TEXT BY RICHARD HARKER

Measuring Turbulent Flow In Reef Tanks

Do you have an opinion on the issues raised in this article? Join in the discussion by going to: Measuring Turbulent Flow In Reef Tanks.

The importance of adequate water motion to the health of corals is well documented. More than 150 years ago, Darwin noted that there was greater coral growth in exposed areas of reefs than in protected areas. Sheppard found that water current and wave action have a significant impact on reef distribution and coral zonation (Sheppard 1982). Dennison and Barnes demonstrated that coral photosynthesis and calcification increase with increased water motion (Dennison and Barnes 1988).

The hobby also recognizes the importance of adequate water motion. Virtually every popular book on reefkeeping mentions the importance of adequate water motion. Dana Riddle examined the issue in depth in a two part article “Water Motion in the Reef Tank” in Aquarium Frontiers (1996, 1997). Unfortunately, none of the authors who stress the importance of adequate water motion offers any suggestion on how to determine whether a reef hobbyist has adequate water motion in his or her tank. The reason is that it is difficult to characterize water motion on a natural reef or in a captive system.

RICHARD HARKER

This photo compares “clod cards” that have never been used to one that has already been in a reef tank.

A directional and relatively stable laminar flow can be measured in a number of ways. The movement of objects such as floats can be monitored. Dyes can be released into the water and videotaped. Flow or current meters can also be used. Unfortunately, currents on an open reef can hardly be characterized as stable laminar flow. Water motion on the natural reef is turbulent and unpredictable. Current in a closed system is turbulent as well. Whether from pumps and pipes or from powerheads, flow in a reef tank is initially directional and stable, but quickly degenerates as it strikes the walls of the tank and objects within it.

The difficulty of measuring current in a captive system using a method designed for measuring laminar flow can be seen in the Riddle articles in Aquarium Frontiers. Mr. Riddle used a sophisticated digital water velocity meter to measure the water output from several powerheads. A water velocity meter has a sensor that is placed in the water, and converts the water flow past the device into electrical current that the meter reads as speed. The problem with this is that the water must be moving in the direction of the sensor — it does not read water motion at any angle other than in line with the sensor.

Water that flows from a powerhead is unidirectional as it leaves the nozzle, so a current meter placed against the nozzle does a reasonably good job of characterizing the velocity of flow at that point. The unidirectional flow leaving the nozzle quickly ceases to be unidirectional, however, and then spreads and mixes with other currents in the tank. What begins as a unidirectional flow soon becomes turbulent, as it does on a natural reef.

Figure 2

The limitations of using a flow meter are illustrated in Figure 2 of part two of Mr. Riddle’s article. This diagram shows water speed versus distance for an Aquaclear 802 powerhead. Water speed at the nozzle is approximately 0.7 meters per second (m/sec), falling linearly until water speed is shown to be zero at approximately 0.62 meter. Anyone who has used a powerhead in a closed system knows this is clearly impossible — The water flowing from the powerhead does not stop at 0.62 meter. It continues on, but it does so in a diffused, unpredictable manner. The current meter is not measuring velocity of the water at 0.62 meters. Rather, it is reading the coherence of the water flow, which is zero at 0.62 meter. Clearly, a water velocity meter is of very limited value in evaluating turbulent flow, and therefore is of little help in evaluating adequate water motion in a reef tank.

Recognizing the need for a device designed to measure turbulent water motion, Muus (1968) proposed using the dissolution of plaster of Paris as a means to characterize water motion. The technique involved placing plaster balls around an area of interest, leaving them for a prescribed period and then using the loss of mass over time as an indicator of water motion. The technique was refined by Doty (1971). He argued that the weight loss of “clod cards” made of plaster of Paris held in moving water could be compared to weight loss of identical cards held in calm water — the difference being the impact of water motion.

This approach was examined by two noted marine biologists (Jokiel and Morrissey 1993) and subsequently validated. The authors wrote, “The clod card measurement appears to integrate both turbulent agitation and unidirectional flow into a single number related to mass transfer...Linear and oscillatory flow both appear to contribute to weight loss in the clods and their effects appear to be additive in a complex fashion.”

RICHARD HARKER

This shot of a Heteractis species in the Great Barrier Reef indicates that it needs a great deal of water motion. It has positioned itself vertically in a high flow zone.

