The following article was featured in the 2023 NLRS Biennial Report.
Between the 1980–96 and 2015–19 reporting periods, nitrate-nitrogen loads in the Illinois portion of the Rock River between Rockton and Joslin more than doubled, resulting in the largest increase of any major Illinois river. Over the same period, nitrate-nitrogen loads from a major tributary to the Rock River — the Kishwaukee River monitored at Perryville — increased by only 5%. Thus, most of the increased load in the Rock River came from the section of the river downstream of Rockton and Perryville, referred to here as the Lower Rock River.
Nitrate-nitrogen yield from the Lower Rock River Basin during the 1980–96 baseline period was only 4.2 pounds of nitrogen per acre per year, which was much lower than other northern and central Illinois watersheds with similar land cover during the same baseline period. For instance, nitrate yields from the Kishwaukee River and Green River during this period were 17.1 and 12.2 pounds of nitrogen per acre per year, respectively. In the 1950s and ’60s, when less nitrogen fertilizer was applied, nitrate-nitrogen yields from several northern Illinois watersheds monitored by the Illinois State Water Survey (Harmeson et al., 1973) generally ranged from 3–9 pounds of nitrogen per acre per year.
More recently, in 2015–19, nitrate-nitrogen yield from the Lower Rock River was 21.5 pounds of nitrogen per acre per year, a yield similar to other watersheds with comparable corn-soy acreage. Additionally, after major droughts that reduced corn yields (e.g., in 1988 and 2012), there was little change in nitrate yields in the Lower Rock River. However, other watersheds’ nitrate yields were clearly elevated in years following major droughts (Lucey and Goolsby, 1993; Gentry et al., 2009; Loecke et al., 2017). A possible explanation is the existence of long groundwater flow paths that may have created long lag times between changes in watershed nitrogen inputs in the Lower Rock River nitrate loads. Long lag times can result from the presence of a large aquifer that interacts with the Lower Rock River system.
Researchers at the Illinois State Water Survey Groundwater Section calibrated the MODFLOW groundwater model to groundwater elevations measured in the Lower Rock River Basin starting in 1995. Based on this calibration and a variety of assumptions, average groundwater transit times varied across the basin, ranging from 5–50 years, with the shorter times occurring near the Lower Rock River and longer times occurring in remote sections of the watershed. While these transit times support a lag in nitrate reaching the river, the calibration might not adequately account for drier conditions in the 1980s, when groundwater levels were not measured. Periods of low rainfall would reduce recharge to the aquifer, result in longer groundwater transit times, and allow for increased nitrate loss through denitrification.
MODFLOW was also used to simulate nitrate flows to the Rock River using assumed concentrations in recharge water and denitrification rates. Initial results suggested that nitrate reaching the river through the groundwater could be relatively low and not explain the large increase in riverine nitrate load. However, this estimate was based on assumptions that may not accurately reflect conditions in the watershed and aquifer. Additional simulations may be conducted with a range of assumptions about aquifer recharge and denitrification rates. Additional groundwater and river sampling is needed to identify model assumptions that are most reflective of watershed and aquifer conditions and processes.
Increased precipitation and water yield were likely factors promoting increased nitrate load in the Rock River Basin from the 1980–96 baseline period to the 2015–19 period (Table 3.6). During 1980–96, water yields across the basin ranged from 10.7 inches per year at Rockton to 11.7 inches per year for the Kishwaukee River at Perryville. During 2015–19, water yields were between 2.6 and 8.1 inches per year greater than during 1980–96. The largest increase occurred in the Lower Rock River. This may have been due to the additional 8 inches of precipitation per year in this section of the watershed, the expansion of irrigated area from approximately 1.3% to 6.6% of the land area, greater groundwater discharge to the river, or a combination of these factors.
Table 3.6. Average annual water yields and the percentage change (water year basis) for several locations and subwatersheds of the Rock River Basin for 1980–96 and 2015–19
Watershed outlet or section | Water yield (in/yr) | Change (%) |
1980-96 | 2015-19 | Change |
Rockton | 10.7 | 15.7 | +5.0 | +45% |
Kishwaukee R. | 11.7 | 16.9 | +5.2 | +44% |
Elkhorn Creek | 11.5 | 14.1 | +2.6 | +23% |
Joslin | 10.9 | 16.5 | +5.6 | +52% |
Joslin - Rockton | 11.2 | 18.3 | +7.1 | +64% |
Joslin - Rockton - Kishwaukee | 10.9 | 19.0 | +8.1 | +75% |
Additional factors that may have contributed to the increased nitrate-nitrogen load were also considered, such as a larger human population and reduced in-stream denitrification due to increased streamflow; however, these appeared to contribute very little additional nitrogen to the river loads. Estimated nitrogen in livestock manure in the counties draining to the Lower Rock River peaked around 1960 and have since declined (Falcone, 2021). However, these county-level livestock manure nitrogen estimates would not capture the potential impact of spatially concentrated livestock operations that have occurred in recent decades. Finer spatial resolution of livestock production and manure application locations are needed to better understand whether this plays a role in the higher river nitrate loads.
