Skip Links

USGS - science for a changing world

Scientific-Investigations Report 2009–5199

National Water-Quality Assessment Program

Development and Application of Regression Models for Estimating Nutrient Concentrations in Streams of the Conterminous United States, 1992–2001

By Norman E. Spahr, David K. Mueller, David M. Wolock, Kerie J. Hitt, and JoAnn M. Gronberg


Thumbnail of and link to report PDF (10.1 MB)

Data collected for the U.S. Geological Survey National Water-Quality Assessment program from 1992–2001 were used to investigate the relations between nutrient concentrations and nutrient sources, hydrology, and basin characteristics. Regression models were developed to estimate annual flow-weighted concentrations of total nitrogen and total phosphorus using explanatory variables derived from currently available national ancillary data. Different total-nitrogen regression models were used for agricultural (25 percent or more of basin area classified as agricultural land use) and nonagricultural basins. Atmospheric, fertilizer, and manure inputs of nitrogen, percent sand in soil, subsurface drainage, overland flow, mean annual precipitation, and percent undeveloped area were significant variables in the agricultural basin total nitrogen model. Significant explanatory variables in the nonagricultural total nitrogen model were total nonpoint-source nitrogen input (sum of nitrogen from manure, fertilizer, and atmospheric deposition), population density, mean annual runoff, and percent base flow.

The concentrations of nutrients derived from regression (CONDOR) models were applied to drainage basins associated with the U.S. Environmental Protection Agency (USEPA) River Reach File (RF1) to predict flow-weighted mean annual total nitrogen concentrations for the conterminous United States. The majority of stream miles in the Nation have predicted concentrations less than 5 milligrams per liter. Concentrations greater than 5 milligrams per liter were predicted for a broad area extending from Ohio to eastern Nebraska, areas spatially associated with greater application of fertilizer and manure. Probabilities that mean annual total-nitrogen concentrations exceed the USEPA regional nutrient criteria were determined by incorporating model prediction uncertainty. In all nutrient regions where criteria have been established, there is at least a 50 percent probability of exceeding the criteria in more than half of the stream miles.

Dividing calibration sites into agricultural and nonagricultural groups did not improve the explanatory capability for total phosphorus models. The group of explanatory variables that yielded the lowest model error for mean annual total phosphorus concentrations includes phosphorus input from manure, population density, amounts of range land and forest land, percent sand in soil, and percent base flow. However, the large unexplained variability and associated model error precluded the use of the total phosphorus model for nationwide extrapolations.

First posted February 10, 2010

For additional information contact:

Director, USGS Colorado Water Science Center
Box 25046, Mail Stop 415
Denver, CO 80225

Part or all of this report is presented in Portable Document Format (PDF); the latest version of Adobe Reader or similar software is required to view it. Download the latest version of Adobe Reader, free of charge.

Suggested citation:

Spahr, N.E., Mueller, D.K., and Wolock, D.W., 2010, Development and application of regression models for estimating nutrient concentrations in streams of the conterminous United States, 1992–2001: U.S. Geological Survey Scientific Investigations Report 2009–5199, 22 p.





Correlation between Nutrient Concentrations and Basin Characteristics

Development of Models for Total Nitrogen

Nationwide Prediction of Total-Nitrogen Concentrations

Development of Models for Total Phosphorus

Comparison to Results from other Models

Summary and Conclusions


References Cited

Accessibility FOIA Privacy Policies and Notices logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Contact USGS
Page Last Modified: Monday, March 07, 2016, 01:43:43 PM