Dissolved Oxygen Variations Research Articles

Using Machine Learning Models for Short-Term Prediction of Dissolved Oxygen in a Microtidal Estuary

This paper presents a comprehensive approach to predicting short-term (for the upcoming 2 weeks) changes in estuarine dissolved oxygen concentrations via machine learning models that integrate historical water sampling, historical and upcoming 2-week meteorological data, and river discharge and discharge metrics. Dissolved oxygen is a critical indicator of ecosystem health, and this approach is implemented for the Neuse River Estuary, North Carolina, U.S.A., which has a long history of hypoxia-related habitat degradation. Through meticulous data preprocessing and feature selection, this research evaluates the predictions of dissolved oxygen concentrations by comparing a recurrent neural network with four other models, including a Multilayer Perceptron, Long Short-Term Memory, Gradient Boosting, and AutoKeras, through sensitivity experiments. The input predictors to our prediction models include water temperature, turbidity, chlorophyll-a, aggregated river discharge, and aggregated wind based on eight directions. By emphasizing the most impactful predictors, we streamlined the model-building processes and built a hindcast system from 2015 to 2019. We found that the recurrent neural network model was most effective in predicting the dissolved oxygen concentrations, with an R2 value of 0.99 at multiple stations. Different from our machine learning hindcast models that used observed upcoming meteorological and discharge data, an actual forecast system would use forecasted meteorological and discharge data. Therefore, an actual operational forecast may have lower accuracy than the hindcast, as determined by the accuracy of the predicted meteorological and discharge data. Nevertheless, our studies enhance our understanding of the factors influencing dissolved oxygen variability and set the basis for the implementation of a predictive tool for environmental monitoring and management. We also emphasized the importance of building station-specific models to improve the prediction results.

Physics-guided spatio–temporal neural network for predicting dissolved oxygen concentration in rivers

The prediction of river water quality is key in water resource management. Data-driven machine learning models have been widely used for predicting river water quality. However, these models seldom consider the physical mechanisms of water quality variation, which degrades the accuracy and stability of the prediction results. Hence, we develop a physics-guided spatio–temporal neural network (PGSTNN) model to predict a critical parameter for water quality assessment, i.e. dissolved oxygen. Physical information regarding spatio–temporal interactions in a hydrological network is explicitly considered to construct the architecture of PGSTNN. Two physical rules of dissolved oxygen variation (i.e. Henry’s law and power-scaling law) are established for the loss function of PGSTNN to guarantee the physical consistency of the prediction results. Experiments on the 2020–2021 water quality dataset in Atlanta, USA show that PGSTNN outperforms seven baseline neural network models in terms of prediction accuracy and stability. PGSTNN typically brings at least 10% accuracy (e.g. root mean square error and mean absolute error) improvement over the comparison methods. The proposed PGSTNN may not only improve the emergency response ability of water resource management, but also provide useful ideas for integrating scientific knowledge with machine learning.

Simulating implications of fish behavioral response for managing hypoxia in estuaries with spatial dissolved oxygen variability

Hypoxia, or low dissolved oxygen (DO), is a widespread water quality problem affecting estuaries and coastal waters around the world. Water quality criteria for DO have been established for every estuary in the US and are an important part of the regulatory response to nutrient pollution and associated anthropogenic eutrophication. Experimental studies examining effects of low DO exposure have been to quantify outcomes based on hypoxia effects observed in individuals, such as increased mortality or growth impairment. Although laboratory exposure tests provide useful benchmarks for policy development, most of those considered in policy development did not consider behavioral responses to low DO. However, experimental research has shown that behavioral responses occur, and that behavior modifies exposure to low DO conditions. Here we begin development of a spatially explicit individual based model (SEIBM) intended to project behavioral outcomes of exposure to spatially variable hypoxia in estuaries. Our goal is to consider the responsiveness of an SEIBM to both different behavioral hypotheses, as well as realistic spatial patterns in hypoxia. A sensitivity analysis was used to explore responsiveness based on two movement strategies: avoidance and behavioral switching. We tested the sensitivity of a suite of movement parameters to changes in spatial patterns representative of an index estuary. The sensitivity analysis demonstrated that model responses to changes in movement strategies include biologically meaningful changes in site occupancy and movement distance centered on individual behavior near a normoxic–hypoxic boundary. Further, the model demonstrated important sensitivity to realistic changes in movement parameters, including the size and shape of the individual neighborhood describing knowledge useful for movement decisions. These results support the utility of the developed SEIBM for exploring behavioral responses of fish to hypoxia in estuaries. The sensitivity analysis also demonstrates parameter values that must be set based on empirical data and are sensitive to data quality. These results will be used to further develop the model and to plan field and laboratory studies to support model parametrization. The end goal is a model framework that can inform policy decisions regarding hypoxia resulting from anthropogenic nutrient loading in estuaries.

