The Utility of Probability-based Sampling in Pennsylvania's Water Quality Program: Monitoring, Assessment & Methods Developement
Pennsylvania Department of Environmental Protection, Water Standards & Facility Regulation, Harrisburg, PA
Pennsylvania’s first experience with random sampling design on a large scale was in 1993 and 1994 as a participating member of EPA Region III States’ Mid-Atlantic Highlands Assessment (MAHA) - an EMAP pilot project to assess water quality conditions on a large, regional scale. With the exception of limited participation in the National Wadeable Stream Assessment, Pennsylvania has had no involvement with probabilistic sampling design since the MAHA project. However, while Pennsylvania’s water quality monitoring programs have traditionally employed targeted sampling designs, new and evolving water quality program issues call for probability-based sampling implementation . For wadeable streams assessment and monitoring purposes, Pennsylvania has transitioned from the targeted assessment approach of its first statewide assessment cycle to a more rigorous, probability-based sampling design to provide wadeable stream information on a regionalized scale for Cycle 2 assessments. With assistance from EPA’s Corvallis NHEER Laboratory, the first probability-based application was designed to review non-impairment decisions made during Cycle 1 assessments. Subsequent probability-based Cycle 2 sampling will assess Pennsylvania’s remaining basins on a rotating basis. Concurrent with incorporating probabilistic sampling in wadeable stream assessments, Pennsylvania is applying probabilistic sampling in other assessment projects that include a methods comparison project, Wetland Assessment pilot project, benthic macroinvertebrate and fish IBI development projects, and the National Lakes Survey. Participation in the future National Large Rivers Survey is under consideration.