Marine ecosystems influence atmospheric chemistry and climate by exchanging volatile organic compounds (VOCs) with the atmosphere, yet the chemical diversity and the environmental controls of the exchanged compounds remain poorly constrained at ecosystem timescales. This gap persists because most marine observations are targeted, short-term, and narrow-suite measurements that seldom deliver chemically broad, continuous fluxes. Here we show that marine VOC exchange is chemically diverse and changes in magnitude and composition on event‑to‑season timescales. Using continuous flux measurements of 48 compounds in the Baltic Sea, we observe a pronounced reorganization of net exchange: early organosulfur emissions diminish as cyanobacterial biomass declines (indicating water-side source limitation), while oxygenated and nitrogen-containing VOCs exhibit mixed behaviors ranging from deposition to episodic emissions (often driven by kinetic factors). Together, these results demonstrate that chemically diverse marine VOC exchange is closely coupled to wind-induced mixing, water‑column structure, and phytoplankton community succession, highlighting the need to resolve these dynamic ecosystem states to predict future atmospheric composition.

The decreasing water transparency of the Baltic Sea over the last century has been documented by in situ measurements of the Secchi disk depth ( z SD ) as well as by satellite measurements over recent decades. While the decrease in transparency for the Baltic Sea in general has become negligible from about 2013 onwards, satellite-detected light attenuation of the violet and blue light ( k d 412) has increased, particularly from 2013 onwards. In the Gulf of Bothnia (particularly in the Bothnian Bay), eastern Gulf of Finland and eastern Gulf of Riga, k d 412 has increased significantly, indicating a dramatic increase in the concentration of colored dissolved organic matter (CDOM). These findings are partly supported by in situ CDOM data. The increase in CDOM concentration is probably due to climate change induced shifts in hydrology and greening of the land, all related to global warming. The increase in CDOM concentration and its effect on light availability may cause significant ecological changes and deterioration of the aquatic food chains.

Summer filamentous cyanobacterial blooms strongly influence the biogeochemistry of the Baltic Sea. In surface water, they sustain organic matter (OM) production via N2 fixation despite the depletion of dissolved inorganic nitrogen (DIN). In deeper water, remineralization of this OM contributes to O2 depletion and H2S formation, thereby promoting Baltic Sea hypoxia/anoxia. Primary production by cyanobacteria also drives a characteristic summer drawdown of total dissolved CO2 (CT). While this biogeochemical signal is thought to reflect N2 fixation, direct observational evidence supporting this link remains limited. In addition, key questions remain regarding the factors regulating bloom initiation and intensity.This study applied a multi-method approach, combining continuous high-resolution surface water measurements of biogeochemical and biological variables made from the ship of opportunity Finnmaid traveling between Travemünde (Germany) and Helsinki (Finland), with modeled physical parameters, to investigate a cyanobacterial hotspot identified in summer 2023. During the 2.5-week CT drawdown (∼1500 to ∼1325 μmol/kg) in June, at the entrance to the Gulf of Finland, 70% of the decrease occurred in the final five days and coincided with a sharp increase in filamentous cyanobacterial biomass and pronounced N2 undersaturation. Our observations indicate that once the cyanobacterial biomass has been accumulated during favorable environmental conditions, their impacts can be drastic to the carbon system. Our results link N2 fixation with the CT decline in surface water, while also demonstrating that ecosystem productivity is enhanced by co-occurring phytoplankton species, which likely benefit from the nitrogen released by cyanobacterial N2 fixation.

Phytoplankton undertake daily vertical migration through the water column to optimize light and nutrient access while avoiding predators. However, diel vertical migration (DVM) patterns remain poorly characterized for many taxa due to limitations of labor-intensive traditional microscopy. Here, we employed high-throughput in situ imaging flow cytometry to investigate DVM. An Imaging FlowCytobot (IFCB) was deployed to continuously profile the vertical water column for ~10 weeks (August-October 2016) at a location in the Skagerrak, eastern North Sea. This revealed significant DVM for several morpho-taxonomic groups, including taxa belonging to ciliates, dinoflagellates, and diatoms, shifting median depth by 2-6 m between night and day. The analysis also revealed that DVM can be inferred from diel pulses in surface water biomass, which we leveraged to study DVM in an extensive IFCB time-series dataset from the central Baltic Sea (June-October in 2020 and 2021). Migratory taxa accounted for 77% and 79% of total phytoplankton biomass (size range <10-150 μm) in the Skagerrak and Baltic Sea, respectively, underscoring the ecological significance of DVM. Most populations peaked near the surface at midday, although other patterns were also observed. While many taxa displayed consistent migration behaviors across both regions, others differed-likely due to population-specific traits or local environmental conditions. Seasonal changes in migration patterns suggest a role for community turnover and shifting environmental conditions. This study highlights the prevalence of DVM in phytoplankton and showcases the power of automated, high-throughput imaging technologies to advance our understanding of plankton ecology.

