Hydrobiologia 379: 135���145, 1998. �� 1998 Kluwer Academic Publishers. Printed in the Netherlands. 135 Effects of acidic pH on benthic macroinvertebrate communities in stream microcosms Lisa A. Courtney & William H. Clements Department of Fishery and Wildlife Biology, Colorado State University, Fort Collins, CO 80523, U.S.A. Received 29 January 1998 in revised form 29 May 1998 accepted 12 June 1998 Key words: Stream acidification, macroinvertebrate communities, aquatic insects, stream microcosms Abstract The effects of acid (HNO3) on drift and survival of benthic invertebrate communities were assessed in stream microcosms over a 7-day exposure period. Communities were obtained from the Cache la Poudre River, Col- orado, using artificial substrates colonized in the stream for 30 days and then transferred to stream microcosms. Streams receiving the highest acid concentration (pH 4.0) contained significantly fewer individuals (F = 378.42, p 0.0001) and taxa (F = 7.8, p = 0.0123) at the end of the experiment compared to the other two treatments (pH 5.5, 6.5) and the control (pH 7.4). Reduced macroinvertebrate abundance resulted primarily from reduced abundance of mayflies (Ephemeroptera) which were particularly sensitive. Comparisons of Plecoptera, Trichoptera, and Diptera abundances showed no statistically significant differences among treatments. Analysis of invertebrate drift samples, collected after 2, 6, 18, and 42 h exposure, revealed that percent drift in the most acidic streams was nine times that of control streams. Ephemeroptera was the only aquatic insect order to exhibit a significant drift response, and timing and magnitude of responses varied among mayfly taxa. Differences in sensitivity to acid among aquatic insect orders observed in our experimental streams were similar to those reported from field studies in other regions. Effects of acid on drift and survival of benthic invertebrate communities were also similar to effects of heavy metals, one of the primary water quality concerns in the Rocky Mountain region. These results suggest a general pattern of responses to chemical stressors in benthic communities from Rocky Mountain streams. Introduction Acidification has severely impacted freshwater re- sources in Eastern North America and Europe, and the effects on aquatic biota have been studied exten- sively in these regions (Haines, 1981 Baker et al., 1991 Herrmann et al., 1993 Wigington et al., 1996). Analyses of benthic macroinvertebrate communities in acidic streams show a correlation between pH and in- vertebrate density and diversity and variation among taxa in response to acidic conditions (Simpson et al., 1985 Hall & Ide, 1987 Weatherly & Ormerod, 1987 Feldman & Connor, 1992 Rosemond et al., 1992). Experiments conducted at a range of spatial scales, from laboratory experiments to whole-stream manipu- lations, demonstrate adverse effects of acidic pH on benthic invertebrate communities and differences in sensitivity among aquatic insects (Bell, 1971 Hall et al., 1980 Hall & Ide, 1987 Allard & Moreau, 1987 Hopkins et al., 1989). Studies also show dif- ferences in sensitivity among aquatic insect life stages (Bell, 1971 Camargo, 1995) and document behavioral responses to acid exposure (Hall et al., 1980 Hop- kins et al., 1989 Bernard et al., 1990 Pennuto and deNoyelles, 1993). Although acidification of freshwater environments in the Rocky Mountain region has received less at- tention than in the eastern United States and Europe, studies have addressed the potential for acidification in the region. Gibson et al. (1983) examined geology and water chemistry data from watersheds in Rocky Mountain National Park and concluded that while there is no evidence of acidification, geochemistry data indicate that some areas in the region are acid- sensitive. Studies in Boulder (Lewis & Grant, 1980) hy83.tex 6/10/1998 18:38 p.1 Article: hy83 GSB: 7011886 Pips nr. 183234 (hydrkap:bio2fam) v.1.1

