Skip to Content

Manure Matters: IA 2020 Nitrate Summary

Posted on March 1st, 2021

Preface: I hit the wrong button and accidentally posted this for about 5 minutes before I was done, and realizing my mistake, took it down. What you see here is the final product. You may have noticed the mistake. Witold Krajewski once told me aging is something that is impossible to understand until it actually happens to you. I can confirm this, and part of that includes clicking the wrong the button on your computer. I had titled this “Farming in Skinny Jeans” so I could save drafts while I worked on it, without really intending to give it that title upon finishing. I usually come up with the title after I’m done. “Farming in Skinny Jeans” is not a bad title, but not a very good overview of what I have written here. So I changed it.

You can’t beat a good title. I wrote a data summary like this a year ago and titled it “Iowa is Hemorrhaging Nitrogen“. That was a good title. 

I haven’t posted anything in two months, and part of the reason is that I have been consumed with aggregating the data that is presented here. There a lot of maps and graphs and charts in this one that I fear may rheumy-up your eyes, but I’ve tried to sprinkle some No-Doz pills here and there which you can pick up and hopefully stay awake.

Before I go any further, I want to say my main man, Dan Gilles from IIHR, makes beautiful maps and he helped me with these. I also should mention Tom Stoeffler and Jordan Barnett (and before him, Jason McCurdy) whose diligent work in the field and shop keeps water quality data flowing in to the University of Iowa.

The data presented here were generated from the IIHR real-time water quality sensor network that Tom and Jordan maintain, USGS water quality sensors, and the Iowa DNR ambient water monitoring program. All the data is for nitrate/nitrite reported “as nitrogen”, which is customary in the U.S. Nitrate is a regulated drinking water contaminant, helps drive the harmful algae and cyanobacteria blooms that plague Iowa’s lakes and streams and drives the Dead Zone in the Gulf of Mexico. It’s an environmental consequence of producing crops and food animals the way we do. I also used USGS stream discharge data for this essay.

The period analyzed here is the “water year” (WY) of 10/1/19 to 9/30/2020. I quantified various aspects of nitrate (as N) loss for 50 Iowa watersheds, and used data from 15 of them to calculate statewide nitrate loss. I also quantify these aspects for areas draining to the Missouri River, and areas draining to the Upper Mississippi River. The 15 watersheds, along with the “divide” between the Missouri and Mississippi, are shown in the first map below. Some notable subwatersheds within the larger 15 are noted in smaller font.

Watersheds in the large font were used to calculate statewide N loading. Some notable sub watersheds are shown in the smaller font.

I’m missing data from a couple of watersheds I might otherwise have, and you can see these gaps in the above map: in Eastern Iowa, the Maquoketa River, and in Southern Iowa, the Thompson Fork of the Grand River. USGS removed their Maquoketa nitrate sensor a couple of years ago; we finally had to pull the IIHR equipment from the Thompson Fork last year because of repeated vandalism of the site. All in all though, we still have extensive areal capture (78%) of water leaving the state.

Data Sources

The water quality (nitrate) concentration data is an aggregation of the IIHR, USGS and Iowa DNR  datasets. This water quality data is combined with USGS discharge (volume, streamflow) data to generate loads, e.g. the total mass of nitrate transported by a river over time (in this case, 10/1/19 to 9/30/20). Loads are divided by watershed area to get yields, e.g., load per unit area. This enables comparison of N loss or yield in an apples-to-apples way from watersheds of different sizes. Total discharge is divided by watershed area to get runoff, or water yield. This allows me to compare water discharge in an apples-to-apples way from watersheds of different sizes. Water yields are estimated for watersheds without a discharge gaging station by looking at nearby discharge gages.

East vs West

The map and the table below illustrate Iowa’s division between the two major basins and how the stream nitrate data varies for each. The total mass (load) of nitrate-N leaving Iowa last year was 626 million pounds, with 435 million leaving on the Mississippi side and 191 million on the Missouri side. More is lost on a per acre basis from the Mississippi side, although the concentration of nitrate (mass per unit volume of water) is greater on the Missouri side. This is consistent with historical patterns, with the greater loss per acre in the east but higher concentration in the west. I will elaborate more on this later.

Nitrate-N loading and runoff from Iowa for the 2020 Water Year (10/1/19-9/30/20).

