Concentrations, loads, and yields of select constituents from major tributaries of the Mississippi and Missouri Rivers in Iowa, water years 2004-2008

Abstract

Excess nutrients, suspended-sediment loads, and the presence of pesticides in Iowa rivers can have deleterious effects on water quality in State streams, downstream major rivers, and the Gulf of Mexico. Fertilizer and pesticides are used to support crop growth on Iowa's highly productive agricultural landscape and for household and commercial lawns and gardens. Water quality was characterized near the mouths of 10 major Iowa tributaries to the Mississippi and Missouri Rivers from March 2004 through September 2008. Stream loads were calculated for select ions, nutrients, and sediment using approximately monthly samples, and samples from storm and snowmelt events. Water-quality samples collected using standard streamflow-integrated protocols were analyzed for major ions, nutrients, carbon, pesticides, and suspended sediment. Statistical data summaries of sample data used parametric and nonparametric techniques to address potential bias related to censored data and multiple levels of censoring of data below analytical detection limits. Constituent stream loads were computed using standard pre-defined models in S-LOADEST that include streamflow and time terms plus additional terms for streamflow variability and streamflow anomalies. Streamflow variability terms describe the difference in streamflow from recent average conditions, whereas streamflow anomaly terms account for deviations from average conditions from long-to short-term sequentially. Streamflow variability or anomaly terms were included in 44 of 80 site/constituent individual models, demonstrating the usefulness of these terms in increasing accuracy of the load estimates. Constituent concentrations in Iowa streams exhibit streamflow, seasonal, and spatial patterns related to the landform and climate gradients across the studied basins. The streamflow-concentration relation indicated dilution for ions such as chloride and sulfate. Other constituent concentrations, such as dissolved organic carbon and suspended sediment, increased with streamflow. Nitrogen concentrations (total nitrogen and nitrate plus nitrite) increased with low and moderate streamflows, but decreased with high streamflows.

Temporal changes in nitrogen and phosphorus concentrations with comparisons to conservation practices and agricultural activities in the Lower Grand River, Missouri and Iowa, and selected watersheds, 1969–2015

“…3). Model bias was computed for WRTDS TN and TP concentration models by comparing the sum of estimated concentrations on days when samples were collected to the sum of measured sample concentrations for each site, similar to methods described in Stenback and others (2011), and used in Garrett (2012), Sprague and others (2011), and Sprague and Gronberg (2013). The absolute bias in the nitrogen and phosphorus WRTDS models was not greater than 3.3 percent at any site (table 2).…”

Occurrence and transport of selected constituents in streams near the Stibnite mining area, Central Idaho, 2012–14

“…Ninety-five percent confidence intervals around estimated mean annual loads and FWCs quantify uncertainty around estimates and generally were larger for constituents with greater variability in measured loads (tables 4 and 5). Recent research has shown that LOADEST models can produce biased estimates of long-term average constituent loads (Stenback and others, 2011;Garrett, 2012;Hirsch, 2014), and LOADEST has since been modified to facilitate residual analysis and bias identification (Runkel, 2013). Long-term bias in model estimates was less of a concern because the modeling period was only 3 years and predictor variables describing trends in time were not used.…”

“…This type of model has been used to estimate constituent concentrations for periods when sample data were not available (Gilroy and others, 1990), and to estimate a basin flux of water-quality constituents (Goolsby and others, 1999). Additional predictor variables describing streamflow variability were tested for significance in LOADEST models according to methods described by Garrett (2012).…”

Sediment transport and evaluation of sediment surrogate ratings in the Kootenai River near Bonners Ferry, Idaho, Water Years 2011–14

“…The explanatory variable dQ k was evaluated to represent streamflow variability or hysteresis and is described in Garrett (2012). Hysteresis occurs when the value of a physical property (constituent concentration or load) changes at a different rate than the effect assumed to be causing it (streamflow).…”

“…Use of a streamflow hysteresis term might eliminate the need to develop a different regression between streamflow and load depending on whether samples were collected on the rising or the falling limb of the hydrograph. The streamflow hysteresis term was defined as the difference between mean streamflow (Q) on day i and the mean streamflow of the previous k days, given as equation 8 but described in more detail in Garrett (2012):…”