Link to USGS home page.
PUBLICATIONS—Scientific Investigations Report 2004-5235

Scientific Investigations Report 2004-5235

Version 2.00, December 12, 2005

Phosphorus and Suspended Sediment Load Estimates for the Lower Boise River, Idaho, 1994-2002

Prepared in cooperation with the Idaho Department of Environmental Quality

By Mary M. Donato and Dorene E. MacCoy


The U.S. Geological Survey used LOADEST, newly developed load estimation software, to develop regression equations and estimate loads of total phosphorus (TP), dissolved orthophosphorus (OP), and suspended sediment (SS) from January 1994 through September 2002 at four sites on the lower Boise River: Boise River below Diversion Dam near Boise, Boise River at Glenwood Bridge at Boise, Boise River near Middleton, and Boise River near Parma. The objective was to help the Idaho Department of Environmental Quality develop and implement total maximum daily loads (TMDLs) by providing spatial and temporal resolution for phosphorus and sediment loads and enabling load estimates made by mass balance calculations to be refined and validated.

Regression models for TP and OP generally were well fit on the basis of regression coefficients of determination (R2), but results varied in quality from site to site. The TP and OP results for Glenwood probably were affected by the upstream wastewater-treatment plant outlet, which provides a variable phosphorus input that is unrelated to river discharge. Regression models for SS generally were statistically well fit. Regression models for Middleton for all constituents, although statistically acceptable, were of limited usefulness because sparse and intermittent discharge data at that site caused many gaps in the resulting estimates.

Although the models successfully simulated measured loads under predominant flow conditions, errors in TP and SS estimates at Middleton and in TP estimates at Parma were larger during high- and low-flow conditions. This shortcoming might be improved if additional concentration data for a wider range of flow conditions were available for calibrating the model.

The average estimated daily TP load ranged from less than 250 pounds per day (lb/d) at Diversion to nearly 2,200 lb/d at Parma. Estimated TP loads at all four sites displayed cyclical variations coinciding with seasonal fluctuations in discharge. Estimated annual loads of TP ranged from less than 8 tons at Diversion to 570 tons at Parma. Annual loads of dissolved OP peaked in 1997 at all sites and were consistently higher at Parma than at the other sites.

The ratio of OP to TP varied considerably throughout the year at all sites. Peaks in the OP:TP ratio occurred primarily when flows were at their lowest annual stages; estimated seasonal OP:TP ratios were highest in autumn at all sites. Conversely, when flows were high, the ratio was low, reflecting increased TP associated with particulate matter during high flows. Parma exhibited the highest OP:TP ratio during all seasons, at least 0.60 in spring and nearly 0.90 in autumn. Similar OP:TP ratios were estimated at Glenwood. Whereas the OP:TP ratio for Parma and Glenwood peaked in November or December, decreased from January through May, and increased again after June, estimates for Diversion showed nearly the opposite pattern — ratios were highest in July and lowest in January and February. This difference might reflect complex biological and geochemical processes involving nutrient cycling in Lucky Peak Lake, but further data are needed to substantiate this hypothesis.

Estimated monthly average SS loads were highest at Diversion, about 400 tons per day (ton/d). Average annual loads from 1994 through 2002 were 144,000 tons at Diversion, 33,000 tons at Glenwood, and 88,000 tons at Parma. Estimated SS loads peaked in the spring at all sites, coinciding with high flows.

Increases in TP in the reach from Diversion to Glenwood ranged from 200 to 350 lb/d. Decreases in TP were small in this reach only during high flows in January and February 1997. Decreases in SS, were large during high-flow conditions indicating sediment deposition in the reach. Intermittent data at Middleton indicated that increases and decreases in TP in the reach from Glenwood to Middleton were during low- and high-flow conditions, respectively. All constituents increased in the reach from Middleton to Parma, particularly in May 1995. Periods of large increases in SS were interpreted as times of active erosion of sediment in the channel and (or) addition of sediment to the river from tributaries.

Statistically significant downward temporal trends in load were determined for SS at Glenwood, for OP at Middleton, and for TP, OP, and SS at Parma. A significant upward trend in TP was determined at Diversion.

Calculated annual flow-weighted concentrations highlighted the strong interaction between flow and particle-associated constituents such as TP and SS and enabled concentrations to be assessed despite the large range in flows. At Parma and Glenwood, where the OP:TP ratio was high, the flow-weighted concentration of TP was inversely related to load because of dilution and the strong effect of flow on the load calculation. At Diversion, where the OP:TP ratio was low, load and concentration were directly related because particulate matter was increased, not diluted, during high flows. Relations were similar for SS.

LOADEST average daily TP load estimates indicated that reductions in load of 24 to 75 percent would have been necessary to meet the proposed goal of 565 lb/d set forth in the Snake River-Hells Canyon TMDL. Estimated average daily loads of SS at Parma from 1994 through 2002 exceeded the current lower Boise River TMDL load allocation of 101 ton/d at Parma except in 2001.

The LOADEST model provided spatial and temporal resolution to help refine mass-balance spreadsheet calculations used for TP TMDL allocations between Diversion and Parma. LOADEST results compared favorably with previous mass-balance spreadsheet load estimates for TP and have an advantage over spreadsheet estimates in that they allow more detailed examination of loads at different timescales, whereas spreadsheet estimates are static.


Method for Estimating Phosphorus and Suspended Sediment Loads
Input Data
Loadest Modeling Results
Monthly Average Increases and Decreases of Contituent Loads within Reaches
Temporal Trends
Comparison with Mass Balance Spreadsheet Load Estimates
Annual Flow-Weighted Concentrations
Implications for Total Maximum Daily Load Implementation
Limitations and Advantages of the Loadest Model
References Cited

This report is available online in Portable Document Format (PDF). If you do not have the Adobe Acrobat PDF Reader, it is available for free download from Adobe Systems Incorporated.

Download the report (PDF, 1.09 MB)

Revision history

Document Accessibility: Adobe Systems Incorporated has information about PDFs and the visually impaired. This information provides tools to help make PDF files accessible. These tools convert Adobe PDF documents into HTML or ASCII text, which then can be read by a number of common screen-reading programs that synthesize text as audible speech. In addition, an accessible version of Acrobat Reader 5.0 for Windows (English only), which contains support for screen readers, is available. These tools and the accessible reader may be obtained free from Adobe at Adobe Access.

Send questions or comments about this report to the authors, M.M. Donato , (208) 387-1350, and D.E. MacCoy, (208) 387-1354.

For more information about USGS activities in Idaho, visit the USGS Idaho Water Science Center home page.

FirstGov button  Take Pride in America button