Dissolved oxygen is the main limiting factor in pond aquaculture production, yet the relative contributions of its various sources and sinks remain poorly quantified in eutrophic earthen pond systems. Photosynthetic oxygen production, surface exchange, sediment oxygen demand and water respiration were assessed in greenwater earthen fishponds using a combination of in situ measurements, bottle incubation experiments and laboratory batch trials. Dissolved oxygen levels followed a diel cycle with near-anoxic conditions before sunrise and hyperoxic conditions in the late afternoon. Nocturnal respiration rates exhibited oxyconformity, declining >5-fold from hyperoxic to moderately hypoxic conditions. Photosynthetic oxygen production was confined to a shallow photic zone, and gross primary production rapidly declined with depth, becoming negligible at 50 cm, with a compensation depth of 21 cm. Photosynthesis increased O2 saturation by up to 16% min−1 at high light intensity, but oxygen evolution was suppressed under hyperoxic conditions regardless of irradiance. Sediment oxygen consumption ranged from 110 to 288 mg O2 m−2 h−1 and was correlated with sediment organic carbon content. Oxygen consumption of the pond water was the dominant nocturnal oxygen sink. A theoretical oxygen budget for a 1000 m2 pond indicated that daytime photosynthesis generated sufficient oxygen to meet pond requirements, but lack of mixing, atmospheric losses and nocturnal respiratory demand deplete all available oxygen, resulting in pre-dawn anoxia. These findings underscore the importance of phytoplankton biomass management for high light penetration, and the potential benefits of installing mixing devices to eliminate stratification.

The African catfish, Clarias gariepinus, is an air-breathing fish that possesses well-developed gills and a specialised air-breathing organ composed of supra-branchial chambers with arborescent organs. The different types of air-breathing behaviours exhibited by C. gariepinus were documented, and air-breathing frequency (fAB) was assessed in a series of trials where individual C. gariepinus (average mass: 300 g; n = 60) were subjected to either avian predation threats, changes in water turbulence and flow, water depth, or environmental noise. Diurnal changes in fAB were also monitored. Three distinct air-breathing behaviours were observed, distinguishable by the pattern of bubble release through the opercular slits or from the buccal cavity during the ascent to perform air-breathing. Photoperiod clearly modulated fAB in C. gariepinus, as nocturnal fAB was almost twice as high as daytime fAB. Water depth (27, 54, and 81 cm) had a significant effect on fAB (p = 0.021), with fAB being inversely proportional to water depth. Similarly, flow velocity and turbulence had significant effects on fAB. Under low turbulence conditions, C. gariepinus performed 1.65 ± 0.18 air-breaths min−1, which was 2.3 times higher than fAB in water without turbulence. In the presence of an avian predator, fAB was nearly halved and significantly reduced compared to when the predator was absent. While the concentrations of dissolved gases are known to be the main drivers of air-breathing in many air-breathing fish species, including C. gariepinus, the data generated from this study highlight the sensitivity of the species to hydrodynamic alterations and aerial predation threats.

The global shrimp processing industry generates substantial amounts of solid waste (head, abdominal exoskeleton, and tail), with a considerable part currently being disposed of in landfills. It holds significant potential as an alternative ingredient in aquaculture feed due to its relatively high crude protein content, balanced amino acid profile, and presence of bioactive compounds. However, one of the main challenges with shrimp processing waste is its rapid spoilage. Consequently, shrimp solid processing waste must undergo further refinement to produce shrimp-derived products suitable as aquaculture feed ingredients, such as meal, hydrolysate, or silage. This literature review describes the nutritional value, applications, challenges, and prospects of these shrimp-derived products in aquaculture feed. Among the investigated shrimp-derived products, shrimp hydrolysate has the highest nutritional value considering the high crude protein content, balanced amino acid profile, low chitin content, and low ash content. However, producing shrimp hydrolysate requires extensive processing, which can be costly, limiting its applications to high-value aquaculture species. On the other hand, shrimp meal and shrimp silage, which require less energy-intensive processing, may be more suitable for lower-value aquaculture species that naturally consume feeds high in ash and chitin. The prospects for using shrimp-derived products in aquafeed are promising, with advances in processing technologies showing potential to reduce costs, improve nutritional value, and enhance product quality and safety. Ultimately, shrimp-derived products could replace current aquafeed ingredients while simultaneously utilizing current shrimp solid waste streams, provided that quality and safety measures are carefully considered.