Individual Watersheds

The next map illustrates data from each of the 15 watersheds that figure into the statewide data. The black number is the N loss per total watershed area in pounds per acre. The red number is the N loss per crop area (corn + soybean) in pounds per acre. The blue number is the runoff in inches. Yes I know that not all nitrate comes from crop ground, but in Iowa it’s about 90-95% and probably higher in sparsely populated watersheds. A couple of things stand out on this map, the first being the extraordinary N loss in the three NE Iowa watersheds of the Upper Iowa, Yellow, and Turkey Rivers, with loss per crop acre exceeding 50 pounds in each and topping out at 78 pounds in the Yellow River watershed. We can see from the runoff numbers that this was the wettest part of Iowa; still 78 pounds is 78 pounds, and the Yellow River was one of the only watersheds in Iowa to have more N loss in ’20 than in ’19 (Yellow River runoff was less in ’20, 24.3″, vs 27.8″ in ’19).

N loss in pounds per acre is shown in black. N loss in pounds per crop acre is shown in red. Runoff in inches is shown in blue.

Very large losses of nitrogen in NE Iowa have been apparent for a few years now. Why? Some thoughts: thin topsoil in this unglaciated part of Iowa (Driftless Area, Pleistocene Plateau) provides very little storage of water or nitrogen. Karstic bedrock near the surface provides an easy and rapid pathway to the stream network for water and everything it can dissolve, essentially acting like drain tile. Corn yields can be very good here, but the environmental cost of row crop in this area of Iowa is very, very high.  Yes, when they try really hard and suck everything in, agriculture can squeeze the intense corn/soy/CAFO production model into this area of Iowa. Kind of like an old guy trying to get into a pair of skinny jeans. In theory, maybe it can be done, but……

The other thing that stands out in this second map is the Floyd River watershed in NW Iowa. One of Iowa’s driest in the past year (9.6″ of runoff), but still an N loss per crop acre of nearly 36 pounds. Compare the Floyd to the Wapsipinicon watershed of Eastern Iowa that had a similar N loss per crop acre (35.1 pounds) with nearly two times the amount of runoff as the Floyd. Lesson to be learned: Manure Matters. Watershed Floyd has about the same number of permitted CAFOs as it does square miles (~900), and it has been the only sensored watershed both of the last two years with an average flow-weighted nitrate concentration greater than 10 mg/L. Floyd is also the name of the pig trophy that Gophers and Hawks fight over each fall. Coincidence?

2020 vs. Previous Years

Statewide 2020 nitrate-N loads were lower than last year. There’s the good news. Of course we had a dry growing season in much of Iowa last year, and since nitrate is water soluble, it stands to reason more water = more nitrate in the streams. But it is interesting that statewide, the average flow weighted concentration of nitrate actually increased a little from last year. Think of it like this: you have a cup of sugar and a gallon of water. That gallon of water dissolves half of the cup of sugar. Now jump to the next year. We have another cup of sugar but only a half gallon of water. That half gallon of water dissolved not a quarter cup, but rather a third cup. This illustrates how nitrate pollution is driven by both supply and transport limitations. The supply is the amount of nitrogen vulnerable to loss every year; transport is the rain and river discharge. This supply of loss-vulnerable nitrate seems to be higher overall in areas draining to the Missouri than areas draining to the Upper Mississippi.

To account for year-to-year variations in precipitation, we often look at the 5-year Running Annual Average (RAA) load, i.e. the previous five years averaged as one. This is shown in the next graph. Although the RAA inched down in 2020, it’s still increased quite a lot over the past 20 years, and especially in areas draining to the Missouri River.

5-Year Running Annual Average Nitrate-N Loading, with the % increase since 2003 (Data begins in 1999). Iowa areas draining to the Upper Mississippi are shown in blue; Iowa areas draining to the Missouri are shown in green, and Iowa in total is shown in red.

Another Way of Looking At It

If you haven’t noticed, industry apologists like to rationalize the nitrate problem by handcuffing it to wet weather because, well, I guess they think it must look bad when you kill off part of an ocean and impair the drinking water supply of the state’s largest city for 50 years. But that’s just me speculating. Sort of. Anyway, I thought I would try to get creative with this idea. So I looked at all 50 of my monitored watersheds, determined what portion (%) of Iowa’s nitrate came from each one, what portion (%) of Iowa’s water came from each one, and divided the first number by the second number. So a value of 1.0 would be equivalent to the state as a whole; above 1 = worse than average; below one = better than average. This data is shown in the table below.

