Print This Page

Research Comparing Clear Water RAS and Biofloc Systems

(and other topics)


In response to a research report that concluded clear-water recirculating systems were more productive that biofloc systems, Daniel Gruenberg, a shrimp farmer and consultant in Thailand, posted to The Shrimp List:


Daniel Gruenberg ( That’s something I have been saying for many years.  Biofloc theory says that feed conversion ratios (FCRs) should be lower in floc systems due to nitrogen recycling, but in practice, this is never what we find.  This research further confirms our experience.


I routinely get FCRs near one and often get FCRs of 0.8 or less in my “Natural Pond Management” (NPM) systems.  So I still contend that at least a derivative of NPM will always be more efficient than floc for commercial production.


We haven't done much with super-intensive farming, but we think by partitioning a NPM system, it could produce super-intensive crops.


I think the high nitrite levels in floc systems harm the shrimp’s metabolism, outweighing any benefits they may get by consuming flocs.


Martin J.M. Guerin ( Daniel, thank you for your comments on the management of nitrite levels in RAS and NPM systems.


Being a nutritionist and not a farmer, I acknowledge that proper farming systems, good water quality and feed management should be the preferred approaches to maintaining optimum water quality, but I can't help thinking what, as a nutritionist, I can do to improve water quality or help shrimp deal with adverse water conditions.


Therefore, with regard to NO2 levels in the water, I could not resist mentioning my recent research conducted in China at Jimei UniversityIt demonstrated that feeding 0.3ppm organic selenium (as hydroxy-selenomethionine) helped increase survival of Penaeus vannamei exposed to a nitrite challenge trial.


As to formulating feeds that lead to lower NO2 levels in the water, they are normally and usually based on formulating, well-balanced, high-digestibility and nutritionally-complete diets that promote rapid growth combined with minimum FCR and maximum net protein utilization—to minimize nitrogen released into the water by the feed, while maximizing harvested nitrogen/protein.


I will be happy to share the details of the research offline.


Jorge Cordova ( Martin, what was the LD–50 [the dose required to kill half the members of a tested population after a specified test duration.] of nitrite you used in the challenge?


Martin J.M. Guerin ( Jorge, the study tested five concentration groups by equal logarithmic intervals, including concentrations of 33.2, 59.0, 105.0, 186.7 and 332.0 mg/L Nitrite-N.  Each group had two parallel tanks.  One hundred P. vannamei of the same size from the temporary-keeping tank were allotted equally to 10 tanks, in which nitrite was added to corresponding concentration.  Then P. vannamei surviving in 24 hours were counted and recorded.  During the experiment, the water temperature was controlled at 30 ±2°C and pH at 7.9.  Finally, LC50 (24h) was 290.4 mg/L, which was calculated according to the linear interpolation method.  This was the dose used in the challenge test with various organic selenium supplementation levels.


Initially, I was a bit surprised that such a high concentration was needed to kill 50% of the shrimp, however when reviewing the literature, I found that it was more or less in line with other studies.  The fact that our study was run at 30ppt salinity, 30°C, pH 7.9, with large shrimp (15-gram), and in well-aerated tanks and measured toxicity over 24 hours, not 48 hours, probably explains the high LC50-24h dose we found.


Dallas Weaver ( Daniel, Martin and others: Nitrite toxicity for catfish is a function of the chloride concentration.  The higher the chloride, the lower the toxicity.


“The formation of nitrite from ammonia via bacterial oxidation utilizes relatively slow growing bacteria (relative to heterotrophic bacteria feeding on carbon sources like molasses) and the second group of bacteria that convert nitrite to nitrate and eliminate the toxicity issue are also relatively slow growers and can’t grow at all until you have some nitrite in the system.”


“In biofloc systems, what is bio-chemically happening in the deepest part of the biofloc particles becomes critical to both nitrite formation and destruction.  With high enough oxygen levels and low enough metabolism rates of the biofloc (older floc), you can get full conversion to nitrate.  However, an increase in floc size, increase in floc metabolism rate (pulse of sugars) or a decrease in oxygen can lead to hypoxia and low ORP in the central part of the floc.  Under these very low DO and near zero ORP conditions, any oxygen diffusing into that area can convert ammonia into nitrite and other bacterial groups can remove oxygen from nitrate producing nitrite.  That nitrite can then diffuse towards the outside of the particle while oxygen is diffusing in where some of that nitrite can be converted into, but a lot of it can diffuse out into the water.”


“The same thing happens in thick biofilms in RAS and pond bottoms, and nitrite can become an issue in all these non-photosynthetic thick film systems.”


“With biofloc systems, the nutritional value of biofloc particles decreases as the particle ‘matures’ from young fast growing living organisms to one containing decayed and refractory organic materials.  This is the same process that occurs in sewerage, and the organic materials mature from ‘putrescible’ to ‘mature sludges’ or ‘stabilized sludges.’”


