Coralmania by Eric Borneman

The Need to Breathe in Reef Tanks: Is it a Given Right?

This article deals with oxygen dynamics in some reef aquaria. It is the first of what will be a two-part set of articles, not only because of the length and breadth of material covered, but also because I have not yet finished collecting and analyzing all of the data for the second article. I began to think about oxygen in reef aquaria quite some time ago and only recently began to explore it in some depth. I have been investigating the possibility of hypoxia acting as a trigger for cellular death in corals in my own work. During the course of background searches and the use of a chambered aquarium I designed to test hypotheses regarding the effects of hypoxia on corals, I became interested in the issues from an aquarist's viewpoint.

Certainly there has been some discussion regarding oxygen in aquaria over the years. During a recent trip to Atlanta for Saltwater U. (, my hosts generously gave me a set of antique aquarium magazines. In the September 1932 issue of The Aquarium, an article was devoted solely to the topic of oxygen in aquaria (Timm, 1932). Typically, the first time aquarists become concerned with oxygen is when shipping and transporting fish from either their home on tropical reefs, or from a livestock vendor. Packing fish and coral in bags with a topping of pure oxygen is testimony to the concern with oxygen levels in small volumes of water. If fish are packed with only air, the usual drill is to drive home as quickly as possible to get the new stock into quarantine or an aerated tank as quickly as possible. This is somewhat less of a concern with corals and other invertebrates, but the situation remains similar. Those who have tried to ship livestock without a topping of pure oxygen will recognize the disturbingly high mortality that can occur during transit.

In terms of aquariums, I had always heard that using air bubblers kept tank water oxygenated. I believed it, too, until someone pointed out that the diffusion across an air bubble as it rises to the surface and breaks is so low as to be negligible. It has also been pointed out that most of the gas exchange occurs across the water's surface, and bubbling or any type of surface stirring that mixes the water and creates a greater surface area for gas exchange is far more effective than merely bubbling air into a tank using an airstone and air pump. Many aquarists also have stated that protein skimming dramatically increases "aeration," and that low oxygen levels in highly skimmed reef aquaria are rarely a concern. I was always unsure how much oxygenation occurred in a closed chamber with most of the bubbles being forced to the surface prior to their return to the tank. Still, it made sense, too, that water in contact with such a frothy mix must certainly gain oxygen during its pass through a foam fractionation device. In fact, so much banter exists around this topic that I began to wonder if anyone had really measured reef tanks' oxygen levels, or if any data really existed to support any of the statements commonly made and accepted.

The Oxygen Environment of Earth

Complex life on our fragile planet Earth would probably cease to exist, or may never have come to exist at all, without oxygen. About two billion years ago, a significant buildup of oxygen began to occur after life first arose approximately 3.85 billion years ago. This buildup of oxygen was largely due to primitive photosynthetic cyanobacteria. A period of intense diversification and distribution, known as the Cambrian Explosion, began 542-544 million years ago when the Earth contained a level of oxygen roughly equivalent to today's atmosphere, representing approximately 21% of the gases present.

Bacteria were probably the first life forms. Since bacteria can be anaerobic or chemoautotrophic [def = being autotrophic and oxidizing an inorganic compound as a source of energy], many do not require oxygen at all and to some bacteria oxygen is a death sentence. In the early years of earth's history, there was no oxygen. Without oxygen, there is no ozone. Without ozone, high levels of UV radiation, a potent mutagen, would strike the planet. At some point, it is believed that a mutation for oxygen tolerance arose in the archaebacteria that allowed for the evolution of cyanobacteria, sometime in the first billion years of life. The cyanobacteria were able to take up water and release oxygen as a metabolic waste product.

As oxygen levels rose in a previously anaerobic world, adaptations had to occur in order for the anaerobes to persist in aerobic environments. At some point, a bacterium engulfed another bacterium, and the relationship became one of symbiosis. The primitive relationship eventually led to eukaryotic life and the engulfed microbe became the oxygen-utilizing powerplant of cells, the mitochondrion. Mitochondria still retain their own DNA, even after a billion years of evolution. The choroplasts of plants are also a result of an early invasion and subsequent symbiosis that allowed for photosynthesis.