You can see Boss Hogg is the Floyd River Watershed. No surprise there. Sharing the same sty as Floyd is the very swiny South Fork of the Iowa River watershed. Going down the list, every one of the top 10, except #3, is either home to a lot of swine, or very near one. Also, if you need evidence that we aren’t going to “project” (i.e. watershed project) our way to clean water, look no further than South Fork. They’ve had an ongoing project since the Clinton Administration (1999), and it’s still one of Iowa’s worst for nitrate pollution. By the same token, South Fork is also evidence we aren’t going to “science” our way to clean water–it’s the topic of countless published scientific papers, including at least 10 in the last year alone. Now that’s some exciting momentum!

Table showing a watershed’s share of Iowa nitrate loading divided by the watershed’s share of Iowa’s runoff (2020 water year).




Tags: , , , , , , , , ,

24 Responses

  1. Marie DeVries says:

    Chris–You’re the only writer who can make me laugh — and cry — at the same time. Keep it up, please.

  2. cjones says:

    Thanks Marie!

  3. Chris Lee says:

    Another fine bit of data, sir! I love reading your posts. I wonder if you can help me understand something… If I remember correctly, Iowa’s first version of the Nutrient Reduction Strategy came out in 2012. Looking at your graph with the 5 year running average of total N, it would appear that we were making good progress between 2012 and 2015. Then the trend turned around dramatically and has flown upwards since, which is seriously disturbing. But, since that’s a 5-year running average, post-NRS progress wouldn’t be fully captured until 2017, right? In other words, any point on the graph includes the average of the five years prior, so any point prior to 2017 would include at least one year prior to 2012 when the NRS was first published. If that’s the case, why isn’t anyone (other than folks like us) asking why, despite having this strategy in place, are we worse off than, well, ever?

    • cjones says:

      the nutrient strategy objective is 45% reduction from a 1981-1996 baseline which comes from the Gulf of Mexico Hypoxia Task Force. We did indeed see some downward nitrate trends in several Iowa rivers up to about 2015. Since then, things have gotten worse. My own feeling is that the 2012 drought was really mid 2011-2014 with two exceptionally wet months in (Apr and May) in 2013. 2015 and 2016 were very high loading years on some rivers.

  4. David J Hilgemann says:

    Great Map. Regarding NE Iowa and your numbers. I suggest you check the runoff (21.6″) for the Boyer River in SW Iowa. I think you have an error. (I’d like to know your calculation methodology for runoff as well). Note that the runoff values you have for the 3 NE Iowa Rivers are the highest. For having a drought, it has been wet (we got 10″ of rain this fall near Garber) and it has been wet for the last 30 years. That anount of rain leeches the N out of the bottom ground planted to corn. I know as I had 21 acres still in corn next to S. Cedar Creek and the yields got hammered the last years. (I am sure the karst plays a role in this as well, particularly on the Yellow River as it has the most sinkholes and the highest N loss. I have seen a sinkhole drain a field of spring snowmelt in a couple minutes and was awestruck). On a positive note, there won’t be corn planted in my bottom ground this spring. I may be slow, but I do learn. By the way, can you distinguish between N from applied ammonia and manure? Kep u0p your good work.

    • cjones says:

      yes you were right. Boyer was incorrect. Should be 9.8 inches. to calculate, I sum up the total daily discharges for the year in cubic meters and then divide by watershed area to get runoff in mm, then convert to inches.

  5. David J Hilgemann says:

    I have another question about NE Iowa N numbers in red (N pounds/acre of crop ground). What is the N loading of woodland? Since there is more woodlands in NE Iowa, the ratio of total area of the watersheed to crop ground would be highest for this area. Could the N loading of the woodlands impact your numbers significantly?

    • cjones says:

      there is nitrate loss from forested landscapes, especially after they have been disturbed (logging, fire), and post-disturbance loss can be substantial. On the whole, however, N loss from forest/woodland is mostly linked to deposition amounts, and forests tend to be sinks for nitrogen deposited from the atmosphere.

  6. Dennis Goemaat says:

    Good presentation of data, Chris. I like the title, but the skinny jeans title would have also worked well for me. As an old guy with a not-so-skinny build, the analogy conveys the difficulty of what we’d like to have with the reality of the situation. I appreciate your ongoing work on this subject and your ability to present facts in a straightforward way.

    • cjones says:

      thanks Dennis for your kind words, and for continuing to follow the blog. Hope you and your crew are well.

