FIELD DATA COLLECTION
Macroinvertebrate samples are collected each fall using methods employed by the DWM for assessing the condition of macroinvertebrate communities in Massachusetts streams (Nuzzo 2003). These methods are based on the US EPA Rapid Bioassessment Protocols (RBPs) for wadeable streams and rivers (Barbour et al. 1999). Sampling activities are conducted in accordance with the Quality Assurance Project Plan (QAPP) for the DRWA benthic macroinvertebrate monitoring program (Cole and Walk 2011). Macroinvertebrates are collected from each site using kick-sampling, a method by which organisms are sampled by disturbing streambed substrates and catching dislodged organisms in a net. At each sample site, ten kick samples of approximately 0.46 m x 0.46 m are collected and composited for a total sampled area of approximately 2 m2. Sampling targets fast-water areas with coarse substrate within each of the sample sites. Samples are labeled and preserved in the field with denatured 95% ethanol for later processing and identification in a laboratory.
SAMPLE SORTING AND MACROINVERTEBRATE IDENTIFICATION
Samples are sorted to remove a 100-organism subsample from the original sample using procedures described in Nuzzo (2003). Samples are first distributed in gridded pans. Macroinvertebrates are sorted from randomly selected grids until 100 organisms (±10%) are removed. The remainder of the unsorted grids are then scanned for large/rare organisms that were not encountered during the 100-organism subsampling. These organisms are then removed and placed in a separate “large/rare” organism vial. Specimens are identified to the lowest practical taxonomic level (generally genus or species) as allowed by specimen condition and maturity. We use an experienced taxonomist to perform the identification work.
Macroinvertebrate taxonomic data are analyzed using DWM’s modification (Nuzzo 2003) of EPA’s Rapid Bioassessment Protocol III multimetric scoring and analysis (Barbour et al. 1999) to determine the condition of macroinvertebrate communities. Multimetric analysis employs a set of metrics, each of which describes an attribute of the macroinvertebrate community that is known to be responsive to one or more types of pollution or habitat degradation. Because a number of biological attributes are simultaneously evaluated, the multimetric approach is a robust assessment tool and a deficiency in any one metric should not invalidate assessment results (Barbour et al. 1999). Each attribute value is first calculated from the taxonomic data and then converted to a standardized score by comparison with the reference site score (Table 2). Standardized scores of all metrics are then summed to produce a single multimetric score that is a numeric measure of overall biological integrity. DWM currently employs a 7-metric set for use with fast-water samples from streams.
A Quality Assurance Project Plan (QAPP) was developed and written for this project (Cole and Walk 2011). The QAPP included all required state and federal elements and was approved by MA DEP and the US Environmental Protection Agency prior to the beginning of this assessment. Elements of the QAPP included the project background, site selection rationale, measurement quality objectives, training, documentation, sampling design, protocols, quality control requirements, instrument/equipment testing and maintenance, data management, data review, and data validation. Although the details of the QAPP are too lengthy to present in the context of this methods overview, several of the critical elements of the QAPP are as follows.
Volunteers collecting field samples and data are trained on the day they assisted in the field and worked closely at all times in the field with Dr. Michael Cole. Duplicate samples are collected at 2 or 3 sample sites in each year to ascertain repeatability and reliability of field and laboratory methods. All macroinvertebrate identifications are performed by Michael Cole, a professional aquatic entomologist. Representative specimens of each taxon encountered are labeled and saved as vouchers for later reference and verification, as needed. Sorted macroinvertebrate samples are preserved in 95% ethanol and archived. Unsorted fractions of all samples are also preserved and are archived for two years following project completion. All data entered into spreadsheets are checked for transcription errors and outliers before analyses were performed. Analyses are also checked for errors in formulae used and results. In 2011, the DRWA developed a single relational database to archive all of our macroinvertebrate data.