Long-term nutrient loading and warmer, longer summer temperatures have promoted summer cyanobacteria-dominated phytoplankton blooms in the Baltic Sea, shifting the annual chlorophyll maximum toward peak summer. In turn, organic matter production is increasing, altering the carbon cycle by shifting the bioavailable carbon pool to later in the season and towards microbial heterotrophy. These ecosystem changes may have consequential impacts on the production of trace gases, such as volatile organic compounds (VOC). Enhanced stratification and reduced vertical mixing may further regulate VOC water-air exchange. In the coastal zone, significant changes to macroalgae communities have been observed in association with persistent eutrophication. Shifting coastal dynamics, along with increased warming and, consequently, increased decomposition of organic material, will likely impact VOC production. Therefore, the aim of this study is to evaluate the influence of a summertime phytoplankton bloom on the composition and concentrations of VOCs in seawater, and to examine differences between distinct coastal habitats.Summer sampling was conducted on Utö Island (59º 46'50N, 21º 22'23E; Archipelago Sea), and samples were processed at the Utö Atmospheric and Marine Research Station. Seawater VOCs were collected using the purge and trap method four times across three habitat types along the open coast—open water (250 m off shore; 4.5 m depth), a cove (15 m off shore; 0.5 m depth), and a vegetated beach (on shore; surface). Samples were stored in stainless steel absorbent cartridges and analyzed with Thermal Desorption Gas Chromatography Mass Spectrometry. Phytoplankton community composition and abundance were captured using an Imaging FlowCytobot, complemented by bacterial abundance from flow cytometry and microscopy.Preliminary results indicate clear temporal variability in open water VOC concentrations. Some compounds such as isoprene were persistently detected throughout the summer whereas other compounds, e.g. toluene and dimethyl disulfide, varied across the season in association with changes in phytoplankton and bacterial abundance. Taxa-specific links between VOCs and phytoplankton composition, as well as the potential influence of abiotic drivers, including dissolved organic matter and vertical mixing, is still under investigation. Further analysis indicates that VOC concentrations are highly dependent on coastal habitat type, with composition and concentration of VOCs from the vegetated beach showing approximately 10-fold higher values as well as a more unique VOC blend, suggesting contributions from macroalgae and sediment processes. In contrast, the cove was highly dominated by bromoform, comprising >50% of the measured proportional VOC signal throughout the summer.

Phytoplankton, as a primary producer and first level of food webs, is an important component of all aquatic ecosystems. Phytoplankton communities are very dynamic having fast turnover rates, from half a day to a few days, which makes monitoring of the phytoplankton communities challenging. Classically phytoplankton communities are followed with manual sampling and light microscopy counts. Although providing detailed taxonomic information the method is labor intensive restricting the frequency of data collected. Long-term monitoring is often executed with only a few samples per year targeting different seasons, even weekly sampling monitoring stations being rare. Alternative methods to follow the phytoplankton community composition and succession traditionally include optical fluorescence detection of different phytoplankton pigments, such as chlorophyll a, phycocyanin, and phycoerythrin, and satellite remote sensing. Fluorometers provide a cost-effective solution to follow the community in high temporal resolution, but the taxonomic composition stays unresolved. Satellite remote sensing allows monitoring of large spatial scales, but similarly, lacks taxonomic information and relies on clear skies. Within recent decades novel methods, such as eDNA and pulse shape and imaging flow cytometry, have been emerging providing more detailed information on the phytoplankton communities. While eDNA cannot be yet used in high temporal resolution, high-frequency in situ imaging and cytometry provide autonomous and automated alternatives to follow the community composition and dynamics, however, their downside is the slightly lower taxonomic resolution. The long-term monitoring programs and data series collected using light microscopy methods are the key reference to understand the shifts in phytoplankton communities in the changing climate. The rapid uptake of the novel methods allows collection of more fine-grained datasets, but demands an understanding of how they compare to the existing methods. Here we present the results of the comparison of different methods, with an emphasis on the in situ instrument Imaging FlowCytobot. The results suggest that in situ imaging is an excellent addition to monitoring phytoplankton communities and to studies for enhancing our understanding of community dynamics and succession in relation to environmental forcing. However, novel methods should be adopted as an additional way of collecting information and used in parallel with ongoing monitoring activities to understand how the different methods describe the communities also in the long run.

