Tonight I'll be presenting a paper from the scientific literature, called "Coral reef mesocosms and microcosms -- successes, problems, and the future of laboratory models," by Luckett, Adey, Morrissey, and Spoon.
Before I get started with the paper, I want to explain how what I'm doing tonight could fit into a broader context, and then I'll give an introduction to some earlier writings of one of tonight's paper's author's, Walter Adey.
A lot of what we reef hobbyists do amounts to rediscovering in an ad-hoc way what the scientific community has long known, through more rigorous investigation and documentation. There's a lot we could learn from the scientific literature on coral reefs, but that literature is not particularly accessible to most hobbyists, because it's availability is usually restricted to academic research libraries, and because there are certain barriers to our understanding the papers. Some of the papers assume a technical background that most of us, myself included, don't have. And they all use a language rich with buzzwords that the reader is assumed to know.
With effort, normal people like you and me can overcome those difficulties, and learn a lot from the scientific literature. Many times, the concepts are not hard, just the language and the many references to other parts of the scientific literature web are what's hard (to us). For a while I've had the idea of having a periodic seminar that meets in fishroom. At each seminar meeting, one member of the community would present a paper from the scientific literature. I got this idea from various seminars I've participated in in graduate school. The idea would be to capture the essential ideas and present them in a more accessible language, possibly omitting certain boring details like the "cover-thy-butt" style of writing that makes up so much of academic argumentation. That style is necessary for making a difficult point among academic researchers, but it's largely useless information to us hobbyists who aren't familiar with most of the various cited works.
With that introduction out of the way, I'll move on to my second introduction.
Some of you know the name Walter Adey. He (and Karen Loveland) published a book called Dynamic Aquaria in 1991. Dynamic Aquaria serves as a general introduction to aquatic ecology, and to building functional models of aquatic ecosystems based on ecological principles. That is, each ecological aspect of the wild ecosystem is represented in the model, either by that same thing (e.g., tangs in nature are modeled by tangs in the model), or by some functional replacement of it (e.g., water motion comes from pumps and dump buckets rather than from tides and wind). The point is that we deliberately look to natural ecology to determine how to set up our tanks, and as much as possible, our tanks work on the same principles as nature. At least that's the claim.
But virtually everyone in the hobby remembers Dynamic Aquaria differently. The purpose of Dynamic Aquaria, and Walter Adey, was to introduce and push the Algal Turf Scrubber (TM), which is a horrible device that produces yellow water and kills corals. At least that's the claim.
The Algal Turf Scrubber is basically a refugium for algae. You encourage algae to grow on screens (like fiberglass window screen), by brightly illuminating the screens, isolating them from grazers, and providing them with a lot of wave energy. Periodically, you remove the screens from the system, scrape off the algae, and discard it. In this way, you export nutrients and other things from the system, This includes the macronutrients, nitrogen, phosphorus, and carbon, as well as other minor elements including heavy metals. Its fundamental purpose is similar to that of a protein skimmer, in that they are both nutrient export machines. I don't want to press the similarity beyond that right now.
In fact, the Algal Turf Scrubber (ATS) was a relatively minor point in Dynamic Aquaria, and there's still a lot of valuable information in the book even if you totally reject the idea of the ATS. Also in fact, some tanks managed with the ATS as the primary method of water purification can support stony corals. A tank's ability to support stony coral growth seems to be as dependent on a host of other aspects of the tank's management as it is on whether an ATS is used, and in all cases of ATS tanks I know of where stony corals don't do well, I can point to numerous other aspects of the tank management that are not in agreement with how successful stony coral aquarists run their tanks. But the ATS is the most visible characteristic of Adey tanks, so it alone gets the blame, at least from casual commentators.
I do not mean to be advocating Algal Turf Scrubbers in this talk. If I seem to be defending the idea right now, it's because I think Adey and his coauthors deserve to be taken seriously, and not completely discredited for one idea which may in fact not be God's gift to aquarium keeping, and the only valid way to manage water quality in an ecosystem model, as they seem to think it is. There are many interesting things about their systems, whether we like the ATS or not.
Adey has several other strong agenda items that are not shared by most reef hobbyists. It's worth covering these in advance of looking at the article, so that you understand the frame of mind from which he and the other authors are writing.
First, they hold (or at least Adey holds) the assumption that the proper way to set up an ecosystem model is to model as closely as possible as many aspects of the natural ecosystem as possible. For example, a reef ecosystem would have a lagoon section, a backreef, a reef crest, and a forereef, but it would not include the open ocean because that's not possible. Instead, it would provide some technical replacement for the open ocean, that provides some or all of the effects of the missing component. In their case, the open ocean is replaced by the Algal Turf Scrubber (for nutrient export) and a small amount of food input (to replace the small amount of plankton that washes in from the ocean).
It would also be possible to concentrate on one part of a reef, like the backreef, and treat the other aspects of the ecosystem at a higher level of abstraction (i.e., replacing them with engineering, as Adey does for the open ocean). There's nothing wrong with trying to directly include as many parts of the ecosystem in the model as possible, as Adey does, but you should keep in mind that other possibilities exist and are valid.
