By Dallas Weaver*
Once the oxygen issue previously discussed is solved, the next problem confronting an aquaculturist is often nitrogen in the form of ammonia as a waste product from the fish or shrimp being cultured. This rapidly becomes a very complex subject dependent upon the details and will take more than one column to adequately discuss.
When any external fish/shrimp feed goes into a culture system, the culture species consumes and metabolizes this food. Its ability to convert all the protein into its own growth is far from 100%. Proteins are made from amino acids that contain fixed nitrogen containing molecules required for life and growth, however excess amino acids ingested by the animals are metabolized to energy and ammonia. In our bodies, we convert ammonia + carbon dioxide into urea, which is much less toxic to cells than the ammonia. Being non-toxic, we can excrete urea at high concentrations in our urine, but this conversion requires metabolic energy. Aquatic animals, breathing water, effectively have about a million kg of water passing over their gills for every kg of food metabolized. This opens up the options of directly dumping waste ammonia into the water. Un-ionized ammonia can pass directly from the blood through the gills into the water, as long as the un-ionized ammonia levels in the water are less than the blood.
As a rule of thumb, somewhere around 3 to 4% of the input feed will show up as waste ammonia (as N) going into the water. A 100 mg/l feed rate input to your system will result in an elevated TAN reading (total ammonia nitrogen – what a test kit usually measures as N). The addition to the water would be 3 to 4 mg/l as N.
Ammonia in water can exist as both ionized ammonia (NH3+) and un-ionized ammonia (NH3). The relative amounts depend upon the pH of the water. The relationship can be describes as:
NH3= (TAN) / (1+10 (pKa – pH))
where: [TAN] is the measured concentration of total ammonia nitrogen (mg/L); pKa , the acidity constant for the reaction (9.40 at 20 ºC); and pH, the pH of the water. Many calculators and spread sheets are available on the internet for doing these calculations.
Note that the pH dependence is very strong (exponential) and a TAN in the 3 mg/l range has low (non-toxic) levels of un-ionized ammonia at a pH of 7: a pH of 9 will kill the fish. One pH unit change will make an almost tenfold change in the un-ionized ammonia concentration and corresponding toxicity. If you keep the un-ionized ammonia < 20µg/l, there appears to be few health issues.
The question now becomes how the aquaculturist handles this ammonia. Flow-through aquaculture and net pens simply depend upon high flow rates to remove the ammonia. The ammonia problem goes elsewhere. With huge amounts of water such as found in offshore net pens, the ammonia is diluted to way below background levels. In non-offshore situations where the volumes of water are smaller, consideration must be given to the potential eutrophication effects of high nutrient water going into the environment.
In flow-through systems using pure oxygen to solve the oxygen problem, the CO2 produced by the fish metabolism can decrease the pH and the un-ionized ammonia allowing higher ammonia levels in the discharge. This allows for higher feed/flow ratios. If this water is then used for agriculture, the ammonia is effectively recycled as fertilizer for more plant protein production.
Photosynthetic based systems ranging from green water ponds to aquaponics growing emergent plants all use the solar energy to produce plant proteins utilizing the ammonia and other forms of fixed N from the water. In the case of algae based systems or submerged aquatic plants, which also use CO2 from the water along with the ammonia, the carbon dioxide removal will increase the pH and the concentrations of un-ionized ammonia. This can make a delicate balancing game where sunlight increases the pH and the ammonia becomes highly toxic with pH level > 9 in the afternoon. Depending upon the alkalinity and water depth, you can have a situation where your TAN is going down but the pH is going up fast enough with photosynthesis to dramatically increase the un-ionized ammonia toxicity (the opposite of the pure O2 flow through case above). Higher alkalinity per unit area (water alkalinity times water depth) will decrease the pH daily swing caused by photosynthesis, while higher alkalinity will increase the average pH.