136 and Gunnison (Harte et al., 1985) Counties, Colorado, indicate that acidification from atmospheric deposi- tion has not occurred, but stress the importance of monitoring changes in water chemistry. More recently, Vertucci & Corn (1996) examined water chemistry data from central Colorado and southern Wyoming for evidence of episodic acidification and the potential ef- fects on amphibians. Consistent with other studies in the region, they found that episodic acidification may occur in high elevation habitats but found no evidence that episodic acidification has led to long-term acidic conditions. Although the literature suggests that acidification of aquatic ecosystems in the Rocky Mountain region has rarely occurred, experimental studies would in- crease our understanding of how aquatic communities in the region respond to acid stress. One question is whether responses of benthic invertebrate commu- nities to acidic pH in the Rocky Mountain region are similar to communities in other regions. Also of interest is whether the response of the benthic inver- tebrate community to acid exposure is unique or is a predictable, general response to chemical stressors. In this experiment, we examined the effects of acidic pH on the benthic invertebrate community from the Cache la Poudre River, Colorado. Benthic inverte- brate assemblages were obtained from the river using artificial substrates and exposed to a range of pH levels in stream microcosms. Our objectives were to assess the effects of acidification on behavior (invertebrate drift) and survival, to compare the responses among taxa, and to compare the responses to results from experimental studies in other regions. Methods Aquatic insect assemblages were obtained from the Cache la Poudre River, a fourth order stream located 50 km west of Fort Collins, Colorado. We collected aquatic insects using artificial substrates (10 �� 10 �� 6-cm plastic trays filled with pebble and small cob- ble) which were colonized in the stream for 30 days. Wooden racks were secured to the bottom of the river in a riffle area, and trays were secured to the racks with plastic ties. After the colonization period, 48 trays were removed from the river and transferred to the Stream Research Laboratory at the Colorado State University Foothills Campus. Each tray was removed by placing a net downstream to minimize loss of or- ganisms, and placed into a water-filled cooler (four trays per cooler). Aeration was provided throughout the transfer to laboratory and microcosms. The four trays in each cooler were assigned randomly to a stream microcosm. The oval, 13-L stream microcosms are flow-through systems (76��46��14-cm), with water drawn from the hypolimnion of Horsetooth Reservoir. A paddle wheel generated a current of 30 cm s-1. The procedure of using colonized substrates in the stream microcosms has been used successfully in previous ex- periments (Clements et al., 1988 Kiffney & Clements, 1994) and allows us to examine the effects of a stressor on an intact community in an experimental setting. After a 48-hour acclimation period, the streams were randomly assigned to one of four treatments: control, pH 4.5, pH 5.5, or pH 6.5, with three repli- cates of each treatment. Peristaltic pumps delivered nitric acid solution to each treated stream from a 20-L carboy at a rate of 10 ml min-1. Each carboy received an appropriate volume of HNO3 solution to obtain the target pH for each treated stream. Physical and chemical measurements were taken in the field when the trays were collected and in the stream microcosms. Field measurements included wa- ter temperature, dissolved oxygen, conductivity, pH, current velocity, and depth. pH and dissolved oxy- gen were measured daily in stream microcosms. Water samples were taken from the river and from experi- mental streams at the beginning and end of the exper- iment for analysis of alkalinity and hardness based on U.S. EPA guidelines (USEPA, 1983). Drift nets (mesh size 500 ��m), were placed in each stream when dosing began and remained in the streams for 42 hours. Organisms and debris collected in the drift nets were removed after 2, 6, 18, and 42 hours, preserved in 100% ethanol, and sorted and identified in the laboratory. After dosing for 7 days, the four trays in each stream were pooled and washed into a 355 ��m sieve. All samples were preserved in 100% ethanol and sorted in the laboratory. Organisms were counted, and identified to the lowest practical taxonomic unit. Because we expected acid exposure to have greater effects on smaller individuals, we measured head cap- sule widths of the most abundant mayfly, Rhithrogena hageni (Heptageniidae), in the control and the most acidic (pH 4.0) streams, at the end of the experiment. We measured all R. hageni in the pH 4.0 streams (average of 16 individuals per stream) and a random subsample of 50 individuals from each of the con- trol streams to the nearest 0.1 mm using an ocular micrometer. hy83.tex 6/10/1998 18:38 p.2