February 15-18, 2009 Washington State Convention Center Seattle ...

WATER QUALITY AND TREATMENT EFFICIENCY IN REPLICATED RECIRCULATING
SYSTEMS OPERATED WITH LOW FLUSHING CONDITIONS WITH AND WITHOUT
OZONE
John Davidson*, Christopher Good, Carla Welsh and Steven Summerfelt
The Conservation Fund’s Freshwater Institute
1098 Turner Road
Shepherdstown, WV 25443 USA
j.davidson@freshwaterinstitute.org
Anecdotal evidence from previous un-replicated experiments at the Freshwater Institute indicates that rainbow trout mortality
increases and health declines with low water exchange rates and feed loading rates of 1.3-2.0 kg/d per m 3 /d of make-up water
flow. The decline in fish health appeared to be ameliorated by ozonation and was unrelated to an infectious disease. Typical
water quality parameters were also within safe limits. To further investigate, a series of controlled studies are being conducted
in six replicated recirculating systems (RAS) using rainbow trout.
This study compared rainbow trout performance and water quality criteria between three RAS operated with ozone and three
RAS without ozone. All systems were operated at low water exchange rates. Rainbow trout were randomly stocked, 1000
fish/system. To begin the study ozone was turned on and low water exchange rates (0.26 % of the total recycled flow) were
established. At this time fish in the ozone and no ozone RAS were 294 ± 1 and 296 ± 2 g, respectively. Each system recycled
380 L/min of water, which allowed a 15 min tank turnover rate. Each system contained a fluidized-sand biofilter, CO 2 stripper,
low head oxygenator (LHO), LHO sump, single 5.3 m 3 culture tank, drum filter, particle trap, and a pump sump. A 24-hr photoperiod
was provided. During a one week period when fish were at maximum feed levels (6.6 kg/day/tank) and densities (80
kg/m 3 ), water samples were collected across all unit processes to compare water quality and unit process removal efficiencies.
Mean feed loading rates for all systems were approximately 4.66 kg/d per m 3 make-up water flow. All tanks were fed equal
portions during the first month of the study and during the water sampling event. Thereafter, feeding rates were adjusted based
on satiation and waste feed observations.
Ozone addition resulted in substantially lower BOD, TSS, color, total particles, heterotrophic bacteria, higher UV transmittance,
and also appears to have removed dissolved copper (Table 1). Removal efficiencies across all unit processes were also
better for the ozone treatment. Statistical analysis for water quality is pending and will be presented. Ozone provided conditions
for improved growth. After five months, rainbow trout in the ozone and no ozone RAS were 1161 ± 6 and 993 ± 12 g, respectively
(p = 0.001). Cumulative mortality for ozone and no ozone was 7 ± 2 vs. 17 ± 5 morts, respectively (p = 0.145). Survival
was not statistically different but was better for the ozone treatment.
3