Posts Tagged 'methods'

« Previous Entries
Next Entries »

New capability in autonomous ocean carbon observations using the autosub long-range AUV equipped with novel pH and total alkalinity sensors

Published 11 April 2025 Science Closed
Tags: , ,

The development of marine autonomous platforms has improved our capability to gather ocean observations at fine spatial scales and high temporal frequency, which can be used to better measure, characterize, and model ocean carbon. As part of the OCEANIDS program, novel carbonate sensors were integrated into the Autosub Long-Range (ALR) autonomous underwater vehicle (AUV) and deployed in the Celtic Sea. Autonomous Lab-On-Chip (LOC) sensors measured pH and total alkalinity (TA) while onboard the ALR. Using interpolation, the ALR-sensor data set is compared against CTD co-samples. The average differences between the LOC sensor and co-sample pH range from −0.011 to −0.015. The TA sensor data agrees with co-samples within 1–2 μmol kg–1 on average. Biogeochemical water properties differing between CTD and ALR observations reveal correlations to carbonate parameter variations. The LOC sensors enabled the characterization of the marine carbonate system from autonomous subsurface measurements for the first time. Sensor pH and TA data were used to calculate dissolved inorganic carbon (DIC), partial pressure of CO2 (pCO2), and aragonite saturation state (ΩAr) and are compared with CTD co-samples with mean residuals of 4–7 μmol kg–1, 10–17 μatm, and −0.03 to −0.06, respectively. Future perspectives on sensor deployment and analysis are discussed.

Continue reading ‘New capability in autonomous ocean carbon observations using the autosub long-range AUV equipped with novel pH and total alkalinity sensors’

Evaluating data quality of coastal spectrophotometric pH measurements: implications for ocean acidification and ocean alkalinity enhancement research

Published 10 April 2025 Science Closed
Tags: ,

pH, which reflects the thermodynamic balance of acid-base systems in seawater, serves as a key indicator of the interplay between acidic and basic components in marine environments. When combined with another parameter, such as TA, DIC, or pCO2 the entire inorganic carbon system can be derived. However, each parameter presents methodological challenges that may introduce random or systematic errors, which then propagate through subsequent calculations. In coastal and estuarine environments, errors can become more pronounced, as standard operating procedures (SOPs) developed for open-ocean conditions may not adequately address the complexities unique to these regions. Measuring more than two parameters enables further insight into systematic errors through the evaluation of internal consistency, where existing data products often reveal pH-dependent offsets between measured pH and pH calculated from measured TA and DIC. These offsets may arise from errors in pH measurements, TA and DIC measurements, or the equilibrium constants used in the calculations, and are therefore difficult to tease apart. Comparing measurements from different research groups can help identify the specific measurement biases responsible for these offsets; however, the lack of inter-comparison studies, particularly in field settings, hinders our understanding. This work advocates for integrating internal consistency and inter-comparison studies in field conditions, as conducting them at sea provides a realistic evaluation of reproducibility between research groups. Chapter 2 utilizes this method by comparing at-sea spectrophotometric pH (pHspec) measurements from two research groups aboard the R/V Coriolis in June 2022 in the Gulf of St. Lawrence and the Lower St. Lawrence Estuary during the Tracer Release Deep Experiment 2 (TReX2) cruise. This combined analysis of reproducibility and internal consistency highlights how even minor methodological differences can substantially affect data quality, and in turn, shape data interpretation. These impacts are particularly pronounced when estimating potential bias from unidentified, excess components of TA (TAx), expected to be non-negligible in estuarine environments, where the two groups had notably different estimates. Chapter 3 draws conclusions from the discussion of data quality in estuarine environments from Chapter 2, focusing on the potential role of pHspec in Monitoring, Reporting, and Verification (MRV) frameworks for ocean alkalinity enhancement (OAE), a proposed marine carbon dioxide removal (mCDR) strategy. It incorporates insights from OAE field trial work in the Bedford Basin, Halifax, a fjord-like estuarine system, to assess the quality of pHspec, TA, and DIC data, offering an assessment of the reliability of these measurements for interpreting potential carbon dioxide removal. This chapter also includes suggestions for a future protocol for observational components of MRV frameworks.