  7. Kenneth Elgersma says:

    Great post! I’m really interested in your creative approach to scoring watersheds, in particular because the majority (about 2/3?) of watersheds have a score above 1.0. Like Lake Wobegon, it appears we are all above average. This implies that the below-average watersheds tend to be further below average than the above-average watersheds are above that average. Can you find any reason for this? What makes some of these watersheds so very “bad” at losing N? You’ve shown in this post and others some pretty clear evidence for why those top-ranked areas are hemorrhaging N; are there any additional patterns (besides just lack of CAFOs) that help explain why the low-ranked watersheds are lower than expected creating the imbalance between below-average and above-average?

    • cjones says:

      More watershed are above 1.0 because there is no area weighting here. Many of the large watersheds (i.e. Des Moines River) are below 1.0 and thus there are fewer that are below 1. What makes a watershed “bad”: lots of manure, lots of tile, being on the Paleozic Plateau.

      • Kenneth Elgersma says:

        Ah, yes the area weighting makes sense. Thanks. What I actually meant by “bad” was “good at retaining N” (i.e. “bad” at exporting N, not good at exporting it). Sorry my question was unclear, but with the area-weighting, the skewed distribution makes sense and I wouldn’t expect any additional factors (other than little manure, little tile, and not being on the Paleozoic Plateau) to be important.

  8. John Norwood says:

    Thoughtful post. I have been advocating for looking at nutrient loading as a more meaningful management tool than concentrations of pollutants at any given time so glad to see you provided some numbers! Not sure I understand the runoff metric? Can you explain in a little more detail. Also, realize it might be harder to assemble the data, but I wonder about a metric that includes resident hogs in each watershed. Down the road, we might have a multi-regression predictive model that can help explain the results we see, factoring in animal units, crop acres, no till acres, cover crop acres, and so forth. I’d like to see us understand the loads that flow across county lines so we have a better understanding of where we need to focus our efforts and what tools we have at our disposal to impact loading. County Government will play an important role in managing the tile drainage infrastructure in particular plus county parks might handle drainage system wetlands, buffers, flood zones, etc. Thanks again Chris.

    • Cindy Hildebrand says:

      John, I’d be nervous about those last ideas unless counties and county conservation boards were given plenty of money for that work. I know that Polk County has had the good sense and fortune to establish a natural resources fund.

      But in many if not most other counties, CCBs are struggling to do what they’ve already been assigned to do. And in certain counties, including mine, new money is badly needed for more public conservation land to accommodate growing populations of people who want more outdoor recreation.

      It is not the responsibility of counties, including CCBs, to pay for dealing with farm pollution, unless the costs of doing that are covered by new money, not current budgets. And unfortunately, providing adequate money to public agencies for new responsibilities hasn’t always happened in Iowa in recent years.

      • John Norwood says:

        I share your concerns about funding and unfunded mandates. I’m not suggesting tasking counties with those. What I am suggesting is we the state needs a framework, dedicated funding, and a process for prioritizing the modernization of the drainage plumbing to meet ROI criteria and include missing pieces of infrastructure like wetlands and floodplain storage. In my vision, I see this happening voluntarily by drainage districts who understand their systems are in need of repair and can access funding help provided they meet system design requirements (system based approaches, bundled and delivered with cost efficiency) that include important public benefits. Recreation is one benefit. Others include water filtration, aquifer recharge, flood storage, pollinator habitat, hunting, etc. Some of this land could be county managed by the county parks department such as is already done in Floyd County. One glaring omission of the so called Nutrient Reduction Strategy (really more of a menu since it lacks a specific plan of action, measurable goals and resources) was that the financial analysis conducted by ISU did not monetize any of these benefits, nor did it place value on the very long life of this infrastructure. It only values nitrogen reduction which skewed the attractiveness of certain options. Well, most of us realize these systems require the delivery of more than just one service like nitrogen. We need clean water, bees, etc. So let’s wake up and look at holistic solutions including the repurposing of 2 million of the 23 million acres to deliver missing land use pieces wetlands, buffers, etc. and higher value activities including new forms of indoor agriculture that can employ the next generation and help revitalize communities that have been hollowed out by a race to the bottom (cost driven) commodity agriculture that is out-of-balance. Best, J

        • Cindy Hildebrand says:

          John, thank you for taking time to respond. I have a lot of thoughts in response to yours, but I won’t take up more space on Chris’s good blog at this time except to say that I remain concerned that the very limited agency funding in Iowa that is used to help native biodiversity and the management of conservation land might be tapped to help deal with water pollution caused by the owners and operators of private farmland. If my concerns turn out to be unwarranted, that will be great. But I’ve recently heard a few things that will keep me concerned for the time being.