Introduction Phytoplankton are microscopic organisms that form the foundation of aquatic food webs. Accurate identification and classification of phytoplankton species are crucial for monitoring all aquatic ecosystems, from marine to freshwater, understanding ecological dynamics, and assessing environmental changes. Traditional methods of phytoplankton identification, which rely on manual microscopy, are time-consuming and require expert knowledge. Recent advancements in machine learning, particularly Convolutional Neural Networks (CNNs), offer promising solutions for automating this process. This abstract explores the application of pre-trained CNNs in recognizing phytoplankton species, highlighting their advantages, methodologies, and potential impacts. Methodology We present three approaches from a marine site, the Gulf of Venice site of the LTER-Italy network (DEIMS.ID https://deims.org/758087d7-231f-4f07-bd7e-6922e0c283fd), which includes the 'Acqua Alta' Oceanographic Tower (AAOT) (Fig. 1), the brackishwater site Utö Atmospheric and Marine Research Station (ResNet-18, located at 59°46.84’ N, 21°22.13’ E) https://en.ilmatieteenlaitos.fi/uto, and the freshwater site the IGB-LakeLab in Lake Stechlin NE Germany (DEIMS.ID https://deims.org/2223bc9c-12b2-49fe-af73-4299f553e054). Three different architectures of CNN were used: VGG16 for the Gulf of Venice, ResNet-18 for the Finnish station and a YOLOv11-cls for the German Lake Stechlin LakeLab station. These CNN models were pre-trained on the ImageNet dataset and subsequently fine-tuned with specific datasets for the respective geographic areas. These CNNs were chosen for their ability to autonomously extract features from images without external assistance, making them efficient, fast tools for analyzing large amounts of data and due to their specificity regarding the characteristics of the observational site. The process involves several steps: Data Collection and Preprocessing : several public datasets are available (Ciranni et al. 2024), where each image is annotated according to its class. Each model is structured to require input images in a specific format, so depending on the chosen model, it is necessary to preprocess the images accordingly. With an Imaging Flow Cytobot (IFCB, an in-situ automated submersible imaging flow cytometer that generates images of particles in-flow taken from the aquatic environment.), the produced images are of good quality (Fig. 2), and the main modification applied is resizing the images to fit the model requirements; Transfer Learning : Transfer learning allows the weights of a pre-trained neural network to be retained and updated (only if specified) for specific tasks. It has been demonstrated that using pre-trained models leads to significant results, reducing both training time and the amount of data required compared to an untrained model (Maracani et al. 2023); Training and Validation : The modified CNN is trained on the annotated phytoplankton images. Techniques such as data augmentation (to increment the number of images), dropout, and batch normalization are employed to enhance model performance and prevent overfitting. The model's accuracy is validated using a separate dataset; Evaluation Metrics : Performance metrics, including accuracy, precision, recall, and F1-score, are used to evaluate the model. Confusion matrices and receiver operating characteristic (ROC) curves provide additional insights into the model's classification capabilities. Data Collection and Preprocessing : several public datasets are available (Ciranni et al. 2024), where each image is annotated according to its class. Each model is structured to require input images in a specific format, so depending on the chosen model, it is necessary to preprocess the images accordingly. With an Imaging Flow Cytobot (IFCB, an in-situ automated submersible imaging flow cytometer that generates images of particles in-flow taken from the aquatic environment.), the produced images are of good quality (Fig. 2), and the main modification applied is resizing the images to fit the model requirements; Transfer Learning : Transfer learning allows the weights of a pre-trained neural network to be retained and updated (only if specified) for specific tasks. It has been demonstrated that using pre-trained models leads to significant results, reducing both training time and the amount of data required compared to an untrained model (Maracani et al. 2023); Training and Validation : The modified CNN is trained on the annotated phytoplankton images. Techniques such as data augmentation (to increment the number of images), dropout, and batch normalization are employed to enhance model performance and prevent overfitting. The model's accuracy is validated using a separate dataset; Evaluation Metrics : Performance metrics, including accuracy, precision, recall, and F1-score, are used to evaluate the model. Confusion matrices and receiver operating characteristic (ROC) curves provide additional insights into the model's classification capabilities. Results Studies have demonstrated that pre-trained CNNs can achieve high accuracy in phytoplankton classification. In our case, models like ResNet and VGG have shown classification accuracies exceeding 80% on diverse phytoplankton datasets (Fig. 3, Kraft et al. 2022). These models effectively distinguish between species with subtle morphological differences, which are often challenging for human experts. Discussion The use of pre-trained CNNs in phytoplankton recognition offers several advantages: Efficiency : Automated classification significantly reduces the time and effort required for phytoplankton identification compared to manual methods. Scalability : CNNs can handle large volumes of image data, making them suitable for Long Term Ecological Research. Consistency : Machine learning models provide consistent and objective classifications, minimizing human error and variability. Efficiency : Automated classification significantly reduces the time and effort required for phytoplankton identification compared to manual methods. Scalability : CNNs can handle large volumes of image data, making them suitable for Long Term Ecological Research. Consistency : Machine learning models provide consistent and objective classifications, minimizing human error and variability. However, challenges remain. The automatic taxonomic identification level is still not as detailed as that of human expertise. The quality and diversity of training data are critical for model performance. Inadequate or biased datasets can lead to poor generalization. Additionally, the interpretability of CNNs is limited, making it difficult to understand the decision-making process fully. Conclusion Pretrained CNNs represent a powerful tool and a pipeline for phytoplankton species recognition, offering significant improvements in efficiency, scalability, and consistency over traditional methods. Continued advancements in machine learning and the availability of high-quality datasets will further enhance the capabilities of these models. Future research should focus on addressing current limitations, such as data quality and model interpretability, to fully realize the potential of CNNs in marine science. In this work, we will present the results as discussed to demonstrate possible workflows to fully realize the potential of CNNs in marine science and potentially contribute to the Standard Observations (SOs) addressing current limitations. We will also bring a workflow proposal to manage and perform actions related to harmonization, interoperability, quality control and sharing of the data obtained througth the CNNs recognitions following the directives proposed by Torstensson (2025).