Continue reading ‘Evaluating data quality of coastal spectrophotometric pH measurements: implications for ocean acidification and ocean alkalinity enhancement research’

Chapter 13 – Future strategy for a resilient production and certification of seawater reference materials for the carbonate system

Published 1 April 2025 Science Closed
Tags:

Ocean inorganic carbon research is crucial to quantify the global ocean uptake of atmospheric carbon dioxide (CO2), understand its spatiotemporal variability and mechanisms that control this process, and monitor ocean acidification. This requires high-quality measurements of the seawater carbonate system that rely on the availability of reference materials (RMs). The COVID-19 pandemic highlighted the fragility of the production system of the seawater RMs for the carbonate system, currently depending on one single laboratory. With that in mind, a new model for seawater RMs for the carbonate system, centered on regional hubs, is being developed to create a more resilient system. Challenges associated with establishing new production centers, such as funding their startup and ongoing costs, ensuring the quality and stability of the materials, and staff training are discussed. Opportunities to minimize the cost of these RMs and to supply certified or indicative values for currently uncertified carbonate system variables are explored. Additionally, a vision to integrate the new model into the global metrology landscape whereby the materials are comparable and metrologically traceable to the International System of Units is highlighted. As more laboratories are seeking to undertake seawater carbonate system measurements, access to these RMs is ever more critical.

Continue reading ‘Chapter 13 – Future strategy for a resilient production and certification of seawater reference materials for the carbonate system’

Chapter 12 – Development of multiparametric standard seawater (MSSW) for CO2 parameters, dissolved oxygen, and density of seawater

Published 27 March 2025 Science Closed
Tags:

Multiparametric Standard Seawater (MSSW) is being developed based on the technology used to manufacture the reference material for nutrients in seawater (RMNS), without adding mercuric chloride to sterilize it, but by adopting aluminum bottles and plastic inner caps with high impermeability to gas and water vapor. The history, current status, and future plans for the development of MSSW for measurements of Practical Salinity, density, dissolved inorganic carbon, total alkalinity, pH, dissolved oxygen (DO), and dissolved organic matter are discussed. Substances that interfere with the Winkler method for DO determination (nitrite, iodate, and hydrogen peroxide) were evaluated for MSSW. The values of the parameters of interest were relatively homogeneous, but the concentrations of dissolved organic carbon depended on the serial number in the lot tested. Long-term stabilities were good in most cases, but DO concentrations began to gradually decrease immediately after production and appeared to stabilize a few years later. The Practical Salinity also tended to decrease (–0.00015 year–1), and the cause of the decreasing trend urgently needs to be clarified. One possibility is an increasing trend (about +0.00015 year–1) of the Practical Salinity of IAPSO standard seawater for salinity measurements.

Continue reading ‘Chapter 12 – Development of multiparametric standard seawater (MSSW) for CO2 parameters, dissolved oxygen, and density of seawater’

On the measurement of ocean acidity with ambient sound

Published 12 March 2025 Science Closed
Tags: , , , ,

The volume-integrated pH of seawater can be determined from the frequency and depth dependence of wind-generated ambient noise in the ocean. Over the 1 − 10 kHz frequency band, three main processes contribute to the acoustic attenuation in seawater: the chemical relaxation of boric acid and magnesium carbonate (< 3 kHz, related to pH), and magnesium sulfate (> 3 kHz, unrelated to pH). When local winds are strong (> 10 m/s), the ambient noise is dominated by locally generated surface noise, which exhibits a depth-independent directionality, and weak frequency and depth-dependent intensity. By measuring the depth-dependence of the spectral slope, the pH may be estimated from a comparison of the experimental data with an analytical model of ambient noise. Measurements of the depth-dependent ambient noise field were carried out in the Philippine Sea, Mariana Trench, and Tonga Trench from 2009 to 2021. The wideband (5 Hz – 30 kHz) acoustic data were recorded with untethered, free-falling, autonomous instrument platforms known as Deep Sound, equipped with two or four hydrophones. In all the data collected, the power spectral slopes became steeper with depth due to the stronger attenuation of high frequencies compared to low frequencies. Depth-averaged pH values, ranging from 7.68 to 8.35, were obtained from eight instrument drops. The noise spectral method, which has the potential for determining the depth-averaged value of pH, with the averaging depth being adjustable, could be suitable for the long-term passive acoustic monitoring of ocean acidity.