“However, to keep the sludge age (sludge retention time SRT) down in a biofloc systems to the point where the nutritional value is high with a high percentage of live single cell organisms would mean that the slower growing nitrification bacteria couldn’t survive.  To obtain short SRT and nutritious biofloc means lower floc densities (factor of 10 going from 10 day SRT to 1 day SRT) and higher carbon input rates and more oxygen.  This could work with your production animals at very high densities to consume all the feed required to even get a manageable floc density, but this would limit the system to one loading rate with zero nitrification and 100% of the ammonia producing single cell protein using a reasonably non-toxic carbon source like ethanol and tons of oxygen.  This could work for air breathing catfish species and tons of monitoring and automated controls.”


“Instead, biofloc systems use long SRT and produce a mature biofloc about as nutritious as ‘compost,’ but with nitrification capability.  The dream of nitrogen recycling becomes a real dream disconnected from reality.”


“In Daniels system, he uses the nitrogen to grow diatoms, but doesn’t have to keep the SRT up for more than seven days and recycles the nutrients through a copepod food chain.  It would be even better to use an algae that the shrimp can utilize directly like Spirulina, but that would require partitioning to obtain stability.”


Daniel Gruenberg ( Dallas, in our systems, diatoms are continuously removing nitrogen in different forms from the culture water.  Copepods eat the diatoms, and the diatom biomass is continuously turning over.  Dead diatoms sink to the pond bottom, becoming part of the detrital layer where the shrimp graze when not consuming pellets.


It’s the metabolic stress that nitrite puts on the animals that force them to spend energy on things other than growing.  Almost all floc systems end up with high FCRs.


Daniel Gruenberg ( The problem that Dallas pointed out is that with floc systems that can control nitrite, the floc nutritive value is crap.  On the other hand, if you have young flocs with good nutrition, they cannot control nitrite.  Since the cycle of most shrimp cultures is short, most farmers default on the high nutritive value young flocs and live with high nitrite, and this is at least one of the reasons I believe that floc systems never reach their theoretically possible low FCRs because the nitrite wreaks havoc on the nitrogen metabolism in the shrimp.


One of the key reasons to go with flocs was to recycle nitrogen into feeds that the shrimp could eat, but it does not work in the vast majority of biofloc systems.


Martin J.M. Guerin ( Our role as nutritionists is to formulate the most cost-effective feed for the targeted system.  As it is not always possible to formulate feeds for all production systems, it is often best to err on the side of safety and over-formulate your feeds.


RAS or NPM production systems provide “natural free-of-charge” nutrients that complement the feed and therefore can allow to under-formulate feeds and focus on specific macronutrients as well as the impact of the feed on water quality and pond ecosystem.  However, this “natural” supply of essential nutrients is not always consistent, hence some over-formulation is probably still needed.


In addition, whenever the “natural” system fails, which often happens at an unpredicted time, it is helpful to have a complete and well-balanced feed on hand.


Finally, whatever the system, stress always happens.  Extra nutrients, like organic selenium, help shrimp deal with stress.  Selenium is a co-factor of glutathione peroxidase one of the most critical antioxidant enzymes that help shrimp fight oxidative stress.  0.3ppm supplied in the form of hydroxy-selenomethionine has proven to be the optimum level both for growth and resistance to stress.  It is much more effective than selenium from raw materials and very safe to use as it deposits in the muscle, unlike inorganic selenium, which can have toxic effects.  The role of organic selenium on resistance to diseases (bacterial and viral) has also been well demonstrated.  So in case you use feeds specifically formulated for RAS, don't underestimate nutrients like selenium.


Durwood Dugger (, From a gross input standpoint, no other inputs drive the system like feed nutrients.  How those nutrients are metabolized by the system contributes to the economic feasibility of any shrimp production system.


The big weakness of the current terrestrial animal based least-cost, shrimp-formulating algorithms is that the assumptions are based on terrestrial animal digestion and or nutrient synthesis processes.  Yes, gross nutrient levels are fairly well established for shrimp.  However, shrimp are not terrestrial animals, which means that we are far from optimizing shrimp nutrition, especially when you take into consideration the huge role the pond plays in the shrimp nutrition process.


Commercial shrimp feeds tend to follow the tried and true terrestrial feed development process of feeding all the nutrients that make up a shrimp’s body—minus those that we have demonstrated or that we assume that the shrimp can synthesize from other nutrient inputs—plus atmospheric and metabolic gases and seawater chemicals.


Feed formulation is another one of the scaling difficulties for large-scale, recirculating systems.  Until you reach a certain size and can afford to generate your own feeds, you must use whatever commercial shrimp feed formulations are available, rather than those formulated (and adapted as needed) specifically for the design conditions of your specific RAS system.  Until your feed formulation reflects the repeatable conditions of your RAS system design, it is impossible to economically optimize the operating costs (particularly feed, energy and management) of your RAS system.  How significant this lack of economic optimization is varies by system design and the feeds used.  Most shrimp RAS systems to date have not been able to come close to competing in the global shrimp commodity markets.  Shrimp RAS feed economic optimization is one of the hurdles that must be overcome.