The African catfish (Clarias gariepinus) is an extensively researched species in various aquaculture studies, but knowledge about its air-breathing behaviour is generally limited. This study aimed to evaluate changes in air-breathing behaviour in response to environmental stressors by investigating the air-breathing frequency (fAB) under varying O2 tensions, dissolved carbon dioxide (CO2) levels, and temperatures. Varying O2 tensions had a significant influence on the air-breathing behaviour of C. gariepinus. Under normoxic O2 conditions, (15.9 kPa O2), fAB was 0.47 ± 0.11 breaths min−1. Under moderate (7.9 kPa) and severe O2 tensions (0.8 kPa), mean fAB were significantly elevated to 1.40 ± 0.25 and 1.23 breaths min−1, respectively. Increasing water temperatures also significantly increased fAB up to temperatures greatly exceeding thermal optima where there were significant declines in fAB. Subjecting fish to thermally stratified water columns also resulted in significant adjustments in air-breathing behaviour. Under isothermal conditions at 26 ℃, C. gariepinus performed 0.43 ± 0.07 air-breaths min−1. Fish sequentially increased fAB as the temperature of the top water layer increased. C. gariepinus responded to the elevated CO2 levels by sequentially upregulating fAB. There was a threefold increase in fAB at 40 mgL−1 CO2 relative to normocapnic control conditions. This study has demonstrated that the air-breathing behaviour of C. gariepinus is complex and largely driven by fluctuations in dissolved gas concentrations and water temperature. These insights are especially relevant in the context of the increments in surface water temperatures, which typically exacerbate hypoxic conditions in many aquatic ecosystems.

Chemical disinfection in a recirculating aquaculture system (RAS) may affect biofilm-associated bacteria and the nitrification performance in the biofilter units. The biofilm response to chemical disinfectants in RAS remains unclear, but it can be understood using methods to quantify biofilm activity. Here, we compared the effects of two disinfection strategies, continuous ozone at 0.06 mg/L Cl2 equivalent (Ozone group) and peracetic acid (PAA) at 1 mg/L (PAA group) to control group without disinfectant, on biofilm activity on biofilter elements from nine identical experimental RAS with Atlantic salmon parr (Salmo salar) during a four-week trial. Biofilm activities on biofilter elements from the three groups were examined by measuring oxygen consumption rates (OCR) following sequential spiking with pure tap water, and tap water spiked with nitrite or ammonium, as well as oxygen release rate (kor) following hydrogen peroxide (H2O2) addition, to estimate metabolic activities of biofilm related to endogenous respiration and substrate turnover. The results show that the applied PAA dose increased the endogenous respiration activity of biofilm by 39-133%, stimulated the rate of biofilm enzymatic decomposition of H2O2 by 135%, and partially impaired the biofilm metabolism for nitrite oxidation (decrease by 36%), resulting in a significant nitrite accumulation (rise by 59%) in the cultured water, compared to control over experimental period. Ozone treatment caused an enhanced endogenous respiration activity of biofilm at the beginning of the experiment (increase by 45-74%), but dropped to control levels at the end of the experiment. The results indicate that chronic exposure to PAA can alter the metabolic state of biofilm, which can have consequences for biofilter functions, while chronic exposure to ozone improved water clarity without compromising the metabolic status of biofilm. The investigations provided insights into biofilm response to chemical disinfectant in RAS, which would benefit the optimization of an effective and safe use of disinfectants in RAS.

Peracetic acid (PAA) is an effective disinfectant in aquaculture systems to reduce pathogen loads and improve water quality. However, its effectiveness in disinfecting biofilm in recirculating aquaculture systems (RAS) and resetting biofilters between productions remains unknown. This study evaluated the effects of acute PAA exposure on biofilter biofilms from freshwater RAS. Identical types of bioelements were collected from a pilot-scale RAS (without prior PAA treatment) and a commercial RAS (with PAA treatment), and exposed to PAA concentrations of 0, 1, 2, 4, 8, and 16 mg/L for 1 h. Microbial activity and viability of the exposed biofilms were evaluated using respirometry and flow cytometry. Results showed dose-dependent inhibition of biofilm activity and viability in the pilot-scale RAS. Nitrite oxidation was the most sensitive process to PAA, with an IC50 of 1.27 mg/L (the concentration at which PAA inhibited biofilm metabolic activity by 50 %), followed by ammonia oxidation (IC50 = 1.59 mg/L) and endogenous respiration (IC50 = 2.67 mg/L). Microbial activity linked to H2O2 decomposition was least affected (IC50 = 4.68 mg/L). Live cell counts decreased from 9.1 × 107 counts/cm2 to 2.4 × 107 counts/cm2 of bioelement surface, with dead cells proportion increasing from 15 % to 54 %. In contrast, biofilter biofilms from the commercial RAS exhibited significantly lower sensitivity to PAA dosage, with reductions in nitrite oxidation (39 %) and ammonia oxidation (51 %) observed only at 16 mg/L compared to control. These findings suggest that routine PAA exposure, as part of the other operating conditions on the commercial RAS, can enhance the biofilm's sensitivity to PAA. The study provides new insight into the sensitivity of aquaculture biofilm to PAA treatment and its effect on associated microbial processes.