Continue reading ‘On the measurement of ocean acidity with ambient sound’

Assessing benthic invertebrate vulnerability to ocean acidification and de-oxygenation in California: the importance of effective oceanographic monitoring networks

Published 21 February 2025 Science Closed
Tags: , , , , , , , , ,

Greenhouse gas emissions from land-use change, fossil fuel, agriculture, transportation, and electricity sectors expose marine ecosystems to overlapping environmental stressors. Existing climate vulnerability assessment methods analyze the frequency of extreme conditions but often minimally consider how environmental data gaps hinder assessments. Here, we show an approach that assesses vulnerability and the uncertainty introduced by monitoring data gaps, using a case study of ocean acidification and deoxygenation in coastal California. We employ 5 million publicly available oceanographic observations and existing studies on species responses to low pH, low oxygen conditions to calculate vulnerability for six ecologically and economically valuable benthic invertebrate species: red sea urchin (Mesocentrotus franciscanus), purple sea urchin (Strongylocentrotus purpurpatus), warty sea cucumber (Apostichopus parvimensis), pink shrimp (Pandalus jordani), California spiny lobster (Panulirus interruptus), and Dungeness crab (Metacarncinus magister). Further, we evaluate the efficacy of current monitoring programs by examining how data gaps heighten associated uncertainty. We find that most organisms experience low oxygen (<35% saturation) conditions less frequently than low pH ( < 7.6) conditions. It is only deeper dwelling (>75 m depth) life stages such as Dungeness crab adults and pink shrimp embryos, juveniles, and adults that experience more frequent exposure to low oxygen conditions. Adult Dungeness crabs experience the strongest seasonal variation in exposure. Though these trends are intriguing, exposure remains low for most species despite well-documented pH and oxygen declines and strengthening upwelling in the central portions of the California Current. Seasonal biases of data collection and sparse observations near the benthos and at depths where organisms most frequently experience stressful conditions undermine exposure estimates. Herein we provide concrete examples of how pink shrimp and Dungeness crab fisheries may be impacted by our findings, and provide suggestions for incorporating oceanographic data into management plans. By limiting our scope to California waters and assessing the limitations presented by current monitoring coverage, this study aims to provide a granular, actionable framework that policymakers and managers can build from to prioritize targeted enhancements and sustained funding of oceanographic monitoring recommendations.

Continue reading ‘Assessing benthic invertebrate vulnerability to ocean acidification and de-oxygenation in California: the importance of effective oceanographic monitoring networks’

Autonomous sensor for in situ measurements of total alkalinity in the ocean

Published 18 February 2025 Science Closed
Tags: ,

Total alkalinity (TA) is one of the measurable parameters that characterize the oceanic carbonate system. A high temporal and spatial frequency in TA data can lead to better measurements, modeling, and understanding of the carbon cycle in aquatic systems, providing insights into problems from global climate change to ecosystem functioning. However, there are very few autonomous technologies for in situ TA measurements, and none with field demonstrations below 2 m depth. To meet this need in marine observing capabilities, we present a submersible sensor for autonomous in situ TA measurements to full ocean depths. This sensor uses lab-on-a-chip technology to sample seawater and perform single-point open-cell titration with an optical measurement. It can carry multiple calibration materials on board, allowing for routine recalibration and quality checks in the field. The sensor was characterized in the laboratory and in a pressure testing facility to 600 bar (equivalent to 6 km depth) and deployed in a shallow estuary, on a lander at 120 m depth, and on an autonomous underwater vehicle. With a demonstrated precision and accuracy regularly better than 5 μmol kg–1 in field deployments, this sensor has the potential to dramatically expand our ability to perform long-term autonomous measurements of the marine carbonate system.

Continue reading ‘Autonomous sensor for in situ measurements of total alkalinity in the ocean’