Since the first article on using clod cards for characterizing water motion appeared, numerous other articles have made use of the technique. It has been used to examine coral reef distribution at Johnston Atoll (Maragos and Jokiel 1995), to study coral calcification rates (Dennison and Barnes 1988) and to examine growth rates of Acropora in the Philippines (Yap and Gomez 1981) and the Great Barrier Reef (Oliver and Dunlap 1983). The technique has been used to examine boundary layers over coral reefs (Shashar et al. 1996) and to measure circulation in aquaculture ponds (Howerton and Boyd 1992). It seems to have become the method of choice for most water motion researchers.

Most equipment and techniques used by coral reef researchers in the field are beyond the reach of hobbyists. Spectrophotometers, meters for measuring photosynthetically available radiation (PAR), complex chemical analyzers and expensive electronic devices are required in today’s marine biology. In contrast, the clod card approach to measuring water motion is accessible to every hobbyist. It is inexpensive and it is one of the few areas where a hobbyist can directly compare his or her tank to field work done on natural reefs.

In an effort to better understand turbulent water motion in captive systems and compare captive systems to natural coral reefs, I conducted a series of experiments with clod cards. I used them to measure water motion both on natural reefs as well as in reef tanks around the country.

Powerhead Evaluations

Methods and materials

Using the recipe outlined by Doty, I prepared a number of clod cards. I mixed plaster of Paris, poured it into molds, and then dried, mounted and weighed them. I began with an examination of popular powerheads in the hobby. I mounted a powerhead on one side of the test tank filled with artificial seawater adjusted to a specific gravity of 1.024. A test card of known weight was placed on the bottom of the tank opposite the powerhead. The powerhead was operated for 12 hours, after which the card was removed, rinsed briefly with tap water and set aside to dry. Once completely dry, the card was again weighed and the loss of mass noted.

The tests began in a 10-gallon tank for convenience. However, the more powerful powerheads generated so much current that water sloshed out of the tank. The tests were moved to a 40-gallon tank and started over again. One unexpected finding was that dissolution did not vary by tank size. A powerhead that produced a certain level of turbulent flow energy in the 10-gallon tank generated an equal level of turbulence in the larger tank. In other words, a test card that lost 20 percent of its weight in the 10-gallon tank also lost 20 percent of its weight in the 40-gallon tank for a given powerhead. This was not expected, but later confirmed in the hobbyist tests. Little water motion generated in the tank dissipates. Most is conserved as it bounces off the tank walls.

Figure 1

Figure 1 is a plot of the dissolution of each card against the water turnover rate claimed by several popular powerheads. As the output of the powerhead increases, the dissolution of the test card increases. Regression suggests the relationship is linear and has an R-squared of 92.43, a very high correlation. A complete examination of powerheads will be featured in an upcoming product review column.

Water Motion on the Natural Reef

A set of cards were taken to Bonaire, the Netherlands Antilles, Roatan, Honduras in the Caribbean and Kungkungan Bay North Suluwesi, Indonesia. The locations were all areas of relatively low water energy. The Bonaire cards were placed on Bari Reef, a fringing reef slope on the protected leeward side of the island. This area has virtually no tidal currents, but there is periodic surge in shallow areas. The Roatan site is off a small island in a protected sandy reef flat near the barrier reef. There is some minor surge action at the site. The Indonesian site is adjacent to Lembeh Strait, an area populated by soft corals and large-polyped stony corals, such as Euphyllia ancora. There is little surface water motion, although the strait is known for its strong tidal currents. Figure 2 shows the results of two tests — dissolution is plotted against depth for Bonaire and Kungkungan Bay. In general, energy decreases with depth. The greatest dissolution in these areas occurred in surging shallow water.

Based on my measurements on several reefs, as well as published research, a reasonable energy target for the hobbyist keeping large-polyped stonies and lagoonal soft corals should be in the 25 to 30 percent dissolution range. For hobbyists keeping small-polyped stony (SPS) corals and corals from high energy zones, dissolution should approach 50 percent.

Captive Reef Evaluations

I enlisted the assistance of hobbyists around the country to evaluate water currents in their tanks. Nearly 30 hobbyists volunteered to place test cards in their tanks. Each hobbyist was sent a serialized, pre-weighed card, which he or she then placed in his or her tank at an assigned location in the center of the tank near the front glass. After 24 hours the card was taken out and returned to me for weighing.

