Use of dissolved oxygen monitoring to evaluate phosphorus loading in Connecticut streams, 2015–18

The Connecticut Department of Energy and Environmental Protection (CT DEEP) has developed an interim phosphorus reduction strategy to establish water-quality-based phosphorus limits in nontidal freshwaters for industrial and municipal water pollution control facilities. A recommendation in the strategy included the addition of diurnal dissolved oxygen (DO) sampling to the sampling of diatom communities collected by CT DEEP. The chemistry data coupled with biological data will help to examine the effects of phosphorus loading in streams. The U.S. Geological Survey (USGS), in cooperation with the CT DEEP and New England Interstate Water Pollution Control Commission, implemented a summer DO monitoring program from 2015 to 2018 to examine the effects of phosphorus loading in streams. Continuous DO data were collected at 18 sites in streams with varying concentrations of phosphorus throughout the State of Connecticut. Discrete water-quality nutrient data were collected by the USGS at 11 of the 18 sites. All continuous and discrete data collected from June to September for the 4 years were examined for all sites. This report documents a pattern of diurnal DO for monitoring sites across 4 years and presents estimated daily gross primary productivity (GPP), ecosystem respiration (ER), and a standardized rate coefficient for gas exchange for selected streams. Relations of phosphorus concentrations to the diurnal DO response and stream metabolism are described. Interannual variability in average annual total phosphorus (TP) concentrations and maximum daily DO concentrations were evaluated among sites in years of the study. Streams identified as impaired by CT DEEP such as Naugatuck River at Beacon Falls (USGS station 01208500), Still River at Route 7 at Brookfield Center (USGS station 01201487), and Quinnipiac River at Wallingford (USGS station 01196500) had higher TP concentrations (greater than 0.10 milligram per liter [mg/L]) throughout the study. Reference streams considered unimpaired had lower concentrations of TP (less than 0.10 mg/L). The range in daily DO concentrations remained less than 4 mg/L for most of the sites during the study except for Naugatuck River at Beacon Falls and Still River at Route 7 at Brookfield Center. Daily GPP and ER were summarized for 11 sites using the maximum likelihood estimation model of the streamMetabolizer package in the R statistical program. The models indicated that most sites had an estimated negative net primary productivity, based on the daily estimates of GPP and ER, which indicates the systems are heterotrophic and dominated by respiration. The high variation of GPP and ER reported for several sites can be affected by many physical, chemical, and biological factors, including the abundance and community composition of phytoplankton, periphyton, and macrophyte algae present. The variability in mean GPP was similar to the variability in maximum DO concentrations when plotted against annual average TP concentrations for the maximum likelihood estimation model in streamMetabolizer. The concept that phosphorus loading can affect the stream metabolism requires more detailed knowledge of stream geomorphic variables (canopy cover, stream velocity, water depth) and algal communities to help improve the scientific basis for managing phosphorus loading.