Innovations in ocean biogeochemical instrumentation and monitoring

Published 11 February 2025 Science Closed
Tags: , , ,

Anthropogenic activities are driving changes in ocean biogeochemistry, which can be monitored through instruments and sensors deployed across diverse platforms in even the harshest marine environments. Continued monitoring of these changes demands innovations in instrumentation, calibration and quality control to effectively capture dynamic signals and ensure comprehensive ocean coverage. This dissertation focuses on advancements in oceanographic pH sensors, starting with the longest near-continuous ocean pH dataset collected using ionsensitive field effect transistor (ISFET) technology at Scripps Pier. A new in situ calibration approach, based on direct tris buffer injection, was compared to the traditional bottle collection method, yielding a fourfold improvement in repeatability with an uncertainty of 0.006 pH. Additionally, an automated calibration system integrated into the sensor package was evaluated, offering near real-time, self-calibrating capability for ocean acidification and biogeochemical monitoring programs. To continue the discourse of pH sensor technology in the second section of this dissertation, a novel optical pH sensor was evaluated in laboratory settings to establish its accuracy and precision, response time, temperature and pressure sensitivity, and calibration techniques which improved accuracy over factory methods. Field tests of the optical pH sensor across diverse marine environments—deep ocean, dynamic nearshore, and open ocean profiling—provided guidelines for field calibration, correction and optimal field use. In a scaled-up sense, the final section of this dissertation leveraged pH and other biogeochemical sensors on BGC-Argo profiling floats to explore biogeochemical variability in the equatorial Pacific from 2019 to 2024. While the region has extensive physical data, subsurface biogeochemical observations and their links to El Niño and La Niña cycles are sparse. These floats revealed distinct biogeochemical patterns driven by vertical movement of the mixed layer depth, meridional subtropical water transport and primary production shifts associated with ENSO phases. Overall, this work combines new sensor technologies and analytical methods to provide essential data, instrument guidelines and reveal insights into ocean biogeochemical phenomena. Ongoing instrumentation development and monitoring will be critical to expand and deepen our understanding of how human-driven impacts are transforming our oceans.

Continue reading ‘Innovations in ocean biogeochemical instrumentation and monitoring’

Perspectives from developers and users of the GOA-ON in a box kit: a model for capacity sharing in ocean sciences

Published 24 January 2025 Science Closed
Tags: ,

Providing reliable instrumentation that enables collection of high-​quality, comparable data is one of the most challenging aspects of establishing ocean acidification monitoring programs. This is especially true for under-resourced countries, where such data are mostly unavailable. In 2016, The Ocean Foundation (TOF) worked with international bodies, including the International Atomic Energy Agency’s Ocean Acidification International Coordination Centre and the Global Ocean Acidification Observing Network (GOA-ON), and subject matter experts to develop a set of equipment known as the “GOA-ON in a Box” kit (The Ocean Foundation, 2017). This comprehensive kit provides researchers with everything needed—down to specialized rubber bands—to obtain weather-quality carbonate system measurements as defined by GOA-ON (Newton et al., 2015). Data are generated from spectrophotometric measurements of pHT and manual titrations for total alkalinity from discrete samples as well as in situ sensors, the iSAMI-pH and a CTD. The kit’s modular design, composed of nearly 100 unique items, makes it much less expensive than comparable integrated systems and allows for easier troubleshooting and replacement of supplied spare components.

Continue reading ‘Perspectives from developers and users of the GOA-ON in a box kit: a model for capacity sharing in ocean sciences’

An aquaria set-up for long-term, multiple-stressor research in marine organisms

Published 17 January 2025 Science Closed
Tags: , , , , ,
  1. Multiple stressor research in aquaria is a useful approach to better understand the ecophysiology of marine species under different environmental conditions, including global change scenarios. Long-term experiments are helpful to detect the response of sustained exposure to selected environmental conditions. Here, we present an experimental set-up suitable to run long-term experiments, composed of a life support system, a cost-effective aquaria set-up and an open-source controller based on the use of a Raspberry Pi. In this set-up, temperature, pH and dissolved oxygen (DO) are individually manipulated and simultaneously controlled in eight different treatments.
  2. To prove the efficacy of the set-up, we provide an assessment over a nine-month experiment on a deep-sea coral species, combining values from current in situ and IPCC AR5 RCP 8.5 scenarios for the aforementioned parameters. Recorded data from the controllers and independent measurements (e.g. cross-checking with portable multiparameter devices and laboratory analyses) throughout the experimental time have been analysed and results have been discussed.
  3. Overall, the experimental set-up performed well, proving the stability of the parameters over time, both individually and in combination. On average, low and high-temperature treatments varied ~0.4 and 0.3°C, respectively. Low pH treatments were maintained within 0.05 pH units, whereas ambient pH treatments varied ~0.04 pH units. Low DO treatments had a variation of ~0.3 mg L−1, and ambient DO treatments varied ~0.2 mg L−1. No significant differences between scenarios for any parameter were detected (p < 0.05). The resulting programming code to read, control and register the values for these parameters is provided to contribute to its replicability across institutions.
  4. The set-up performed well over extensive periods while providing a customisable controller as a cost-effective alternative. The versatility of the system, allowing to work with different species, environments and scenarios makes it valuable for aquaria experiments where interactions of multiple environmental factors need to be tested.
Continue reading ‘An aquaria set-up for long-term, multiple-stressor research in marine organisms’