    • cjones says:

      John, we really don’t need more models to explain what we are measuring in Iowa waters. While continuing to study these sorts of things is meaningful, uncertainty about causes is not the reason we have the problems that we do. I suggest you read this one if you haven’t already:

      • John Norwood says:

        The point is to have management tools that can help us understand the impact of management options and tradeoffs, some of which you have outlined elsewhere, we will have to make at an operational level of watersheds. We do not have a monitoring and investment system that is localized and or targeted enough in my opinion. So the result becomes, “if everyone is responsible, no one is responsible.” And we sprinkle around investments and hope something good will result. Broad based regulatory solutions of the kind you and others have advocated before have not been adopted or enforced. I think we need a more focused approach and a more advanced monitoring system is an important tool in my opinion for management and modernization of our 23 million acre system.

        • Veronica Lack says:

          Reinstate Adam Lack’s USDA-NRCS “RUSLE3″ or Adam’s Revised Universal Soil Loss Equations Version Three” that was installed under Democratic President Obama’s Administration and then in February of 2017 our RUSLE3 was REDACTED by the Republican’s President Trump’s Administration.

  9. Keith Kuper says:

    Thanks for this article and all you do to show the ongoing disastrous effects of agriculture on water quality.

  10. Veronica Lack says:

    Back in 2000, the greatest Iowa DNR director, Paul Johnson, who pasted in February, was pushed out, after he published his research on Agriculture’s long-lasting man-made NITRITES that were and are being Point Source drained from Worth County and part of Cerro Gordo, Mitchell, and Floyd County in an illegal Flood Drain, that contaminates the Cedar Valley Group Aquifers. This set of aquifers mostly along the west side of the Cedar River, supply the drinking water for over 1/2 Million Iowans. Cedar Falls and Waterloo have lost wells because of their increasing Nitrite and Nitrate levels that they are unable to safely treat or filter to safe levels. I am a small farmer who has tried to get the Iowa DNR to alter their Chlorination Protocol in case CVCV046761, the Iowa DNR is hiding the building levels NITRITES- Ammonia Nitrogen as N, and Ag herbicides chemicals like Roundup’s glyphosate, that we cannot filter from our Iowa Tap Water.

    Iowans especially in the 8 or so Counties in Russ Tell’s mapped out contaminated Plume in the Cedar Valley Group Aquifers need the reinstatement of my son Adam Lack’s USDA-NRCS Revised Universal Soil Loss Equations Version Three or the “RUSLE3” with access to the EPA and NRCS data bases it used to educate farmers and limit the point source Ag Nitrite and Nitrate contamination of U. S. Source Waters, our drinking water Sources.

    These Ag Nitrite levels and Cyanobacteria’s deadly Cyanotoxins are building up to increasingly higher levels in our Aquifers/groundwater. Causing higher animal loss numbers. We are not told to test for the Ag long-lasting Nitrites like Trump’s herbicide brand name “Trump Card”.

    “Ag Nitrite’s tests, that Iowa Tax-Payers funded for Paul Johnson’s research and later for Robert Libra’s are ignored by our Republican Governors, and Senators and are covered up by the huge Chemical Conglomerates that produce and pay researchers to hide the Iowa research data on these AG Nitrites. Iowa and other states DNR agents tells us to drill our wells on small farms to deeper and deeper aquifers, and then yet in about 4-5 years small female farmers like me are told again to drill to yet deeper aquifers. Now we are told to drill our wells thousands of feet deep costing 1/2 a million or more. Some nearby CAFO’s have been seen drilling to deeper wells for their CAFO animals while their neighboring small farmers new wells are only a little over 250 feet or so deep, and the Ag Nitrites are testing well over the Nitrite and Nitrate MCL’s.

  11. Veronica Lack says:

    Reinstate Adam Lack’s USDA-NRCS “RUSLE3″. Adam’s Revised Universal Soil Loss Equations Version Three” were installed under Democratic President Obama’s Administration and then in February of 2017 the RUSLE3 was REDACTED by the Republican’s President Trump’s Administration.

Leave a Reply