Abstract. The Earth system model EC-Earth3 for contributions to CMIP6 is documented here, with its flexible coupling framework, major model configurations, a methodology for ensuring the simulations are comparable across different high-performance computing (HPC) systems, and with the physical performance of base configurations over the historical period. The variety of possible configurations and sub-models reflects the broad interests in the EC-Earth community. EC-Earth3 key performance metrics demonstrate physical behavior and biases well within the frame known from recent CMIP models. With improved physical and dynamic features, new Earth system model (ESM) components, community tools, and largely improved physical performance compared to the CMIP5 version, EC-Earth3 represents a clear step forward for the only European community ESM. We demonstrate here that EC-Earth3 is suited for a range of tasks in CMIP6 and beyond.

Abstract. The Earth System Model EC-Earth3 for contributions to CMIP6 is documented here, with its flexible coupling framework, major model configurations, a methodology for ensuring the simulations are comparable across different HPC systems, and with the physical performance of base configurations over the historical period. The variety of possible configurations and sub-models reflects the broad interests in the EC-Earth community. EC-Earth3 key performance metrics demonstrate physical behaviour and biases well within the frame known from recent CMIP models. With improved physical and dynamic features, new ESM components, community tools, and largely improved physical performance compared to the CMIP5 version, EC-Earth3 represents a clear step forward for the only European community ESM. We demonstrate here that EC-Earth3 is suited for a range of tasks in CMIP6 and beyond.

Abstract. The co-variation of key variables with simulated phytoplankton biomass in the Baltic proper has been examined using wavelet analysis and results of a long-term simulation for 1850–2008 with a high-resolution coupled physical–biogeochemical circulation model for the Baltic Sea. By focusing on inter-annual variations, it is possible to track effects acting on decadal timescales such as temperature increase due to climate change as well as changes in nutrient input. The strongest inter-annual coherence indicates that variations in phytoplankton biomass are determined by changes in concentrations of the limiting nutrient. However, after 1950 high nutrient concentrations created a less-nutrient-limited regime, and the coherence was reduced. Furthermore, the inter-annual coherence of mixed-layer nitrate with riverine input of nitrate is much larger than the coherence between mixed-layer phosphate and phosphate loads. This indicates a greater relative importance of the vertical flux of phosphate from the deep layer into the mixed layer. In addition, shifts in nutrient patterns give rise to changes in phytoplankton nutrient limitation. The modelled pattern shifts from purely phosphate limited to a seasonally varying regime. The results further indicate some effect of inter-annual temperature increase on cyanobacteria and flagellates. Changes in mixed-layer depth affect mainly diatoms due to their high sinking velocity, while inter-annual coherence between irradiance and phytoplankton biomass is not found.

Abstract. State-of-the-art global nutrient deposition fields are coupled here to the Pelagic Interactions Scheme for Carbon and Ecosystem Studies (PISCES) biogeochemistry model to investigate their effect on ocean biogeochemistry in the context of atmospheric forcings for pre-industrial, present, and future periods. PISCES, as part of the European Community Earth system model (EC-Earth) model suite, runs in offline mode using prescribed dynamical fields as simulated by the Nucleus for European Modelling of the Ocean (NEMO) ocean model. Present-day atmospheric deposition fluxes of inorganic N, Fe, and P into the global ocean account for ∼ 40 Tg N yr−1, ∼ 0.28 Tg Fe yr−1, and ∼ 0.10 Tg P yr−1. Pre-industrial atmospheric nutrient deposition fluxes are lower compared to the present day (∼ 51 %, ∼ 36 %, and ∼ 40 % for N, Fe, and P, respectively). However, the overall impact on global productivity is low (∼ 3 %) since a large part of marine productivity is driven by nutrients recycled in the upper ocean layer or other local factors. Prominent changes are, nevertheless, found for regional productivity. Reductions of up to 20 % occur in oligotrophic regions such as the subtropical gyres in the Northern Hemisphere under pre-industrial conditions. In the subpolar Pacific, reduced pre-industrial Fe fluxes lead to a substantial decline of siliceous diatom production and subsequent accumulation of Si, P, and N, in the subpolar gyre. Transport of these nutrient-enriched waters leads to strongly elevated production of calcareous nanophytoplankton further south and southeast, where iron no longer limits productivity. The North Pacific is found to be the most sensitive to variations in depositional fluxes, mainly because the water exchange with nutrient-rich polar waters is hampered by land bridges. By contrast, large amounts of unutilized nutrients are advected equatorward in the Southern Ocean and North Atlantic, making these regions less sensitive to external nutrient inputs. Despite the lower aerosol N : P ratios with respect to the Redfield ratio during the pre-industrial period, the nitrogen fixation decreased in the subtropical gyres mainly due to diminished iron supply. Future changes in air pollutants under the Representative Concentration Pathway 8.5 (RCP8.5) emission scenario result in a modest decrease of the atmospheric nutrients inputs into the global ocean compared to the present day (∼ 13 %, ∼ 14 %, and ∼ 20 % for N, Fe, and P, respectively), without significantly affecting the projected primary production in the model. Sensitivity simulations further show that the impact of atmospheric organic nutrients on the global oceanic productivity has turned out roughly as high as the present-day productivity increase since the pre-industrial era when only the inorganic nutrients' supply is considered in the model. On the other hand, variations in atmospheric phosphorus supply have almost no effect on the calculated oceanic productivity.

Long sea level records with high temporal resolution are of paramount importance for future coastal protection and adaptation plans. Here we discuss the application of machine learning techniques to some regression problems commonly encountered when analyzing such time series. The performance of artificial neural networks is compared with that of multiple linear regression models on sea level data from the Swedish coast. The neural networks are found to be superior when local sea level forcing is used together with remote sea level forcing and meteorological forcing, whereas the linear models and the neural networks show similar performance when local sea level forcing is excluded. The overall performance of the machine learning algorithms is good, often surpassing that of the much more computationally costly numerical ocean models used at our institute.