An environmentally adaptive CRO-SL algorithm based on dynamic agents for the channel assignment problem in wireless networks

Published 15 January 2025 Science Closed
Tags: , , , , ,

In recent decades, metaheuristic algorithms have emerged as indispensable tools for addressing complex optimization challenges, particularly in several engineering fields, where NP-hard problems are prevalent. A common NP-hard problem in communication engineering is the Channel Assignment Problem (CAP) for wireless access points (APs), with a determined number of stations (STAs) connected to them. The performance of the complete network depends on the interference and noise among the different clusters of devices and the obstacles or elements placed in the physical transmission space. To address the CAP, a new environmentally adaptive approach is proposed for the Coral Reefs Optimization with Substrate Layers (CRO-SL) algorithm, introducing new environmental agents: algae (representing tabu positions) and ocean water acidification (lowering fitness thresholds). The Environmentally Adaptive CRO-SL (EnvAdapt-CRO-SL) implementation aims to improve solution exploration, enhancing computational efficacy in generating new candidate solutions within the coral reef population. An exhaustive comparative analysis of four configurations of the proposed EnvAdapt-CRO-SL variant assesses the impact of each environmental agent on the algorithm’s performance. Additionally, external benchmarks against four different metaheuristics, along with an analysis of the influence of pseudorandom number generators on initialization and search operators, and a robust optimization case study, provide deeper insights. The results show that incorporating the new environmental agents into the EnvAdapt-CRO-SL workflow significantly boosts throughput while reducing the computational time required to obtain optimal solutions.

Continue reading ‘An environmentally adaptive CRO-SL algorithm based on dynamic agents for the channel assignment problem in wireless networks’

Prediction technology of ocean acidification based on quantum particle swarm optimization and long short-term memory network

Published 9 January 2025 Science Closed
Tags: ,

In this paper, the ocean acidification prediction problem was studied deeply, and a new prediction model was constructed by using quantum particle swarm optimization algorithm (QPSO) and long short-term memory neural network (LSTM). Firstly, by introducing the quantum particle swarm optimization algorithm, the limitations of the traditional particle swarm optimization algorithm in search accuracy and global search ability are overcome, and the convergence speed and prediction accuracy of the model are improved. Secondly, using the ability of long short-term memory neural network to process time series data, it effectively captures the long-term dependence relationship in ocean acidification data, and further enhances the prediction ability of the model. Finally, through a series of experimental verification and performance evaluation, including root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and coefficient of determination (R2), it is proved that the model has high accuracy and generalization ability in ocean acidification prediction. The model provides a new method for the effective prediction of ocean acidification.

Continue reading ‘Prediction technology of ocean acidification based on quantum particle swarm optimization and long short-term memory network’

Metrological traceability path for marine pCO2 measurement results

Published 6 January 2025 Science Closed
Tags:

Despite the increasing availability of sensors for gathering data on the partial pressure of CO2 (pCO2) in the waters of the world’s seas and oceans, there is a growing awareness within the scientific community of the pressing need for very high quality measurements of the variable to help monitor the Ocean Acidification threat and, more generally, the effectiveness of overall global CO2 emission mitigation strategies. This is because discriminating observed variations ascribable to long-term trends from the ones due to natural fluctuations and meeting established criteria to reach the “Climate goal” for marine carbonate system research and monitoring require extremely accurate and traceable pCO2 measurements to be useable. Here, we present some details on work being done to expand the use of reference-grade CO2 gas mixtures to help calibrate and/or assess the performances of marine sensors employed for measuring seawater pCO2.

Continue reading ‘Metrological traceability path for marine pCO2 measurement results’

Ocean acidification trends and carbonate system dynamics across the North Atlantic subpolar gyre water masses during 2009–2019 (Update)

Published 20 December 2024 Science Closed
Tags: , , ,

Key points

  • During the 2010s, the subpolar North Atlantic experienced a 50 %–86 % increase in anthropogenic CO2, accelerating by 7 %–10 % the acidification.
  • Anthropogenic CO2 contributed to acidification by 53 %–68 % in upper layers and > 82 % in the interior ocean.
  • The acidification trends (0.0006 and 0.0032 units yr−1) declined the ΩCa and ΩArag by 0.004–0.021 and 0.003–0.0013 units yr−1, respectively.

