Environ. Sci. Technol. 2006, 40, 5132-5138
Implications of Pulsed Chemical Exposures for Aquatic Life Criteria and Wastewater Permit Limits J E R O M E M . D I A M O N D , * ,† STEPHEN J. KLAINE,‡ AND JONATHAN B. BUTCHER§ Tetra Tech, Inc., 400 Red Brook Boulevard, Suite 200, Owings Mills, Maryland 21117, Department of Biological Sciences, Clemson University, 509 Westinghouse Road, Pendleton, South Carolina 29670, and Tetra Tech, Inc., P.O. Box 14409, Cape Fear Building, Suite 105, 3200 Chapel Hill-Nelson Highway, Research Triangle Park, North Carolina 27709
Subacute effects of pulsed copper, zinc, or ammonia exposures were examined, including a range of pulse concentrations, durations, frequencies, and recovery times between pulses, using short-term chronic Pimephales promelas and 21-d Daphnia magna tests. Sublethal effects were rarely observed independent of mortality. Effects were observed only at concentrations near the species continuous exposure 48 h LC50 for each chemical. Daphnia often rebounded from temporary reproduction effects, meeting or exceeding control responses by the end of the test. Effects of 24 h ammonia or copper pulses were diminished soon after the pulse was removed, while 24 h zinc pulses caused continued effects for several days following removal of the pulse, indicating a slower uptake and/or depuration rate for zinc. D. magna exhibited less mortality as copper pulses were spaced further apart, while fish were equally or more affected with longer recovery times between copper pulses, indicative of different adaptation mechanisms between the two species. Responses were not predictable based on either average concentration or a combination of duration and concentration. Chronic water quality criteria and effluent permit limits, expressed as a 4- or 30-d average concentration, respectively, may not be appropriate for protecting against effects of pulsed exposures, depending on the frequency, magnitude, and duration of pulses, as well as the recovery period between events.
Introduction Environmental regulatory bodies throughout the world rely on water quality criteria to set legally enforceable water quality standards. The process used to develop these criteria is based on results of controlled laboratory tests, in which test organisms are exposed to a constant concentration of a chemical. Through a number of fairly standardized procedures, test results for a given chemical (spanning many species in some cases) are converted to criteria with a fixed * Corresponding author e-mail:
[email protected]; phone: 410-356-8993; fax: 410-356-9005. † Tetra Tech, Inc., Owings Mills, MD. ‡ Clemson University. § Tetra Tech, Inc., Research Triangle Park, NC. 5132
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 16, 2006
duration and frequency regardless of the type of chemical and its mode of action (1). A major concern regarding the current paradigm for development of water quality standards is that it uses continuous exposures to develop limits protective of episodic, variable exposures typical in the environment. Role of Frequency in Chemical Exposures. Toxicologists have long recognized that the frequency, magnitude, and duration of chemical exposure determines organism response and that the results of the interplay of these three factors can be difficult to predict (2-6). Furthermore, studies examining nonconstant or pulsed chemical exposures have shown either greater (e.g., 7, 8) or lesser (e.g., 9, 10) effects on test organisms than would be predicted based on results using a constant exposure at similar concentrations, depending on the chemical, test species, and test design used. Although the effects of pulsed exposures have generated much research, a clear understanding of sublethal or chronic effects from such exposures is currently lacking because: (1) frequency, magnitude, and duration could be varied simultaneously, resulting in potentially large, expensive, experimental designs and complex models, and (2) the objectives addressed in pulsed experiments have varied widely (e.g., academic toxicology questions, stormwater exposures, pesticide applications) making it difficult to integrate results and construct usable models (6). Need to Incorporate Frequency into Regulatory Decision-Making. Results of our previous work (11) suggested that sublethal effects (effects on fish growth or daphnid reproduction) of pulsed exposures involving acutely toxic chemical concentrations (i.e., high magnitude over durations of 48-96 h) could be reasonably predicted using current models (12-15). These are one-compartment kinetic models that assume a fairly simplistic relationship between accumulation of a contaminant inside the organism and effects (i.e., mortality). Time lags between the accumulation of a contaminant and an effect, chemicals that require longer depuration periods, and effects of multiple pulses over time are not generally addressed by these models (5, 16-18). A major information gap, we observed, was the relative lack of data from pulsed exposures using subacute concentrations; i.e., concentrations substantially greater than the continuous chronic effect level but less than the continuous acute effect level (e.g., 48 h or 96 h LC50), or the effects of multiple pulses. Such exposures appear to be especially relevant to treated wastewater discharges but may be applicable to many nonpoint source exposures as well. This research evaluated sublethal effects of pulsed exposures with the goal of providing useful information for water quality criteria applications, and, specifically, wastewater discharge permit limit development in the United States. We examined a range of magnitudes, durations, frequencies, and recovery time periods between pulses, for three different contaminants (copper, ammonia, and zinc) and two indicator test species. This paper presents representative results that illustrate the major findings of this research. More detailed information can be found in Diamond et al. (19).