Europe is one of the most studied areas related to biogenic volatile organic compound (BVOC) emissions. However, our knowledge of these atmospheric reactive compounds is still quite limited even there. Total hydroxyl radical (OH) reactivity studies indicate that half of the atmospheric reactive compounds are still unknown especially in the forested areas (Yang et al. 2016) and OH and ozone reactivity studies of our group have shown high fractions of reactivity from biogenic emissions (Praplan et al. 2020 and Thomas et al. 2023).Globally, isoprene is the primary emitted BVOC. While boreal forests in Northern Europe are mainly considered as monoterpene emitters, Central Europe is expected to be dominated by isoprene (e.g. Messina et al. 2016). However, our results from a campaign at 17 stations over Europe in summer 2022 indicated that BVOC mixing ratios are highly variable and some areas also in Central Europe may be dominated by monoterpenes.Sesquiterpenes and diterpenes have very high potential for secondary organic aerosol formation, but much less is known on their emissions and atmospheric concentrations. Our studies show that birches and spruces may be strong sesquiterpene emitters. We have also found that some urban trees in Montreal and wetlands in Lapland known as isoprene emitters may also release significant amounts of sesquiterpenes. Additionally, forest floor represents a potential source of sesquiterpenes.Compared to terrestrial sources very little is known on the marine emissions of BVOCs. There are studies on dimethyl sulphide, but our recent results on an island in Baltic Sea suggest that other sulphuric compounds, like methanethiol, may be important too and could have strong impacts on SO2 production and therefore also on new particle and cloud formation. Furthermore, our recent campaign at the coast of Baltic Sea indicates that phytoplankton and macrophytes could be a source of isoprene and monoterpenes (Thakur et al., 2024 publication under prep).Compounds classified as BVOCs (e.g. monoterpenes) can also be emitted from anthropogenic sources, such as construction sites (e.g. from wooden material), as well as cleaning and personal care products. Our studies in a street canyon in Helsinki in 2022 indicates that they strongly impact local atmospheric chemistry even in wintertime. Messina, P., Lathière, J., Sindelarova, K., Vuichard, N., Granier, C., Ghattas, J., Cozic, A., and Hauglustaine, D. A.: Global biogenic volatile organic compound emissions in the ORCHIDEE and MEGAN models and sensitivity to key parameters, Atmos. Chem. Phys., 16, 14169–14202, https://doi.org/10.5194/acp-16-14169-2016, 2016Praplan, A. P., Tykkä, T., Schallhart, S., Tarvainen, V., Bäck, J., and Hellén, H.: OH reactivity from the emissions of different tree species: investigating the missing reactivity in a boreal forest, Biogeosciences, 17, 4681–4705, https://doi.org/10.5194/bg-17-4681-2020, 2020.Thomas, S. J., Tykkä, T., Hellén, H., Bianchi, F., and Praplan, A. P.: Undetected biogenic volatile organic compounds from Norway spruce drive total ozone reactivity measurements, Atmos. Chem. Phys., 23, 14627–14642, https://doi.org/10.5194/acp-23-14627-2023, 2023.Yang, Y., Shao, M., Wang, X., Nölscher, A. C., Kessel, S., Guenther, A., and Williams, J.: Towards a quantitative understanding of total OH reactivity: A review, Atmos. Environ., 134, 147–161, https://doi.org/10.1016/j.atmosenv.2016.03.010, 2016.

