KI Nitrifier – Nitrifying Oxidizing Ammonia

Nitrifying Oxidizing Ammonia and Nitrate

Texas A&M AgriLife Research Mariculture Laboratory at Flour Bluff

Use of KI-Nitrifier to Inoculate Nursery Raceways – Production Season 2014

Report Prepared for Keeton Industry

High nitrite concentrations (up to 57 mg L-1 NO2-N) were recorded during the nursery phase of Litopenaeus vannamei culture in super-intensive, biofloc-dominated raceways during the 2013 production season. This had a negative effect on shrimp performance, causing significant mortality. These high nitrite concentrations occurred due to the extended period of time required for nitrite-oxidizing bacteria (NOB) to become established in the system. Therefore, in an attempt to avoid the same problem occurring again in the 2014 production season, a commercial product designed to inoculate the raceways with nitrifying bacteria was tested. KI-NitrifierTM (Keeton Industries, Inc., Wellington, Colorado) is a “super concentrated microbial blend for biofiltration activation designed to quickly restart difficult biological filters by providing an immediate source of nitrifying microorganisms for biochemically oxidizing ammonia and nitrite.” One month before shrimp postlarvae (PL) were stocked into raceways (RWs), a trial was conducted in independent 10,000 L tanks to test the efficacy of KI-NitrifierTM at oxidizing nitrite. This trial had the secondary purpose of inoculating media with nitrifying bacteria to be added to RWs to accelerate the development of nitrifying bacteria in these systems once PL were stocked.

Materials & Methods

Water was drawn from Laguna Madre (salinity – 45 ppt) and sterilized with 10 ppm chlorine and then neutralized with sodium thiosulfate. This was pumped into nineteen 10,000 L flat-bottomed fiberglass tanks (diameter – 3.65 m) with a central standpipe and the salinity in each was reduced to 25-26 ppt with municipal water. The final volume in each tank was 4,000 L. Each tank had 10 air stones evenly distributed around the outer walls.  This water was aerated for three days, after which sand was added to some tanks. Aeration was continued for the duration of the trial. Enough sand was added to cover the base of each tank to 8.2 mm depth. This was made up of 20% local intertidal beach-sand (unsterilized) obtained from South Padre Island and 80% bagged sand (Sakrete All-Purpose Sand – Medium coarse sand for general use. Product # 15160. A graded, washed, dried & sanitized sand). Water samples were taken immediately before and 6 h after adding sand. The experimental treatments were:

– Water only (Control) (W) – 4 replicates

– Sand only (S) – 5 replicates

– KI-NitrifierTM 100% of recommended dosage rate (A) – 4 replicates

– KI-NitrifierTM 50% of recommended dosage rate (B) – 4 replicates

– KI-NitrifierTM 200% of recommended dosage rate (AB) – 4 replicates

Two days later, all tanks containing sand, had all the sand in them thoroughly suspended with a broom. A separate broom was used for each treatment and was rinsed in municipal water between each use. This process occurred daily throughout the trial to prevent sand accumulating in certain areas of the tank and creating anoxic zones. One hour after the initial sweeping, 10 mg L-1 NO2 (76.28 g KNO2 97%) was added to each tank. Each tank was agitated with a bucket to mix the NO2 as it was added and again after 30 minutes. After one hour the bacterial treatments were inoculated: KI-NitrifierTM at 100%: 1.32 g / 4,000 L (80 g per 8,000 gallons), 50%: 0.66 g / 4,000 L, and 200%: 2.8 g / 4,000 L of the recommended dosage. The KI Nitrifier was mixed with 1 L of tank water and then poured into the tanks. This inoculation was repeated after 3, 6 and 9 more days. Once the NO2 concentration in an individual tank reached 0 mg L-1, 5 mg L-1 NO2 (38.14 g KNO2¬ 97%) was added to that tank.

Temperature, salinity, pH and dissolved oxygen (DO) were recorded twice daily in each tank (08:00 and 15:00) with a hand-held meter (650 YSI Inc., Yellow Springs, Ohio). This meter was rinsed in municipal water between each treatment to avoid cross-contamination. Water samples were taken daily from each tank to measure NO2-N and alkalinity. TAN and NO3-N were measured weekly. TAN, NO2-N and NO3-N concentrations were measured with a FIAlab auto-analyzer and alkalinity by titration (Standard Method #2320 B).

At the conclusion of the inoculation period (28 days after sand addition), the water from each of the tanks, except treatment W, was partially mixed and pumped into six 40 m3 and two 100 m3 RWs to be stocked with shrimp PL. This constituted approximately 10% of the total initial water volume in each RW. PLs were stocked into RWs and the nursery trial commenced the following day. In addition, KI Nitrifier was added to each RW at a rate of 1 g / 3,785.41 L (10.57 g / 40 m3, 26.42 g / 100 m3) on days 1, 4, 7, 10, and 32 of the nursery.   

Results and Discussion

Only three tanks depleted 100% of the NO2 during the trial – one in Treatment A (100%) on day 22 after sand addition, and two in Treatment B (50%) on days 18 and 25. Five mg L-1 NO2 was added to these tanks on days 23 and 25. Therefore the final day of the trial was counted as day 23.