The authors' emphasis on plankton preservation is a second point where they differ from common hobbyist mentality. The importance of plankton on reef communities is undisputable. Without it, most animals can't reproduce, and many animals can't even live as adults without a planktonic food source. But in most hobbyists' tanks, plankton is either ignored, or it takes a back seat to other important aspects of reef husbandry, like protein skimming or convenient and inexpensive centrifugal pumps (including powerheads). Adey and his camp don't use protein skimming because of its tendency to remove plankton, and they only grudgingly use centrifugal pumps, preferring where possible to use "plankton friendly" pumps such as diaphragm pumps, bellows pumps, or Archimedes screw pumps. I have a lot of admiration of Adey for his emphasis on plankton. However, I question the wholesale assumption that avoiding protein skimming and centrifugal pumps will make a significant improvement in plankton population, in the face of other aspects of reef tanks that make it difficult to maintain a functional plankton population. We'll revisit this in more detail at the end of the talk.
All right, this marks the end of the introductory material, and the beginning of my presentation of the paper. Comments I make are generally taken from the paper, and I mean to present them without necessarily agreeing with what they say. In some cases I won't be able to resist giving my opinion about something, and I'll try to mark that with words like "I think . . . " or "My feeling is . . . ."
The Marine Systems Laboratory (MSL) at the Smithsonian Institution has created five small to moderate sized coral reef ecosystem models, and operated them for a total of 31 ecosystem-years. The Great Barrier Reef Aquarium (GBRA) has a 2.5 million liter mesocosm that has been operating for 7 years. Space Biospheres Ventures' Biosphere 2 project has a 3.4 million liter mesocosm that has been operating for 3 years. The Smithsonian group (MSL) was in close cooperation with the other two groups.
In this article, we will discuss and compare four of these systems, and try to gain insight into the following questions: 1) How accurately do these systems model the wild analog? 2) What are the uses of these models? 3) What can we learn about natural reefs and the global ecosystem from these models? I'm going to spend a lot more time on point 1) than on the other two.
The authors point out that algal turf scrubbers have a structural similarity to natural nutrient flushing to the open ocean, and allow the maintenance of nutrient levels, pH, and oxygen at wild reef levels, all the while not significantly removing plankton. Algal scrubbers don't provide the planktonic input that the ocean provides, so they have to substitute for this with simulated plankton input (feedings).
Maintenance of this system involves "minor manipulation" at weekly to monthly intervals. Several species of algae have gone through periods of abundance and dominance, when they required removal or "hand grazing." At other times these algae back off to an inconspicuous level. The authors attribute this to a combination of altered grazing intensity and reproductive success of various life in the tank. This is analogous to wild ecosystems' natural tendency to form patches -- areas where certain species dominate, while in other areas, different species dominate.
The highest predator is a rare reef lobster (Enoplometopus occidentalis). It is typically fed mussels. About every six months, a fish falls victim to the lobster. The frequency of this is related to the frequency of manual feeding (sort of like with my cat). Fish were replaced once in 1993, to make up for this predation.
Coralline algae growth is good in this system, and the growth rate is close to natural levels, according to informal observation.
There are seven stony coral species in five genera. The paper lists growth rates for Porites astreoides and Dichocoenia stokesii. Three species have sexually reproduced and recruited viable colonies, as I mentioned earlier. Six coral colonies have died, three of Dichocoenia stokesii and three of Porites astreoides. Sometimes these deaths were due to obvious predation. In other cases, they say, "one colony of a species can be shrinking while another, 20 cm away, is growing rapidly." So obviously water quality alone is not the explanation in those cases.
The tank sustains populations of mollusks, fan worms, and sponges, demonstrating some potential for filter feeding. The mollusks in the tank included the easy to keep Tridacna clams, but they did not say whether there are other mollusks. Also, I would like to note that some filter feeding worms and sponges do well in typical hobbyist aquaria, while other types do very poorly. It's not clear from the article whether they are truly experiencing something special here, in this respect.
There are two grazing fish (Zebrasoma flavescens and Pomacentrus coeruleana), and a variety of predators of the abundant small invertebrates. There is a rock beauty angel (Holacanthus tricolor) that "preys heavily on the 'weedy' anemone, Aiptasia pallida." That's an interesting observation for those of you with reef tanks large enough to support one of the larger angels.
Algal behavior in this system is similar to in the 500 liter system, with species going through periods of abundance and then decline. Halimeda opuntia and the crustose coralline algae, Neogoniolithon soluble, are the dominant algae in the parts of the mesocosm that are most heavily grazed, though the more turbulent reef crest supports a wider variety of algae.
One very interesting feature of this system is that there is a "lagoon" section, which is in a separate tank, connected to the main tank by a pair of 4" PVC tubes. The article doesn't say, but I believe that there is mesh over these tubes so that fish are not free to move between the two systems. The lagoon is densely planted in the seagrass, Thalassia testudinum, which flowers regularly. Grazing in the lagoon is minimal, and blade lengths of the seagrass often exceed 50 cm.