The impact of chromophoric dissolved organic matter (CDOM) on the spectral underwater light field within photosynthetically active radiation (PAR), and the related responses of spectral phytoplankton light absorption are poorly documented in the mixed layer of lakes. We examined how CDOM influences the spectrum of lake optical properties, underwater light field, photons absorbed by phytoplankton, photoacclimation, and chromatic adaptation of phytoplankton in 127 boreal lakes. In lakes with increasing CDOM content, light intensity decreased steeply with depth, but mixed layer depth and mean light intensity of the mixed layer decreased only moderately. Increasing CDOM content red‐shifted the central wavelengths of lake optical properties, underwater light fields, and the photons absorbed by phytoplankton. In study lakes with increasing CDOM content, the highest light availability shifted stepwise from 580 to 650 nm and close to 700 nm. The ratio of chlorophyll a concentration to phytoplankton biomass decreased slightly in lakes with higher CDOM content. The absorption coefficient of phytoplankton at blue relative to red wavelengths decreased with increasing CDOM content, indicating decreased blue‐absorbing pigmentation in response to decreased availability of blue light. Surprisingly, the number of red photons (600–700 nm) absorbed by phytoplankton increased with CDOM content below the mid‐point of the euphotic zone. Over the entire water column, red light (600–700 nm) accounted for > 50% of PAR absorption by phytoplankton in 69% of the lakes. The high contribution of red photons to the absorption of PAR by phytoplankton may have photobiological consequences, which are poorly understood and require further study.

Several methods to monitor cyanobacteria exist, based partly on characteristics that differentiate cyanobacteria from other phytoplankton, such as pigmentation and morphology. However, it is not certain whether all methods give similar insights into the development and properties of a cyanobacterial bloom. It is important to understand the level of consistency of measurements, especially in the case of novel methods. In situ imaging flow cytometry provides community composition information at high frequency but has been little used for filamentous bloom-forming cyanobacteria. To understand if different methods agree, we compared multi-year biomass data collected with Imaging FlowCytobot (IFCB), CytoSense (CS), phycocyanin (PC) and chlorophyll (Chl) a fluorescence, and turbidity sensors, light microscopy, and satellite-based Frequency of Cyanobacteria Accumulations (FCA). Continuous high-throughput data was recorded at Utö Atmospheric and Marine Research Station in the Baltic Sea during summers 2018-2022, along with samples for light microscopy and adjacent satellite observations. The IFCB cyanobacteria biomass pattern most closely resembled those of CS and PC fluorescence. IFCB also described the blooms similarly to FCA, and to some extent to turbidity, but differed from Chl a fluorescence. IFCB and light microscopy agreed on the bloom development and species composition but differed concerning exact biomass. Our study demonstrates that both IFCB and CS are suitable for studying filamentous cyanobacteria and that in situ imaging flow cytometry provides valuable support for cyanobacteria monitoring by yielding detailed high-frequency taxon-specific information. Still, the best overall coverage of rapid biological processes such as bloom development is achieved with the parallel use of multiple observation techniques.