Partial nitrification response to dissolved oxygen variation and aerobic starvation: Kinetics and microbial community analyses

Partial nitrification (PN) has great significance in promoting both anaerobic ammonium oxidation (anammox) and advanced nitrogen removal. However, the kinetics of PN have not been sufficiently studied. Deterioration of PN often occurs due to nitrite-oxidizing bacteria (NOB) outgrowth, for which the dissolved oxygen (DO) is a critical factor. This study investigated the kinetic parameters of ammonia-oxidizing bacteria (AOB) and NOB by varying the DO concentration. The partial nitrification-anammox (PNA) process was operated under sequentially low (0.3–0.8 mg/L) and high (4.0–5.0 mg/L) DO. The results show that AOB could not adapt to low DO, resulting in a significant decrease in the maximum AOB activity (Vmax,AOB) and a slightly altered oxygen half-saturation constant of AOB (KO,AOB). Conversely, KO,NOB decreased and Vmax,NOB maintained, indicating that the main NOB genus, Nitrospira, was more adaptable to DO changes. Under high DO, AOB growth was weaker than NOB, leading to a low Vmax,AOB – Vmax,NOB value; KO,AOB increased and was larger than KO,NOB and PNA was eventually destroyed. The aerobic starvation was then adopted to restore PN. AOB recovered faster and dominated, with Vmax,AOB greatly exceeding Vmax,NOB. Finally, NOB were inhibited and PN was recovered. During this process, KO,NOB increased and was higher than KO,AOB. Changes in the kinetic parameters, especially the Vmax,AOB -Vmax,NOB value, have specific guiding significance for DO regulation during PNA. This work helps better understand the effect of DO variations on the nitrifying microbial community, which influences oxygen affinity and maximum activity.

The role of submesoscale filaments in restratification of the surface mixed layer and dissolved oxygen variability in large Lake Geneva: field evidence complemented by Lagrangian particle-tracking

The role of submesoscale filaments in restratification of the surface mixed layer and dissolved oxygen variability in large Lake Geneva: field evidence complemented by Lagrangian particle-tracking

Microfluidic chip-based microbial metabolism-indexed BOD sensor for rapid determination of biochemical oxygen demand

Reliance on five-day biochemical oxygen demand (BOD5) as an indicator of water quality has inhibited the development of efficient process control due to the associated uncertainty and lag time. In the work, we proposed a novel strategy for rapid on-site BOD5 determination by a microfluidic chip-based microbial metabolism-indexed BOD sensor (MCB-MMI BOD sensor). The rapid on-site BOD5 detection was achieved by quantifying the equilibrium concentration of tryptophan involved in aerobic biodegradation for alternating the traditional 5-day monitoring of dissolved oxygen variation. The regulated metabolized tryptophan was identified by fluorescence spectrum and converted into BOD5 value. The prominent performance for MCB-MMI BOD sensor BOD5 determination was accessible in the field application, such as decreasing detection time from 5 days to 10 min, lowering sample volume from 250 mL to 2.5 mL, diminishing the working area, extending the detection range (without dilution) from 6 mg/L to 27 mg/L, featured with 83% accuracy, and free-labor automation, which exhibited the infinite potential for MCB-MMI BOD sensor on rapid on-site BOD5 monitoring. We believe the MCB-MMI BOD sensor would broaden its practical applications in BOD5-contained multiple-parameters water quality sensing, and would favorably supply a new tool for sewage-treatment process control, environmental monitoring, and aquaculture.

