Is Your Sampling System Causing Effluent Toxicity?
The Sacramento Regional Wastewater Treatment Plant (SRWTP), located in Sacramento, CA (USA), conducts quarterly short-term chronic whole effluent toxicity (WET) tests with the water flea (Ceriodaphnia dubia), fathead minnow (Pimephales promelas), and green algae (Selenastrum
capricornutum) on dechlorinated final effluent (DFE) discharged to the Sacramento River. These bioassays are performed according to methods described in the USEPA 3rd Edition WET test protocols. In April, 2004, elevated toxicity was found in the C. dubia bioassay and
follow-up testing was required to verify and determine the persistence of this toxicity. A toxicity reduction evaluation (TRE) was initiated, as required by the plant's National Pollutant Discharge Elimination System permit, after toxicity was confirmed in the follow-up bioassay.
goal of the TRE was to identify and reduce any source of toxicity to achieve compliance with permitted toxicity level. SRWTP was also required to continue monthly accelerated C. dubia toxicity testing until showing six consecutive months with low toxicity. Steps for identifying and
eliminating toxicity are described by a tiered approach set forth in the USEPA TRE Guidance: 1) Information and Data Collection, 2) Facility Performance Evaluation, 3) Toxicity Identification Evaluation (TIE), and 4) Toxicity Source Evaluation. These TRE phases are not necessarily conducted
in sequence, and when possible can be conducted in parallel.
Intermittent toxicity, the quick loss of toxicity strength over time, and a complex effluent matrix complicated this TRE. Once toxicity was observed in a sample it was often too late to conduct further tests to identify the cause.
Therefore, TIEs, which identify the cause of toxicity by applying treatments to selectively remove specific toxicant classes, were conducted on initial samples without any guarantee that toxicity would be present. These TIEs had to be repeated until toxicity did occur and the treatments could
identify the class of toxicant. Despite these challenges, the TRE quickly ruled out conventional toxicants (e.g. ammonia, metals, chlorine, etc.) through TIEs and treatment process evaluations. Standard Phase I TIE treatments (e.g., zeolite, SPE columns, EDTA, aeration, and pH manipulation)
accounted for only a small background portion of toxicity associated with ammonia, non-polar organics, and metals. Surfactants and polymers were also ruled out as sources of toxicity, yet filtration TIEs implicated particle-associated toxicity. Statistical analysis (e.g., principal component
and discriminant analyses) of daily influent and effluent chemical parameters paired with daily sample toxicity tests in 2006, did not identify any specific toxicant. During this period, chronic C. dubia toxicity ranged from 1.2 to 50 TUc (100/concentration of effluent that causes
a 25 percent effect), while neither algae nor fathead minnows showed any effects.
Bacteria counts (heterotrophic plate counts) and a detailed particle analysis (counts of the 2–5 μm fractions) suggested that live bacteria and not particles themselves were causing toxicity. Order
of magnitude increases in bacteria were documented along the DFE lines following dechlorination (100s to 1×106 CFU), despite relatively high levels (20 mg/L) of effluent chlorination that effectively sterilized the treated wastewater. However, coliform counts and the absolute
number of bacteria did not correlate well with toxicity. Daily toxicity variability, the loss of toxicity over time, and high variability between test replicates supported the hypothesis of a biological toxicant. Non-standard TIE treatments such as sterilization (e.g., heating with autoclave
or microwave, UV exposure, and chlorination/ dechlorination) and centrifugation consistently removed up to 95% of toxicity. These data comprised a compelling weight-ofevidence that bacteria was the principle toxicant.
A source evaluation study focused on sample toxicity throughout
the effluent collection system found that post-dechlorination bacteria growth in the DFE composite autosamplers was the cause of toxicity. Confirmation studies successfully isolated and re-introduced several toxicity causing bacteria from the effluent matrix at concentrations relevant to effluent
samples. Two bacteria (Brevundimonas sp. and Pseudomonas sp.) were isolated and confirmed as contributors to toxicity at ambient concentrations. Changes to the sampling protocols were implemented in January, 2007, to reduce the potential for bacteria growth in composite samples.
These controls included new autosamplers with improved refrigeration, more frequent sample line changes, and a revised configuration to improve sample line purging efficiency. Toxicity was effectively reduced to background levels and the TRE ended after six months with low toxicity.
reproduction impairment to C. dubia was experienced at SRWTP from April, 2004, through September, 2007, and a TRE was conducted to identify and reduce this effluent toxicity. The TRE quickly ruled out conventional toxicants (e.g. ammonia, metals, chlorine, etc.) through TIEs, plant
process evaluations, and statistical analyses. The cause of toxicity was shown by weight-of-evidence approach to be from bacteria growth in DFE composite autosamplers. This indicated that elevated toxicity was an artifact of sampling and not a characteristic of effluent discharged to the receiving
water. To our knowledge this is the first known case of bacteria causing chronic toxicity to C. dubia from treated wastewater effluent. As a result of this TRE, the focus of toxicity removal shifted from additional effluent treatment and source controls to implementation of improved
sampling protocols to limit biological growth during sampling. TRE managers could benefit from considering this possible source of toxicity in their investigations.
More about this publication?