Oceans alleviate the accumulation of atmospheric CO2 by absorbing approximately a quarter of all anthropogenic emissions. In the deep oceans, carbon uptake is dominated by aquatic phase chemistry, whereas in biologically active coastal seas the marine ecosystem and biogeochemistry play an important role in the carbon uptake. Coastal seas are hotspots of organic and inorganic matter transport between the land and the oceans, and thus they are important for the marine carbon cycling. In this study, we investigate the net air-sea CO2 exchange at the Ut & ouml; Atmospheric and Marine Research Station, located at the southern edge of the Archipelago Sea within the Baltic Sea, using the data collected during 2017-2021. The air-sea fluxes of CO(2 )were measured using the eddy covariance technique, supported by the flux parameterization based on the pCO(2) and wind speed measurements. During the spring-summer months (April-August), the sea was gaining carbon dioxide from the atmosphere, with the highest monthly sink fluxes typically occurring in May, being -0.26 mu mol m-2 s(-1) on average. The sea was releasing the CO(2 )to the atmosphere in September-March, and the highest source fluxes were typically observed in September, being 0.42 mu mol m(-2) s(-1) on average. On an annual basis, the study region was found to be a net source of atmospheric CO2, and on average, the annual net exchange was 27.1 gC m(-1) yr(-1), which is comparable to the exchange observed in the Gulf of Bothnia, the Baltic Sea. The annual net air-sea CO2 exchanges varied between 18.2 (2018) and 39.1 gC m(-1) yr(-1) (2017). During the coldest year, 2017, the spring-summer sink fluxes remained low compared to the other years, as a result of relatively high seawater pCO(2 i)n summer, which never fell below 220 mu atm during that year. The spring-summer phytoplankton blooms of 2017 were weak, possibly due to the cloudy summer and deeply mixed surface layer, which restrained the photosynthetic fixation of dissolved inorganic carbon in the surface waters. The algal blooms in spring-summer 2018 and the consequent pCO(2) drawdown were strong, fueled by high pre-spring nutrient concentrations. The systematic positive annual CO(2 )balances suggest that our coastal study site is affected by carbon flows originating from elsewhere, possibly as organic carbon, which is remineralized and released to the atmosphere as CO2. This coastal source of CO2 fueled by the organic matter originating probably from land ecosystems stresses the importance of understanding the carbon cycling in the land-sea continuum.

Plankton recognition provides novel possibilities to study various environmental aspects and an interesting real-world context to develop domain adaptation (DA) methods. Different imaging instruments cause domain shift between datasets hampering the development of general plankton recognition methods. A promising remedy for this is DA allowing to adapt a model trained on one instrument to other instruments. In this paper, we present a new DA dataset called DAPlankton which consists of phytoplankton images obtained with different instruments. Phytoplankton provides a challenging DA problem due to the fine-grained nature of the task and high class imbalance in real-world datasets. DAPlankton consists of two subsets. DAPlankton_LAB contains images of cultured phytoplankton providing a balanced dataset with minimal label uncertainty. DAPlankton_SEA consists of images collected from the Baltic Sea providing challenging real-world data with large intra-class variance and class imbalance. We further present a benchmark comparison of three widely used DA methods.

Planktonic organisms including phyto-, zoo-, and mixoplankton are key components of aquatic ecosystems and respond quickly to changes in the environment, therefore their monitoring is vital to follow and understand these changes. Advances in imaging technology have enabled novel possibilities to study plankton populations, but the manual classification of images is time consuming and expert-based, making such an approach unsuitable for large-scale application and urging for automatic solutions for the analysis, especially recognizing the plankton species from images. Despite the extensive research done on automatic plankton recognition, the latest cutting-edge methods have not been widely adopted for operational use. In this paper, a comprehensive survey on existing solutions for automatic plankton recognition is presented. First, we identify the most notable challenges that make the development of plankton recognition systems difficult and restrict the deployment of these systems for operational use. Then, we provide a detailed description of solutions found in plankton recognition literature. Finally, we propose a workflow to identify the specific challenges in new datasets and the recommended approaches to address them. Many important challenges remain unsolved including the following: (1) the domain shift between the datasets hindering the development of an imaging instrument independent plankton recognition system, (2) the difficulty to identify and process the images of previously unseen classes and non-plankton particles, and (3) the uncertainty in expert annotations that affects the training of the machine learning models. To build harmonized instrument and location agnostic methods for operational purposes these challenges should be addressed in future research.