In time, oxygen levels in the oceans rose, as gases diffused between air and liquid. This happened very slowly, and primarily only after all of the exposed rock on the planet was oxidized. The oceans are still not oxygen-saturated and there is a lot of evidence that until the mid to late Mesozoic they were only oxygenated to a depth of a couple of hundred meters. Liquids, including water with its relatively high solubility, can, however, dissolve only so much gas. Compared to the atmosphere, seawater is oxygen-poor.

Oxygen in the Oceans

In the oceans, oxygen exists in seawater as a result of exchange at the air/water interface and due to photosynthesis (primarily by marine plants, algae and phytoplankton). The measure of primary productivity is roughly approximate to photosynthesis, and is measured in gC/m2/yr (where C = carbon). However, direct measurements of photosynthesis are often taken by measuring in the currency of oxygen using bottles that are either exposed to light or left in the dark. This is important, because what such experiments measure are the relative rates of respiration versus photosynthesis. The former consumes oxygen; the latter produces it. This small-scale measurement is a microcosm of what happens in the ocean.

The amount of gas that seawater will hold is a reflection of several physico-chemical factors. The first is the partial pressures of the gases in the atmosphere (largely a function of atmospheric pressure), which is roughly 14.7 psi (760 Torr) at sea level. Oxygen partial pressure at sea level is roughly 150 Torr. The second factor is temperature; solubility of oxygen is inversely proportional to temperature. Cooler water can dissolve more oxygen than warmer water, a fact that is occasionally mentioned when discussing optimal temperatures for aquaria. A third factor involves salinity, again representing an inverse relationship. Seawater of lower salinity can hold more oxygen than a similar seawater sample of higher salinity. This attribute is the longstanding rationale for keeping marine fish in water with a lower salinity than seawater (e.g., 29-30 ppt or 1.022 SG), the argument being that reduced salinity makes it "easier for the fish to breathe." This argument, while perhaps true in terms of oxygen, is preposterous. Marine fish, while some are tolerant of variable salinity, have evolved to exist in a marine environment where the salinity is usually between 34-36ppt. Long-term exposure to hyposalinity may have negative effects that outweigh the benefits afforded by increased oxygen solubility, a point which will become increasingly apparent throughout this and the next article. A fourth factor is pressure, with deep water under enormous pressure able to dissolve more gas than surface waters. Since most of our coral reef species are from relatively shallow waters, and our aquaria are hardly under great pressure (though our floorboards might be), this factor may be less important in aquaria than in the ocean.

Salinity (ppt)
Temp (°C)
Table 1. Oxygen saturation levels in water (mg/l) at different salinities and temperatures (from Adey and Loveland, 1991). Highlighted cells represent those parameters representative of most reef aquaria.

Other factors influence seawater's oxygen content. The amount of gas exchange occurring at the air/water surface interface is contingent on the properties of both air and water. Air circulation correlates positively with increased dissolution, especially at the surface. Wind, for example, creates higher oxygen values in surface waters than occur during periods of calm weather. Water circulation is equally important, and the mixing that occurs involves waves, currents, upwellings and circulation patterns. Another factor, perhaps obviously, is the production of oxygen by photosynthesis, which is now viewed as a major contributor to oxygen saturation states in the waters of the photic zone. Obviously, respiration is also important in terms of oxygen consumption. The depth where oxygen consumption by respiration equals oxygen production by photosynthesis is called the compensation depth. Because of mixing, there is a depth slightly below the compensation depth, called the critical depth, where total plant production equals total plant respiration. Below this depth is the aphotic zone, where there can still be abundant life despite the lack of photosynthesis. Animals, such as fishes and plankton, may vertically migrate seasonally or diurnally to take advantage of the productive areas above them.