Despite degassing efforts in recirculating aquaculture systems (RAS) and short hydraulic retention time in rearing units, carbon dioxide (CO2) concentrations reach a hypercapnic steady state. Furthermore, as CO2 excretion is correlated with the oxygen consumption of fish and bacteria, CO2 levels in RAS may undergo fluctuations, of greater or smaller magnitude, depending on systems design and operation. Experimental approaches to assess the effects of CO2 on fish usually subject fish to constant CO2 concentrations, which might not reflect rearing conditions on an industrial production scale. Here, we compare the effects of oscillating against constantly elevated CO2 levels on the appetite, growth, feed utilization, and mineral deposition in three separate growth trials. Two trials were conducted in freshwater (FW) and one in seawater (SW). In each trial, rainbow trout (Oncorhynchus mykiss) were subjected to four different CO2 treatments: either constant levels of 10 mg/L CO2 (pCO2 = 3.86 mmHg in FW / 4.57 mmHg in SW), 25 mg/L CO2 (pCO2 = 9.65 mmHg in FW / 11.42 mmHg in SW), fluctuating between these two concentrations over 24 h, or normocapnic control conditions of <= 3 mg/L CO2 (pCO2 <= 1.16 mmHg in FW / 1.37 mmHg in SW) for at least 5 weeks on fixed daily rations of 1.3 % of the tank biomass. In one of the freshwater trials, the diet contained high levels of phosphorus (1.8 %) to assess if elevated dietary phosphorus concentrations promoted mineralization in kidney tissues (nephrocalcinosis) under hypercapnic conditions. In both freshwater trials, fish at all CO2 levels accepted the offered feed, while in seawater daily feed intake was reduced by 35 % at exposure to 25 mg/L CO2 (pCO2 =11.42 mmHg). Despite accepting the full, albeit restricted ration, fish reared in freshwater showed that CO2 affected appetite, evidenced by changes in mRNA expression of appetite-regulating peptides in the hypothalamus and liver of the fish. This finding was confirmed by a maximum voluntary feed intake test, showing that fish consumed less food at higher CO2 concentrations. Despite consuming similar daily ration sizes, the specific growth rate (SGR) and feed conversion ratio (FCR) were significantly affected in the 25 mg/L freshwater group (pCO2 = 9.65 mmHg) but not at 10 mg/L (pCO2 = 3.86 mmHg) or when oscillating between those concentrations. In the seawater trial, however, SGR and FCR of the trout were already significantly reduced at 10 mg/L of dissolved CO2 (pCO2 = 4.57 mmHg) compared to the control group, but even more so in the 25 mg/L CO2 group (pCO2 = 11.42 mmHg). While the CO2 regimes applied in this study and high dietary phosphorous did not result in clear macroscopic pathologies indicative of nephrocalcinosis, there was calcium deposition in trout kidneys was highly elevated in the 25 mg/L group (pCO2 = 9.65 mmHg), indicating the potential onset of this pathology, which was supported by histopathological examinations. Overall, the results of this study show that while fish were affected by CO2 in all trials, the severity depends on water chemistry.