Dallas Weaver ( Martin, good comments.  One of the major differences between all the nutritional research on terrestrial animals and aquatic animals is related to having water as a medium and having some of the required nutrient components, including selenium as selenate (soluble ionic form), and all the water soluble vitamins being capable of diffusing out of the diet when they hit the water.


Terrestrial animals eat what we feed them.  What you put in the water for slow feeding animals, like shrimp, is not what they eat.  The water-soluble component can diffuse or leach out of the diet in very short time scales.  Just because a diet holds together mechanically doesn’t mean the chemistry doesn’t change.


For those who are curious about aquatic diets, you just need a magnetic stirrer, a beaker, a conductivity meter and a stopwatch to see the leaching rates on water-soluble ionic components in the diets.  Add the diet to the distilled water and measure the conductivity as a function of time.  Soluble vitamins (like C and the Bs), pure amino acids, taurine and many minerals (Na, K, and selenite) will diffuse out very rapidly.  You can loose half the soluble components in a one-millimeter particle in one minute.


Diffusion can make an “over-formulated” diet into a “under-formulated diet” in minutes.


Your selenium example is instructive.  If you look at the solubility of the newer organic selenium compounds such as 2-Hydroxy-4-MethylSelenoButanoic Acid, they are very stable and fully soluble in water.  This makes them stable enough to go through extruders and be mixed very homogeneously in the feed (important when using trace components that are both required and highly toxic with little room between the two concentration levels).  However, the literature may say it is good for aquatic diets, but in reality for larval, small or slow/messy feeding animals that 300 ppb in the diet may equilibrate with that near zero ppb in the water (discharge standards for industrial and mining discharges are down around 5 ppb total selenium — usually selenite and selenite).  Note: Selisseo® 2% Se liquid is a pure source of hydroxy-selenomethionine (OH-SeMet) and the only organic selenium soluble in aqueous media.


Shrimp diets that contain live organisms and fresh whole diets seem to work better than artificial diets because the nutrients the fresh diets are packaged within cell walls in living and fresh tissue.  Even if you start with the right nutrition and bind it with a hydrophilic binder, the animal may not consume the right nutrition.  The devil is really in the details on this issue, and the research is very far from complete and very expensive.  Results dependent upon “French cooking” type impacts where how and what order you mix ingredients is highly relevant.  The same ingredients mixed in a different order with different times/temperatures will result in different diffusion/leaching rates.


Being able to keep the water-soluble components in a prepared diet is one of the areas I have been researching for a long time.  I think I may have found a way to do it.


Eric Muylder (, Everybody, there are so many ways to utilize bioflocs and recirculation systems that one study is certainly not enough to draw conclusions.  I have done the floc/RAS comparison several times and mostly found better results with flocs, but the last time I compared them, RAS won.  If you have mature systems, the nitrite and nitrate are as high in RAS systems as they are in biofloc systems.  If, however, you use additional carbon in a biofloc system, nitrite and nitrate would be lower.  Normally, once your system is mature, nitrite is never a problem.  And nitrite is much less toxic in salt water than in fresh water.  I have seen nitrate levels as high as 60 ppm and 10 ppm ammonia (pH 8) without observing mortality.  But for sure, the shrimp were effected in some way.


I agree with Daniel that a system based on algae will surely outperform a system based on bacteria.  But in intensive systems, algae is not an option.


Daniel Gruenberg ( Hi Eric, I want to point out two things.


First, bioflocs can control nitrite, but the mature flocs that control nitrite contain much less nutrition.  Forget about the nutrition and go for the best possible water quality.  FCRs that I have seen in real floc systems are always higher than those in RAS or algae systems.  RAS can provide oligotrophic water quality (low nutrient levels) and better FCR if designed properly.


Second, algae can be used in super-intensive systems if you use partitions, and we are working on that!  Still too early to say, but I think it can be done.


Dallas Weaver ( Daniel, when talking about super-intensive systems and photosynthetic partitioned aquaculture systems (PAS), the culture part of the system may be super intensive, but the overall system is still limited in production per hectare by photosynthesis in the 10 grams of carbon fixation/m2/day range and the NPK uptake required for that growth.


However, the part of the system outside the high-intensity area with the desired animals can be shallow with lower costs per hectare along the lines of Dave Brune’s excellent research on PAS (partitioned aquaculture systems).  That water treatment area can also have low DO conditions at night and extreme high pH during the day, but with oxygen control on the water going into the culture system area and the CO2 generated by the high-intensity animals, you don’t have to pay the energy cost of maintaining high DO in the treatment area with algae that can take very low DO at night.


Are you familiar with Dave Brune’s work?  If not, there is a lot of information and experience there.


Sources: The Shrimp List (a mailing list for shrimp farmers).  Subjects: Update to The Shrimp on the Treadmill Video, Plus Research Comparing RAS and Biofloc Systems and Research Comparing RAS and Biofloc Systems.  August 18 to 21, 2017.  2. Bob Rosenberry, Shrimp News International, August 21, 2017.

Print This Page