Abstract

The CO2–carbonate system dynamics in the North Atlantic subpolar gyre (NASPG) were evaluated between 2009 and 2019. Data were collected aboard eight summer cruises through the Climate and Ocean: Variability, Predictability and Change (CLIVAR) 59.5° N section. The ocean acidification (OA) patterns and the reduction in the saturation state of calcite (ΩCa) and aragonite (ΩArag) in response to the increasing anthropogenic CO2 (Cant) were assessed within the Irminger, Iceland, and Rockall basins during a poorly assessed decade in which the physical patterns reversed in comparison with previous well-known periods. The observed cooling, freshening, and enhanced ventilation increased the interannual rate of accumulation of Cant in the interior ocean by 50 %–86 % and the OA rates by close to 10 %. The OA trends were 0.0013–0.0032 units yr−1 in the Irminger and Iceland basins and 0.0006–0.0024 units yr−1 in the Rockall Trough, causing a decline in ΩCa and ΩArag of 0.004–0.021 and 0.003–0.0013 units yr−1, respectively. The Cant-driven rise in total inorganic carbon (CT) was the main driver of the OA (contributed by 53 %–68 % in upper layers and > 82 % toward the interior ocean) and the reduction in ΩCa and ΩArag (> 64 %). The transient decrease in temperature, salinity, and AT collectively counteracts the CT-driven acidification by 45 %–85 % in the upper layers and in the shallow Rockall Trough and by < 10 % in the interior ocean. The present investigation reports the acceleration of the OA within the NASPG and expands knowledge about the future state of the ocean.

Continue reading ‘Ocean acidification trends and carbonate system dynamics across the North Atlantic subpolar gyre water masses during 2009–2019 (Update)’

Advanced deep learning technique for estimating global surface ocean calcium carbonate saturation (Ωcal)

Published 18 December 2024 Science Closed
Tags: , , ,

Highlights

  • The study employed FFNN, RF, and TabNet ML models to estimate Ωcal using in-situ measurements and satellite-derived data.
  • The TabNet model outperformed with significantly low errors (RMSE=0.39, MNB=0.0058, MRE=0.019) and high accuracy (R2=0.96).
  • SST strongly correlated with Ωcal (r=0.98), SSS moderately (r=0.49), and Chla showed a weak negative correlation (r=-0.27).
  • The study observed seasonal Ωcal variability, with higher values in summer months, notably in temperate and polar regions.
  • Emphasized a declining trend of Ωcal from 2012 to 2022, likely influenced by changing oceanic and atmospheric conditions.

Abstract

The accurate estimation of surface ocean calcium carbonate saturation (Ωcal) is crucial for understanding the impacts of ocean acidification (OA) on marine ecosystems, particularly for calcifying organisms. This study investigates the estimation of global surface ocean Ωcal using machine learning (ML) models and satellite-derived data. Three ML models such as feed-forward neural networks (FFNN), random forests (RF), and Tabularnet (TabNet) were employed to estimate Ωcal, utilizing in-situ and satellite measurements of sea surface temperature (SST), sea surface salinity (SSS), and Chlorophyll-a concentration (Chla). Among these, the TabNet model exhibited superior performance, with a root-mean-square error (RMSE) of 0.39, mean relative error (MRE) of 0.019, mean normalized bias (MNB) of 0.0058 and coefficient of determination (R2) of 0.96. SST showed a strong positive correlation with Ωcal (r = 0.95), while SSS and Chla exhibited moderate positive (r = 0.49) and weak negative (r = −0.27) correlations, respectively. The study revealed significant spatiotemporal variability in Ωcal, driven by seasonal changes and ocean circulation patterns. Sensitivity analysis highlighted the robustness of the TabNet model, maintaining high predictive capability despite variations in SST, SSS, and Chla. The TabNet model high accuracy provides a valuable tool for monitoring and forecasting changes in ocean chemistry, informing conservation efforts and policy-making. This study emphasizes the importance of advanced ML models in marine science and their potential for enhancing our understanding of global oceanic processes.