The system includes filter feeders such as several bivalves (not Tridacna), fan worms, and sponges. At the system start, there were three large urchins of the species Diadema antillarum. These urchins died shortly after the biological introduction of 1993, presumably because that introduction brought in whatever it was that killed so many Diadema urchins in the Caribbean around that time. Two more Diadema were introduced after a 9 month quarantine, and they remain active as of the time of the article. There are three other urchin species, and numerous ophiuriods, holothuroids, and asteroids, and these were all unaffected by the 1993 introduction.
The grazing fish in the system consist of five dusky damselfish (Stegastes fuscus), three striped parrotfish (Scarus iserti), a blue tang (Acanthurus coeruleus), two ocean surgeons (Acanthurus bahianus), and a seaweed blenny (Parablennius marmorcus). The damsels have taken over the back reef area, and through their aggression, basically don't let other fish graze there. Other grazers mostly reside in the deeper fore-reef environment, and the standing crop of algae there is low, due to the heavy grazing and the lower light intensity.
The "browser" fish are a queen angelfish (Holacanthus ciliarus) and a rock beauty angelfish (Holacanthus tricolor). These fish keep the main reef almost completely devoid of Aiptasia anemones, while those anemones are quite common in the isolated but connected lagoon.
The remaining fish are predators, and eat larvae, polychaetes, amphipods, isopods, and added dry food.
No fish or corals have successfully reproduced, although the damselfish have spawned many times. The authors attribute this lack of reproductive success to high predation, filter feeding, centrifugal impeller pumps, the relatively small volume of the system, and the short turnover rate (meaning more frequent trips through the pumps). They say, "The impeller pumps appear to have strong selective effects, passing algal spores and some invertebrate larvae, but effectively removing fish and coral larvae." I personally think it's more than the pumps that have such a selective effect. Some reef dwellers simply have less of a requirement for a long planktonic life for their reproduction.
The daily pH range of the tank is 7.80-7.99. The authors attribute this low pH to several factors. One is the so-called "low light levels," produced by the mere 12 kilowatts of MH light they have over this system. This produces 1650 uE/m2/s of light, which is not much below maximum daily irradience in the wild. However, due to public expectations of a particular "look," they have put blue filters over the lights to provide a more appropriate spectrum. These cut the light level roughly in half. That's still brighter than almost all hobbyist tanks. Most recently, they have replaced the 4500 K lamps with 6000 K lamps that don't require the blue filters. I'm skeptical that the reason for the low pH is the so-called "low" light level alone, though I would expect an even higher light level to increase the pH, and the paper confirms this.
Coral growth has been minimal to absent in this system, and the authors attribute this to the low pH. Actual coral loss in the system has been low (less than 10%), and the authors say this cannot be attributed to the low pH. Other factors are predation by fish and invertebrates (remember that they include many corallivores that hobbyists omit). System pH has risen with the addition of the 6000 K lamps, and they expect coral growth to increase as well.
This system is directly open to full natural sunlight.
The GBRA has approximately 800 colonies of hard and soft corals, from 100 species. There are over 200 species of fish in the mesocosm. The system includes 104 square meters of algal turf scrubbers. A sand filter is also used, to help meet public expectations of water clarity (30 meters morning visibility). The authors consider the presense of the sand filter unfortunate, because it substantially reduces plankton. The bottom-associated planktonic community is similar to in the wild analog, but at 1/10 the density. Low levels of ozone are used to break down water coloring compounds. The authors stress that the sand filter and ozone treatment are to meet public viewing expectations, not because turbid water has displayed any negative ecological effects in this model or in others.
In the first year of operation, the system suffered algae blooms. The operators addressed this by adding dozens of Diadema urchins. Over time, the initial algae bloom and the urchins have been replaced by a more diverse algae and herbivore community.
Present collecting efforts are designed to introduce particular target species in an effort to produce an ecological community with a balance similar to the wild.
Many algae and invertebrates reproduce successfully. Echinoderms and mollusks spawn regularly. Some fish display courtship behavior, and the catfish, Plotosus anguillaris, has successfully reproduced in the tank. Some hard corals have spawned synchronously with corals on the real reef on the other side of the cement wall of the tank. The article doesn't say this, but I read elsewhere that those corals hadn't been in the tank very long before the synchronous spawning events. I don't know if they've had more serious success since that time.
Some corals do better than others. Mussids, favids, and Turbinaria do well all the time. Acroporid corals suffer the highest mortality. Soft corals do great, naturally. The hydrozoan, Millepora, does well. More than two thirds of the coral mortality occurs in summer months, due to higher temperatures and strong rains. The temperature range for this tank is listed at 27.5-33.8 degrees C (81.5-92.84 degrees F). This is the highest for all of the four systems considered in this paper. I'll add my opinion that the high temperature alone is probably not the reason for the coral mortality, because the temperature occasionally gets even higher in the natural reef. I don't have information about how long the tank was at the high temperature range compared to how long the wild reefs are at or above that temperature. I should also point out that the mean temperature listed for this tank is 25.9 degrees, which is lower than the so-called minimum temperature. This brings into question the validity of any of these numbers. I'm sure there's a reasonable explanation (possibly a misprint, possibly missing information about how the max and min temperatures were measured), but it's not clear from the paper. Some corals are killed by predation, particularly by the parrotfish, which are no longer collected. Existing parrotfish have been removed.
