USGS Home
Contact USGS
Search USGS

Explanation of Methods Used for this Analysis

The USGS provided nutrient flux and yield estimates for six water-quality parameters (dissolved ammonia, dissolved nitrite plus nitrate, total ammonia plus organic nitrogen, dissolved orthophosphate, total phosphorus, and dissolved silica) for the downstream end of the Mississippi-Atchafalaya River Basin from 1968 through June 2002 in July of 2002. These loads were posted on a USGS website and were updated annually through the end of June every July through 2005, so that nutrient flux estimates for the previous 12 months (July through June) were provided in a timely manner and coordinated with surveys of the Gulf of Mexico hypoxic zone, which occur in mid-July.

Nutrient fluxes were estimated for the Atchafalaya River using water-quality data from the Atchafalaya River at Melville, LA (USGS station id 07381495) and flow data from the Atchafalaya River at Simmesport, Louisiana (USACE station id 03045). Nutrient fluxes were estimated for the Mississippi River using water-quality data from the Mississippi River near St. Francisville, Louisiana (USGS station id 07373420) and flow data from the Mississippi River at Tarbert Landing (USACE station id 01100). The data set used to construct flux estimates consists of instantaneous sample data only. Results from composite samples, collected prior to October 1967 are not used. Half the detection limit was substituted for concentrations indicated to be less than the detection limit.

Nutrient fluxes were estimated using ESTIMATOR (Gilroy et al., 1990; Cohn et al., 1992), a regression-model approach (also known as a rating-curve method). The regression models had terms for the log of flow and flow-squared, seasonal sine and cosine functions, and a linear time trend. The Mississippi River flux models were modified such that the natural flow of the Mississippi River was used, which is the sum of the Mississippi River at Tarbert Landing plus the flow diverted upstream by the Old River Outflow Channel near Knox Landing, Louisiana (total outflow; USACE station id 02600). Note that the fluxes are later scaled such that it represents only the flow going down the lower Mississippi River, not the flow diverted by the Old River to the Atchafalaya River. The Mississippi River flux models were further modified to include flow terms from two upstream stations, the Mississippi River at Thebes, Illinois (USGS station id 07022000) and the Ohio River at Metropolis, Illinois (USGS station id 03611500). Upstream flows were lagged 10 days to account for travel time between the upstream and downstream sites and both log of flow and flow-squared terms were included.

The first 10 years of flux estimates are calculated using the first 10 years (July through June) of data. Each subsequent year of load estimates are calculated using samples from the current year and the previous 9 years. This "moving window" approach allows a sufficient number of samples in each model run to represent the full range of flow and nutrient concentration conditions.

The resulting nutrient flux and yield estimates differ slightly from those calculated for the Task Force Science Assessment (Goolsby et al., 1999). This report used the same method (ESTIMATOR) to estimate the fluxes, but the model structure and calibration periods were changed. The models used here do not include the time-squared trend term used in the CENR report. Different upstream flow terms are used for the Low Mississippi River model than were used in the CENR report. The CENR models did not use a 10-year moving window approach for the calibration period.

All flux estimates are in metric tons as N, P, or SiO2 and all yield estimates are in kilograms per square kilometer per month or year. Discharge estimates are given in cubic feet per second.

Cohn, T.A., Caulder, D.L., Gilroy, E.J., Zynjuk, L.D., and Sommers, R.M., 1992. The validity of a simple statistical model for estimating fluvial constituent loads: An empirical study involving nutrient loads entering Chesapeake Bay. Water Resources Research 28:2352-2363.

Gilroy, E.J., Hirsh, R.M., and Cohn, T.A., 1990. Mean square error of regression-based constituent transport estimates. Water Resources Research 26: 2069-2077.

Goolsby, D.A., W.A. Battaglin, G.B. Lawrence, R.S. Artz, B.T. Aulenbach, R.P. Hooper, D.R. Keeney, and G.S. Stensland. 1999. Flux and sources of nutrients in the Mississippi‑Atchafalaya River Basin. Topic 3 Report for the Integrated Assessment on Hypoxia in the Gulf of Mexico. NOAA Coastal Ocean Program Decision Analysis Series No. 17. Silver Spring Maryland: National Oceanic and Atmospheric Administration.