Nitrogen Availability from Liquid Swine and Dairy Manure: Results of On-Farm Trials in Minnesota
Michael Russelle, Kevin Blanchet, Gyles Randall, and Les Everett
Copyright © 2013 Regents of the University of Minnesota. All rights reserved.
On this page
- Nitrogen sources and their availability in manure
- Current predictions of manure N availability
We appreciate the many University of Minnesota Extension personnel who provided local assistance for this research, including Regional Extension Educators: Liz Stahl, David Nicolai, Ryan Miller, and Jodi DeJong-Hughes, and local Extension Educators Tim Dolan, Dan Martens, Jerry Tesmer, David Pfarr, Nathan Winter, and Brad Carlson.
The research would not have been possible without the cooperation of the following 13 Minnesota farm families – we deeply appreciate their involvement. Farmer cooperators and county: Gregory Read, Mower; Michael Heim, Winona; Kevin Lindeman, McLeod; Dennis Coleman, Watonwan; Brian Waage, Steele; Karl/Jim Dieball, Sibley; Mike Krenik, LeSueur; David Kolb, Stearns; Donald Schliep, Goodhue; Andy Schiller, Redwood; Brian Hanson, Rice; Craig Mensink, Fillmore; David Malakowsky, Blue Earth.
This research and outreach effort was supported by funds from the Minnesota Pollution Control Agency-USEPA Federal 319 Program, the University of Minnesota Extension, and USDA-ARS.
Selected nitrogen forms and transformations
Nitrate – a molecule (NO3-) that is dissolved in water. It is formed in soil by bacteria that oxidize ammonium. It can move long distances in soil with water, because its negative charge repels it from soil clays and organic matter. This movement makes nitrate more available to crops, but also increases the risk of leaching out of the root zone.
Urea – an organic molecule [(NH2)2CO] that is very soluble in water, moves in soil because it is not charged, and is rapidly disassociated into ammonium and carbon dioxide in the environment. Urea is the main nitrogen component in urine.
Dinitrogen – a molecule (N2) that is an uncharged gas. It is very stable and benign. About 78% of the atmosphere near the Earth's surface is dinitrogen. It can be converted into other nitrogen compounds by biological and industrial processes.
Nitrous oxide – an uncharged gas molecule (N2O). Nitrous oxide is a potent greenhouse gas (nearly 300 times more effective than carbon dioxide in trapping heat), and is a major contributor to global climate change. In the upper atmosphere, it reacts with ozone, increasing the amount of harmful ultraviolet radiation reaching the Earth's surface.
Denitrification – the conversion of nitrogen to two gases, nitrous oxide and dinitrogen. This process occurs mainly when nitrate is present in warm, wet soils. It also can occur in well aerated soil during nitrification.
Nitrogen (N) availability from liquid swine and dairy manure was measured by corn yield response on 13 Minnesota farms, and compared to predicted N availability based on current University of Minnesota guidelines (UM Extension Bulletin 03553, 2007). The predictions of N availability were most reliable for injection application. For broadcast-incorporated manure, nearly one-half of the predictions were more than 30 lb N/acre higher or lower than the measured N availability. This variation likely was due to ammonia losses from manure that were higher or lower than average due to weather conditions (rainfall, temperature, windspeed) after application and before incorporation. Direct injection by knives or sweeps is recommended to get the best value from manure N.
In terms of its nutrient content, manure is increasingly valuable. For both short-term decisions on fertilizer purchases and for long-term nutrient management plans, farmers and their advisors need reliable predictions of nutrient availability from manure. This is a report of on farm research designed to test the accuracy and precision of current University of Minnesota predictions of nitrogen (N) availability from dairy and swine manure slurry.
The vast majority of dairy and swine manure is stored and applied as slurry. About 80% of the P and 90% of the K in manure slurry is available for crop uptake during the first growing season after application. Predicting N availability is more difficult, because losses can occur before plants absorb the N and because N is present in both inorganic, rapidly-available forms as well as organic, slowly available forms in manure.
Readers who are already familiar with nitrogen forms, availability, and losses from manure may wish to go directly to the section titled “Research methods” where the methods and results of our on-farm research begin. The following information and glossary are provided for those new to the topic or wishing a review.