Experimental Section Observations of Organism Response to Chemical Exposures. Pimephales promelas (fathead minnow) and Daphnia magna (water flea, Cladocera) were used in toxicity testing. Both species were cultured and tested using moderately hard synthetic water (20). Water hardness and alkalinity ranged between 94 and 106 mg/L and 58 and 66 mg/L as CaCO3, 10.1021/es0604358 CCC: $33.50
2006 American Chemical Society Published on Web 07/14/2006
FIGURE 1. (A) Schematic depicting the general experimental design for a given set of tests for a single chemical and test species, and (B) example pulsing regimes used in experiments involving one or two pulses and different recovery times between pulses. Magnitudes (chemical concentrations) used were controls (no chemical exposure) and fractions of the 48-h continuous exposure median lethal concentration (LC50). respectively, and pH was between 7.6 and 7.9 in this study and never varied more than 0.2 pH units within a test. The 21 d Daphnia test method (18) was used in D. magna testing, which is a standard freshwater test method used for criteria development in the United States and yields both survival and reproduction endpoints. D. magna were cultured at the Institute for Environmental Toxicology, Clemson University (Pendleton, SC) and were 0.25 mg and D. magna reproduction was always
>40 offspring/female, in accordance with their respective test acceptance criteria. Copper as copper sulfate pentahydrate, ammonia as ammonium chloride, and zinc as zinc chloride were selected as contaminants because they are common treated effluent constituents; they represent a range of modes of toxic action to aquatic life and, therefore, potentially elicit different types of organism responses to pulsed exposures. Stock chemical solutions were prepared using reagent grade chemicals in laboratory culture water. Culture water had negligible concentrations of these chemicals (50% mortality from a single 24 h pulse, based on toxicity test data reported by USEPA (22-24) as well as data from screening tests conducted prior to definitive testing in this research. Pulse concentration was constant for a given treatment when multiple pulses were applied during a test. In addition to pulsed exposures, some continuous exposure tests were also used in this study to help determine appropriate pulse concentrations given a desired duration, and to compare organism responses under the two types of exposure regimes. 5134
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 16, 2006
Continuous exposure tests were conducted in the same manner as pulsed exposure tests described above. Chemical concentrations were verified at the beginning and end of each pulse as well as at least once under nonpulsed conditions (i.e., culture water only) using standard analytical methods (25), along with test blanks, spiked samples, and duplicate analyses (19). More detail on quality control can be found in ref 26. Focus on Number of Pulses and Recovery Time Between Pulses. This research focused on two experimental factors: frequency and recovery period between pulses. These were found to be particularly important for predicting effects of pulsed exposures, based on our previous work (6, 11). Pulse frequency ranged from one to three pulses in a given test
FIGURE 2. Cumulative reproduction of surviving D. magna in 21 d tests in response to pulsed ammonia exposures. Arrows indicate initiation of pulse. No effects on D. magna survival were observed in these treatments (mean survival >85% in all treatments; ANOVA; p > 0.10). and recovery time between pulses ranged between 24 and 120 h (or 192 h for D. magna; Figure 1). In a given treatment (i.e., combination of pulse concentration, duration, and frequency), the recovery time used between pulses was constant. Rationale of Statistical Analysis. We analyzed statistical significance of main effects (i.e., frequency, duration, recovery time) that were varied within a given experiment using ANOVA. For most of this research, pulse magnitude and duration were fixed factors and experiments focused on frequency, recovery time, and the interaction between these two factors. Duncan Multiple Range (DMR) Test (p < 0.05) was used to test for significant differences in organism response among treatments in a given experiment if ANOVA indicated significant effects. Separate ANOVA and a posteriori DMR tests were conducted for each type of response (survival, reproduction, or growth) and each species for each experiment. All of the data generated in this research can be obtained from the Water Environment Research Foundation (www.werf.org; research project 02-WSM-3).
Results and Discussion Mortality is the Dominant Organism Response. Most of the treatments did not elicit measurable sublethal effects on early life stage fish growth or D. magna reproduction, independent of effects on survival (Table 1; ANOVA; DMR test). In most cases, between 20 and 60% mortality was observed in 24 h pulsed exposures (19, 26). Responses of test organisms to e24 h pulses, even multiple pulses a day or two apart, were observed only at chemical concentrations 5-10 times higher than continuous exposure chronic values reported for either species (Table 1; 19). Effects occurred only using pulse concentrations at or near the species 48 h LC50 for either copper, zinc, or ammonia in these experiments (Table 1; 19). D. magna often demonstrated the ability to rebound from pulsed exposures, and meet or exceed control responses in reproduction if given sufficient time (typically a few days; Figure 2), even in the absence of effects on organism survival (ANOVA, p > 0.30). Preliminary results in this research suggested that fish growth was also capable of rebounding
FIGURE 3. Comparison of daily P. promelas and D. magna mean survival following a single 24 h pulse of either 300 µg/L zinc, 40 µg/L copper, or 50 mg/L ammonia. Pulses were administered on Day 0. 95% confidence intervals for survival were e20% of the mean values shown for all data points. following a brief pulse, even when fish survival was unaffected (19). These results may help explain why few sublethal effects were observed in this research independent of mortality effects. This type of resiliency to pulsed chemical exposures has been reported in natural populations (e.g., 27). Mortality effects in pulsed tests were usually completed within 48 h of pulse termination for copper and ammonia (Figure 3). Similar results were reported by others using pulsed exposures of these chemicals and either test species (2, 28, 29). Zhao and Newman (18), however, reported continued mortality of the amphipod Hyalella azteca up to 64 h following a 48 h continuous exposure to copper sulfate. Our research used pulses no longer than 24 h in duration, which may explain some of the difference in results with this chemical. Other researchers noted that the closer the magnitude and duration of the pulse is to a continuous exposure acute LC50 concentration, the more likely that latent mortality might occur (3, 7, 18). Continued mortality effects were observed using 24 h zinc pulses and both species in our experiments (Figure 3). This result suggested either a slow depuration rate or a slow organism response time for zinc. The latter hypothesis was further suggested in D. magna testing. Single VOL. 40, NO. 16, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5135
TABLE 2. Significance (p Values from ANOVA) of Recovery Time, Frequency, and the Interaction between the Two Factors on Survival and Either Growth (P. Promelas) or Reproduction (Daphnia) in Experiments Using a 24 h Pulse Duration and a Single Pulsed Chemical Concentration for Each Speciesa survival test
recovery time
frequency
ammonia zinc
7d 14 d 7d 7d
