A Comparison of Two Methods for Benthic Macroinvertebrate Sampling and Analysis:

USEPAís Rapid Bioassessment Protocol II vs. West Virginia Save Our Streams

Cacapon Institute, High View, WV

Project Overview

In recent years, the science of using animals to assess the vitality of a river ecosystem has gone public. Volunteer monitoring programs, such as the Isaak Walton League's pioneering Save Our Streams (SOS) program, have sprouted up around the country. The SOS and other volunteer methods are similar in general design to the methods used by professionals, but tailored to the capabilities of non-professionals. Such programs make the link between causes and effects of pollution using hands-on, in-stream activities, and make the societal benefits of clean water immediate and real.

Cacapon Institute (CI) received funding from the WVDNR Non-Game Program to compare results from WVís volunteer SOS monitoring and the more scientifically rigorous Rapid Bioassessment Protocol (RBPII) stream assessment methods used by WVís Division of Environmental Protection (WVDEP). Both methods assess stream health using benthic macroinvertebrates, the small animals without backbones (invertebrates) that live on the river bottom (benthos) and are visible without magnification (macro). Our study had two primary goals:

  • The first was to confirm a study done by Virginia Tech, which found that the SOS "score" tends to overestimate the health of a stream when compared to a professional assessment. Their data indicated that the difference was not a result of the physical method of collection (i.e. net size, etc.), but how the data was analyzed.
  • The second goal was to "tweak" the information gained by SOS field methods to obtain a result similar to WVís professional RBPII method, without making the procedure too difficult for the SOS volunteer. For this reason, we tried to modify SOS protocols not in physical collection methods, but in tallying methods and calculations.

A central assumption of this project was that the biological assesment protocols utilized in the RBPII protocol represent a high professional standard and therefore provide a suitable tool to measure the success of volunteer methods.


WVDEP and CI collected benthic macroinvertebrates at 20 sampling sites in WVís Cacapon River watershed during the summer of 2000. WVDEP used standard RBP II collection and identification procedures, while CI used WVís SOS method. The main differences between the two techniques are in the details:

  • The RBP II method uses a smaller mesh net size than SOS to collect samples at eight sites rather than three;
  • RBPII preserves samples in the field for later "picking" of a subsample under slight magnification in the laboratory, while the SOS method involves fieldpicking live organisms;
  • The RBPII counts organisms and identifies them to the family level using trained taxonomists while the SOS method does not use counts and calls for identification in the field using a visual key that groups organisms in general categories (mayfly, stonefly, aquatic worm, etc.).

The professional RBP data analysis uses the number of different kinds of organisms to calculate a set of six standard Indices, which can be used to indicate various attributes of the stream site. These indices are then combined into the WV Stream Condition Index (WVSCI) Score, which is used to rate the site. In the SOS method, the streamís biological health is described using a single numeric indicator, the SOS score; abundance of organisms does not factor into this rating.

CI modified the standard SOS protocol slightly. Each of the three samples per site was collected using a timed, two-minute effort, in order to obtain a catch per unit effort. After handpicking live in the field, organisms were preserved in alcohol and brought to the lab for identification to the standard SOS level. We also counted these organisms, and noted visually distinct "kinds" within each SOS taxonomic grouping. This allowed us to calculate additional indices similar to those used in the professional RBPII method.

Some overall conclusions

SOS scores as currently calculated don't provide stream assessment data comparable to professional RBPII results because they lack abundance data and thereby lack critical information. As the Virginia Tech study found, SOS scores tend to overrate sites the RBPII stream score found to be impaired. However, the SOS method provides a conservative stream assessment tool Ė although it may miss some impaired streams, if the SOS score says itís bad, it really is.

The identification level used with the SOS method can provide a stream assessment comparable to professional methods, based on our analysis. The main drawback, is that SOS simply categorizes the organisms using A = 1-9, B = 10-99 and C = 100 or more, and these "count" categories are then not used in the calculation of the stream score. Our study indicates that actual counts of the organisms collected are essential in order to properly weight the importance of each organism. These counts allow volunteers to calculate variants of the biological indices used in rapid bioassessment protocols (RBP II), and also calculate an overall stream index similar to DEP's West Virginia stream condition index (WVSCI). The indices can be easily calculated using a computer spreadsheet, developed by Cacapon Institute, where volunteers enter organism counts and the number of different "kinds" and the computer does the rest.
Our study also found that the SOS method of fieldpicking live organisms tends to disproportionately miss some small organisms in specific groups, like midges and blackflies, when compared to picking preserved samples in the lab. This difference appears to be of paramount importance. We believe it is unlikely that field picked samples analyzed under any method (SOS or professional RBP) will produce results consistent with laboratory picked samples because certain critical groups are likely to be seriously underrepresented in field picked samples.

Suggested modifications to WVSOS methodology

to improve validity of volunteer collected data

1. Count the organisms collected by SOS level identification categories.

2. Samples should be preserved in alcohol in the field and picked in the lab (or on the kitchen counter) under slight magnification with good lighting. This will eliminate disproportionate "under-picking" of important groups such as midges and blackflies and will reduce the overestimation inherant in SOS methods. While this is a separate step for volunteers, it is actually easier (and more comfortable) than the current field picking - and more rewarding because the results will have greater validity and, hopefully, achieve greater acceptance by government agencies.

3. Note visually different "kinds" of organisms in SOS level identification categories, to obtain a measure of species richness, or diversity.

4. Supply an automated spreadsheet for volunteers to enter raw data and generate a composite metric (based on a number of metrics) similar to the WV Stream Condition Index, which relates well to the professional stream quality rating score.


Based on our study, it is possible for non-professional, volunteer conducted, benthic stream assessments to obtain results that compare favorably to professional assessments. The proposed method utilizes the same level of identification skill currently required of SOS volunteers and the same collection technique. It differs by requiring samples to be preserved in the field for "picking" under slight magnification and good lighting at home, in counting the organisms obtained, and in noting different "kinds" within each of the current SOS identification categories.

A detailed report on this project in PDF format is available here.  

Many thanks to the good folks at WV DEP's Watershed Assessment Program.  This project would not have been possible without their cooperation. 

For more information, contact Neil Gillies at

Cacapon Institute, Rt. 1 Box 326, High View, WV 26808, pcrel@mountain.net, (304) 856-1385