Red Pivots and Nitrate

An exploration of visual water quality indicators

Sometimes local knowledge generates the best, curiosity-driven research.

We don't have much groundwater nitrate in that area with the red pivots.

We know that irrigation wells are a direct connection to the groundwater below our feet. Could the visual appearance of center pivots supplied by those wells give us some indication of water quality conditions in the aquifer?

We decided to test this new idea that red, rust-stained pivots could be correlated with lower groundwater nitrate concentrations. First, we looked at Google Earth imagery in the Phelps, Kearney, and Adams County area. Surprisingly, we could do a fairly good job of classifying pivots as "full rust", "part rust" or "no rust". Almost 90% of the pivots observed on Google Earth could be classified with reasonable confidence.

Is this pivot stained a reddish rust color? Mikaela and Bethany spent many hours viewing similar Google Earth imagery to classify center pivots by color.

After classifying pivot colors in Google Earth, our team drove through parts of Phelps, Kearney, and Adams Counties to ground truth those classifications. On the ground, we viewed and classified the color of 250 center pivots. Overall, ground-based and Google Earth observations matched up 82% of the time.

Mikaela and Bethany drove through parts of the three counties to ground-truth Google Earth classification of pivot colors. Pivots with drop nozzles were better identified through ground observation, because staining is mostly on the lower parts of towers and not the spans visible from above.

With 250 pivots classified from ground-based surveys, the next step was to see whether red (rust-stained) pivots were correlated with groundwater nitrate concentrations. Thanks to many samples collected over the years by Tri-basin and Little Blue Natural Resources Districts, and the data publicly available in the  Nebraska Groundwater Quality Clearinghouse , this task was not too difficult.

https://clearinghouse.nebraska.gov/

The Nebraska Groundwater Quality Clearinghouse was used to obtain nitrate data for the study area.

Our original question was whether "rusty" red pivots were correlated with lower nitrate concentrations. But a correlation doesn't explain why two things are related. Is there a reasonable explanation for connections between lower nitrate concentrations and red pivots?

One possible explanation we thought of was redox chemistry. Redox reactions are chemical reactions that involve the transfer of electrons and are commonly facilitated by microorganisms as a way of gaining energy. Microbes that carry out these reactions are present in soils and in the geological materials that store and transport groundwater.

In order to carry out redox reactions, microbes need two things: electron acceptors and electron donors. Organic carbon is often the most available electron donor in aquifers. Microbes can use a variety of electron acceptors, however. And microbes selectively use these electron acceptors in a very specific order based on the amount of energy they get from the reaction.

The first and most favorable electron acceptor is dissolved oxygen. Next in line is nitrate. Only after nitrate is consumed do the microbes eventually interact with iron, which can then stain rusty pivots. Check out the full sequence below.

So, the idea of rust-stained pivots being correlated with lower nitrate does have a plausible explanation. If iron is staining the center pivots, there's a good chance that nitrate has already been consumed from that same groundwater.

We then looked at some groundwater nitrate data compared to center pivot classifications to see if this would all pan out, at least for the three-county area in this study.

In this study area, groundwater nitrate concentrations were significantly lower for rust-stained center pivots compared to pivots with no rust. Partially rusty pivots also tended to have lower nitrate concentrations. Could this correlation hold in other parts of Nebraska? We share a few caveats below. (Figure modified from Cherry et al. (2022))

When we plotted up the nitrate data based on pivot appearance, it was apparent that rust-stained pivots consistently yielded groundwater with low nitrate concentrations. Most of part-rust pivots had lower nitrate, and no-rust pivots produced groundwater with a wide range of nitrate concentrations. So, a pivot with no rust does not necessarily indicate high nitrate in groundwater. But a rust-stained pivot does tend to indicate lower nitrate concentrations, most of which are below 10 ppm.

We were curious how this might look on a map of center pivots and groundwater nitrate concentrations. The interpolated maps of center pivot appearance and groundwater nitrate concentrations show reasonable agreement.

Interpolated maps showing (A) center pivots classified as being at least partially rust-stained and (B) groundwater nitrate concentrations in the study area. The maps are not perfectly correlated, but a diagonal swath of lower nitrate concentrations appears to align with center pivots with at least some rust staining. (Figure modified from Cherry et al. (2022))

As with any study, we have thought a lot about uncertainties and caveats. Below are a few questions we have. Perhaps you can think of a few more.

  • This study was in a three-county area in south-central Nebraska. Will the correlation hold in other study areas?
  • What if some older pivots are actually rusting, rather than stained with rust?
  • Some pivots are painted. This is obvious during ground-based classification, but could this create uncertainty in Google Earth assessments?
  • Irrigation wells have long screened intervals. Could there be a mixture of low-nitrate and high-nitrate groundwater in some wells?

In conclusion, this study shows significantly lower groundwater nitrate concentrations associated with center pivots stained with rust. Shiny, no-rust pivots yielded groundwater with a wide range of nitrate concentrations.

We believe this new method warrants broader testing as a visual method for assessing groundwater quality in irrigated areas. In addition to applications in Nebraska, the method could be used internationally in areas with new irrigation development. Observations of rust staining on new center pivots could serve as an initial assessment of aquifer vulnerability to nitrate contamination.


Acknowledgements

This story map is based on a peer-reviewed scientific journal article: Cherry, M.L., Gilmore, T.E., Messer, T., Li, Y., & Westrop, J. (2022). A pivotal new approach to groundwater quality assessment. Environmental Science and Technology Water [Open Access:  https://pubs.acs.org/doi/full/10.1021/acsestwater.2c00121 ]

We are very grateful to Tri-Basin Natural Resources District, Little Blue Natural Resources District, and Hastings Utilities for their encouragement and support during this project.

We gratefully acknowledge support for PhD student Mikaela Cherry from the Nebraska Environmental Trust.

Thank you also to those who reviewed this story, including Jeffrey Westrop (UNL/CSD), Marty Stange, and François Birgand (NCSU).

This project is based on research that was partially supported by the Nebraska Agricultural Experiment Station with funding from the Hatch Act (Accession Number 1015698) through the USDA National Institute of Food and Agriculture.

The Nebraska Groundwater Quality Clearinghouse was used to obtain nitrate data for the study area.

In this study area, groundwater nitrate concentrations were significantly lower for rust-stained center pivots compared to pivots with no rust. Partially rusty pivots also tended to have lower nitrate concentrations. Could this correlation hold in other parts of Nebraska? We share a few caveats below. (Figure modified from Cherry et al. (2022))

Interpolated maps showing (A) center pivots classified as being at least partially rust-stained and (B) groundwater nitrate concentrations in the study area. The maps are not perfectly correlated, but a diagonal swath of lower nitrate concentrations appears to align with center pivots with at least some rust staining. (Figure modified from Cherry et al. (2022))