Below the critical depth, however, oxygen levels continue to drop until a point is reached at approximately 500-1000m called the oxygen minimum layer, or oxygen minimum zone. Here, oxygen levels are at their lowest as the organisms are respiring, exhausting the oxygen replaced by air and photosynthesis. Some organisms do live in the oxygen minimum zone, although they have evolved special adaptations to exist in the hypoxic conditions found there. Below the oxygen minimum zone and including the very deepest depths, oxygen levels again increase. This increase, sometimes reaching saturation but rarely reaching levels present in the shallow photic zone, occurs because of the cold temperatures and the relatively low density of deep-sea organisms respiring in the vast volume of deep ocean water. The deep sea's oxygen is provided mainly by downwelling currents near the poles where cold surface water sinks to the deep ocean where there is less mixing.

Figure 1. The oxygen minimum zone of the tropical eastern Pacific (dashed line) and tropical Atlantic (solid line), modified from Nybakken (2001).

Shallow coral reef waters are often supersaturated with oxygen. This is because of the high productivity resulting from algal and coral photosynthesis, coupled with shallow water and often-strong tropical air and water circulation patterns. Despite the general notion that shallow tropical waters are well oxygenated, it has long been known that areas of lower oxygen exist, especially where there is comparatively little water exchange, such as lagoons, some reef flats cut off from circulation at low tide, and even shallow reefs, in general, during periods of doldrums.

Oxygen and Coral Reef Fishes

In one of the more interesting studies I have found on the subject, Nilsson et al. (2004) examined hypoxia tolerance in the coral-dwelling goby, Gobiodon histrio. More on G. histrio can be found in another article in this magazine by Henry Schultz. This species spends its whole life inside branches of Acropora, showing a preference for A. nasuta. The authors suspected that the oxygen environment of the coral could be highly variable, being exposed to air at low tide, and also possibly becoming hypoxic at night. They specifically mention calm nights when the respiration of the coral and associated organisms and lack of water mixing created a hypoxic environment for the goby (discussed in more detail below). In fact, they found that such conditions did exist and that G. histrio was extremely hypoxia tolerant, only showing equilibrium loss at water oxygen levels of approximately 3% of air saturation.

Nilsson and Östlund-Nilsson (2004) expanded this report by examining hypoxia tolerance in 31 species (7 families) of fish in the shallow lagoon of Lizard Island on the Great Barrier Reef. They found that hypoxia tolerance was variable but widespread, with all species maintaining normal respiration rates in water down to 20-30% of air saturation levels, with most species unaffected until about 10% of air saturation levels. In an earlier work, they had found two species of cardinalfish (Apogon leptacanthus and A. fragilis) to have a critical oxygen level of approximately 20% of air saturation, at which point they begin to rely on anaerobic metabolism. The effect on mouth-brooding and on cardinalfish is expanded in Östlund-Nilsson and Nilsson (2004). Aquarists who are breeding or considering breeding cardinalfish should read that article, and perhaps pay attention to this article and next month's article, as well. Blennies and gobies, in general, were the most tolerant of hypoxia. Surprisingly, damselfishes were a mixed lot, with some showing much higher tolerance than others. Some damselfish were more sensitive than even the cardinalfish to hypoxia. It should be noted that this study used small species or those that seek refuge within corals or the coral reef framework at night. No mention is made of fishes that tend to remain in more well-oxygenated waters at night, most notably for aquarists, the surgeonfishes. The general groups examined were the cardinalfish (9 species), the damselfish (14 species), the gobies (3 species), the blennies (2 species), the filefish (one species), the breams (1 species) and the wrasses (2 species). Hypoxia tolerance has also been found in the epaulette shark (Routley et al. 2002).

Oxygen and Corals

The investigation of hypoxia and corals is only slightly greater in scope than that of fishes, and is also comparatively recent. As students of coral reef literature may find unsurprising, C.M. Yonge investigated this aspect of coral physiology in the 1930's along with, it seems, virtually every other modern topic of coral study (Yonge et al. 1932) After a forty year hiatus, it was mostly assumed (correctly) that the zooxanthellae provided copious oxygen to corals and similar anthozoans that are generally incapable of moderating flux across their surfaces. The existence of diffusive boundary layers was known to impede gas exchange across surfaces (Dennison and Barnes 1988, Patterson and Sebens 1989), and the first study to suggest the existence of a hypoxic environment at the tissue surface was for an anemone (Shick and Brown 1977). Studies were extended to zoanthids and corals by Shick (1990), who demonstrated that various species of anthozoan polyps may be diffusion limited with respect to gas exchange, and that water flow across a colony and the production of oxygen by zooxanthellae resulted in reduced diffusion limitation and a potentially hyperoxic environment in light conditions. Edmunds and Davies (1988) had previously shown Porites porites to increase its respiration rate to a mean of 39% above its pre-illumination rate within three hours of exposure to light equivalent to light levels at 10m depth, and increased 58% above pre-illumination respiration rates after 80 minutes when nubbins were exposed to subsaturating light levels. Thus, water flow and photosynthesis were beginning to be seen as the primary modulators of oxygen uptake in corals, similar to the larger scale processes occurring on reefs.