Nitrogen and phosphorus levels are routinely checked, and the levels are close to wild levels. The strontium level has not been checked. The authors seem concerned that strontium may be of importance to coral growth. Interestingly, at this point in the paper they chose to cite Ron Shimek's Aquarium Frontiers article, in which he argued that there's no evidence that strontium is of any importance to coral growth.
pH in this system ranges from 7.90 to 8.55, with a mean of 8.18. Mean dissolved oxygen is 92.8% of saturation (mean of readings taken at 10 am, which is not when dissolved oxygen would be highest). The measured range is 84.5% to 126.9% of saturation. The water volume is turned over through the pumps once every 6 hours. The water change rate is 2% per month, and evaporation topoff is 3.3% of system volume per month.
This system is similar to the others in the basic engineering choices made. However, the basic structure of the reef comes from a local marble rather than reef forms of carbonate. A significant amount of oolitic aragonite sand was used in the system, but much of the carbonate surface consists of calcite or other carbonates that have little buffering capability (low solubility, I assume is what they mean).
10% of the system was filled with water from the Pacific ocean. This probably included some microbiota intact. Most of the life, including all of the fish, are from the Caribbean. There are a few Pacific invertebrates, including Tridacna clams, and numerous biota from the Sea of Cortez. These non-Caribbean species are a minor component of the system.
The early years of this mesocosm suffered from poor documentation, but there were management changes in 1994, and the new management intends to make publishable scientific observations.
The location of Biosphere 2 has naturally high light, but this is reduced about 55% by the overhead structure. Since the time that the structure was sealed in September 1991, the atmosphere has experienced a yearly, light controlled cycle of CO2 concentration. The winter high is 3000 ppm, and the summer low is about 1000 ppm (for 1992). By comparison, natural unpolluted atmosphere has about 350 ppm CO2 (from Dynamic Aquaria, not mentioned in the paper). The reef water pH varies with this CO2 cycle, showing a daily cycle of 0.3 pH units corresponding to the 400 ppm daily fluctuation in atmospheric CO2. The pH range of this system is 7.65 to 8.10, with a mean of 7.9. In past winters, sodium carbonate, sodium bicarbonate, and calcium chloride were added to limit the low pH to 7.65. What they're saying is that they had to artificially increase alkalinity in order to stablize pH.
When the system was originally established, water quality maintenance was achieved through 24 algal turf scrubbers totaling 48 square meters of area. Later, protein skimmers, plate filters, rotary drum filters, and diatom filters were added, and the algal turf scrubbing was reduced. The paper continues, "To replace the trace nutrients lost to the various filtering procedures, a mixture of 27 trace elements (mainly magnesium, potassium, calcium, strontium, and boron) is added monthly. This newer filtration system has not been successful due to excessive plankton removal and poor macronutrient control, and currently the number of ATS units is being increased and the filters removed."
The paper doesn't elaborate on how they know that all these added filtration gadgets are what is responsible for those element depletions, rather than natural bioassimilation or geological storage within the tank, or assimilation by scrubber algae which is deliberately removed. One would hope that they did tests of depletion rates before and after the filtration was changed, but considering their comment about the poor state of documentation of the system in the early years, I have doubts that they did those tests. I suspect that they're just assuming that the evil filters are responsible for their problems, because they have an axe to grind against traditional aquarium maintenance procedures, particularly protein skimming.
This mesocosm has a microbiota exceeding 500 species, all of which reproduce. The plankton suffers from the impeller pumps and protein skimming (again, no evidence or elaboration is given for this point). The bottom dwelling life includes amphipods, tanaids, ostracods, limpets, snails, brittle stars, rotifers, sponges, tunicates, nereids, spirorbids, terebellids, nematodes, rhabdocoels, acoels, and many other invertebrates and protists. The system generally lacks specialist predators, so there are periodic blooms of some species, especially opportunistic generalists (I assume this would include amphipods, for example.).
The food web is self sustaining. At least, no food is added, though there are incidental nutrient inputs from other biomes in the form of falling leaves and insects, etc.
During the first 6 months of operation, before any conventional filters were installed, several damselfish species reproduced, and hundreds of juveniles survived the first few months. That means a lot to me, considering how much micromanagement it takes to rear the larvae of marine fish in more conventional settings. This means that there was a sufficient planktonic population to support the larvae during their early lives. Striped parrotfish (Scarus iserti) and sergeant majors (Abudefduf saxitilis) were also successful at reproduction during this period.
In time, traditional filtration was added and the carnivorous fish grew, and at this point fish reproduction stopped. Even after the removal of most of the piscivorous (fish-eating) fish, no subsequent fish reproduction has been verified.
In the Spring of 1993, at the same time that the algal turf scrubbing was reduced, the bottom of the mesocosm developed a profuse growth of algae, dominated by the red algae Hypnea and Acanthophora. The management took care of this problem by early 1994 by adding Mithrax crabs and tangs. In 1994, the red calcareous alga, Amphiroa, began to overgrow corals, and the management responded with manual harvesting.
In October 1993, a survey of corals was conducted. They found 966 individual corals, including 87 new coral recruits. In May 1994, they performed plankton sampling, and they found several species of coral larvae. Also in 1994, they found recruits of the corals Agaricia, Favia, and Porites, on the walls near the surface.
