AquariumStocking

How to Fix Cloudy Water in a New Tank

Published October 3, 2026. Built by Jeremy Panasuk.

Quick answer: A cloud in a new tank is not a diagnosis. Test total ammonia nitrogen and nitrite before you add a treatment. The Merck Veterinary Manual's management chapter says new tank syndrome usually occurs within the first 6 weeks, and that a tropical biofilter can take up to 8 weeks to establish. Its environmental-diseases chapter uses two other clocks: un-ionized ammonia rising in the first 2–3 weeks and/or nitrite rising after that, and about 6 weeks for a new biofilter to become completely established. Those are separate statements.

What the cloud is not

This page does not call white water a bacterial bloom. Neither cited Merck chapter defines a new-tank cloud that way. What they do define is new tank syndrome: a water-quality failure in a new aquarium, with fish that become lethargic, stop eating, and frequently die. Owners may report that everything was fine for several weeks and then the fish changed suddenly. The management chapter says the water should be tested, and that total ammonia nitrogen, nitrite, or both are often high enough to be toxic.

A cloudy cornea is also a different finding. In the environmental chapter, sublethal chlorine exposure can show up as ragged fins, excess mucus on skin and gills, and cloudy corneas — the eyes — plus lethargy or irritation. Chlorine is adverse to fish at concentrations as low as 0.02 mg/L, and it kills at 0.04 mg/L. No chlorine or chloramine should be detected in the tank. That eye sign is not a description of cloudy water.

Which tests, and which numbers stay separate

Most home kits report total ammonia nitrogen (TAN), not the toxic un-ionized fraction. Under most circumstances, freshwater fish can tolerate TAN concentrations under 1 mg/L. In most home aquariums, a TAN reading of 1.0 is not a huge concern unless pH exceeds 8.5. The same chapter still says the un-ionized fraction should be calculated. Un-ionized ammonia under 0.05 mg/L is not considered harmful, with the caveat that tolerance differs by species and situation. A calculated un-ionized ammonia of 0.05 mg/L is high enough to cause histologic gill changes and is considered harmful. Above 0.05 mg/L, gill tissue can show hyperplasia, hypertrophy of chloride cells, and lamellar fusion. Un-ionized ammonia of 2 mg/L is lethal for many fish. Fish may be lethargic and eat poorly; acute toxicity can look like spinning, disorientation, and convulsions.

A separate table row, labeled ammonia toxicity (new tank syndrome), lists un-ionized ammonia above 1 mg/L and pH above 8, with lethargy, anorexia, spinning, convulsive swimming, darkening, and catastrophic mortality. That row is not the 0.05 mg/L harm threshold and it is not the 2 mg/L lethality sentence. Report them as three different lines.

Nitrite kits often measure nitrite-nitrogen. Actual nitrite in mg/L is 3.3 times the nitrite-nitrogen reading. The same table lists nitrite above 0.1 mg/L with low chloride, fish piping at the surface, and gills and blood a dark brown that can be hard to see. Nitrite forms methemoglobin (brown blood), so red cells do not deliver oxygen. Many ornamental ponds and aquaria are kept with a residual salinity of 1–3 ppt from salt, which raises chloride and makes nitrite uptake less likely. That is a description of systems already run that way. This page does not turn it into a teaspoon dose.

The management chapter's new-tank table says one or both of ammonia and nitrate are higher, and it notes that nitrate is usually 0 mg/L. A nitrate-only test can miss the problem. Total alkalinity at or below 50 mg CaCO3/L (0.998 mEq/L) inhibits the biofilter. Buffer so total alkalinity is at least 100 mg CaCO3/L (1.997 mEq/L).

What the manuals say to change

Management of new tank syndrome, in that chapter's words, includes a decrease in feeding, water changes, the addition of chloride for nitrite toxicity, and evaluation of biofiltration. The environmental chapter is more specific once un-ionized ammonia is already high: if possible, a water change of at least 50% as soon as high un-ionized ammonia is detected; check whether the source water itself contains ammonia or chloramines, because a change can add ammonia directly or release it when chloramine is dechlorinated; if TAN is extremely high (above 5 mg/L) and pH is acidic (under 7), move the fish to a clean system already tempered for pH and temperature, so the rise in pH does not suddenly convert ammonium to un-ionized ammonia. Feeding should be discontinued or substantially reduced until the problem is corrected.