Kuhl et al. (1995) used oxygen microsensors to measure the oxygen, pH and light in tissue and near the boundary layer of several species of Favia and Acropora in both light and dark conditions. They found that intracellular oxygen levels were hyperoxic (250% of air saturation values) after a few minutes of exposure to sunlight. They also found that upon initiation of dark conditions, oxygen at the tissue surface was less than 2% of air saturation values within five minutes. They discovered a boundary layer from 200-300 µm in thickness with flow velocities from 5-6cm/sec, and from 500-600 µm in thickness with flow velocities from 1-2 cm/sec. Oxygen in coral tissues decreased with the increasing thickness of boundary layers, and became anaerobic in stagnant water. Other oxygen studies using microsensors confirmed the findings that coral tissue becomes hypoxic at night (Shasar and Brown 1992, Shasar et al. 1993, Jones and Hoegh-Guldberg 2001).

A few years later, Gardella and Edmunds (1999) measured oxygen levels directly adjacent to tissue in the coral Dichocoenia stokesii, and obtained similar results, with hyperoxia occurring during the day, and hypoxia to anoxia at night, and they also made measurements at different flow speeds. In general, higher flow speeds reduce the degree of hypoxia within coral tissues at night, but it was found that oxygen production by zooxanthellae was the major factor providing oxygen to the coral polyps and that respiration is oxygen limited in low light and in darkness. Similar findings have been reported for the epilithic algal community of coral reefs (Larkum et al. 2003).

The most recent paper on this subject (Ulstrup et al. 2005) is perhaps the most telling. Noting the possibility of anaerobiosis, they used Pocillopora damicornis as a test subject and found that oxygen levels at the tissue surface fell from 100 to 35+/- 5% air saturation levels within ten minutes of darkness and that hypoxic environments lower than this level impact Photosystem II, potentially leading to bleaching or higher bleaching susceptibility. These findings were suggested in earlier works by Nakamura and van Woesik (2001) and Nakamura et al. (2003). Finally, Nilsson and team members in the work mentioned above on of gobiid fishes took oxygen level readings within the branches of Acropora species and also determined that a moderately severe hypoxic environment does, in fact, occur within the colony branches at night.

Oxygen and Aquaria

The most complete information I have found regarding oxygen levels and aquaria is in the book, Dynamic Aquaria (Adey and Loveland 1991). They utilized data from the Smithsonian mesocosm, showing that that the system closely approximated oxygen levels found on coral reefs. However, the Smithsonian mesocosm utilized algal turf scrubbers operated on a reverse daylight period to help stabilize oxygen and pH in the system. Thus, photosynthesis was occurring somewhere in the multi-compartment mesocosm throughout each twenty-four hour period. Other sources of aquarium information are, typically, either completely lacking in data or based on conjecture or anecdotal observations.

Because of the insufficiencies that existed in understanding oxygen dynamics in closed systems, and because our lab has a high quality field oxygen meter and probe (YSI-58), I embarked on a mission of discovery that would hopefully shed some light on oxygen levels in various closed systems, and attempt to tease out those factors which play significant or insignificant roles in the oxygenation of our aquaria. These findings will be communicated in my article next month, and will also be the subject of my presentation at the upcoming International Marine Aquarium Conference in Chicago in late June (

If you have any questions about this article, please visit my author forum on Reef Central.