The mixing ratios of highly volatile organic compounds (VOCs) were studied on Utö Island in the Baltic Sea. Measurements of non-methane hydrocarbons (NMHCs) and methanethiol (unexpectedly found during the experiment) were conducted using an in situ thermal desorption–gas chromatography–flame ionization detector/mass spectrometer (TD-GC-FID/MS) from March 2018 until March 2019. The mean mixing ratios of NMHCs (alkanes, alkenes, alkynes, and aromatic hydrocarbons) were at the typical levels for rural/remote sites in Europe, and, as expected, the highest mixing ratios were measured in winter, while in the summertime, the mixing ratios remained close to or below detection limits for most of the studied compounds. Sources of NMHCs during wintertime were studied using positive matrix factorization (PMF) together with wind direction analyses and source area estimates. Shipping was found to be a major local anthropogenic source of NMHCs with a 21 % contribution. It especially contributed to ethene, propene, and ethyne mixing ratios. Other identified sources were petrol fuel (15 %), traffic exhaust (14 %), local solvents (6 %), and long-range-transported background (42 %). Contrary to NMHCs, high mixing ratios of methanethiol were detected in summertime (July mean of 1000 pptv). The mixing ratios followed the variations in seawater temperatures and sea level height and were highest during the daytime. Biogenic phytoplankton or macroalgae emissions were expected to be the main source of methanethiol.

Accurate and traceable measurements are required to understand ocean processes, to address pressing societal challenges, such as climate change and to sustainably manage marine resources. Although scientific and engineering research has resulted in advanced methods to measure Essential Ocean Variables (EOVs) there is a need for cross comparison of the techniques and traceability to recognized standards. Metrological laboratories are experienced in accredited methods and assessment of methodology. An EU INFRAIA-02-2020: Integrating Activities for Starting Communities project MINKE (Metrology for Integrated marine maNagement and Knowledge-transfer nEtwork https://minke.eu) brings European marine science and metrology Research Infrastructures together to identify synergies and create an innovative approach to Quality Assurance of oceanographic data. Quality depends both on the accuracy (that can be provided through the metrology component) and the completeness of the data sets. The collaboration between different Marine Research Infrastructures (RIs) places a fundamental role on assuring the completeness of the datasets, particularly at global scales. The MINKE project encourages enhancement through collaboration of national metrology laboratories and the oceanographic community. Metrological assessment of the accuracy and uncertainties within multidisciplinary ocean observations will provide data that are key to delivering policy information. Objectives across all the RIs are to facilitate ocean observation and build wider synergies. MINKE will investigate these synergies, then introduce metrology to the core of various EOV measurements. Currently the marine RIs cover laboratory and field operations, from the surface seafloor, coastal waters to deep sea, fixed ocean stations to ship and autonomous vehicle operations to ships of opportunity, and flux stations focusing on carbonate system variables. The nexus of these operations is the focal point for coordinated improvement of ocean observing methods. Measurement intercomparisons, traceability and uncertainty assessments should be at the core of the scientific observations. Specifically, MINKE will work with RIs and Metrology Institutes to improve the quality of dissolved oxygen, carbonate system, chlorophyll-fluorescence, ocean sound and current meter measurements, through access to metrology laboratories, Transnational Access and intercomparison studies across existing marine consortia and RIs. MINKE will also promote the development of absolute salinity observation, and improvements in marine litter measurements.