(See Glossary for a description of italicized terms in this section)
Some of the N in manure is present as ammonium, which is directly available to crops but is even more available as nitrate after soil bacteria nitrify the ammonium. However, ammonium also can be lost as ammonia gas. This loss by volatilization is highest when manure is exposed to air, especially at the alkaline pH of manure and many Minnesota soils. Rapid incorporation of manure is the best way to limit ammonia volatilization.
Losses of nitrate occur when it is leached below the crop root zone or is denitrifed to N gases when the soil is too wet. To reduce the potential for nitrate losses, manure should not be applied in autumn until soil temperatures remain below 50° F, which slows nitrification. Corn does not absorb much N until mid-June, increasing the risk of nitrate leaching during spring on coarse-textured or shallow soils. To reduce the risk and cost of nitrate leaching losses on these soils, we recommend postponing application of both manure and fertilizer N until spring.
Organic forms of manure N include organic bedding materials, unconsumed and undigested feed, microorganisms from the animal's intestines, intestinal cells that were shed during digestion, and urea, proteins, and amino acids in urine. It is not likely that undigested feed will break down quickly in soil – after all, it went through the animal's digestive system. However, other forms of organic N can be converted to plant-available ammonium and nitrate over periods of days to weeks. For example, urea is rapidly converted to ammonium and carbon dioxide (even on the barn floor!), whereas microbial cells in the manure are decomposed by soil microorganisms within a few weeks in warm, moist soil.
Organic N continues to supply plant-available N over several years, although at declining amounts. In the first year, perhaps 35% of the organic N in swine manure becomes available (Koelsch and Shapiro, 2006). This declines to about 15% the next year. As a result, repeated application of manure to a field provides cumulative effects. After two years of manure application at the same rate of organic N, available N will equal about 50% of the annual rate of organic N application (35% + 15%).
Broadcast manure application is fast, cheap, and wasteful. Because a large area of manure is exposed to the atmosphere, very large N losses occur by ammonia volatilization. For the same reason, both odors and surface runoff can occur.
Manure N availability is measured in the field as a ‘fertilizer N equivalent,’ expressed as the amount of fertilizer N that can be replaced by applying manure. The current University of Minnesota predictions of N availability from manure were developed in 1999 and include the two most important factors to consider: a) manure composition; and b) method of application (Blanchet and Schmitt, 2007).
Manure N composition is determined by:
- animal species and age, because of differences in the ration and digestive systems; and
- typical storage methods, which affect the separation of liquid and solids and ammonia volatilization losses.
Tables of typical manure composition by species and age are available (Blanchet and Schmitt, 2007). For example, estimated total N concentration in liquid swine manure increases from 15 lb/1000 gal from farrowing sheds to 58 lb/1000 gal during finishing, but manure composition is similar for dairy heifers and dairy cows (32 and 31 lb/1000 gal, respectively). These values should be used when on-farm measurements of manure are not available.
Inadequate coverage of manure during injection, because of excessive application rate, shallow injection depth, or lack of closing disks, exposes manure to the atmosphere and allows ammonia loss and runoff.
Direct injection at moderate manure rates with thorough incorporation requires more passes and greater drawbar power, but conserves manure N while eliminating runoff and minimizing odors.
Sampling manure is not easy, but it is the best way to know what is being applied. It usually is not possible to obtain a representative, well mixed sample of manure except during application. Analyses from those samples can be used after application to determine actual nutrient application rates. Any remaining nutrient deficits can be remedied with commercial fertilizer. The accumulated record of manure analyses can be used to guide future application rates, but the N concentration of manure in the tank does not tell the whole story.
Manure application method also affects N availability to the crop. Broadcast applications, whether by tractor-drawn equipment or through irrigation systems, promote ammonia volatilization. Once the manure is mixed with soil, ammonia released from the manure is dissolved in the soil solution as ammonium, and captured by soil clays and organic matter. Direct injection minimizes N losses, especially when a sweep is used to spread the manure under the soil surface. Injection without a sweep increases the potential of denitrification, because the manure band is not well mixed with soil. Denitrification results in N loss as the gases dinitrogen and nitrous oxide.