The mean oxygen saturation was 95%, but this site was at high altitude, so they added oxygen directly to achieve that. The system pumping turnover time is 38 hours. Since these are potentially damaging centrifugal pumps, the long turnover rate probably had something to do with the early and impressive fish reproductions. I'm becoming more interested in low turnover tanks, as long as some non-pumping method of water motion can be devised. No water changes are performed. The evaporation topoff amounted to 1.5% per month. The temperature varied between 23.5 and 27.5 C (74.3 to 81.5 F), with a mean of 25.5 C (77.9 F). Salinity, nitrogen, and phosphorus are at natural reef levels.
The macronutrients, carbon, nitrogen, and phosphorus, have been fairly steady at or near wild levels. The exception is the excess CO2 in the Biosphere 2 system, due to the high atmospheric CO2, low light levels (half of natural light, which is still way more than we hobbyists use), and reduction in algal turf scrubbers. The low pH of two of these systems is probably largely responsible for the poor performance of stony corals and other calcifiers.
The Biosphere 2 mesocosm saw not only a reduction in calcification due to the low pH and high CO2, but a shift in calcification from aragonite formation (in stony corals and Halimeda algae) to calcite formation (coralline algae). In general, the authors believe that the Biosphere 2 experience shows considerable promise for studying the effects of a global increase in atmospheric CO2.
The issue of calcium and alkalinity maintenance is very interesting, and mysterious, in these systems. The authors discuss calcium concentration, but not alkalinity directly. Their intent is that calcium will be maintained by the natural dissolving of the aragonite sand present in all of these systems. The hobby experience is that this passive process is not sufficient to keep up with calcification demands, but keep in mind that these systems generally have a much deeper sand bed (the ones I've seen anyway), so there's a lot more volume of low pH water in the vicinity of aragonite. Maybe it works better with these deep sand beds.
As noted earlier, the Biosphere 2 mesocosm addressed this problem by the addition of buffer and calcium chloride, which is generally considered one of the poor ways for hobbyists to keep these levels up. The authors also consider this poor practice.
The other systems did display some calcium depletion, and they addressed the problem in the other systems by running the topoff water through a column of oolitic sand. This practice is similar to the calcium reactors that many of us use now, but without the recirculation and without CO2 injection. I think I read somewhere that you can get about 30-50 ppm increase in Ca++ by this practice. I'm surprised that that's enough. However, remember that calcification is slow in these systems (at least among stony corals), so that will make for a smaller demand than we see in the best hobbyist tanks, which calcify at or above wild rates (Our "success" is probably largely due to the recent fad of goosing the alkalinity way above the natural level. We could have another talk on the observed and predicted negative aspects to this practice.).
There is a table in the paper that gives the calcium concentration for each of these systems. They're all in the natural range.
They note that when calcium depletion does occur, different corals respond differently. Some continue to grow, while others regress.
They suggest that the coral mortality in the Great Barrier Reef Aquarium is not due to calcium depletion, but strontium depletion must be happening in that tank, and they suppose that that might have something to do with the coral mortality. Note that they do not claim to have any evidence that strontium is required for corals, but they are taking seriously the debate over its possible necessity. I would say I'm of similar sentiments.
Here is a very interesting comment: "Injections of SrCl (sic) in the smallest systems during the last 6 months of operation have prevented non-predatory stony coral bleaching and loss, and in the 12,000-l model had re-initiated Acropora palmata growth." They don't say anything about what else changed during that time, so I remain skeptical that the strontium addition is what was responsible for the improvement, just as I am in the case of hobbyist folklore.
The smaller units and the Great Barrier Reef Aquarium have shown a consistent depletion (about 15%) of potassium. The authors don't know the cause and effects of that depletion. It might be something we should look into in our own tanks.
Some of the reproductive successes are impressive, particularly the fish reproduction in the Biosphere 2 system and the sexual coral reproduction in the various systems. Where reproduction wasn't as successful as the authors had hoped, they blame the unplanned or non-ideal hardware, such as impeller pumps and mechanical filtration, and the unplanned biota, like excessive predatory fish. It's significant to me that the smallest system, Walter Adey's home aquarium, was also one of the most reproductively successful, and that his system is the only system of the four that uses only non-impeller pumps.
Some of the unplanned changes, such as the addition of mechanical filters, were done to satisfy the public's need to see exceptionally clear, deep blue water. While the open ocean is often very clear, and some reefs are clear some of the time, there is often a significant amount of particulate matter, both living and dead, in the water over healthy reefs. The constant removal of that material is a step away from the natural ecosystems, and it's understandable that the model would suffer from it.
As we can see, there are various negative aspects of these systems that we hobbyists try to avoid. We generally pay more attention to calcium and alkalinity maintenance, at least judging by the paper. We are much more willing to micromanage the life in the tank, particularly with regard to algae and coral placement. The authors of the paper more or less expect things to take care of themselves, and they usually respond to a problem by adding herbivores, removing higher predators, and so on. That's the right way to address these problems, in my opinion, but the effects of those changes will take longer to manifest themselves than a quick manual removal. When the authors resorted to manual removal of algae, I suspect that they did it less often and less completely than we hobbyists do, because of the much larger scale of their models, and because they would rather do it with in-model ecology. The lighting of the Smithsonian tank was for a long time coming from lamps that most aquarists would consider of an inappropriate spectrum for a reef tank. Whether this is a real problem or not is not something I know.