In fish, thermal and hypoxia tolerances may be functionally related, as suggested by the oxygen- and capacity-limited thermal tolerance (OCLTT) concept, which explains performance failure at high temperatures due to limitations in oxygen delivery. In this study the interrelatedness of hyperthermia and hypoxia tolerances in the Nile tilapia (Oreochromis niloticus), and their links to cardiorespiratory traits were examined. Different groups of O. niloticus (n = 51) were subjected to hypoxia and hyperthermia challenges and the O-2 tension for aquatic surface respiration (ASR pO(2)) and critical thermal maximum (CTmax) were assessed as measurement endpoints. Gill filament length, total filament number, ventricle mass, length and width were also measured. Tolerance to hypoxia, as evidenced by ASR pO(2) thresholds of the individual fish, was highly variable and varied between 0.26 and 3.39 kPa. ASR events increased more profoundly as O-2 tensions decreased below 2 kPa. The CTmax values recorded for the O. niloticus individuals ranged from 43.1 to 44.8 degrees C (Mean: 44.2 +/- 0.4 degrees C). Remarkably, there was a highly significant correlation between ASR pO(2) and CTmax in O. niloticus (r = -0.76, p < 0.0001) with ASR pO(2) increasing linearly with decreasing CTmax. There were, however, no discernible relationships between the measured cardiorespiratory properties and hypoxia or hyperthermia tolerances. The strong relationship between hypoxia and hyperthermia tolerances in this study may be related to the ability of the cardiorespiratory system to provide oxygen to respiring tissues under thermal stress, and thus provides some support for the OCLTT concept in this species, at least at the level of the entire organism.

The landlocked Atlantic salmon population “bleke” faces extinction due to environmental acidification (EA) and hydropower expansion in the Norwegian river Otra. Despite of restoration, unexpected mortality has been reported for this population, possibly due to gas bubble trauma (GBT) from gas supersaturation (GSS) downstream of hydroelectric plants, or EA induced aluminum toxicity. In this study, we applied the allostasis concept to investigate interactions between EA and GBT. This concept comprises additive effects of stressors, which can lead to allostatic overload. Stress coping mechanisms become maladaptive in such situations, which can be indicated by an inability to mount a proper cortisol response in fish. Fish were exposed to sublethal levels of simulated EA (SEA), GSS (a total gas pressure; TGP; of 110%) or a combination of these stressors for six days. Effects on allostatic load were subsequently investigated by assessing the cortisol response to an acute stress test. SEA increased cortisol responsiveness and GSS induced clinical signs of GBT, but no interacting effects between GSS and SEA were observed. This suggests that that 110% TGP did not have an additive effect on the allostatic load imposed by SEA.

High levels of dissolved carbon dioxide (CO 2 ) occur nightly in earthen ponds characterized by high respiration rates. Exposure to high CO 2 conditions (hypercapnia) leads to acidosis in fish, which can be compensated by an accumulation of HCO 3 - to recover intra- and extracellular pH levels, with a capacity that appears to be speciesspecific. For Nile tilapia, a freshwater tropical teleost traditionally produced in earthen ponds, little information is available on the tolerance to dissolved levels of CO 2 and associated acid-base disturbances. Here, we investigated first the effects of acute and progressively increasing CO 2 , from normocapnic conditions to 60 mg CO 2 L -1 , on oxygen uptake rates (MO 2 ). This was followed by exposure to three concentrations of CO 2 ; 10, 30, and 60 mg L -1 (equivalent to p CO 2 of 5.4, 16.2, and 32.4 mmHg) against a normocapnic control ( p CO 2 0.3 mmHg), to investigate acute (1 h) or prolonged (24 h) effects on standard (SMR) and maximum metabolic rates (MMR), haematology, and extra- and intracellular acid-base status in adult Nile tilapia (mean BM 435 +/- 16 g +/- SE). Acute exposure to hypercapnia led to concentration-dependent decreases in both SMR and MMR. Fish were able to fully or partially recover MMR and metabolic scope (MS) after 24 h, while depression of SMR persisted at all CO 2 levels. Acute exposure to CO 2 caused intra- and extracellular pH levels to decrease by up to 0.5 units in a concentration-dependent manner. Only the lowest hypercapnic treatment ( p CO 2 5.4 mmHg) was able to fully recover within 24 h. Changes in haematological variables appeared minor, being restricted to increasing haematocrit, haemoglobin concentration, and mean cell volume in the highest CO 2 treatments after 24 h exposure. Although the Nile tilapia is generally considered a species able to tolerate poor water quality, the modest or slow acid-base regulation following hypercapnic exposure suggests sensitivity to hypercapnia.