Current University of Minnesota predictions of manure N availability are based on all these factors, and are presented in Table 1 for the first and second year after application, based on animal species and application method (Blanchet and Schmitt, 2007). The table is simplified so that only a total manure N analysis is required, although losses of inorganic N and mineralization of organic N were considered in the predictions. Our objective in this research was to test on farmers' fields how accurate these predictions are with different manure sources and application methods.
Table 1. Predicted manure N loss and N availability for the first and second year after application of dairy and swine manure.
|Surface broadcast followed by incorporation in||Injection|
|≤ 12 hours||< 4 days||≥ 4 days||Sweep||Knife|
|Type||% Total N|
|Year 1 availability||55||40||20||55||50|
|Year 2 availability||25||25||25||25||25|
|Year 1 availability||75||55||35||80||70|
|Year 2 availability||15||15||15||15||15|
|≤ = equal to or less than; < = less than; ≥ = equal to or more than|
Factors affecting ammonia volatilization loss
Ammonia moves from areas of high concentration to low. The higher the ammonia concentration in manure or soil water, the greater the risk of ammonia loss.
Concentration in water increases with:
- amount of ammonium
- high pH (alkalinity)
- coarser soil texture (holds less water than finer textures)
- drying (Similarly, the lower the concentration in the air, the more ammonia loss occurs.)
Concentration in air decreases with:
Ammonia loss also is increased by:
- longer exposure time to the air
- shorter distances between water and air
- larger surface area (such as occurs with broadcast application or with small droplet sizes from sprinkler systems)
Field experiments were conducted on 13 Minnesota farms from 2004 through 2006 (Figure 1). The previous crop was soybean in most trials, corn in the rest. Selected fields had no alfalfa crop within the previous 5 years and no manure application within the previous 3 years. Corn was the test crop planted in the spring following manure application in late autumn at each location. Farmers agreed to avoid fertilizer N application on the plot area. P and/or K fertilizer was applied before spring tillage to two sites where soil tests indicated a deficiency.
Figure 1. Approximate locations of on-farm research sites.
Manure slurry from lactating dairy herds or liquid swine manure from finishing barns was applied at 10 sites using the farmer's equipment, which had been calibrated immediately before the application. A certified commercial applicator applied the manure at the other 3 sites. Strip widths (12 to 16 rows) were determined by available field equipment, and lengths (500 to 2480 feet) were determined by field dimensions. Manure samples collected during application were frozen and later analyzed by a certified commercial laboratory. Details of manure composition, application method and application rates are in Table 2.
Dairy manure slurry from the five farms contained an average of 24 lb total N and 13 lb ammonium N per 1000 gallons. Ammonium N concentration was nearly constant, but the proportion of ammonium N to total N ranged from 41 to 81%. Swine manure from eight farms was more variable in composition than the dairy manure slurries, and both ammonium N and total N varied with solids content. In contrast to dairy manure, the proportion of total N that was present as ammonium N in swine manure was less variable, averaging 79% and ranging only from 72 to 89%.
Manure application rates were established using table values for the various N sources and predicted first-year N availability (Table 1). Actual N application rates were calculated after manure analyses were received. Manure treatment strips were replicated 3 times. Rates were selected to provide: 1) the full recommended N rate to corn; and 2) one-half that amount. In some cases, the farmer's equipment could not apply as low a rate as was desired, highlighting a logistical problem for achieving optimum manure management. Application rates ranged from 2,900 to 11,900 gal/acre for dairy manure and from 1,190 to 6,000 gal/acre for swine manure.
Figure 2. Schematic diagram of the layout of one replicate of all treatments. Manure was applied in late autumn to long strips, and subplots of urea fertilizer treatments were spring-applied to smaller areas near one end of the strips. Rates of urea are expressed as lb N/acre.
Small plots for fertilizer N application were located near one end of the strips (Figure 2). Each small plot was four rows wide by 30 feet long, and received a different fertilizer N rate. After planting, urea was broadcast on the soil surface by hand at rates of 0 to 160 lb N/acre in the control (nonmanured plots) and 0 to 120 lb N/acre in the manured plots. The urea was not incorporated.