A total of 27 cards from 19 hobbyists was returned and survived shipping and handling. Each participant estimated tank volume and total current, including powerheads and pumps returning water from the sump, and listed the pumps used.

The average loss for the participants was 35.5 percent±9.8 percent. The highest dissolution was 45.4 percent, while the lowest was 12.4 percent. This means the most turbulent tanks exceeded the Bonaire surge zone, while the majority of tanks more closely resembled back reef or lagoonal energies in their water motion.

Figure 3

Few writers offer recommendations on what constitutes adequate water motion in the reef tank. Those daring souls who offer a suggestion generally express the recommendation based on tank volume turnover. For example, Fosså and Nilsen write, “As a general rule we recommend a water motion at least 10 times the aquarium volume per hour.” The assumption has been that higher turnover rates translate into higher tank turbulent flow. This turned out to be untrue for the hobbyist tanks studied. Figure 3 shows the results of the first wave of testing. Turnover (current divided by volume) is plotted on the X-axis, while dissolution is plotted on the Y-axis. Regression suggests a negative relationship in which dissolution goes down as turnover goes up. This might have been a surprise were it not for the experience with the test tanks. Because dissolution in the powerhead tests was not related to tank size, one could hypothesize that the volume of a hobbyist’s tank would have little to do with energy transfer as measured by dissolution.

Plotting dissolution against circulation (the sum of all pumps in the system) produces a positive correlation. In other words, the more circulation in the tank, the more energy transfer — hence dissolution. The R-squared for the hobbyist test is 55.48 percent, with a probability level of 0.01. In other words, there is a 99-percent chance that the relationship is more than just chance. Removing two raises the R-squared to 84 percent.

In the course of the tests, two hobbyists reworked their water motion regime, giving them an opportunity to evaluate the impact of their changes. Mark Chapman has a 260-gallon tank with a Dolphin pump circulating water through a three-way valve that switches the flow in the tank from side to side. His first test card lost only 12 percent of its initial weight, well below average. Learning of the results of his test, he added a pair of Hagen 802 powerheads. In the next test, his test card lost over 27 percent of its weight, doubling the amount of current in his tank.

Larry Jackson has a 200-gallon reef tank using a pair of Gemini pumps for circulation. After the first round of tests, he replumbed his tank, increasing the flow from the sump and moving the return from behind the rock structure to over it. In the first test, his card lost slightly less than 40 percent of its mass, placing Mr. Jackson’s tank well above average in water circulation. After the changes, he increased his already frothy current by 17 percent. His 125-gallon tank has less circulation, and replumbing the return improved this tank even more, increasing its circulation by 22 percent. My own 300-gallon SPS-dominated reef tank has a dissolution rate of 43.5 percent, using four Turbelle powerheads on Turbelle controllers. Adding two additional Turbelle pumps increased dissolution to 59.3 percent. With this energy level, I am able to keep fore reef corals, such as Acropora palifera and A. humilus, growing rapidly with normal looking morphology.

Placing cubes in different locations within the tank enables a hobbyist to determine how currents differ throughout the tank. During the tests, several hobbyists placed multiple test cubes in their tanks. In general, the test position — just behind the front glass in the center of the tank — had the greatest current. Current at the ends of the tank was generally significantly lower.

In my 300-gallon tank, there is more than 40 percent less energy in the corners. While the center of my tank approaches fore reef turbulence, the corners are lagoons. Gary Russell of Tulsa, Oklahoma, has a 75-gallon reef tank with the rock structure in the center of tank. While Mr. Russell’s tank measures 58.7 percent dissolution in the front, the rear center measures only 30.9 percent. This means that fore reef corals would be at home in the front of the tank, and back reef corals would be at home in the rear of the tank.

Test card dissolution has proved to be a valuable tool in assessing water motion in captive reef systems, and for the first time has enabled hobbyists to compare their tanks to other systems and natural reefs. The tests suggest that most hobbyists have lagoonal energy levels in their tanks and that increasing the water motion would be in order.

Acknowledgments — I wish to thank the many hobbyists who contributed their time and tanks to the study. I extend a special thanks to Gary Russell, who supported my efforts

References