Nitrite concentrations steadily declined in all tanks over the course of the experimental period (Figure 1). In general, daily nitrite concentrations between treatments increased in the order of B < A < AB < S < W. Final nitrite concentrations were significantly lower (P<0.05) with 50 and 100% inoculation (A and B) than with no inoculation or sand (W) (Table 1). This demonstrates that the KI-NitrifierTM was successful at stimulating the production of NOB to oxidize nitrite. However, the dosage rate (50-200% of the recommended dose) made no significant difference (P>0.05) to nitrite concentrations under the conditions of this study. In fact, the lowest inoculation rate (B – 50%) resulted in the lowest nitrite concentrations, and two of the three tanks in which 100% of the nitrite was removed received this treatment.

There were minimal differences in mean temperature and salinity between treatments. Mean afternoon DO was significantly lower (P<0.05) in W than in A, AB, and B, and pH increased from morning to afternoon in all treatments. Maximum pH over the experimental period was lower in W (AM: 8.85; PM: 9.00) than in the other treatments (AM: 9.02-9.10; PM: 9.14-9.25).

Table 1. Mean ± SD NO2 concentration at the conclusion of the trial (Day 23 after sand added) and mean temperature, salinity, DO and pH in the morning and afternoon over the whole trial.


able 2. Mean ± SD alkalinity immediately before and 6-hours after sand was added to the tanks, at the conclusion of the trial and over the whole trial.


These differences may be due to the brown algae blooms that were observed in most inoculated tanks (A, AB, and B) between days 5-10 and 17-21. The blooms crashed and left some foam on the water surface at the conclusion of each of these two periods. The daily pH was closely correlated with the development of the algae blooms. Minimal bloom was observed in the S tanks and none in W tanks, although some foam was observed on the surface of W tanks on days 9-10. Temperatures were low for the majority of the trial period, ranging from 11.33 – 23.25oC in the morning to 13.03 – 24.42oC in the afternoon. This may have slowed the development of nitrifying bacteria and limited differences between treatments. Alkalinity increased significantly in most tanks after the sand was added and increased significantly by the conclusion of the trial (Table 2). Carbonates in the sand may have increased the initial alkalinity and the algal blooms may have contributed to the later increase, offsetting any reduction in alkalinity caused by the nitrifying bacteria. Ammonia concentrations were 0 mg L-1 in the early stages of the trial. Nitrate ranged from 3.33 to 9.00 mg L-1 when the sand was added. Measured ammonia and nitrate concentrations were untrustworthy thereafter due to problems with the FIAlab Auto-analyzer.


In the 2013 shrimp nursery trial, when no nitrifying bacteria were inoculated, NO2-N concentrations peaked at 57 mg L-1 in the 40 m3 RWs and caused PL mortality. In the 2014 nursery trial NO2-N only peaked at 10.93 mg L-1 in one RW (much lower in other RWs) and had no observed negative impact on PL (Figure 2). Similarly, ammonia concentrations only peaked at 4.95 mg L-1 TAN (Figure 3). Note that supplemental carbon (molasses in 2013 and white sugar in 2014) was also used to manage ammonia and nitrite during the nursery. Ammonia and nitrite concentrations in the subsequent grow-out trial, using water and shrimp from the nursery trial, were negligible. This suggests that using KI-NitrifierTM to inoculate RWs was beneficial for accelerating the development of nitrifying bacteria and limiting the accumulation of ammonia and nitrite to toxic concentrations. Whether this improvement was due to inoculation with water from the earlier trial (described above) or due to direct inoculation into the RWs or both is impossible to ascertain based on the data obtained.    


Another interesting point to note was the difference in development of nitrifying bacteria between the two different raceway systems, 40 m3 (RW1-6) and 100 m3 (B1 & B2). Ammonia oxidizing bacteria (AOB) developed sooner, resulting in lower ammonia peaks in the 100 m3 RWs (Figure 3: B1 & B2) than in the 40 m3 RWs. In addition, nitrite concentrations remained close to 0 mg L-1 in the 100 m3 RWs once NOB were established, by 7 June in B1 and 13 June in B2 (Figure 2). In contrast there was much more fluctuation in nitrite concentrations in the 40 m3 RWs once NOB appeared to have been established, from 13 June onwards. The 100 m3 RWs were aerated and mixed with pump-driven a3 nozzles only, while blower driven air diffusers and air lift pumps, and a pump driven Venturi injector and nozzles were used in the 40 m3 RWs. All RWs were also manually mixed. The 100 m3 RWs appeared to be more evenly mixed, biofloc developed sooner and alkalinity declined at a faster rate than in the 40 m3 RWs. The mean am-pm DO over the nursery period in the 100 m3 RWs (6.55-6.79 mg L-1) was also slighter higher than in the 40 m3 RWs (6.36-6.57 mg L-1). This suggests that the design of the 100 m3 RWs, with a3 nozzles, provided a superior environment for the development of nitrifying bacteria, through enhanced mixing and higher DO.


In conclusion, results from both the tank and RWs studies indicate that KI-Nitrifier was beneficial for accelerating the development of nitrifying bacteria and limiting the accumulation of nitrite to toxic concentrations. The dosage rate (50-200% of the recommended dose) made no significant difference to nitrite concentrations. The low temperatures experienced during both trials may have slowed the development of nitrifying bacteria.