Corn was planted by the farmers in April or May in 30- inch rows, except for three sites where farmers used equipment with 22-inch rows. Tillage, corn hybrid, seeding rate, pest control, and other farm operations were determined and conducted by each farmer. Grain yield was measured in small plots by hand harvesting 40 feet of row. All yields were adjusted to 15.5% moisture.
Table 2. Manure characteristics at each farm site
|Manure Type||Crop year||County||DM||Total N||NH4-N||P2O5||K2O||Application Method||Half rate of manure||Full rate of manure|
|2005||Mower||11.4||37||15||16||32||Broadcast, incorporated <12 hrs||2900||6700|
|Stearns||4.0||21||14||11||23||Broadcast, incorporated 12-96 hrs||4800||8400|
|2005||Le Sueur||4.8||55||41||27||31||Broadcast, incorporated <12 hrs||1420||2850|
|McLeod||6.7||67||53||32||33||Broadcast, incorporated <12 hrs||1190||3770|
|Watonwan||3.3||45||37||25||26||Broadcast, incorporated <12 hrs||1900||3000|
|2006||Blue Earth||8.8||65||51||36||43||Broadcast, incorporated <12 hrs||1500||3000|
|Fillmore||5.3||76||65||36||50||Broadcast, incorporated <12 hrs||3200||4000|
|Redwood||2.1||34||29||12||27||Broadcast, incorporated <12 hrs||2800||4400|
|DM = Dry Matter; P2O5 = phosphate; K2O = potash; < = less than.|
Determining the fertilizer N equivalent
Grain yields were plotted by their respective fertilizer N rates for nonmanured plots and a curve was fit to the data (solid line in the graphs). Data for manured plots without N fertilizer were then plotted on the vertical axis. Horizontal lines were drawn to the curve and the corresponding fertilizer N rate was determined for each manure rate. The resulting fertilizer N rate was considered the fertilizer N equivalent of the manure.
In the Steele County example, the half-rate of manure produced yields equal to 59 lb fertilizer N/acre in nonmanured plots, and the full-rate was equivalent to about 130 lb fertilizer N/acre. In the Rice County example, yields of both manured treatments were higher than the fertilizer response curve, so the actual fertilizer N equivalent could not be calculated. Both manure rates were equal to at least 137 lb N/acre, the fertilizer amount required to maximize grain yield.
Corn yield response to fertilizer N
Average grain yield ranged from 83 to 262 bu/acre. Corn showed a yield response to fertilizer N on nonmanured plots at all locations except in Le Sueur County, which averaged 186 bu/acre across all fertilizer N rates on nonmanured plots.
In some cases, manure alone provided sufficient N to maximize corn yield (Table 3). We intended that the full rate of manure would provide adequate N, but actual manure N content often was lower or higher than published table values. In addition, ammonia losses that were greater than average probably reduced N availability.
Two of 13 locations showed evidence of yield benefits from manure that exceeded those from adequate fertilizer N supply (Table 3). In those cases (Blue Earth and Fillmore County), lack of rainfall after the urea fertilizer application may have allowed greater ammonia volatilization losses from urea than occurred at the other field sites, causing a reduced amount of fertilizer N available to the crop. Once urea dissolves, ammonia losses can be reduced by rainfall or irrigation that moves the N into the soil. More than a week passed in both sites before more than 0.4 inch of rain was received. The resulting fertilizer N loss may have reduced potential crop response on fertilizer-only plots. Therefore, we cannot conclude that manure provided yield enhancing benefits other than N supply.
Table 3. Yield of corn with fertilizer at the Maximum Return to Nitrogen (MRTN) rate or with liquid manure alone.
|Manure Application Method||County||Corn grain yield||Fertilizer N rate or predicted manure N availability||Under-or over-prediction by U of M|
|Fertilizer at MRTN||Half rate of manure||Full rate of manure||Fertilizer at MRTN||Half rate of manure||Full rate of manure||Half rate||Full rate|
|bu/acre||lb N/acre||lb N/acre|
|Broadcast/ Incorporated||Le Sueur||186||182||189||0||58||117||–||–|
a Yields at four sites (italics) may not represent the yield potential, because grain yield appeared to be increasing beyond the highest fertilizer N rate applied.
b Yields at two sites (boldface type) with manure were significantly larger than yield with adequate fertilizer N
Of the 26 potential combinations shown in Table 3 (13 farms, two manure rates per farm), 17 provided valid comparisons of predicted vs measured fertilizer N equivalent, and are shown in Figure 3. Predictions of the fertilizer N equivalent at individual sites were more reliable when manure was directly injected, than when it was broadcast. Predicted manure fertilizer N equivalents for injected manure did not differ from the measured fertilizer N equivalent by more than 29 lb N/acre in any individual comparison.
