Limnology and Oceanography: Methods最新文献

We developed a novel benthic photorespirometer (BP) to overcome limitations of existing benthic incubation chambers for in situ photosynthesis–irradiance (P–I) curve analysis, including prolonged measurement times, uneven light distribution, and poor sealing. The BP integrates an artificial light source, an oxygen optode sensor, and a neoprene skirt to enable rapid P–I curve construction through 2-h incubations. Field tests at four coastal sites in Korea validated uniform light delivery and thermal stability during incubation, which are essential for reliable measurements. The compact and rigid design allowed single-operator deployment with minimal habitat disturbance, while the neoprene skirt ensured effective sealing across diverse substrates, including uneven rocky bottoms. Single deployments yielded reproducible oxygen responses and community-specific P–I curves under uniform light conditions, reflecting distinct photosynthetic characteristics among benthic algal communities and demonstrating the BP's capability to assess benthic photosynthesis. These results validate the BP as a reliable, field-deployable tool for high-resolution assessment of benthic primary production in coastal ecosystems.

In 2021, the Ocean Thematic Centre of the European Research Infrastructure “Integrated Carbon Observation System” conducted an international partial pressure of carbon dioxide (pCO₂) instrument intercomparison. The goal was to understand how different types of instrumentation for the measurement of ocean pCO₂ compare to each other. During the two-week long experiment, we installed various instruments in a tank facility using natural sea water (North Sea). These included direct air–water equilibration systems and membrane-based flow-through instruments along with submersible sensors and instruments that are normally installed on buoys and autonomous surface vehicles. In situ instruments were installed inside the tank and the flow-through instruments were fed the same water using a pumping system. We changed the temperature (between 10°C and 28°C) and the seawater pCO₂ (between 250 and 800 μatm) to observe instrument responses over a wide range. Since there is no reference for surface ocean pCO₂ measurements, we agreed on a set of instruments serving as intercomparison reference. All data from the different instruments were then compared against the intercomparison reference during periods of stable temperature and pCO₂. The study provides important information to enhance future ocean carbon monitoring networks, but makes no direct recommendation for the use of any specific sensor. A major finding is that equilibration through direct air–water contact appears to be more consistent and independent of external factors than equilibration through a membrane or photometric detection. We found several instruments with no temperature measurements at the location of equilibration or with uncalibrated temperature sensors introducing significant uncertainty in the results.

The coral host comprises various microorganisms including those living in its skeleton. In coral skeletons, bioeroding microflora (cyanobacteria, algae, and fungi), which play an important role in reefs and coral resilience, produce specific traces (microborings) by actively dissolving the carbonate. To date, only a few highly time-consuming methods that rely on the observer allow microborings' study, limiting the number of samples that can be analyzed. Recently, a machine-learning approach based on the analysis of scanning electronic microscope images via a modified convolutional neural network (CNN) was developed to evaluate accurately microborings abundance along a core of a massive Diploastrea sp. from Mayotte. The aim here was to test this CNN on another massive coral species, Porites sp., to verify that it can be applied to other massive corals. We found that the classification accuracy decreased by 8% while the other metrics dropped significantly down to 26% on average. To improve our CNN (training step especially), we tested diverse loss functions. We also developed a specific CNN for Porites sp. and obtained a similar accuracy (94%) to that for the CNN for Diploastrea sp. (93%). Despite this result, we developed a CNN-Mixed model combining images collected from both coral genera to propose a unique and accurate model. As we obtained an accuracy above 90% when applying the mixed model to either massive coral genus, we strongly suggest that it can be used to better understand the role of microborers in living massive corals and reefs over long term.

TEOS-10 compliant equations are presented for the direct calculation of salinity, density, and Sigma0 from triplets of temperature, pressure, and sound speed for marine and estuarine waters. The 73, 71, and 71 term, respectively, 6^th-order equations, are valid over an environmental range of 0–40°C in situ temperature, 0–6000 dBar pressure, and 0–40 practical salinity. The equations can reproduce a TEOS-10 speed of sound equation–derived reference parameter space to a root mean square (RMS) error of ± 0.002383 g kg⁻¹, ± 0.002814 kg m⁻³, and ± 0.002812 kg m⁻³, respectively. The limitation for practical use is the order ± 0.05 m s⁻¹ accuracy of the TEOS-10 equation for calculating sound speed. Four new in situ reference datasets are subsequently used to improve on this limitation. Adding the reference datasets leads to 70, 72, and 72 term, respectively, 6^th-order equations, to an overall RMS error of ± 0.012251 g kg⁻¹, ± 0.010023 kg m⁻³, and ± 0.010209 kg m⁻³, and extends the temperature validity range to −1.772 to 40.000°C in situ temperature.