Hydrogen sulfide (H2S) related mass mortality events in land-based recirculating aquaculture systems (RAS) are associated with the sudden release of substantial amounts of H2S, resulting in acutely toxic exposure. However, the production and release of H2S can also occur at slower constant rates, leading to continuous exposure to sublethal H2S concentrations, the effects of which are unknown. Here, we examined growth rates, feed conversion ratios, and apparent nutrient digestibility in post-smolt Atlantic salmon, following a 10-day exposure to one of two sublethal concentrations of H2S, compared against a control group. The H2S concentrations of the low exposure (LE) and moderate exposure (ME) groups throughout the trial were 1.8 +/- 0.8 mu g/L and 4.1 +/- 1.9 mu g/L, respectively. Neither exposure regime affected growth, feed utilization, or the apparent nutrient digestibility. The results suggest that Atlantic salmon tolerate exposure to sublethal H2S concentrations for 10 days, without a discernible effect on the production performance. For aquaculture systems, specifically RAS, this means that while higher concentrations of H2S should be strictly avoided, the presence of low H2S concentrations for a limited period might not be detrimental to fish welfare or production. The RAS environment is complex, and the potential combined effects of H2S and factors pertinent to RAS may influence fish performance differently in a production setting.

Ammonia is a respiratory gas that is produced during the process of protein deamination. In the unionised form (NH3), it readily crosses biological membranes and is highly toxic to fish. In the present study we examined the effects of unionized ammonia (UIA), on the resting oxygen consumption (MO2), ventilation frequency (fV), heart rate (HR) and heart rate variability (HRV) in Nile tilapia (Oreochromis niloticus). Fish were either exposed to progressively increasing UIA concentrations, up to 97 µM over a 5h period, or to a constant UIA level of 7 µM over a 24h period. For both treatment groups resting MO2, HR and fV were recorded as physiological variables. Relative to the control group, the fish groups exposed to the incremental UIA levels did not exhibit significant changes in their MO2, HR and fV at UIA concentrations of 4, 10, 35, or 61 µM compared to control fish. Exposure to 97 µM UIA, however, elicited abrupt and significant downregulations (p < 0.05) in all three responses, as MO2, HR and fV decreased by 25, 54 and 76% respectively, compared to control measurements. Heart rate became increasingly irregular with increasing UIA concentrations, and heart rate variability was significantly increased at 61 and 97 µM UIA. Prolonged exposure elicited significant changes at exposure 7 µM UIA. Standard (SMR) and maximum metabolic rate (MMR) were significantly reduced, as was the corresponding fV and HR. It is evident from this study that Nile tilapia is tolerant to short term exposure to UIA up to 61 µM but experience a significant metabolic change under conditions of prolonged UIA exposures even at low concentrations.

This study evaluated the potential of utilizing a lignocellulosic hydrolysate from brewer's spent grain (BSG) as a substrate for amino acid (AA) production by submerged fermentation. The main objective was to explore AA production from BSG hydrolysate using selected microorganisms. Initially, different microorganisms were screened for their growth on BSG hydrolysate, and selected microorganisms were further investigated for AA production by cultivation in shake flasks and bioreactor. From this screening, Saccharomyces cerevisiae and Corynebacterium glutamicum were selected. C. glutamicum produced alanine, proline, valine, and glycine in shake flasks and bioreactor. Highest alanine production (193.6 & PLUSMN;0.09 mg/L) was found in shake flasks after 30 h while production of proline (22.5 & PLUSMN;1.03 mg/L), valine (34.8 & PLUSMN;0.11 mg/L), and glycine (18.7 & PLUSMN;1.30 mg/L) was highest in bioreactor after 4 h (proline and valine) and 8 h (glycine). To enhance AA production by C. glutamicum, a fedbatch fermentation experiment was performed. Except for glycine, no AAs were produced during the fed-batch phase. S. cerevisiae produced alanine, proline, valine, and glutamic acid in shake flask but not in bioreactor. Highest production of alanine (11.8 & PLUSMN;1.25 mg/L), proline (11.8 & PLUSMN;1.06 mg/L), and valine (4.94 & PLUSMN;1.01 mg/L) was obtained after 50 h while glutamic acid production (66.2 & PLUSMN;0.49 mg/L) peaked after 60 h. This study demonstrates the production of several AAs from BSG by submerged fermentation; however, further optimization is needed to improve the productivity.

Changes in physiological processes can reveal how individuals respond to environmental stressors. It can be difficult to link physiological responses to changes in vital rates such as growth, reproduction and survival. Here, bioenergetics modelling can aid in understanding non-intuitive outcomes from stressor combinations. Building on an established bioenergetics model, we examine the potential effects of parasite infection on growth rate and body condition. Parasites represent an overlooked biotic factor, despite their known effects on the physiology of the host organism. As a case study, we use the host-parasite system of Eastern Baltic cod (Gadus morhua) infected with the parasitic nematode Contraceacum osculatum. Eastern Baltic cod have during the past decade experienced increasing infection loads with C. osculatum that have been shown to lead to physiological changes. We hypothesized that infection with parasites affects cod growth negatively as previous studies reveal that the infections lead to reduced energy turnover, severe liver disease and reduced nutritional condition. To test this, we implemented new variables into the bioenergetics model representing the physiological changes in infected fish and parameterized these based on previous experimental data. We found that growth rate and body condition decreased with increased infection load. Highly infected cod reach a point of no return where their energy intake cannot maintain a surplus energy balance, which may eventually lead to induced mortality. In conclusion, parasite infections cannot be ignored when assessing drivers of fish stock dynamics.