In contrast, predictions of N availability from broadcast/incorporated manure were less reliable, with nearly one-half being more than 30 lb N/acre higher or lower than the measured values in the field. For example, at the Stearns Co. dairy site (28 and 47 lb N/acre under-prediction using Table 1), actual ammonia volatilization likely was reduced by very wet soil and cool, moist weather after the manure was broadcast (average air temperature was 34° F during the 4 days after broadcast application). In contrast, higher than expected ammonia losses may have occurred from some of the sites with swine manure, which would have caused an overprediction of N availability, but we do not have evidence to support this supposition. Therefore, we could not explain the discrepancies between predicted and measured fertilizer N equivalents for the sites with broadcast/incorporated swine manure.
Using the predicted fertilizer N equivalent is easy. The first step is to estimate the fertilizer N rate that will maximize economic returns to N (called the MRTN), using the new Minnesota guidelines that are based on the ratio of fertilizer N price to crop value (Rehm et al., 2006; Sawyer et al., 2006). The MRTN does not consider N supplied by manure. Subtract the fertilizer N equivalent of manure from the MRTN to determine how much fertilizer N to buy and apply. For example, on a highly productive soil and a 0.10 ratio of N price to crop value, a TOTAL is 140 lb N/acre is needed for corn after corn. If the manure application had a fertilizer N equivalent of 59 lb N/acre, only 140-59 = 81 lb of SUPPLEMENTAL fertilizer N would be needed. This lower fertilizer rate will maximize profit, reduce over-application of N, and reduce losses of N to water and air.
These results emphasize both the strength of current (2007) University of Minnesota predictions for injected manure and the difficulties in predicting results for broadcast manure. Direct injection techniques, however, conserve more of the inherent nutrient content of manure, simultaneously reducing the risk of nutrient losses to the environment. With the high cost of commercial fertilizer and increasing scrutiny of manure management, this is a good time to upgrade application equipment or to utilize custom applicators – either of these can deliver more manure value to the crop.
Figure 3. Difference between Table 1 predictions and measured manure N availability, expressed as fertilizer N equivalent (FNE). Each column represents one manure rate at one farm site. Columns above the zero line indicate that more manure N was available than predicted by Table 1; those under the zero line indicate that Table 1 over-predicted N availability.
- Fact Sheet #1: Nitrogen Availability from Liquid Swine and Dairy Manure: Results of On-Farm Trials in Minnesota ( 414 K PDF)
- Fact Sheet #2: Nitrogen Availability from Liquid Swine and Dairy Manure: Can we trust small plot studies? (251 K PDF)
- Fact Sheet #3: Nitrogen Availability from Liquid Swine and Dairy Manure: Ear leaf chlorophyll measurements (199 K PDF)
Blanchet, K., and M.A. Schmitt. 2007. Manure management in Minnesota. University of Minnesota Extension WW-03553.
Koelsch, R., and C. Shapiro. 2006. Determining crop available nutrients from manure. University of Nebraska–Lincoln Extension. NebGuide 494.
Randall, G., M. Schmitt, J. Strock, and J. Lamb. 2003. Validating N rates for corn on farm fields in Southern Minnesota. Univ. of Minnesota Extension MI-07936.
Rehm, G., G. Randall, J. Lamb, and R. Eliason. 2006. Fertilizing corn in Minnesota. 2006. University of Minnesota Extension FO-3790-C.
Sawyer, J., E. Nafziger, G. Randall, L. Bundy, G. Rehm, and B. Joern. 2006. Concepts and rationale for regional nitrogen rate guidelines for corn. Iowa State University Extension PM 2015.