Fishless cycling, from the management chapter, means the tank is set up with no fish and ammonia is added to at least 1–5 mg/L. Test ammonia, nitrite, and nitrate. Add fish only once no ammonia or nitrite is present. The other method named there — slowly adding several fish over several months — can still produce high ammonia or nitrite if it is not done carefully, and the chapter says that practice can be inhumane.

The environmental chapter's maturation picture is daily TAN and nitrite checks plus frequent water changes until the filter cycles. Maturation shows when TAN and nitrite have peaked and dropped, and nitrate then rises. Beginning aquarists, it says, are likely to overstock and overfeed, which is how those spikes happen.

SRAC Publication No. 452 is a March 1999 revision about managing recirculating aquaculture systems, not a home-aquarium day count. In that sheet, a biofilter that can handle the ammonia and nitrite produced at normal feeding rates requires at least 1 month of activation, and stocking and feeding during activation should be greatly reduced. Conversion of ammonia to nitrite, by bacteria the sheet calls Nitrosomonas, takes place about 2 weeks into activation. At the first sign of high ammonia, feeding should be stopped. Overfeeding during activation is how that sheet describes fish dying.

Worked example on a new 10-gallon tank

Bioload points, watts per gallon, and filter turnover on this site are planning estimates for a cycled freshwater tank, not a published standard. The arithmetic below is the calculator's own functions, with the inputs written out. Open the stocking calculator and enter the same numbers.

10 gallons, normal setup, 15 small fish, no medium, large, or XL fish: 15 points against a capacity of 10, which is 150% and overstocked.

Points are 15 × 1 = 15. Capacity is round(10 × 1.0) = 10. Percent is round(15 / 10 × 100) = 150. Status is overstocked because 150 is above 100. That is the overstock the environmental chapter warns beginners about, expressed as this site's function. It is not a count of how many fish a university says a 10-gallon tank holds.

A first group of 4 small fish on the same tank is 4 points, capacity 10, round(4 / 10 × 100) = 40%, and comfortable because 40 is at or under 80. Comfortable applies after the tank is cycled. The management chapter still says fish go in only once ammonia and nitrite are absent.

Filter, 10 gallons, normal level: 5 turnovers an hour, ideal flow 50 gallons per hour, then a 20% derating rounded up to the next 5 gallons per hour. ceil(50 × 1.2 / 5) × 5 = 60 gallons per hour. That rating does not grow the bacteria. It is the flow to plan once they exist. Heater, same 10 gallons, room 68°F, target 78°F: the rise is 10°F. The function uses 4 watts per gallon when the rise is above 9°F and at most 15°F, so the raw load is 40 watts. The next standard size at or above 40, from 25, 50, 75, 100, 150, 200, 250, 300, 400, and 500, is 50 watts.

A common mistake

Treating the cloud, or deciding it is harmless, before ammonia and nitrite are tested. The management chapter says use of prophylactic medication in the absence of diagnostic testing is strongly discouraged, and that it may contribute to resistant bacterial infection and other complications. The other mistake is filling the 10-gallon to the 15-fish load above on day one. The calculator's overstocked result and the manuals' new-tank warning are the same error seen from two sides.

Limits, and when to call a veterinarian

If fish are lethargic, not eating, spinning, or dying in a tank set up in the last several weeks, that matches the chapters' description of new tank syndrome. This page cannot tell you which fish is which, or whether the cloud is algae, a precipitate, or something else. An aquatic veterinarian can examine the fish and the water.

Old tank syndrome is a different failure. The management chapter ties it to small, infrequent water changes, or to topping off evaporated water with no exchange, often in tanks with large carnivorous fish. Total ammonia can be high but non-toxic because pH is low. If the fish are still alive, return the water toward normal with daily small water changes, to avoid pH shock and the ammonia toxicosis that can occur as pH rises. After that, use regular large water changes. One large change into an old, acidic, high-ammonia tank is the opposite of that instruction.

Related pages: how to keep pet fish alive, how to add fish, how to treat sick fish, and what size filter a 20-gallon tank needs.

Shop an aquarium liquid test kit on Amazon

Affiliate Affiliate disclosure: as an Amazon Associate, we earn from qualifying purchases — at no cost to you. The kit is listed because both Merck chapters say the water has to be tested. This site has not reviewed a brand.

Compiled with AI assistance from the sources below. Spot an error? Email jpanasuk@gmail.com.

Sources