The objective of this study was to conduct a multibasin reconnaissance survey to determine the relative importance of chemical properties, land-use, and hydrology to agricultural chemical contamination of streams in northern Missouri. In 1994 and 1995, samples were collected from 140 sites on 95 different streams and rivers throughout northern Missouri. Samples were collected under preplant and postplant conditions and analyzed for common herbicides and dissolved nutrients. Atrazine, the most frequently detected herbicide was detected in all postplant samples and 99% and 90% of the preplant samples in 1994 and 1995, respectively. The study area has significant variations in soils, hydrology, and land-use (row-cropping intensity). The hydrology is largely determined by the soils, as reflected by soil hydrologic groups. Nitrate and herbicide concentrations showed opposite trends across the study region. Streams draining watersheds with runoff-prone soils had the highest herbicide concentrations, while streams draining watersheds with more groundwater recharge had low herbicide concentrations but the highest NO3−N concentrations. Current data are sufficient to develop a conceptual framework for assessing watershed vulnerability based on three key factors. The primary factor is the chemistry of the compound, which determines the potential hydrologic transport pathways for that chemical to be lost from the soil. Nitrate can potentially be leached or lost in runoff. Moderately sorbed compounds, such as atrazine, are more likely to be lost in runoff or degraded within the soil than leached. The hydrology of a region is the secondary factor, as it determines the relative importance of the leaching and runoff transport pathways. The third factor then is the land-use, which includes the percentage of a watershed that is cropped, the locations within the watershed that are cropped, and the chemicals applied. Management practices to improve water quality must be designed in accordance with the dominant problems and transport pathways of a watershed.

Herbicide contamination of streams has been well documented, but little is currently known about the specific factors affecting watershed vulnerability to herbicide transport. The primary objectives of this study were (1) to document herbicide occurrence and transport from watersheds in the northern Missouri/ southern Iowa region; (2) to quantify watershed vulnerability to herbicide transport and relate vulnerability to soil properties; and (3) to compute the contribution of this region to the herbicide load of the Missouri and Mississippi Rivers. Grab samples were collected under baseflow and runoff conditions at 21 hydrologic monitoring stations between April 15 and July 15 from 1996 to 1999. Samples were analyzed for commonly used soil-applied herbicides (atrazine, cyanazine, acetochlor, alachlor, metolachlor, and metribuzin) and four triazine metabolites (deisopropylatrazine, deethylatrazine, hydroxyatrazine, and cyanazine amide). Estimates of herbicide load and relative losses were computed for each watershed. Median parent herbicide losses, as a percentage of applied, ranged from 0.33 to 3.9%; loss rates that were considerably higher than other areas of the United States. Watershed vulnerability to herbicide transport, measured as herbicide load per treated area, showed that the runoff potential of soils was a critical factor affecting herbicide transport. Herbicide transport from these watersheds contributed a disproportionately high amount of the herbicide load to both the Missouri and Mississippi Rivers. Based on these results, this region of the Corn Belt is highly vulnerable to transport of herbicides from fields to streams, and it should be targeted for implementation of management practices designed to reduce herbicide losses in surface runoff.