References and Additional Reading

Adey WH, Loveland K. 2001. Dynamic Aquaria: building living ecosystems. Academic Press, New York: 185-192. Dennison WC, Barnes DJ. 1988. Effect of water motion on coral photosynthesis and respiration. J Exp Mar Biol Ecol 115: 62-77.

Gardella DJ, Edmunds PJ. 1999. The oxygen microenvironment adjacent to the tissue of the scleractinian Dichocoenia stokesii and its effects on symbiont metabolism. Mar Biol 135: 289-295

Jones RJ, Hoegh-Guldberg O. 2001. Diurnal changes in the photochemical efficiency of the symbiotic dinoflagellates (Dinophyceae) of corals: photoreception, photoinactivation and the relationship to coral bleaching. Plant Cell Environ 24: 89-99

Kuhl M, Cohen Y, Dalsgaard T, Jorgensen BB, Revsbech NP. 1995. Microenvironment and photosynthesis of corals studied with microsensors for O2, pH and light. Mar Ecol Prog Ser 117: 159-172.

Larkum AWD, Koch E-MW, Kuhl M. 2003. Diffusive boundary layers and photosynthesis of the epilithic algal community of coral reefs. Mar Biol 142: 1073-1082.

Nakamura T, van Woesik R. 2001. Water-flow rates and passive diffusion partially explain differential survival of corals during the 1998 bleaching event. Mar Ecol Prog Ser 212: 301-304.

Nakamura T, Yamasaki H, van Woesik R. 2003. Water flow facilitates recovery from bleaching in the coral Stylophora pistillata. Mar Ecol Prog Ser 256: 287-291.

Nilsson GE, Östlund-Nilsson S. 2004. Hypoxia in paradise: widespread hypoxia tolerance in coral reef fishes. Proc Roy Soc Lond (B) 270: S30-33.

Nilsson GE, Hobbs J-P, Munday PL, Östlund-Nilsson S. 2004. Coward or braveheart: extreme habitat fidelity through hypoxia tolerance in a coral-dwelling goby. J Exp Biol 207: 33-39.

Nybakken JW, ed. 2001. Marine Biology: an ecological approach. Benjamin Cummings, San Francisco: 6.

Östlund-Nilsson S, Nilsson GE. 2004. Breathing with a mouth full of eggs: respiratory consequences of mouthbrooding in cardinalfish. Proc Roy Soc Lond (B) 271: 1015-1022.

Patterson MR, Sebens KP. 1989. Forced convection modulates gas exchange in cnidarians, Proc Natl Acad Sci USA 86: 8833-8836.

Routley MH, Nilsson GE, Renshaw GMC. 2002. Exposure to hypoxia primes the respiratory and metabolic responses of the epaulette shark to progressive hypoxia. Comp Biochem Physiol 131A: 313-321.

Shasar N, Cohen Y, Loya Y. 1993. Extreme fluctuations of oxygen in diffusive boundary layers surrounding stony corals. Biol Bull 185: 455-461.

Shasar N, Stambler N. 1992. Endolithic algae within corals: life in an extreme environment. J Exp Mar Biol Ecol 163: 277-286.

Shick JM. 1990. Diffusion limitation and hyperoxic enhancement of oxygen consumption in zooxanthellate sea anemones, zoanthids and corals. Biol Bull 179: 148-158.

Ulstrup KE, Hill R, Ralph PJ. 2005. Photosynthetic impact of hypoxia on in hospite zooxanthellae in the scleractinian coral Pocillopora damicornis. Mar Ecol Prog Ser 286: 125-132.

Shick JM, Brown WI. 1977. Zooxanthellae-produced O2 promotes sea anemone expansion and eliminates oxygen debt under environmental hypoxia. J Exp Zool 201: 149-155.

Timm, JA. 1932. Science and our hobby - Part II. The Aquarium 1: 131-132.

Yonge CM, Yonge MJ, Nicholls AG. 1932. Studies on the physiology of corals VI. The relationship between respiration in corals and the production of oxygen by their zooxantellae. Sci Rep Great Barrier Reef Exped 1: 213-251.

Reefkeeping Magazine™ Reef Central, LLC-Copyright © 2008

The Need to Breathe in Reef Tanks: Is it a Given Right? by Eric Borneman -