Abstract

The Eucha-Spavinaw basin is the source of water for Lake Eucha and Spavinaw Lake, which are part of the water supply for the City of Tulsa. The City of Tulsa has received complaints of taste and odor in the finished drinking water because of deteriorating water quality. The deterioration is largely because of algal growth from the input of nutrients from the Eucha-Spavinaw basin. The U.S. Geological Survey, in cooperation with the City of Tulsa, implemented a continuous, real-time water-quality monitoring program in the Eucha-Spavinaw basin to better understand the source of the nutrient loading. This program included the manual collection of samples analyzed for nutrients and the collection of continuous, in-stream data from water-quality monitors.

Continuous water-quality monitors were installed at two existing continuous streamflow-gaging stations—Spavinaw Creek near Colcord, Oklahoma, and Beaty Creek near Jay, Oklahoma, from October 2004 through September 2007. Total nitrogen concentrations for manually collected water samples ranged from 2.08 to 9.66 milligrams per liter for the water samples collected from Spavinaw Creek near Colcord, Oklahoma, and from 0.67 to 5.12 milligrams per liter for manually collected water samples from Beaty Creek near Jay, Oklahoma. Total phosphorus concentrations ranged from 0.04 to 1.5 milligrams per liter for the water samples collected from Spavinaw Creek near Colcord and from 0.028 to 1.0 milligram per liter for the water samples collected from Beaty Creek near Jay. Data from water samples and in-stream monitors at Spavinaw and Beaty Creeks (specific conductance and turbidity) were used to develop linear regression equations relating in-stream water properties to total nitrogen and total phosphorus concentrations. The equations developed for the Spavinaw and Beaty sites are site specific and only valid for the concentration ranges of the explanatory variables used in the analysis. The range in estimated and measured phosphorus is not representative for the range of historic streamflow at the Beaty site and that regression equation would benefit from more high flow and high turbidity samples. In addition, all three study years had below average annual precipitation for the area, and streamflow was especially low in Water Year 2006. Average nutrient concentrations from October 2004 through September 2007, which were drier than others, may not be a good indication of conditions in future wetter years.

The equations for the Spavinaw and Beaty sites may be used to estimate instantaneous nutrient concentrations, which can be used to compute loads and yields in real time in order to better characterize the effect of land-management practices in these watersheds on the transport of nutrients to Lake Eucha and Spavinaw Lake. The methods used in this study show promise for monitoring future effectiveness of implemented best management practices, development and monitoring of total maximum daily loads, early detection of taste-and-odor occurrences, and to anticipate treatment needs for water suppliers.

Revised February 13, 2009

First posted January 29, 2009