Photosynthesis and respiration exert strong opposing effects on dissolved gases in earthen ponds. During daylight this provides oxygen to the water while removing carbon dioxide. At night, fish, plankton, and sediment respiration consumes the available O2 and produces CO2. This leads to daily conditions that fluctuate between being hyperoxic and hypocapnic during the day, while becoming severely hypoxic and hypercapnic at night. The effects of hypoxia on Nile tilapia have been examined in some detail, but the interacting effects of hypoxia and hypercapnia and fluctuating conditions, have not received any attention. Here we evaluated the effects of daily variations in dissolved O2 and CO2 on the appetite and feed utilization in Nile tilapia using groups reared under normoxic -normocapnic conditions (control, C), diurnal hypoxia (HO), diurnal hypercapnia (HC), or combined diurnal hypoxia and hypercapnia (HO x HC) in a digestibility system. We show that hypoxia and hypercapnia exerted strong individual effects on appetite, specific growth rate, and feed conversion, and that simultaneous hypoxia and hypercapnia amplified these effects. Appetite depression induced by nocturnal hypoxia or hyper-capnia was restored at mid-day, but simultaneous exposure resulted in a day-long loss of appetite. Hypoxia and hypercapnia reduced specific growth rates by up to 40% and increased feed conversion ratio by up to 80%, while combined exposure to hypoxia and hypercapnia reduced specific growth rate by >60%. Surprisingly, the di-gestibility of dry matter, protein, and lipid was improved in groups exposed to single or combined diurnal variation in dissolved oxygen and carbon dioxide. It is unknown whether this is the result of the change in feed intake or represents an adaptive mechanism to satisfy an increase energy demand caused by environmental stress. Overall, we conclude that although Nile tilapia is considered resilient to environmental stress, feeding and feed utilization are strongly influenced by daily fluctuations in dissolved gases.

Despite the importance of Atlantic salmon in marine aquaculture production systems, remarkably little is known about the effects of hydrogen sulfide (H2S) on the physiology of the species. In recent years, mass mortalities of Atlantic salmon have been reported in recirculating aquaculture systems (RAS) due to acute H2S exposure. This highlights the importance of obtaining a better understanding of tolerance thresholds and metabolic responses to this toxic gas. The toxicity of H2S is exerted at the level of the mitochondria, where impairment of the enzyme cytochrome c oxidase inhibits cellular respiration. Because H2S depresses oxygen uptake (MO2), intermittent flow-through respirometry, a common method for assessing the metabolic response to various stressors in fishes, is a suitable method to determine concentration thresholds for when H2S affects the metabolism of Atlantic salmon. During exposure trials, 3 size groups (range similar to 100-500g) of fish were acclimated to control conditions to obtain baseline measurements, whereafter they were exposed to progressively increasing H2S concentrations (0.53 +/- 0.14 mu M h-1) until MO2 decreased below the standard metabolic rate or loss of equilibrium occurred, which we considered to be the critical H2S concentration (H2Scrit). Fish were then allowed to recover in H2S free water to determine the excess oxygen consumption (EOC) following H2S exposure. The results show that Atlantic salmon have a lower tolerance to H2S than previously estimated, with a mean H(2)Scrit of 1.78 +/- 0.39 mu M H2S, which was independent of size. During recovery, the estimated EOC greatly exceeded the accumulated oxygen deficit (DO2) in all groups, and the small salmon had a significantly larger EOC. While the magnitude of the EOC was greater for small salmon, it did not differ in duration (recovery time) among the different sizes of fish. The larger EOC showed that H2S exposure had a greater effect on the recovery phase of the small salmon, and exposure to H2S may leave the fish more vulnerable to other stressors post-exposure. This study provides specific values that underline the sensitivity of Atlantic salmon to acute H2S exposure and emphasizes the importance of the aquaculture industry to implement mitigating strategies for the occurrence of H2S at production facilities.