Utilizing Swine Effluent for Sprinkler- Irrigated Corn Production

Utilizing swine effluent for sprinkler- irrigated corn production

M.M. Al-Kaisi and R.M. Waskom

ABSTRACT: The rapid expansion of large swine production facilities in northeast Colorado prompted a need to evaluate the impact of swine effluent applied on irrigated corn grown on sandy soil. The objectives of this study were I) to evaluate the use of swine effluent as a nutrient source for irrigated corn production, 2) to evaluate the response of irrigated corn grown on sandy soils to different application rates, and 3) to evaluate N movement through the soil profile under swine effluent and commercial-N fertilizer for irrigated conditions. The three year study started in 1995 on a 14.5 ha (36 ac) sprinkler-irrigated (center pivot) Valent sand field, (Mixed, mesic Ustic Torripsamments) planted to grain corn (ZEA MAYS L.). Both swine effluent and commercial-N fertilizer treatments were applied at four N rates labeled control, low, agronomic, Copyright © 2002 Soil and Water Conservation Society. All rights reserved.

and high. All treatments were replicated three times in a randomized complete block (RCB) Journal of Soil and Water Conservation design. Approximately go% of the total nitrogen from the two-stage lagoon effluent was in ammoniacal form, and the total dry matter content of the effluent was only 0.1 - 0.2% by volume. Corn yields increased with the increase of both swine effluent and commercial-N fertilizer rates. In contrast to the swine effluent treatments, significant soil-N buildup was observed at the 1.5 - 3.0 m (5 - 10 ft) depths for the commercial-N fertilizer treatments. Higher total N and P plant removal for the swine effluent treatments resulted in little N accumulation below the root zone. As the swine effluent application rate increased, the plant N and P removal and recovery rate increased, even at rates of 50 kg N ha-' (45 Ib N ac-l) above the recommended agronomic rate. An increase in extractable P in the top 15 cm (6 in) of the soil was observed in the effluent-treated soils. The results indicate that managing swine effluent-N becomes very similar to managing commercial-N fertilizer under irrigated conditions. 57(2):111-120

Keywords: Corn yield, irrigation, N and P recovery, N and P removal, and swine effluent

Animal wastes produced by confined crop fields is the most common way to utilize www.swcs.org swine feeding operations can be a vatu- these materials. Over-application of lagoon able source of nutrients for crop produc- effluent, combined with irrigation or precip- tion. However, when manure is used under itation in excess of crop evapotranspiration, has been implicated in NOJ-N leaching

irrigated conditions, especially on sandy soils, an increased potential for nutrient runoff or below the root zone. Burns et al. (1985) leaching occurs. Recent expansion of con- found that soil-N accumulation increased centrated swine production facilities in east- with an increase in the swine effluent appli- ern Colorado has increased concerns about cation rate to bermudagrass, but that soil potential nitrate (NOJ-N) contamination of organic matter did not increase over a six year the Ogallala Aquifer, the sole source of water period. At effluent rates up to 1,340 kg N for drinking and irrigation in the area. ha-' (1,197 lb N ac-'), the crop recovered only Concentrated swine production facilities in the area commonly udze one- or two- stage lagoon systems where effluent must Mahdi M. Al-Kaisi is an assistant professor in the Agronomy Department at Iowa State University, be removed fiom the lagoon periodically to Ames, Iowa, and Reagan M. Waskom is a research prevent overflow. However, effluent fiom scientist in the Department of Soil and Crop second-stage lagoons is used for flush water. Sciences at Colorado State University, Fort Collins, Sprinkler application of swine effluent to Colorado.

I MIA 2002 VOLUME 57 NUMBER 2 I 111 1 44% of the applied N. After 11 years, Burns et lines at the recommended rate of 112 kg N suggests a greater potential for environmental al. (1990) reported that low rates of effluent ha-' (100 lb N ac-'), was minimal as compared losses of nutrients in swine emuent. on bermudagrass [335 kg N ha-' (299 lb N to the check treatment. Nitrate-N loss and These studes indicate that swine emuent ac-')I did not pose a groundwater hazard, accumulation in the top 2 m (6.5 f?) of soil can be managed to minimize nutrient losses while medium and high rates [670 and 1,340 increased as the application rate increased if applied at rates that approximate crop kg N ha-' (598 and 1,197 lb N ac-')I resulted (Gast et al. 1978). Maximum N03-N accu- removal. However, if crop requirements are in elevated N levels at the 2 m (6.5 fi) depth mulation in the soil profile occurred at a exceeded, the potential that excess N accu- in the soil profile.They found that high rates depth of 1.0 m (3.3 fi) with little evidence of mulation will occur is significant, perhaps [1,340 kg N ha-' (1,197 lb N ac-')I caused movement below 2.3 m (7.5 fi). Gangbazo et resulting in groundwater contamination. unstable grass stands, and eventually produced al. (1995) compared the effect of chemical A three year study was conducted to eval- forage that had elevated N03-N concentra- fertilizer and hog manure on potential water uate the potential impacts of swine effluent tions. In addition, quantities of certain ele- contamination. Fall spreading of large application on irrigated sandy soils in eastern ments in the soil environment had increased amounts of hog manure on cornfields result- Colorado. The objectives of this study were to levels that had the potential to become soil ed in two to three times the NH4-N losses 1) to evaluate the use of swine effluent as a and water pollutants, especially N, P, K, C1, compared to the check treatment of chemical nutrient source for irrigated corn production, and Na (Burns et al. 1990). ferthzer at 180 kg N ha-' (160 lb N ac-'). It 2) to evaluate the response of irrigated corn Sweeten et al. (1994) found that N removal was concluded that applying hog manure in grown on sandy soils to different application by forage plants increased as the dairy effluent the fall resulted in greater N loss and water rates, and 3) to evaluate N movement though application rate increased. Lysimeter studies contamination, and that hog manure should the soil profile under swine effluent and com- on forage plots receiving effluent applications be applied during the growing season at rates mercial-N fertilizer for irrigated conditions. at two and four times the agronomic N rate that approximate crop requirements. Hegg et