Thirty years of a bottom oxygen depletion-renewal cycle in the coastal yet deep environment of the West Saronikos Gulf (Greece): Its drivers and the impact on the benthic communities

In the period 1987–2017, a series of physical and chemical measurements related to oxygen variability at a trough area with a maximum depth of ~420 m in the West Saronikos Gulf, reveal the following: In the early 90s, deep winter mixing occurred resulting in an oxygenation down to ~420 m followed by an oxygen decline. This decline reached near-bottom hypoxic conditions (O2 < ~62 μM (μmol/L)) after 1998, while a denitrification phase occurred after 2000 and a complete bottom anoxia in 2005. In June 2012, an oxygenation down to ~350 m was detected that most likely occurred in winter 2012. The 2012 oxygenation raised the until-then anoxic bottom concentrations to hypoxic ones in the years towards 2017 via vertical diffusive oxygen transfer. Observations of the benthic communities during the hypoxia, severe hypoxia (O2 < ~15 μM) and oxygen recovery phases showed a peak of opportunists in the hypoxia and a long faunal depletion in the severe hypoxia phases. A reversal in the benthic community structure appeared after the oxygenation of 2012 with the (re)appearance of opportunists while, in 2017, the community showed signs of retreat to earlier stages. The main anthropogenic pressure that could tentatively affect the oxygen concentration in the study area is posed by the Athens treated-sewage outfall at ~40 km away from the trough, which inputs organic matter into the Saronikos Gulf through effluent water of reduced salinity that, in addition, may alter the stratification opposing the vertical mixing. We show that the treated sewage output had no influence on a) the stratification, b) the particulate and dissolved organic carbon and c) the sewage-derived organic matter. Instead, the long-term dissolved oxygen variability with the deep renewal events was mostly driven by the large-scale atmosphere-ocean conditions (heat exchange and evaporation-minus-precipitation budget) that determine the hydrographic characteristics and the winter mixing.

Disturbance of primary producer communities disrupts the thermal limits of the associated aquatic fauna

Environmental fluctuation forms a framework of variability within which species have evolved. Environmental fluctuation includes predictability, such as diel cycles of aquatic oxygen fluctuation driven by primary producers. Oxygen availability and fluctuation shape the physiological responses of aquatic animals to warming, so that, in theory, oxygen fluctuation could influence their thermal ecology. We describe annual oxygen variability in agricultural drainage channels and show that disruption of oxygen fluctuation through dredging of plants reduces the thermal tolerance of freshwater animals. We compared the temperature responses of snails, amphipods, leeches and mussels exposed to either natural oxygen fluctuation or constant oxygen in situ under different acclimation periods. Oxygen saturation in channel water ranged from c. 0 % saturation at night to >300 % during the day. Temperature showed normal seasonal variation and was almost synchronous with daily oxygen fluctuation. A dredging event in 2020 dramatically reduced dissolved oxygen variability and the correlation between oxygen and temperature was lost. The tolerance of invertebrates to thermal stress was significantly lower when natural fluctuation in oxygen availability was reduced and decoupled from temperature. This highlights the importance of natural cycles of variability and the need to include finer scale effects, including indirect biological effects, in modelling the ecosystem-level consequences of climate change. Furthermore, restoration and management of primary producers in aquatic habitats could be important to improve the thermal protection of aquatic invertebrates and their resistance to environmental variation imposed by climate change.

Dissolved Oxygen Variation on the Steps with a Quarter Circle End Sill for Flows over the Stepped Spillways

Determining the aeration efficiency of the stepped spillways is important because the Dissolved Oxygen (DO) concentration helps indicate the water quality. This study investigated the effects of varying step shape and chute slope on the aeration efficiency for stepped spillways. The measured parameters were DO, the inception point of the free surface, and the water surface profile above the crest to evaluate the geometry variation impacts. Several experiments were conducted on a six-step configuration over a stepped spillway with chute angle (θ = 26.6°, 21.8°, and 8.9°). The discharges up to 0.055 m3/s. The step configurations were including flat step, normal end sill, and quarter circle end sill. The results showed when the chute angle changed from 26.6° to 8.9°, the aeration efficiency of E20 improved with 11.51% at the lowest discharge and 6.05% at the highest discharge for the flat step model with 10 steps. Also, E20 improved 11.39% at the lowest discharge and 6.50% at the highest discharge for the flat step model with 6 steps. The performance of the steps with the quarter circle end sill model in terms of aeration efficiency increased by 10%.

Leverage and Cook distance in regression with geostatistical data: methodology, simulation, and applications related to geographical information

Regression is often conducted assuming independent model errors. The detection of atypical values in regression (leverage and influential points) assumes independent errors. However, such independence could be unrealistic in geostatistics. In this article, we propose a methodology based on least squares and geostatistics to identify such values in spatial regression. Our procedure uses the hat matrix to detect leverage points. A modified Cook distance is employed to confirm whether these points are influential. The methodology is evaluated with stationary and non-stationary geostatistical data. We apply this methodology to real georeferenced data related to depth, dissolved oxygen, and temperature. First, an autoregressive model is fitted to depth data. Second, a regression between oxygen and temperature is estimated. In both models, spatial correlation is assumed to determine the parameters, leverage, and influential observations. Our methodology can be used in regression with geographical information to avoid misinterpreted results. Not considering this information may under- or over-estimate geographical indicators, such as the mean depth, which can affect the circulation of water masses or dissolved oxygen variability. Our results reveal that including spatial dependence to identify high leverage points is relevant and must be considered in any geostatistical analysis.