Continue reading ‘Advanced deep learning technique for estimating global surface ocean calcium carbonate saturation (Ωcal)’

Polymeric membrane ion-selective electrode based on potential-modulated ion transfer: ultrasensitive measurement of oceanic pH

Published 2 December 2024 Science Closed
Tags: ,

The application of a potentiometric pH electrode in ocean acidification observation is still a challenge due to its poor sensitivity to small pH changes. Herein, a simple approach to remarkably improve the detection precision of a polymeric membrane ion-selective electrode is proposed based on the potential-modulated ion transfer mechanism. The present sensing strategy displays highly sensitive responses to small pH changes for seawater analysis with a precision of 5 μpH, which is 2 orders of magnitude lower than that of the conventional pH electrode.

A simple approach to improve the precision of potentiometric sensors is proposed based on potential-modulated ion transfer without memory effects.

Continue reading ‘Polymeric membrane ion-selective electrode based on potential-modulated ion transfer: ultrasensitive measurement of oceanic pH’

A neural network algorithm for quantifying seawater pH using Biogeochemical-Argo floats in the open Gulf of Mexico

Published 29 November 2024 Science Closed
Tags: , , ,

Within the Gulf of Mexico (GOM), measurements of ocean pH are limited across space and time. This has hindered our ability to robustly monitor and study regional carbon dynamics, inclusive of ocean acidification, over this biogeochemically variable sea. The 2021 launch of Biogeochemical-Argo (BGC-Argo) ocean profiling floats that carry five sensors represented the entry of this particular ocean observing technology into this region. The GOM BGC-Argo floats have vastly increased the number of oxygen, nitrate, pH, chlorophyll-a fluorescence, and particulate backscattering profile observations within the “open GOM” region (>1,000 m water column depth). To circumvent a set of uncertainties associated with the collected sensor pH data, regionally trained neural network algorithms were developed to skillfully predict GOM pH (total scale, in situ temperature and pressure), which served as a secondary QC and sensor performance assessment tool. The GOM neural network pH (GOM-NNpH) algorithms were trained using a selection of climate quality CTD and bottle data (temperature, salinity, oxygen, nitrate, pressure, and location) collected as a part of NOAA’s Gulf of Mexico Ecosystems and Carbon Cruises (GOMECC). Neural network pH estimates were generated using the newly developed GOMNNpH algorithm and two widely used, globally trained neural network algorithms (Empirical Seawater Property Estimation Routines (ESPER) and CArbonate system and Nutrients concentration from hYdrological properties and Oxygen using a Neural-network (CANYON-B)) to compare algorithm performance against validation data. The results demonstrate the advanced skill of the GOM-NNpH in capturing water column variability and robustly reconstructing GOM pH profiles. Using a combination of concurrent float-measured seawater values of pressure, temperature, salinity, and oxygen, a GOM-NNpH algorithm was applied to two years of BGC-Argo float data. Resulting data were used to diagnose the performance of float pH sensors and to generate a time series of neural network estimated pH based on the collected float profiles. These algorithms emphasize the value of regionally-trained neural networks and their utility by the BGC-Argo community. Further, the GOM-NNpH algorithms can also be applied by a variety of users to estimate pH values in the open GOM in the absence of direct pH observations.

Continue reading ‘A neural network algorithm for quantifying seawater pH using Biogeochemical-Argo floats in the open Gulf of Mexico’

Effortless alkalinity analysis using AI and smartphone technology, no equipment needed, from freshwater to saltwater

Published 29 November 2024 Science Closed
Tags: ,

Highlights

  • A new, simple, equipment-free method for alkalinity analysis was developed.
  • This method can analyze a wide range of water matrices.
  • It combines commercial reagents and smartphones with machine learning models.
  • A free, user-friendly website is available for the models.

Abstract

Alkalinity is a crucial water quality parameter with significant environmental and engineered system applications. Various analysis methods exist, from traditional titrations to advanced spectrophotometric and electrochemical techniques, each with specific benefits and limitations. Developing simple, affordable techniques for alkalinity analysis is essential to facilitate extensive and reliable water quality monitoring, empowering citizen scientists, and overcoming financial barriers in traditional monitoring programs. In this work, we developed an equipment-free, user-friendly alkalinity analysis approach accessible to a broad demographic. Specifically, we employed low-cost commercial reagents to generate color changes in response to alkalinity levels in various freshwater and saltwater samples. These images were captured with a smartphone and processed using machine learning models to correlate color intensity with alkalinity levels. After examining the effects of container type, lightning condition, ML algorithms, and sample size, we obtained the best models with R2 values of 0.868 ± 0.024 and 0.978 ± 0.008, and root-mean-square-error values of 29.5 ± 2.6 and 14.1 ± 2.0 for freshwater and saltwater, respectively. Five inexperienced users utilized this method for alkalinity analysis and achieved comparable results in performance. Additionally, we developed a user-friendly website where users, without prior experience, can upload images to obtain alkalinity readings for their water samples. This AI-powered, equipment-free technology represents a significant milestone in water quality monitoring, deviating from the trend of developing increasingly advanced analytical techniques and serving as a foundation for developing similar methods across various water quality parameters and broader analytical applications.

