Evaluation of Measurements Collected with Multi-Parameter Continuous Water-Quality Monitors in Selected Illinois Streams, 2001-03

Eight streams, representing a wide range of environmental and water-quality conditions across Illinois, were monitored from July 2001 to October 2003 for five water-quality parameters as part of a pilot study by the U.S. Geological Survey (USGS) in cooperation with the Illinois Environmental Protection Agency (IEPA). Continuous recording multi-parameter water-quality monitors were installed to collect data on water temperature, dissolved-oxygen concentrations, specific conductivity, pH, and turbidity. The monitors were near USGS streamflow-gaging stations where stage and streamflow are continuously recorded. During the study period, the data collected for these five parameters generally met the data-quality objectives established by the USGS and IEPA at all eight stations. A similar pilot study during this period for measurement of chlorophyll concentrations failed to achieve the data-quality objectives. Of all the sensors used, the temperature sensors provided the most accurate and reliable measurements (generally within ?5 percent of a calibrated thermometer reading). Signal adjustments and calibration of all other sensors are dependent upon an accurate and precise temperature measurement. The dissolved-oxygen sensors were the next most reliable during the study and were responsive to changing conditions and accurate at all eight stations. Specific conductivity was the third most accurate and reliable measurement collected from the multi-parameter monitors. Specific conductivity at the eight stations varied widely-from less than 40 microsiemens (?S) at Rayse Creek near Waltonville to greater than 3,500 ?S at Salt Creek at Western Springs. In individual streams, specific conductivity often changed quickly (greater than 25 percent in less than 3 hours) and the sensors generally provided good to excellent record of these variations at all stations. The widest range of specific-conductivity measurements was in Salt Creek at Western Springs in the Greater Chicago metropolitan area. Unlike temperature, dissolved oxygen, and specific conductivity that have been typically measured over a wide range of historical streamflow conditions in many streams, there are few historical turbidity data and the full range of turbidity values is not well known for many streams. Because proposed regional criteria for turbidity in regional streams are based on upper 25th percentiles of concentration in reference streams, accurate determination of the distribution of turbidity in monitored streams is important.

Digital data from all five sensors were recorded within each of the eight sondes deployed in the streams and in automated data recorders in the nearby streamflow-gaging houses at each station. The data recorded on each sonde were retrieved to a field laptop computer at each station visit. The feasibility of transmitting these data in near-real time to a central processing point for dissemination on the World-Wide Web was tested successfully.

Data collected at all eight stations indicate that a number of factors affect the dissolved-oxygen concentration in the streams and rivers monitored. These factors include: temperature, biological activity, nutrient runoff, and weather (storm runoff). During brief periods usually in late summer, dissolved-oxygen concentrations in half or more of the eight streams and rivers monitored were below the 5 milligrams per liter minimum established by the Illinois Pollution Control Board to protect aquatic life. Because the streams monitored represent a wide range in water-quality and environmental conditions, including diffuse (non-point) runoff and wastewater-effluent contributions, this result indicates that deleterious low dissolved-oxygen concentrations during late summer may be widespread in Illinois streams.