Seasonal controls on nearshore dissolved oxygen variability and hypoxia in a coastal embayment

Declining dissolved oxygen (DO) is emerging as an increasingly important stressor in nearshore ecosystems, and there is a growing need to better understand DO dynamics and hypoxia risk in this highly variable environment. In this study, we collected data from monthly cruises on the inner shelf, continuous nearshore moorings inside and outside a small coastal upwelling embayment (San Luis Obispo Bay in Central California), and weekly phytoplankton measurements inside the bay during the upwelling season. Nearshore DO was generally dominated by low-frequency synoptic variability, with increased DO variance near the surface relative to the bottom and inside the bay compared to outside. Two nearshore hypoxic regimes were identified. In the first regime, which occurred during periods of strong upwelling in the spring across all nearshore sites, the nearshore bottom water temperature-DO (T-DO) relationship was aligned with that found offshore, suggesting hypoxia was driven by the direct advection and cross-shelf exchange of low DO subthermocline waters from the shelf. This period also coincided with minimal water-column stratification, small vertical DO differences, and a diatom-dominated phytoplankton assemblage. In the second regime, which occurred during summer months and was characterized by weaker upwelling, strong stratification, and dinoflagellate-dominated phytoplankton assemblage, the near-bottom T-DO relationship inside the bay deviated significantly from that on the shelf offshore. These hypoxic events inside the bay were likely driven by localized respiration and lack of ventilation of bottom waters due to strong stratification. Collectively, these observations reveal a shift in the strength and magnitude of physical versus biological processes driving nearshore DO dynamics. The high spatiotemporal variability of DO dynamics in upwelling bays means that they are likely to be at the forefront of ecosystem impacts of and adaptions to climate change, and may act as sentinel systems or “canaries on the coast.”

First observations of seasonal bottom water deoxygenation off the Gironde estuary (Bay of Biscay, North East Atlantic)

Although not systematically considered as oxygen minimum zones, River-dominated Ocean Margins are sensitive to oxygen depletions. For example, the continental shelf off the Gironde, which flows into the Bay of Biscay, had not yet been studied from this perspective although recent simulations suggested that this area should already have experienced deoxygenations in recent decades. To fill this gap, profiles of temperature, salinity and dissolved oxygen were performed in the water column of the continental shelf off the Gironde during seven cruises distributed over the different seasons between 2016 and 2021. Turbidity, chlorophyll-a and pH were also measured during some of these cruises. In winter, the water column was slightly stratified due to high river flows. Then, a seasonal thermal stratification was present from spring to autumn. Similarly, dissolved oxygen showed a seasonal dynamic with: a well-oxygenated water column in winter, an oxygen oversaturation in the first 20 meters during the spring bloom, and then a progressive oxygen depletion in bottom waters until reaching an oxygen saturation minimum down to 45% in autumn. These deoxygenations are explained by the seasonal stratification that isolates the bottom waters from spring to autumn, and are likely enhanced by the advection of deoxygenated waters from the north of the Bay of Biscay and the settling of the organic matter produced in surface waters. A better understanding of these processes in the context of global warming undoubtedly requires better documentation of dissolved oxygen variability through the implementation of a long-term and continuous in situ monitoring.

Bioavailability of metals in coastal lagoon sediments and their influence on benthic foraminifera

Coastal lagoons experience removal and leaching of metals due to seasonal fluctuation of salinity coupled with pH and dissolved oxygen variability. Benthic foraminifers are susceptible to seasonal variation in physicochemical conditions. Assessment of bioavailable fraction of selected metals in sediments along the salinity gradient of the largest brackish lagoon in Asia- Chilika was carried out. Further, population density and abnormality index of foraminifera in the sediments were calculated to understand the influence of bioavailable metal on benthic foraminifera. Metal concentrations were higher in the low salinity regions and decreased towards the high salinity regions. This suggests river discharge controls the metal input and distribution in the lagoon. Overall, metal concentrations are high in the residual fraction (F4) followed by interplay within the reducible (F2) or oxidizable (F3) fractions, except for Mn. The seasonal fluctuations of pH and cyclic oxygen deficiency intensified the leaching of Fe, Cr, Cu, Pb, and Zn in the bioavailable fraction in the low salinity region of the lagoon. The presence of metals in bioavailable fractions impacted the normal growth of the abundant Ammonia species. Despite low bioavailable metal concentrations, higher morphological abnormalities were observed in the high salinity regions due to higher energy conditions near the sea. Therefore, abnormalities in the benthic foraminifera are attributed to bioavailability of metals from sediments and natural stress conditions in the coastal lagoon environment.