Solar energy systems can only produce about 20 gm of biomass growth per m2/day, which, with a 30% protein algae, would use about 1 gm/m2/day of N. This can be translated into about 20-30 g of feed / m2 /day (300 kg/Ha) being near the maximum feed rate per area that can go into a photosynthetic based system of a pond or an aquaponics system (based upon growing area of photosynthetic plants). If the fish/shrimp consume the algae/plants, the yield of fish per area will increase as reflected in the effective feed conversion ratio (FCR), but the overall system is still limited on N or feed input per area.
Instead of solar energy, chemical energy, in the form of carbohydrates can be added to the water. The carbohydrates feed fast growing heterotrophic bacteria, which utilize the ammonia and other fixed N sources in the water to synthesize amino acids and proteins necessary for growth. This approach using chemical energy as sugars or other carbohydrates is often referred to as biofloc technology because the bacteria biomass being produced tends to flocculate into small particles. Chemical energy driven systems are limited only by the rate of oxygen addition necessary to supply the bacteria and by the species and the social behavior of the animals.
Both solar and carbohydrate driven ammonia removal systems don’t actually remove the ammonia (N), they just store the ammonia N as proteins in the biomass. This biomass could be consumed by the fish, or could be used to feed a food chain whose higher links are consumed by the fish (with corresponding internal ammonia production), or removed from the system.
Both solar and carbohydrate systems have a variety of bacterial species: slower growing species of bacteria (relative to algae or heterotrophic bacteria feeding on sugars) that convert ammonia (NH3) to nitrite (NO2- ) and another group of bacteria that further oxidize nitrite to nitrate (NO3- ). The conversion of ammonia to nitrate makes the waste nitrogen relatively non-toxic to fish/shrimp and most aquatic organisms. Nitrifying bacteria, with their slow growth rates, can’t survive in any system in which the average age of the biofloc is less than about a week.
Growing these ammonia-oxidizing bacteria, usually on surfaces, becomes the key process in recycle aquaculture systems (RAS) where the water is highly recycled and the fish are produced in very high intense systems. Instead of being limited to the 1 gm of ammonia that can be used by photosynthesis / m2 / day in an algae pond or producing 10 gm or more of bacterial biomass per g of ammonia sequestered in a biofloc system, the autotrophic bacteria living on converting ammonia to nitrite and nitrate only use a fraction of the ammonia for protein synthesis and growth, with most ammonia being used for energy production.
The nitrification bacteria seem to perform better in surface biofilms with abilities to handle up to about 0.4 g (N) / m2 / day. This is the same range as photosynthesis, but unlike solar energy driven system, these RAS biofilms can be 3-dimensional with very high surface areas per m3 (from 100’s of m2/m3 for moving beds, trickling filters, submerged filters, RBC’s, etc. to many thousands of m2/m3 for fluidized bed systems).
The biofilters for RAS can be viewed as habitats for bacteria on surfaces with the ability to deliver both oxygen and ammonia to the bacteria, where the bacteria will react to the two chemicals forming, ultimately, nitrate. Of course converting ammonia (a form of alkalinity -- 1 meq per mmole) to nitrate (nitric acid using 1 meq/mmol of alkalinity to be neutralized) results in the loss of 2 meq /mmol of alkalinity. In more common units, oxidizing 1 mg/l of ammonia to nitrate in the biofilter uses 7.14 mg/l of alkalinity (as CaCO3). As the ammonia added by the fish was only a temporary addition to the alkalinity pool of the water, you can really say that you only lost and have to make up 3.57 mg/l of alkalinity.
In the main part of a recycle system and biofilter you usually don’t actually eliminate the fixed N, you just convert it to a much less toxic form. In the next column, I hope to go into how you can convert that fixed N into ordinary atmospheric N2 and remove it from the system allowing true zero discharge recycle aquaculture systems to be built along with producing anammox bacteria that live on ammonia and nitrite as their energy supply.
Dallas Weaver, PhD, started designing and building closed aquaculture systems in 1973 and worked for several engineering/consulting companies in the fields of air pollution, liquid wastes, and solid wastes until 1980. Today, he’s the Owner/President of Scientific Hatcheries.