of 10 mg L-' (10 ppm) N03-N (Evans et al. effluent at daily application rates of 2.5 and The study started in March of 1995 on a Journal of Soil and Water Conservation 1984). Based on this study, the authors pre- 5.0 cm (1 and 2 in) increased the accumula- swine production facility and grain farm in dicted that effluent rates of up to 1.25 times tion of P, K, Cayand Mg in soils. Newton et Yuma County, Colorado, and continued the agronomic N could be applied before al. (1994) observed an increase in NO3-N at through the 1997 growing season. The study leachate concentrations exceeded 10 mg L-' soil depths of 1.6 m (5.2 fi) under swine site was located on a 14.5 ha (36 ac) center- (10 ppm) for N03-N. In another study where effluent application rates of 450 and 900 kg N pivot irrigated circular field with aValent sand groundwater-monitoring wells were installed, ha-' (402 and 804 lb N ac-') on a bermuda- (Mixed, mesic Ustic Torripsamments) (Table no increase in groundwater N was observed grass-fescue field. 1). The field was located near a swine pro- over two years of swine effluent applications Sutton et al. (1982) broadcast-applied and duction facility with a 4,000-head annual at rates up to 874 kg N ha-' (780 lb N ac-') soil-injected swine effluent at rates up to 857 capacity. In 1995 and 1996, the facility was a on bermudagrass (Harvey et al. 1996). kg N ha-' (765 lb N ac-') on corn, and they However, the authors reported a sigmficant found that the application method had a 57(2):111-120 increase in N03-N concentration in tile- greater impact on yield and plant N content I ment site in Yuma Countv. CO. drainage water as the swine effluent rate than did the application rate. Leaching losses increased, implyng that groundwater would were more likely on fields where swine efflu- i Parameters Description be contaminated over time. ent was soil injected because volatilization Texture Sand

Westerman et al. (1985) reported that on a losses were much less compared to broadcast Slope (%) 5 - 15 www.swcs.org grass crop system NO3-N movement to applications. King (1981) investigated N PH 6.6 - 7.8 groundwater at the medium and high appli- recovery from municipal sludge, swine Clay content (%) 3-10 cation rates [670 and 1,340 kg N ha-' (598 manure, and chemical ferthzer, and found Bulk density (Mg m-3) 1.6 - 1.7 and I,197 lb N ac-')I would be a greater con- that N recovery by grass was greatest from Organic Matter (g kg-l) 5 - 10 cern than the contaminated runoff caused by fertilizer, followed by municipal sludge. PermeabiI ity (cm h-l) 15 - 50 excess rainfall. Soil N03-N concentration Municipal sludge supplied 50% more NH4-N Drainage Excessive after the 11th year of effluent application than did the swine manure Also, King (1981) indicated that low rates [335 kg N ha-' (299 reported a slight decrease in the percent of N Ib N ac-')I did not pose a hazard to the recovery with the increase of the application swine finishing unit, and in 1997 switched to groundwater, but the high rate [1,340 kg N rate. In a comparison of ammonium nitrate breeding sows. The waste generated froni the ha-' (1,197 lb N ac-')I &d (King et al. 1990). (NH4NO3) ferthzer and swine effluent at animals was stored in a two-stage anaerobic This conclusion is consistent with the results sidar rates, Liu et al. (1997) found that the lagoon system. A 189 liter (50 -gal) per of their earlier study (Kmg et al. 1985). N source did not impact bermudagrass yield minute commercial well was used to supply In another study using chemical fertihzer at the agronomic N rate. Yield was not water mainly for animal use and limited (urea) on continuous corn at 20, 112, 224, improved by N application in excess of crop flushing of the animal waste to the first and 448 kg N ha-' (18,100,200, and 400 lb requirements, and the N recovery rate lagoon. However, effluent from the second N ac-'), the increase in NO3-N accumulation decreased as the application rate increased. lagoon recycled every eight hours to flush the in the soil profile or in losses from the tile Low N recovery at high application rates animal waste to the first lagoon.

