The effectiveness of using volunteers for biological monitoring of streams

An increase in public environmental awareness and a decrease in resources within government regulatory agencies have led to a larger interest in volunteer biomonitoring programs. Government agencies are currently using volunteer data for official purposes with increasing frequency, but questions have been raised about the validity of the data collected by volunteers who have only limited training and experience. Therefore, we conducted a detailed study to assess, modify, and validate the Virginia Save-Our-Streams (SOS) program, which is a volunteer organization collecting macroinvertebrate data. Sites were sampled using professional methods concurrently with volunteers who utilized the SOS protocol. The volunteer samples were retained for further laboratory analysis. In addition, numerous sites previously sampled by volunteers were re-sampled using professional methods. The data were statistically analyzed to determine if the results of volunteers and professional aquatic biologists were correlated and if they arrived at the same conclusions about ecological condition. It was determined that the Virginia SOS method, and probably other similar volunteer methods, consistently overrate ecological condition. This means that streams impaired by pollution could go unreported, if they are monitored exclusively by volunteers. The cause of this overestimation was determined to be the overly simplistic SOS metric, which is based solely on the presence or absence of taxa. The SOS protocol for data analysis was made more quantitative by developing a multimetric index that is appropriate for use by volunteers. The SOS sampling protocol was modified slightly to obtain actual counts of the different kinds of macroinvertebrates, which allowed for calculation of metrics. Sorting effort and taxonomic level of identification were not changed so that currently participating volunteers would not be excluded because of the need for expensive equipment or advanced technical training. The modified SOS protocol was evaluated by a different set of concurrent samples taken by volunteers and professionals, but using the same statistical techniques. The modified SOS protocol proved to be feasible for volunteers. The new SOS multimetric index correlated well with a professional multimetric index. The conclusions about ecological condition derived from the volunteer multimetric index agreed very closely with those made by professional aquatic biologists. This study demonstrated that volunteer biomonitoring programs can provide reliable data, but every volunteer program needs to be thoroughly validated by statistical comparisons to the professional methods being used in that area.