Continue reading ‘Effortless alkalinity analysis using AI and smartphone technology, no equipment needed, from freshwater to saltwater’

Ocean acidification thresholds for decapods are unresolved

Published 26 November 2024 Science Closed
Tags: , , , , ,

A published analysis of ocean acidification thresholds for decapod crustaceans highlights data showing the negative effects of low pH on many species. However, the methods used in the paper have substantial flaws that call into question the proposed thresholds. The quantitative metrics calculated for the meta-analysis are uninformative with respect to pH sensitivity, which raises concerns about the validity of the thresholds developed by the expert opinion process. We recommend against using the published thresholds and for a reanalysis of the data to identify new thresholds.

Introduction

Identifying thresholds of biological response to environmental stressors can provide valuable information for marine resource managers (Heinze et al., 2021). In the field of ocean acidification (OA), meta-analysis of species pH exposure experiments have been used to develop thresholds or relative risk metrics for multiple taxa (Bednaršek et al., 2019; Bednaršek et al., 2021b; Hancock et al., 2020; Cornwall et al., 2022). Decapod crustaceans are an important ecological and economic component of marine ecosystems, and species in this group are potentially vulnerable to OA. Using quantitative analysis based on published data from species pH exposure experiments and expert opinion, Bednaršek et al. (2021a) developed pH exposure thresholds for a number of biological responses of decapods. The Bednaršek et al. (2021a) thresholds have been used in other analyses to characterize potential risk of decapod species to OA (Alin et al., 2023; Alin et al., 2024; Hamilton et al., 2023; Zeldis et al., 2022; Siegel et al., 2022). Although there is no doubt that many decapod species are sensitive to elevated CO2 and exhibit thresholds in their sensitivity, the quantitative analysis in Bednaršek et al. (2021a), which informs the expert opinion, has several methodological issues which raise questions about the utility of the recommended thresholds. This brief research report describes the methodological issues and discusses the implications for our current understanding of pH thresholds for decapods. It does not recommend a specific pathway for re-evaluation of the data, but does provide some relevant references that could guide a re-evaluation.

Continue reading ‘Ocean acidification thresholds for decapods are unresolved’

Coupled decline in ocean pH and carbonate saturation during the Palaeocene–Eocene Thermal Maximum

Published 20 November 2024 Science Closed
Tags: , , , ,

The Palaeocene–Eocene Thermal Maximum, a climate event 56 million years ago, was characterized by rapid carbon release and extensive ocean acidification. However, our understanding of acidification and the evolution of ocean saturation states continues to be hindered by considerable uncertainties, primarily stemming from the limited availability of proxy data. Under such conditions, data assimilation allows for an internally consistent assessment of atmospheric CO2 changes, ocean acidification and carbonate saturation state during this period. Here, we present a reconstruction of the Palaeocene–Eocene Thermal Maximum carbon cycle perturbation by assimilating seafloor sediment CaCO3 and sea surface temperature proxy data with simulations from an Earth system model, which includes a comprehensive carbonate system. Our reconstructions indicate a substantial increase in atmospheric CO2 from 890 ppm (95% credible interval: 680–1,170 ppm) to 1,980 ppm (1,680–2,280 ppm), coupled with a notable decline in pH (0.46 units, ranging from 0.31 to 0.63 units) and surface-water calcite saturation state, decreasing from 10.2 (7.5–12.8) in the pre-event period to 3.8 (2.8–5.1) during the thermal maximum. Carbonate undersaturation intensified substantially in high-latitude surface waters during the Palaeocene–Eocene Thermal Maximum, paralleling the current decline in Arctic aragonite saturation driven by anthropogenic CO2 emissions.

Continue reading ‘Coupled decline in ocean pH and carbonate saturation during the Palaeocene–Eocene Thermal Maximum’

Subscribe

Search

  • Reset

OA-ICC Highlights

Resources