Using Timescales of Deficit and Residence to Evaluate Near‐Bottom Dissolved Oxygen Variation in Coastal Seas

AbstractWe identify the local, vertical and lateral processes that cause bottom oxygen variation, and characterize timescales for each process using oxygen budget analysis based on a coupled physical‐biogeochemical model for the coastal seas in east China. Local oxygen deficit often occurs in the summer season due to the faster local consumption than the vertical replenishment in seasonal timescale. Lateral transport of oxygen‐rich ambient water replenishes local deficit of bottom dissolved oxygen. Competition between local deficit and lateral exchange determines seasonal hypoxia formation and sustainment. Short local consumption timescale is favorable for hypoxia formation, and transient hypoxia often forms when the local deficit and lateral exchange processes act on comparable timescales, such as the East China Sea in which the bottom hypoxia usually lasts for days. Extremely long lateral exchange timescale suggests that dissolved oxygen variation is predominantly controlled by local processes, and short local consumption timescale often causes increasingly severe seasonal hypoxia until the onset of the relaxation of local deficit, such as the bottom hypoxia in the Bohai Sea. Using these timescales to evaluate local deficit relative to vertical and lateral residence time and their variability is a convenient and potentially powerful general mechanistic framework to evaluate strategies to mitigate coastal hypoxia worldwide.

A Benthic Monitor for Coastal Water Dissolved Oxygen Variation: Mn/Ca Ratios in Tests of an Epifaunal Foraminifer

AbstractAn appropriate proxy could help to better understand dissolved oxygen variations in the past, helping to predict potential outcomes of future environmental changes. In the Changjiang Estuary (China), the foraminifer Cribrononion subincertum (C. subincertum) shows a distinct population maximum in the topmost sediment, an indication of an epifaunal species. Therefore, the geochemical composition of C. subincertum tests could record changes in the region’s bottom water chemistry. Our results showed that Mn/Ca ratios in tests of living (Rose‐Bengal stained) C. subincertum analyzed by LA‐ICP‐MS were responsive to variations of bottom water dissolved oxygen concentrations, with average foraminiferal Mn/Ca ratios three times higher during low‐oxygen period than in winter. In the uppermost centimeters of sediment, wider ranges of foraminiferal Mn/Ca occurred in summer compared to winter ranges. Our results imply that this epifaunal benthic foraminiferal species could serve as a useful benthic monitor with the Mn/Ca ratios representing a reliable proxy of hypoxia in the past.

Dissolved oxygen variability of Indonesian seas over decades as detected by satellite remote sensing

The dissolved oxygen (DO) decrease in the ocean is a notable issue because of its potential impacts on marine biogeochemical cycles and ecosystem services. Satellite remote sensing application to support in-situ measurement is a time and cost-saving on wide scales DO monitoring. This study aims to determine the DO variability from 1993 to 2020, identify the potential areas to experience deoxygenation, and investigate the correlation between DO and other ocean parameters in Indonesian seas. The validation between in-situ and satellite-derived DO shows the determination coefficient of 0.73, indicating the satellite dataset reliability for the entire analysis. The multiple regression analysis among the long-term satellite-derived ocean parameters shows that the in-situ DO can be estimated by the combination of the potential temperature, total chlorophyll-a, and salinity. The potential temperature was statistically identified as the parameter with the highest correlation and influence on DO. The results of DO variability analysis show the overall decreasing trend with significant decreases in 1998, 2010, and 2016. There is a distinct difference in DO’s seasonal patterns in the southwestern and northeastern regions. The potential of ocean deoxygenation is detected in western Sumatra waters and the Arafura Sea at the 200–1,000 meters depth.