My feeling is that there are good and interesting aspects of these modeling efforts, and then there are things that I would try to avoid (based on these authors' experiences, so those experiences were valuable because they suggest what might not work very well). We can talk all day about these things before even coming to the emotionally charged issue of the algal turf scrubbers. The scrubbers might be responsible for some of the problems that these tanks have, but it's not clear to me that that's so. I know of several private reef tanks that use scrubbers and that do quite well, supporting stony coral growth similar to what you'd expect from a typical "Berlin"-type reef tank.
Craig Bingman has tested the water from the Smithsonian's reef exhibit, and has made some comments about that system. You can see some of what he's said at the article http://www.dejanews.com/getdoc.xp?AN=174968565
The relevant part of that article is: "ATS systems do support corals. I recently tested some water from the Smithsonian tank, and it looks on par with water from Berlin aquaria with no additional chemical filtration. The water was not incredibly yellow, contrary to popular belief. Overall it looked fine. I've seen the tank within the last 6 months, and it looked much better than I had been led to believe it would. It supports stony corals. To the extent that I took exception to the tank, it was primarily on minor aesthetic points, I would have been a little more aggressive in removing some organisms from some points in the tank. But that is in the land of shoestyles and haircuts. I don't see any Major deficiency in that methodology, especially if one were willing to do a bit more trace element supplementation and run a bit of carbon on the system. How well an ATS system would handle a SPS dominated tank with the major lipid production that is typical of these systems is an open question. A skimmer can help with that."
As I said before, the authors' emphasis on a plankton population, particularly the reproductive stages of animals that live on the substrate as adults, is very interesting to me. However, I wish they would provide more documentation of "the plankton problem." They are quick to state that skimmers and centrifugal pumps remove or destroy plankton, but they don't have (or don't share) research that shows where the plankton is actually going. How much plankton is being destroyed by pumps, and how much is falling victim to predation by fishes or filter feeders? The ratio of filter feeders and fishes to water volume is much higher in these microcosms and mesocosms than in wild reefs.
My experience in my own tank, though admittedly a very small one compared to the ones we've talked about today, is that live brine shrimp will virtually disappear from the water within 30 minutes, even if I turn off the pumps and provide water motion via air bubblers. This is true even in my reef tanks that don't have any fish in them. Of course, live brine shrimp in a bare culture tank will sustain themselves for as long as you feed them (to oversimplify things a bit).
It's undisputable that protein skimmers will remove particles, including living plankton, and I find it at least plausible that conventional pumps kill a lot of types of plankton (despite the common anecdotes that show that one particular type of small critter can survive at least one pass through a pump without dying). But I wonder if the impact of skimmers and pumps is a rather moot point so long as there are other significant barriers to maintaining a large plankton population, namely the filter feeders, other predators, and passive mechanical traps such as the reef structure itself. The one mesocosm in which damselfish, with their pelagic larvae, survived to the juvenile stage, is the largest mesocosm of the lot.
I'm going to end the talk here, and turn this over to your questions and comments. My intent in presenting this article is not to suggest that we run out and put algae scrubbers on our tanks, but rather to keep an open mind about ecosystem models of all kinds, and to learn what we can from what others have spent their time and energy to share with us.
The PA system is still on in this room. That means that you can't directly talk. If you would like to ask a question or make a comment, please raise your hand with the "raise" command, and I will hand you a microphone. The microphone will allow you to speak.
Thank you for your attention this evening.
Cap gives the mic to thomas.
Thomas says: You mention hobbiest light level being much lower than the target tanks. What are natural levels?
Cap says: Of course it varies a lot, but natural peak level in the tropics is around 2000 uE/m2/s. Richard Harker's recent Aquarium Frontiers article gave comparative measurements for the output of a 400 watt MH bulb at various depths under the water. Unfortunately I don't have the numbers handy, but I believe it was generally less than 1/4 of natural peak light.
Cap says: Thomas, if that answers your question, please give the mic to larryr by typing "give mic to larryr"
Thomas gives the mic to larryr.
Larryr asks: Can a skimmer be made that doesn't destroy plankton? Like the Tunze model for example (ie do they work)?
Cap says: Those are really two different questions. A skimmer that doesn't remove plankton is one that doesn't remove particles, since most plankton can be assumed to be passive particles (larger fish larvae excluded, but that's a small part of natural plankton).
Hermit raises his hand.
Cap says: I've heard mixed reviews of the Tunze skimmers. By personal observation, I would have to say that they do remove particles, like any other decent skimmer.
Cap says: So I would say that yes the Tunze skimmers work, but no they are not harmless to plankton.
Cap says: That leaves the question of whether a skimmer can be made that doesn't remove plankton.
Cap says: And I think the answer to that lies in how you draw the water, not how the skimmer is constructed. If you were to draw the water slowly, over a very very large area fine filter, you would be able to pull water without particulates, and without piling up a bunch of dead plankton against the filter material.