Suspension-feeding organisms play a pivotal role in the cycling of carbon in the oceans. They filter large amounts of water, filter out organic matter, remineralize it, and release respiratory CO₂ back into the water column. Measuring emissions of respiratory CO₂ in situ from suspension feeders poses the challenge of detecting small changes in dissolved inorganic carbon (DIC). To address this issue, we propose a method for measuring CO₂ excretion rates directly and continuously for undisturbed pumping suspension-feeders in the lab and in situ. This technique involves using miniature optodes to measure the pH of the water inhaled and exhaled by suspension-feeders. Dissolved inorganic carbon concentrations are calculated using the pH data along with ancillary measurements of ambient total alkalinity, temperature, salinity, and pressure. The DIC mass flux can be determined by multiplying the difference in DIC concentrations between the inhaled and exhaled water by the pumping rate of the target organism. The method was tested for sponges in situ. pH-based, continuously measured DIC mass flux rates were found to be higher but comparable to those calculated from discrete DIC measurements and slightly lower than concurrently measured oxygen consumption rates. Assuming that total alkalinity remains constant throughout the brief (seconds) passage of the water through the filtration system, this technique provides a reliable and continuous assessment of DIC fluxes from suspension-feeders. Furthermore, pH-optodes can be coupled with O₂-optodes to provide measurements of the respiratory quotient (RQ) that is widely used in ecology but rarely measured continuously, let alone in the field.

Phytoplankton species are essential bioindicators for evaluating the status of freshwater ecosystems in accordance with the EU Water Framework Directive. However, manual identification of phytoplankton is time-consuming and requires taxonomic expertise. Deep learning (DL) offers promising tools for automating the identification, but challenges remain due to imaging biases, morphological diversity, and the lack of validated benchmark datasets. In this study, we trained a DL model on microphotographs of controlled laboratory strains from 20 phytoplankton species and tested its performance on independent environmental image datasets. We assessed which species are suitable for cross-dataset classification and explored whether computer vision–based image representations (DL features) reflect species similarity across datasets. Additionally, we combined shape analysis with DL features to determine whether feature-based species distances correspond to morphological similarity. The model trained on strain images achieved reliable cross-dataset classification for over half of the species. Classification performance declined with increasing feature/domain shifts between training and test images but improved when environmental images enriched the training set. Morphologically distinctive species, such as star-like forms and those with lobes or bristles, exhibited higher classification rates, whereas rectangular or roundish forms posed greater challenges. DL features consistently clustered species across datasets, and the distances in DL feature space aligned with those in simplified shape space. Our findings demonstrate that using strains as references in DL models enables effective cross-dataset classification while capturing morphological patterns. Integrating taxonomic expertise with computer vision is crucial for developing robust, interpretable phytoplankton bioindicator systems for ecological monitoring and biodiversity research.

Release of methane, as gas bubbles or in the dissolved phase, from the seafloor has been observed in coastal waters (< 200 m) and deep ocean basins (> 1000 m). Methane dissolution within the water column affects the geochemistry of the surrounding water, leading to localized oxygen loss and potential escape to the atmosphere, particularly from shallower sites. Traditional methods for detecting and quantifying dissolved methane rely on collecting discrete water samples for ship- or land-based ex situ analysis and post processing. Here, we report on the use of a reduced response time, in situ methane sensor, the Sensor for Aqueous Gases in the Environment (SAGE), for detecting and quantifying dissolved methane concentrations in a wide range of seafloor environments. During a Fall 2022 research cruise on the R/V Thomas G. Thompson in Puget Sound, SAGE was integrated onto a towed conductivity/temperature/depth rosette and deep-sea camera system with live-stream 1 Hz telemetry and used to spatially map the concentration of methane approximately 1 m above the seafloor. The site had been previously identified as an active methane plume field characterized by gas bubbles, fluid venting, and a faulted seabed. The widespread background dissolved concentration of methane measured by SAGE was 83 nM, and a range of 78–670 nM was observed throughout the survey. The results highlight the capacity of SAGE to map the spatial and temporal variability of dissolved methane concentrations in situ and to identify and localize sites of variable methane emissions from the seafloor.