A computer-based tool to simulate raceway photobioreactors for design, operation and control purposes

Microalgae cultivation in photobioreactors is a complex process where environmental, biological and/or design variables affect the system. Due to this great complexity, it is important to have tools to analyze the system in a clear and simple way. This work presents a new software tool for the simulation of microalgae production in raceway photobioreactors. The tool allows to simulate the process on different seasons of the year, using different microalgal strains and considering different reactor designs. Moreover, open-loop and closed-loop control strategies can be analyzed, such as On/Off or PI controllers for pH and dissolved oxygen variables, in order to increase the system productivity. Thus, estimations for productivity and gas consumption are also given to compare several simulated scenarios. Real data for the weather variables are taken from a semi-industrial raceway to be used as inputs to the simulation tool.

Dissolved oxygen variation in the North China Plain river network region over 2011–2020 and the influencing factors

The 2011–2020 data of dissolved oxygen (DO) and nutrients—total phosphorus (TP), solution reactive phosphorus (SRP), total nitrogen (TN), ammoniacal nitrogen (NH3–N), and chemical oxygen demand (CODCr)—in the North China Plain river network region were analyzed. Moreover, the DO variation trend and the influencing factors were investigated. The results showed that between 2011 and 2020, the DO concentration steadily increased from <3 mg L−1 to >5 mg L−1. Negative relationships were observed between the DO and CODCr (R = −0.34, p < 0.01), TN (R = −0.41, p < 0.01), NH3–N(R = −0.40, p < 0.01), TP (R = −0.28, p < 0.01), and SRP (R = −0.19, p < 0.01), indicating that a reduction in the nutrient input promoted the increase in the DO concentration in the past decade. The DO concentration in the rainy season was lower than that in the dry season, which suggests that nonpoint-source pollution caused by heavy storm runoff was the main factor affecting the water quality. The average DO concentration in the suburban rivers (4.88 mg L−1) was higher than that in urban rivers (3.41 mg L−1). Furthermore, comprehensive analysis indicated that the loss of riparian buffer, river solidification, pollutant input, and sluice dam operations are the main factors affecting DO concentration decrease and water quality deterioration. Finally, measures for water improvement and DO recovery in the river network region, namely pollutant input control, reduction in the impact of water conservancy projects, and river ecological restoration, were examined. Overall, this study shows that water quality has improved over the past decade. The study results provide a reference for the continuous improvement of water quality and the continuous recovery of DO in river network regions.

Observations of dissolved oxygen variability and physical drivers in a shallow highly stratified estuary

Evaluating the role of long-term anthropogenic changes on dissolved oxygen (DO) in coastal systems is often a challenge due to variability in a system and measurement sensitivity to spatial and temporal trends. Physical processes can significantly modify the DO variability and act as a major driver of uncertainty when it comes to quantifying the bio-chemical changes associated with the DO budget. Stratification has been observed to be a dominant factor in controlling the hypoxic conditions that can develop in Mobile Bay, AL, a shallow highly stratified estuary prone to episodic hypoxia. Using CTD transect data and long-term water quality monitoring stations the variability of DO was examined throughout Mobile Bay. This study examined the physical drivers of these trends and highlighted the role stratification, temperature, and advection play in driving the bay wide variability. Under stable conditions, the spatial trend in Mobile bay will reflect the along-estuary gradient. When stable conditions don't occur due to random episodic mixing events and cross estuary exchange, the Bay can be driven by a number of factors: the along-estuary gradients, time since the previous mixing event, level of stratification, and biochemical oxygen demand. The combination of these elements provides an increased understanding of the complex dynamics driving low DO in this system. Long-term trends show the DO is decreasing in Mobile Bay based on changes in DO in the shipping channel, northern region, and Bon Secour region.

DIY OCEANOGRAPHY • A Simple and Inexpensive Method for Manipulating Dissolved Oxygen in the Lab

Changes in dissolved oxygen concentration can cause dramatic shifts in chemical, biological, and ecological processes in aquatic systems. In shallow coastal areas, this can happen on short timescales, with oxygen increasing during the day due to photosynthesis and declining at night due to respiration. We present a system controlled by an Arduino microprocessor that leverages the oxygen-consuming capacity of sediments to manipulate dissolved oxygen in an aquarium tank to planned concentrations. With minor adjustments to the Arduino code, the system can produce a variety of dissolved oxygen patterns, including a diel cycle. Designed to be user-friendly and scalable if needed, the system uses easily acquired, low-cost electronic and aquarium components. Its simplicity and accessibility permit deeper exploration of the effects of dissolved oxygen variability in aquatic systems, and the use of Arduino code and basic electronics makes it a potential tool for teaching experimental design and instrument fabrication.