Cap asks: Does anyone else have a comment about this question, before we move on to something else?
Fergy raises his hand.
Larryr exclaims: Interesting answer. Thanks!
Cap says: Larry, pass the mic to Fergy if you're ready.
Cap says: "give mic to fergy"
Larryr gives the mic to Fergy.
Cap says: I think I need to implement a "queue" helper object for these talks, so that we can keep track of who put their hand up in what order. Maybe for the next talk . . .
Fergy says: that partially defeats the purpose of the skimmer in a system running in the Berlin setup, since part of the trend is not to use a mechanical filter in any way, other than the skimmer. If you are using the skimmer with a mesh prefilter, and it draws from a large area, then it will not be rmoving any of the particulate matter from the water, and will be serving as a chemical filter only
Fergy says: so then you will need to add some sort of mechanical prefilter again
ChrisG raises his hand.
Fergy gives the mic to chrisg.
Cap says: I think Hermit was next.
ChrisG asks: One thing I wonder about is their evaporation rate. Slow circulation and low evaporation rate (3.3% per _month_) implies low gas exchange, and with intensely lit separate tanks of algae, I wonder if there might be problems with not enough Co2 (for the symbiotic algae in the corals for instance)?
Cap says: Well, okay. Let's give the mic to hermit next.
ChrisG gives the mic to hermit.
Thomas raises his hand.
Cap says: I'll comment on Fergy's comment first. I agree that removing dead particulate matter is a useful functionality of conventional skimmers, and that you would lose that benefit if you filtered out all particles. However, removing only dissolved organics, and possibly the growth of an algae scrubber, is plenty of nutrient export, so we shouldn't be so concerned about that. I ran my tank for close to a year with no skimmer at all, and the amount of particulates in the water wasn't significant (in fact I'd have wished for more, for the sake of the filter feeders).
Cap says: Chris, that's a good observation. I'd like to know their dissolved CO2 concentration, though we should be able to know it from their alkalinity and their pH.
Cap says: Oh, I meant to say that when I had no skimmer on my tank, I didn't have a mechanical filter either (nor have I ever, except for the skimmer that's on there now).
Hermit raises his hand.
Cap says: Hermit I believe you have the mic. Go ahead.
Hermit says: thanks, say i recently set up a an algae scrubber system and in just 4 weeks my coral banded shrimps had 100's ob babies, don't know if it was a coincindence or not but i have had the pair of shrimps for almost 3 years
Cap exclaims: Who knows what induced them to spawn. Often times animals spawn in response to a sudden change in environment, whether good or bad. But congratulations anyway. Don't expect the babies to survive though!
Hermit says: they were eaten by my clowns and anthias, saved a few and put them in my scrubbers,
Cap says: Higher crustacean larvae are planktonic filter feeders, and they have that lifestyle for a couple months or so (vague memories here). They shouldn't survive in a scrubber.
Hermit says: hey, what effects has carbon had on these large public systems with scrubbers
Keef raises his hand.
Cap says: Oh incidentally, I want to make a general comment about scrubbers. Many people consider them as farms for meiofauna (amphipods, copepods). You can use them for that, but that's in conflict with using them for nutrient export, which is the use that Adey presents for them.
Cap says: The "True Adey Way" does not involve the use of any carbon, because it removes trace elements and acts as a mechanical filter if you use it in the traditional way. I've heard that the Smithsonian tank has departed some from the True Adey Way in recent years, and part of that has been to run some carbon. I believe the Great Barrier Reef Aquarium has used carbon too, but that wasn't in the original design.
Cap sees Keef's hand waiting for the mic.
Hermit says: thanks, cap i have experinced a complete elimination of unwanted algae iin the reef, do you think that the current ideas that by growing more algae acturally starves out the bad algae out of the reef. what was your observations of unwanted algae in the public aquaria.
Cap says: I'll comment on that, and then let's give Keef a turn.
Cap says: It's certainly true that algae will grow better when there are more nutrients, so if you're effectively reducing nutrient concentration with the scrubber, then you're making it harder for algae in the main tank to grow as fast. The fact that we harvest the scrubber regularly provides an environment relatively free of grazers, whereas the tank should be loaded with grazers. That's why you can have a lot of visible algae in the scrubber and none in the tank.
Hermit says: thanks for your input cap
Cap says: Many public aquaria that use scrubbers have a visible crop of turf algae. Though in the Smithsonian tank, and in the Carnegie Science Center tank, this is not a green filimentous hair algae, but rather a hetergeneous turf, which looks quite natural to me, though most reef keepers wouldn't stand for it. I think the reason it's there is because those tanks aren't so loaded down with herbivores as ours are (lots of tangs, and way more snails and hermit crabs per area than occur on real reefs).
Cap says: Type "give mic to keef" when you're ready.
Hermit gives the mic to keef.
Keef says: First: Are there any plans on line or in books that we could look at to build our own TAS
Cap says: The Walter Adey and Karen Loveland book, Dynamic Aquaria, tells you the principles of algae scrubbers, and shows some photos. You should be able to figure it out from there. But the book is certainly not intended primarily as a recipe book for making an ATS! As I mentioned, the ATS is a relatively minor component of the book, although an important one.