The depth of the mixed layer is a major determinant of nutrient and light availability for phytoplankton in stratified waterbodies. Ongoing climate change influences surface waters through meteorological forcing, which modifies the physical structure of fresh waters including the mixed layer, but effects on phytoplankton biomass are poorly known. To determine the responses of phytoplankton biomass to the depth of the mixed layer, light availability and associated meteorological forcing, we followed daily changes in weather and water column properties in a boreal lake over the first half of a summer stratification period. Phytoplankton biomass increased with the deepening of the mixed layer associated with high wind speeds and low air temperature relative to the temperature of the mixed layer (T-air-T-mix < 0), whereas heatwave conditions-shallow mixed layer driven by high T-air-T-mix value and low wind speed-reduced the biomass. Improving light availability from low to moderate light conditions increased the phytoplankton biomass, while the highest light availability was associated with low phytoplankton biomass. Our study demonstrates that the climatic impact-drivers wind speed and T-air-T-mix are major drivers of mixed layer depth, which controlled phytoplankton biomass during the early summer stratification period. Our study suggests that increasing air temperature relative to water temperature and declining wind speeds have potential to lead to reduced phytoplankton biomass due to a shallower mixed layer during the first half of the stratification period in non-eutrophic lakes with sufficient light availability.

Climate change is projected to cause brownification of some coastal seas due to increased runoff of terrestrially derived organic matter. We carried out a mesocosm experiment (15 d) to test the effect of this on the planktonic ecosystem expecting reduced primary production and shifts in the phytoplankton community composition. The experiment was set up in 2.2 m3 mesocosm bags using four treatments, each with three replicates: control (Contr) without any manipulation, organic carbon additive HuminFeed (Hum; 2 mg L-1), inorganic nutrients (Nutr; 5.7 μM NH4 and 0.65 μM PO4), and combined Nutr and Hum (Nutr + Hum) additions. Measured variables included organic and inorganic nutrient pools, chlorophyll a (Chla), primary and bacterial production and particle counts by flow cytometry. The bags with added inorganic nutrients developed a phytoplankton bloom that depleted inorganic N at day 6, followed by a rapid decline in Chla. Brownification did not reduce primary production at the tested concentration. Bacterial production was lowest in the Contr, but similar in the three treatments receiving additions likely due to increased carbon available for heterotrophic bacteria. Picoeukaryotes clearly benefited by brownification after inorganic N depletion, which could be due to more effective nutrient recycling, nutrient affinity, light absorption, or alternatively lower grazing pressure. In conclusion, brownification shifted the phytoplankton community composition towards smaller species with potential effects on carbon fluxes, such as sinking rates and export to the sea floor.

Climate change is projected to cause brownification of some coastal seas due to increased runoff of terrestrially derived organic matter. We carried out a mesocosm experiment over 15 days to test the effect of this on the planktonic ecosystem. The experiment was set up in 2.2 m3 plastic bags moored outside the Tvärminne Zoological Station at the SW coast of Finland. We used four treatments, each with three replicates: control (Contr) without any manipulation; addition of a commercially available organic carbon additive called HuminFeed (Hum; 2 mg L−1); addition of inorganic nutrients (Nutr; 5.7 µM NH4 and 0.65µM PO4); and a final treatment of combined Nutr and Hum (Nutr+Hum) additions. Water samples were taken daily, and measured variables included water transparency, organic and inorganic nutrient pools, chlorophyll a (Chla), primary and bacterial production and particle counts by flow cytometry.

Humic substances, a component of terrestrial dissolved organic matter (tDOM), contribute to dissolved organic matter (DOM) and chromophoric DOM (CDOM) in coastal waters, and have significant impacts on biogeochemistry. There are concerns in recent years over browning effects in surface waters due to increasing tDOM inputs, and their negative impacts on aquatic ecosystems, but relatively little work has been published on estuaries and coastal waters. Photodegradation could be a significant sink for tDOM in coastal environments, but the rates and efficiencies are poorly constrained. We conducted large‐scale DOM photodegradation experiments in mesocosms amended with humic substances and nutrients in the Gulf of Finland to investigate the potential of photochemistry to remove added tDOM and the interactions of DOM photochemistry with eutrophication. The added tDOM was photodegraded rapidly, as CDOM absorption decreased and spectral slopes increased with increasing photons absorbed in laboratory experiments. The in situ DOM optical properties became similar among the control, humic‐ and humic+nutrients‐amended mesocosm samples toward the end of the amendment experiment, indicating degradation of the excess CDOM/DOM through processes including photodegradation. Nutrient additions did not significantly influence the effects of added humic substances on CDOM optical property changes, but induced changes in DOM removal.