Sinking particles play a key role in the biological carbon pump. While previous studies have analyzed particulate carbon flux over timescales of days to years, few have been able to resolve flux variability on shorter, hourly scales at multiple depths simultaneously. This study uses an array of upward-facing cameras, built from off-the-shelf components for under $500 each, to visualize particle fluxes at multiple depths during the EXPORTS campaign in 2018 in the North Pacific. This manuscript is the first comprehensive description of this tool, called GelCam, which captures a time-lapse image sequence at 20-min intervals of particles that settle into a polyacrylamide gel layer located at the base of a sediment trap tube. Methods are described for the design and post-processing pipeline, in addition to two proxy methods for estimating the total particulate organic carbon flux. The GelCam-derived fluxes modeled from individual particle images show strong agreement with the ground-truth data obtained from coincident trap measurements. This approach helps address the need for accessible, open-source tools to more broadly observe and quantify the role of episodic particle flux events across the global oceans.

Historical quantification of cyanobacterial harmful algal blooms (cHABs) typically involved labor-intensive manual cell counting. We developed a novel, cost-effective, field-validated system to perform cell counts of six common toxin-producing cyanobacterial genera within 30 s of upload with 10-min sample preparation. Using a portable field microscope, users can quickly evaluate the type and quantity of freshwater cyanobacteria for use in ecological monitoring, human health, and water quality. Participating groups (n = 21) received digital microscopes and sampling equipment and submitted images from 170 cHAB events occurring in 36 US lakes for machine learning (ML) assisted cHAB analysis via a smartphone app. The accuracy of ML identification was compared to human taxonomic identification, while cell concentrations were compared against FluoroProbe (bbe Moldaenke GmbH) blue-green algal (BGA) chlorophyll fluorescence. Machine learning classification of 497 photos containing 4002 colonies performed as well as, or better than, human taxonomic analysis in 94% of cases. There was a weak correlation between BGA chlorophyll and ML-derived cell counts (R² = 0.33), biovolume (R² = 0.13) or total pixel counts (R² = 0.32) across all samples, but there was a strong correlation (R² = 0.76) between ML cell concentrations and BGA chlorophyll in samples not subject to overnight shipping and handling, where stress induced by transport and dark conditioning of cyanobacteria was hypothesized to drive more error in quantification by fluorescence. Cell counting minimizes errors introduced by fluorescence measurements and could improve risk assessment. The described quantification tool is easy-to-use and readily accessible to users with different levels of expertise in cHAB science.

The seawater partial pressure of carbon dioxide (pCO₂) is an essential ocean variable needed to calculate air-sea gas exchange and to identify marine carbon sinks and sources. Recent technological developments support autonomous pCO₂ measurements with sensors that are smaller and cheaper. In July 2021, these differences were highlighted during the Integrated Carbon Observation System—Ocean Thematic Centre laboratory intercomparison exercise. A key message from the intercomparison was the need for further field comparisons. Here we present the results from a field test of two generations of -4H-Jena HydroC CO2-FT membrane-based sensors alongside a General Oceanics equilibrator system. The intercomparisons were done onboard a ship-of-opportunity regularly traveling between Europe and South America. The first phase of the experiment took place in 2021, when the difference between the two instruments was within ± 10 μatm during 53% of the intercomparison time. For the second phase, improvements were made, including the addition of an automated cleaning routine for the membrane-based sensor, the installation of a new sensor prototype with the ability to measure a reference gas, and an updated data processing method. These changes improved the performance and, during the last 2023 journey, the mean difference decreased to 2.0 ± 5.0 μatm, and was within ± 10 μatm during 97% of the deployment time. This experiment revealed that with a suitable deployment approach considering biofouling and reference gas measurements, membrane-based sensors can measure seawater pCO₂ within the Global Ocean Acidification Observing Network weather goal of 2.5% relative uncertainty on autonomous installations.