Keef says: Second I would like to thank you for your time in preparing this excellent forum
Cap says: Rick mentions that John Walch will be speaking at the Western Marine Conference next week on how to build an algae scrubber.
Thomas raises his hand.
Keef gives the mic to Thomas.
Thomas says: Thanks Keef. I have a question/comment about the CO2 & pH levels.
Thomas says: You mentioned that when one of the tanks saw increased CO2 levels, the calcification switched from argonite to calcium (i.e., SPS to coralline). I guess I really don't have a question per say, I just find that strange.
Thomas says: "> Cap says: The Biosphere 2 mesocosm saw not only a reduction in calcification due to the low pH and high CO2, but a shift in calcification from aragonite formation (in stony corals and Halimeda algae) to calcite formation (coralline algae). "
Cap says: The reason for that is that aragonite formation requires a certain pH level that is higher than for calcite. The article made the point specifically regarding Halimeda (which makes aragonite). This was news to me, so I can't elaborate on it much. They said that Halimeda virtually disappeared in the Biosphere 2 system, whereas it's one of the dominant algae in the other systems.
Hermit raises his hand.
Thomas asks: Hmmm, intereseting. You also mentioned that some coralline algae was taking over certain coral species..Due to the switch to calcite maybe?
Thomas gives the mic to cap.
Cap says: Yes. Corals normally have the advantage over algae, but when certain characteristics of the environment get screwed up compared to nature, the balance can shift. That's what's going on when coralline algae starts overgrowing living corals. And generally speaking, that's what's going on when we get cyanobacteria blooms, or green hair algae, etc.
Keef says: If you were to add a ATS to your system would you add it as a seperate sys or in line with your skimmers
Cap says: I'm not sure what you mean exactly. I would probably want the scrubber to be upstream of the skimmer. The scrubbers are claimed to leak organics into the water, and that's what skimmers are good at removing.
Keef says: that is the answer I was looking for thanks
Keef says: any one else like the mic
Keef gives the mic to cap.
Cap asks: Anyone else want the mic?
Cap says: I'm going to try something here.
Cap has released muting.
Cap says: You can all speak freely now.
Magda wiggles her toes and tests the water.
RgrMill is deafened by the roar
Kaz grins.
Keef stands up a begins a round of applause for CAP
Magda cheers for Cap!
Kaz stands up and applaudes Cap!!!
Kaz exclaims: Encore, Encore!!!
RgrMill says: I see you set a new FR attendance record tonight. Congrats.
Kaz says: yep, new record :)
Cap bows.
Kaz grins.
Cap says: Still didn't compare to Albert Thiel's talk for attendance.
RgrMill says: Sorry I missed it.
Cap says: It was on IRC.
Kaz says: hmmm give it time :)
Magda asks: will you be posting this on the Fishroom web page, Cap?
Cap says: No. Karen will be.
Keef says: Still broke attendence records
Kaz says: that was my next question :)
Ericee exclaims: Congrat's Cap! Nice Talk!
Cap says: Thanks Eric, and everyone.
Kaz says: and I most certainly will, now that i have caps permission (copyrighted of course)
Cap exclaims: Thanks for being patient through it all too!
RickM says: darn i missed most of it
Cap says: A log of the talk will be available on the fishroom web page.
RickM says: should have left them pople outta power
Magda asks: so after all this...would you try out an algae scrubber at some point in time?
Cap says: I've already tried one, but I think they're usually unnecessary in a small tank. I've thought about how I'd incorporate one into my dream tank (outdoors, under natural light). One thing I don't like about algae scrubbers is the cost of lighting them. Being able to use natural light would make me more interested in them. But then you lose the "reverse daylight photosynthesis" benefit.
Cap says: Which is a benefit I didn't talk about tonight. It wasn't in the paper, and the talk was too long already.
Magda nods her head in understanding.
Magda says: it was a long talk, but I liked the more "in-depth" perspective that was given
Cap says: OK, good. I hope you weren't alone.
Magda says: it sounds like you have a different attitude towards the public aquariums now than when I first got on Fishroom
Magda says: a kinder, gentler attitude ;)
Cap says: I don't think I've changed my feelings about them.
Magda asks: are they badly mismanaged or just run according to someone else's view of how a reef should be run?
Cap says: I also wouldn't call these people's work typical of that of public aquaria. Most public aquaria are basically amusement parks for the ignorant public. Education is pretty low on the list of motivations, though obviously they can't say that. I think these guys (tonight's guys) are primarily interested in the research, and they're disappointed about the compromises they have to make because they're doing their work tagged onto a public exhibit, with public viewing needs.
Cap asks: Magda, do you mean are these tanks badly mismanaged, or do you mean that about public aquaria in general?
Cap says: I'd say these tanks are managed in a much more interesting way. Most public aquaria have very little science involved. It's all practical engineering, and legacy engineering at that.
Magda says: how do you feel in regards to the tanks specifically mentioned
Cap says: I think tonight's authors did a good job considering the outside constraints. But I also think that some of their views are more religious convictions than supportable scientific positions. I don't like that at all.