured from a clear-cut peatland forest site. The measurements covered the whole year of 2022, which was the second growing season after the clear-cut. The site was a strong GHG source, and the highest emissions came from CO_2, followed by N₂O and CH₄. A statistical model that included information on different surfaces at the site was developed to unravel surface-type-specific GHG fluxes.

Understanding what determines the establishment of dune-building vegetation could help to better predict coastal dune initiation and development. We monitored the establishment of dune-building grasses and dune initiation in a large field experiment. Our results show that dune initiation takes place during peaks in dune-building grass establishment, which depend on favorable environmental conditions. Our findings can potentially be integrated into beach restoration and management strategies.

Our study showed that stress-related crop response to changing environmental conditions can be detected by monitoring crops using satellite images at the landscape level. This could be useful for farmers to identify when stresses occur. Our results also suggest that satellite imagery can be used to discover soil impacts on crop development at farm fields. The inclusion of soil properties in satellite image analyses could further improve the accuracy of the prediction of drought stress on crops.

This study applied the CoupModel to simulate carbon dynamics and ecohydrology for a restored peatland and evaluated the responses of the simulated carbon fluxes to varying acrotelm thickness and climate. The results show that the CoupModel can simulate the coupled carbon and ecohydrology dynamics for the restored peatland system, and the restored peatland has less resilience in its C-uptake functions than pristine peatlands under a changing climate.

From glacial headwaters to 100 km downstream, we found clear organic matter gradients in Canadian Rocky Mountain rivers. In contrast, microbial communities exhibited overall cohesion, indicating that species dispersal may be an over-riding control on community dynamics in these connected rivers. Identification of glacial-specific microbes suggests that glaciers seed headwater microbial assemblages; these findings show the importance of glacial waters and microbiomes in changing mountain systems.

The effect of winter warming on forest CO₂ fluxes has rarely been investigated. We tested the effect of the warm winter of 2020 on the forest CO₂ fluxes across 14 sites in Europe and found that the net ecosystem productivity (NEP) across most sites declined during the warm winter due to increased respiration fluxes. 

Allometric equations predict tree biomass and are crucial for estimating forest carbon storage, thus assessing the role of forests in climate change mitigation. Usually, these equations are selected based on tree-level predictive performance. However, we evaluated the model performance at plot and forest levels, finding it varies with plot size. This has significant implications for reducing uncertainty in biomass estimates at these levels.

Soil organic carbon models are used to predict how soils affect the concentration of CO₂ in the atmosphere. We show that equifinality – the phenomenon that different parameter values lead to correct overall model outputs, albeit with a different model behaviour – is an important source of model uncertainty. Our results imply that adding more complexity to soil organic carbon models is unlikely to lead to better predictions as long as more data to constrain model parameters are not available.

This study analyzes atmospheric CO₂ concentrations and global carbon budgets simulated by multiple Earth system models, using several types of simulations (CO₂  concentration- and emission-driven experiments). We successfully identified problems with regard to the global carbon budget in each model. We also found urgent issues with regard to land use change CO₂ emissions that should be solved in the latest generation of models.

Nitrogen concentrations in tree tissues (leaves, branches, stems, and roots) are related to photosynthesis, growth, and respiration and thus to vegetation carbon uptake. Our novel database allows us to identify the controls of tree tissue nitrogen concentrations in boreal and temperate forests, such as tree age/size, species, and climate. Changes therein will affect tissue nitrogen concentrations and thus also vegetation carbon uptake.