This essay ran in the quarterly newsletter of the Leopold Center, the Leopold Letter in the Spring 1992 issue.
Nitrate in the Des Moines River: Not a new problem?
A number of studies have confirmed that high levels of nitrate exist in surface and groundwater throughout the Midwestern United States. In Iowa, the press has featured letters and articles about nitrate contamination of the Des Moines River, which has been showing high nitrate-N (nitrate as nitrogen) concentrations, especially during the spring rainfall and runoff events. The public is expressing a growing conviction that these high nitrate levels, which find their way into drinking water, must be reduced, and soon.
This concern is based in part on the perception that nitrate contamination of surface and shallow groundwater is related closely to the large increase in nitrogen fertilizer use by Iowa farmers over the past five decades. Scientific studies support this belief by claiming an apparent, obvious connection between fertilizer N use and surface-water nitrate. Few of these studies, however, have used historic (pre-fertilizer use) records to prove such assertions.
To fill this gap, Dennis Keeney and Tom DeLuca of the Leopold Center compared both recent and historic (before commercial fertilizer use) nitrate-N records for the Des Moines River basin in an attempt to better define the major causes of high nitrate-N levels in the Des Moines River and determine how practices might be altered to reduce these levels.
The Des Moines River basin is dominated by cropland and dotted with pasture and forests along drainages. More than 80 percent of this land is devoted to agriculture, and agriculture indeed has had a profound influence on the river.
When the 1945 flow and average nitrate-N (5.0 parts per million) are compared to flow versus nitrate-N data for the last 11 years, the results are remarkably similar. The 1945 annual nitrate-N value alone also compares closely to the average annual nitrate-N concentration of 5.6 ppm for 1980-1990. This similarity suggests two possibilities: that the causes of high nitrate in the river today are not greatly different from those for 1945--or that the respective combinations of changes in nitrate inputs to the river have effectively "canceled each other out."
There is no indication that the 1944-1945 weather led to high nitrate concentrations in the river. Average annual flow in 1945 was 7040 cubic feet per second, again, a value remarkably close to the recent 110-year average of 7136 cfs.
What is the mystery behind these similar sets of results? Part of it may be that mineralization of soil nitrogen (formation of inorganic nitrogen, such as nitrate, from organic sources) appears to be the largest single contributor of mineral nitrogen to cropland in 1945, whereas commercial fertilizers constituted the greatest nitrogen input source in 1990.
Other nitrate sources that may have changed over time include legume nitrogen fixation and manure spreading. Total hay and alfalfa cropland has decreased since 1945, thus reducing potential input from legumes. And total livestock production, primarily hogs, has increased in Iowa since 1945. (However, all of the nitrogen in feed grain and fodder produced within the basin has been already accounted for.)
Urban runoff, a recently cited culprit, apparently does not significantly increase river nitrate-N levels; in fact, recent data indicate that nitrate-N levels are higher above the city of Des Moines than below. However, rural and urban human-waste treatment has also greatly improved since 1945.
It's no mystery that tile-drained row crops are a likely major contributor of nitrate in the Des Moines River basin. Conventional corn and soybean production promotes formation of high soil-nitrate levels by mid to late spring, and tile drains "short-circuit" the downward flow of water and nitrate in soil, carrying it to nearby drainage ditches and rivers. But because 98 percent of Iowa wetlands were drained by 1945, the effect of tile drainage on the river's nitrate levels should be similar in 1945 and 1990.
Nitrate can be removed from the soil by immobilization, plant uptake, leaching, or denitrification. Assessing the annual inputs and outputs of nitrogen from the basin indicates that the soil nitrate not used by crops in 1990 was about three times greater than that for 1945. Yet both years' data show rather inefficient use of native and applied nitrogen (64 percent efficient in 1945 and 53 percent efficient in 1990).
Considerable research is underway to reduce nitrogen losses from agricultural lands. The late-spring soil test can help reduce over-application of nitrogen fertilizer, and alternative cropping strategies and use of fall cover crops may reduce the nitrate available for leaching in the spring as well as enhance soil organic matter. Restoring and enhancing wetlands will also help decrease the amount of nitrate that reaches surface water systems, while buffer and filter strips can reduce the nitrate lost to surface runoff and shallow groundwater.
Regardless of the complex interaction of factors in 1945 as opposed to those in 1990, we know that agriculture in any form will allow some nitrate to move to water supplies. These complete interactions do, however, show that nitrate contamination of surface water in Iowa will not be alleviated by eliminating or changing one component of the system; rather, changes must be made in the total farming system.