Establishment and Field Testing of a Rapid Bioassessment Screening of Rhode Island Freshwater Benthic Macroinvertebrates. Completion of a Research Project for Rhode Island Department of Environmental Management by Mark Gould School of Science and Mathematics Roger Williams University Bristol, RI 02809 December 1993 INTRODUCTION Previous studies by the author (Gould 1991, 1992) resulted in the enumeration of the freshwater macroinvertebrates in Rhode Island over a period of two years. These studies, along with the present study, provide a baseline for further studies and the documentation of freshwater macrofauna present within the state. The 1990 to 1991 sampling of the Rhode Island streams yielded significant information concerning the distribution of the macrofauna. The methodology provided a fast collection and data interpretation device once proper identification in the field was obtained. The 1991 to 1992 study confirmed the methodology, and began to develop trends within the macrofauna populations. The continuation of the study in 1992 to 1993 presented in this paper, further refined the collection and analytical techniques and reports on longer term population structures within the streams of Rhode Island. Stream invertebrates are well-adapted to their environment. Many species exist in the larval stage for a year or more; the adult often emerges for one or two days, mates, and dies. The survival of the species is dependent upon favorable environmental conditions in the water column. If conditions are not conducive at any time for the survival of a particular species, the stream will not support such a population. Species that occur in the freshwater environment are products of long-term environmental success. The density and species composition of freshwater benthic macroinvertebrates are controlled by many factors. These parameters may include substratum, temperature, oxygen concentration, organic content, and water chemistry. During the summer of 1993, the second driest summer on record occurred. Surface waters were at low levels. Many riffle communities were almost non-riffle environments. This produced a "worst case scenario" for the invertebrates within the streams. The presence or absence of an organism can measure the effects of all past, short-, and long-term environmental stresses. Therefore, the use of benthic macroinvertebrates as indicator organisms for the evaluation of the quality of a water body is a valuable tool for monitoring aquatic ecosystems. It is relatively difficult to obtain quantitative samples of benthic invertebrates due to: a) the heterogeneity of the habitat type; b) the depth of the organisms in the substrate; c) the stage in the life of the organism (many insects emerge as winged adults); d) variations in discharge, environmental conditions such as ice, etc.; and e) movement and transport of the organisms. In Rhode Island two major human contributed factors that may affect species density and diversity are the amount of organic enrichment from point and non-point sources. In order to determine whether or not these factors affect water quality, this study continued to evaluate the aquatic macro-invertebrates found throughout Rhode Island. The information was analyzed to determine species composition and ecosystem structure, as well as to determine reference baseline data for freshwater habitats in the state. Rhode Island water quality for freshwaters have been classified into three categories: Class A, Class B, and Class C. Class A waters are suitable for water supply and all other water uses. Class B waters are suitable for bathing, other recreational purposes, agricultural uses, industrial processes and cooling, excellent fish and wildlife habitat, good aesthetic value, and acceptable for Z public water supply with appropriate treatment. Class C waters are suitable for fish and wildlife habitat, recreational boating, industrial processes and cooling, and good aesthetic value. These classifications are the result of standards that primarily reflect point source discharges into the rivers from treatment plants and industry. This survey sampled 24 Class A streams, 14 Class B streams and 2 Class C streams to determine if the macroinvertebrates would confirm these classifications. Further classification of streams according to stream order (first through fifth) was completed. The objectives of this study were: 1. To continue to refine the methodology of a rapid quantitative selection method for the evaluation of benthic macroinvertebrates in the freshwaters of Rhode Island; 2. To compare the results obtained by this research with those obtained by other freshwater studies; 3. To document baseline populations of the macrofauna in Rhode Island waters. Materials and Methods Forty freshwater sites within the state of Rhode Island watershed were selected in the earlier study (Gould 1991) to measure the effect of varying water quality on the distribution of macrofauna. This survey sampled forty five locations with the addition of the Blackstone River at the Manville Dam, the Kickemuit River at Route 6 in Swansea, MA., Lawton's Brook below the Lawton Valley Reservoir in Portsmouth, the Pawcatuck River in Westerly, and Silver Creek in Bristol. Freshwater streams classified as A, B, or C of the first,second, and third orders were chosen; these included drinking water supply streams, known impacted urban streams, and rural streams. All of the selected sites were within the two ecoregions in Rhode Island (Narragansett/Bristol Lowland and Southern New England Coastal Plains and Hills. Within each of the ecoregions the stations selected included different water quality standard types (see Table 1). Water supply streams in the Scituate, Newport, North Kingstown, Bristol, and Pawtucket systems were included in the sampled streams. Impacted urban and suburban streams included those downstream from treatment plants and in urban environments while rural streams were considered to have minimal point source disturbances based on surrounding land uses and historical information. Each site selected was sampled during the summer of 1993. In all instances the samples were taken within 3 meters of the original designated site. This was to minimize potential on-site differences caused by differences of the microhabitat within the stream. Temperature, pH, and conductivity were measured by means of a Hanna Water Meter at each station during the project. This information enabled the researcher to maintain a general physical/chemical picture of the stream sites. Another study conducted at the University of Rhode Island monitored many of the same stations for their chemical constituents. All biological samples were field preserved in approximately 50% ethyl alcohol and transported to the laboratory for identification. Long-term preservation required at least 70% alcohol; samples once identified became reference samples which were properly preserved in mason jars and stored at Roger Williams University. The summer sample was taken during July and August 1993. Samples were taken at the 45 sites. Protocol HI as outlined in Plafkin et al (1989) was utilized for analysis of samples. At any site, a minimum of 100 organisms was desired and when possible collected and identified at each 3 site as per method. Samples were identified in the field. Collection of organisms occurred at each site by collecting until at least 100 organisms were collected in the net samples (for most stations more than 100 specimens were collected within the net; therefore, these were counted in the methodology). Unidentified specimens were returned to the laboratory for identification. A hand-held drift net (source Wildco Supply) with an opening of 45 x 23 cm provided a reliable and repeatable sampling tool for the protocol. Approximately one square meter of riffle area above the net was thoroughly turned by hand and foot for each replicate for at least three minutes. It was determined (by experimentation and consultation with other researchers conducting the protocols) that three samples of approximately 1.0 m2 in the stream were sufficient to generate suitable specimens that would be indicative of the riffle community in the stream. Results and Discussion According to the Protocol (Chapter 5), a delineation of the streams by a series of metrics was necessary. This was accomplished by observations made at each site visits by obtaining samples of the sediment, rock/grain size, and determining other physical parameters. At each station the following were noted or measured: a. predominant surrounding land use b. local watershed erosion c. local watershed nonpoint source pollution d. estimated stream width in meters e. estimated stream depth in meters f. high water mark in meters g. velocity h. dam presence or absence near the site i. channelization j . canopy cover k. sediment odor 1. sediment oils m. sediment deposits n. inorganic substrate components o. organic substrate components According to the protocol (Chapter 5) delineation of the condition/parameters into primary (substrate and in stream cover), secondary (channel morphology), and tertiary (riparian bank structure) occurs. The above metrics are reported in Table 2. One notes that the values for the sites (based on 130 points as outlined in Section 5-6 of the Protocol) range from a low of 30 for the Woonasquatucket River in Providence to a high of 116 for Ashaway River in Hopkinton. Ten sites considered comparable to the reference site were those that had values of ninety percent or greater