The large increase in nitrate-nitrogen yields from the 1980–96 baseline period to the 2015–19 period appears to be due to a combination of factors. These include lag time between watershed land use and river nitrate concentrations, increased rainfall and water yield, expansion of irrigation, and expansion of corn-soybean acres. The low 1980–96 nitrate-nitrogen yields may have reflected land cover and agricultural practices in the 1960s and ‘70s. The 2015–19 nitrate yields in the Lower Rock River are in line with yields from other northern and central Illinois watersheds with similar levels of corn-soybean acres. If nitrate-nitrogen yields continue to increase in the future and exceed the nitrate yields from similar watersheds, this would suggest additional sources of nitrate that have not yet been identified.
Most conservation practices for reducing nitrogen losses that are applied in other corn-soy acres in Illinois (e.g., Maximum Return To Nitrogen, nutrient management, cover crops, etc.) are also applicable in the Rock River watershed. Saturated riparian buffers may be an exception since this practice does not perform well in sandy soils. Wood chip bioreactors would need plastic liners in sandy soils. Irrigation could be helpful in establishing fall cover crops in sandy soils, but applying irrigation water would add cost to establishing cover crops.
In addition to the common conservation practices focused on reducing cropland nitrogen loss, irrigation efficiency improvements could reduce nitrate movement to groundwater, and therefore to rivers, by eliminating unnecessary water application. Irrigators could also measure nitrate in their raw irrigation water and credit that nitrate in their fertilization management. Nitrate-nitrogen loss from seed corn production can be high due to relatively low removal of nitrogen in harvested seed (David et al., 2016; Gentry et al., 1998; Mitchell et al., 2000). Therefore, reducing nitrogen loss on these acres may be especially valuable. Expansion of riparian wetlands that interact with river water and reduce riverine nitrate concentrations could potentially provide nitrate load reductions in the near term regardless of groundwater lag times.
A lag time between activity on the landscape and nitrate loads in the river means that the benefits of nitrate loss reduction from conservation practice application in cropland may not reduce riverine nitrate loads for several years after implementation.
References
David, M.B., L.E. Gentry, R.A. Cooke, and S.M. Herbstritt, 2016. Temperature and substrate controls woodchip bioreactor performance in reducing tile nitrate loads in east-central Illinois. Journal of Environmental Quality 45:822-829.
Falcone, J.A., 2021, Tabular county-level nitrogen and phosphorus estimates from fertilizer and manure for approximately 5-year periods from 1950 to 2017: U.S. Geological Survey data release, doi.org/10.5066/ P9VSQN3C.
Gentry, L.E., M.B. David, F.E. Below, T.V. Royer, and G.F. McIsaac, 2009. Nitrogen mass balance of a tile- drained agricultural watershed in east-central Illinois. Journal of Environmental Quality 38:1841-1847.
Gentry, L.E., M.B. David, K.M. Smith, and D.A. Kovacic, 1998. Nitrogen cycling and tile drainage nitrate loss in a corn/soybean watershed. Agriculture, Ecosystems, and Environment 68:85-97.
Harmeson, R.H., T.E. Larson, L.M. Henley, R.A. Sinclair, and J.C. Neill, 1973. Quality of Surface Water in Illinois, 1966-1971. Illinois State Water Survey, Urbana, Bulletin 56. isws.illinois.edu/pubdoc/B/ISWSB-56.pdf
Loecke, T.D., Burgin, A.J., Riveros-Iregui, D.A. et al., 2017. Weather whiplash in agricultural regions drives deterioration of water quality. Biogeochemistry 133: 7–15.
Lucey, K.J. and D.A. Goolsby. 1993. Effect of climate variations over 11 years on nitrate-nitrogen concentrations in the Raccoon River, Iowa. J. Environ. Qual., 22: 38-46.
Mitchell, J.K., G.F. McIsaac, S.E Walker, and M.C. Hirschi, 2000. Nitrate in river and subsurface drainage flows from an east central Illinois watershed. Trans.Am. Soc. Ag. Eng., 43: 337–342.