112 I JOURNALOF SOIL AND WATER CONSERVATION MIA 2002 Figure 1 Field study layout for swine effluent and commercial-N fertilizer experiments. block design. Swine effluent was applied on Swine Effluent Plots the field using a sprinkler (center pivot) system by pumping effluent from the second cell of the two-stage lagoon through an underground pipe to the center pivot. Effluent from the first cell flowed into the second cell via gravity through a PVC pipe with an inlet at 1.5 m (5 fi) below the water surface [lagoon depth was 6 m (19.7 ft)j.The solid content of the effluent from the second cell was 0.1 - 0.2% by volume. The emuent application rates were estimated based on the agronomic N requirements for irrigated corn, according to Colorado State University recommendations for irrigated grain corn production (Mortvedt et al. 1995).The yield goal for the study site was set at 11,000 kg ha-' (9,823 lb

ac-'), and the N requirement to achieve that Copyright © 2002 Soil and Water Conservation Society. All rights reserved. yield goal was 207 kg N ha-] (185 lb N ac-'). Journal of Soil and Water Conservation Nitrogen credit from several sources (residual soil-N, soil organic matter, and irrigation water) was accounted for in estimating N requirements (Table 2).The low (L) and high (H) rates were 56 kg N ha-' (50 lb N ac') below and above estimated agronomic N rate.The control (C) received only 28 kg N C-Control (84 kg N ha-') ha-' (25 lb N ac-') of commercial-N fertilizer L=Low (151 kg N ha-') as a starter, applied at planting time. The Ag=Agronomic (207 kg N ha-') actual applied liquid effluent rates for low, H = High (263 kg N ha-')

agronomic, and high treatments were 57(2):111-120 estimated by subtracting N credits from the total required N (Table 2).The total amount Field experiments layouts and descriptions. divided into three pie-shaped replications of effluent N for each rate was split into two Experiment I: swine e$luent study. The study site (Figure 1). Each replication contained four to four applications at different growth stages was under continuous corn production prior treatments, including three swine effluent during the growing seasons (Table 3). www.swcs.org to 1995, and was fertilized with commercial- rates plus a control. Swine effluent treatments Nitrogen credits include residual soil-N for N fertilizer only. The site was a circular field and the control were assigned to each replica- the top 1.2 m (3.94 fi) of the soil profile, under sprinkler irrigation (center pivot) tion randomly using a randomized complete N creht fiom mineralized soil organic matter

for the top 30 cm (12 in), and N from irrigation water [2 3 mg L-' (2-3 ppm) of Table 2. Average total available N for each treatment Including N credit. - N03-NI.The average soil-N credit values for N credit sources control plots in Table 2 were considerably Treatments N appllcatlon rate SoICN 0.M.t Water Total N required lower than those for other plot treatments. Effluent kg ha-l These hgh hfferences in soil-N content between the plots can be attributed to the Control 28 31 17 84 previous N management of the field and to Low 52 69 22 151 soil topography and slope (Table 1) differ- Agronomic 105 75 19 207 ences. We assumed that 55% of the NHJ-N High 172 63 20 263 fi-om effluent would be lost due to volatiliza- Commercial-N tion under northeast Colorado conditions Control 28 31 17 84 when calculating each application rate Faskom and Davis 2000). Low 95 31 17 151 In April of 1995, each plot was sampled to Agronomic 151 31 17 207 a 3.0 m (10 ti) depth prior to applying swine H ieh 207 31 17 263 effluent to the field. Three soil cores were +O.M. is soil organic matter. used to make a composite sample represent-

I MIA2002 VOLUME 57 NUMBER 2 I 113 I Table 3. Effluent application rates and actual N applied during the growing season. Growth N application rate? N treatment (kg ha") stage 1995 1996 1997 (257-356 ppm) total N, depending on the kg ha'l (cm) type of swine being raised and the time of Low rate (151) 3-leaf - - - year (Table 4).To achieve the desired applica- 6-leaf - 33 (1.25) 26 (1.25) tion rates through the center-pivot sprinkler 12-leaf 44 (1.25) 29 (1.25) 24 (1.25) system, a computer aided management system (CAMS) was used. The CAMS was Tassel - - - installed on the center pivot, allowing depth Total 44 (1.25) 62 (2.5) 50 (2.5) of application to be as low as 0.25 cm (0.098 Agronomic rate (207) 3-leaf 40 (1.25) 40 (1.25) 34 (1.25) in). The time schedule for applying swine &leaf 47 (1.25) 33 (1.25) 26 (1.25) effluent was pre-determined to be at 3-leaf, 6-leaC 12-leaf, and tassel growth stages (Table 12-leaf - 29 (1.25) 24 (1.25) 3). Plots that did not receive effluent during a Tassel - 24 (1.25) 17 (1.25) given application period received fresh irriga- Total 87 (2.5) 3.26 (5.0) 101(5.0) tion water equal to the amount of effluent High rate (263) 3-leaf 40 (1.25) 61 (2.0) 50 (2.0) that was applied on other plots. All treat- &leaf 47 (1.25) 49 (2.0) 39 (2.0) ments, including the control, received the same total amount of water during the grow- 12-leaf 44 (1.25) 44 (2.0) 36 (2.0) ing season.

Tassel 45 (1.25) 36 (2.0) 26 (2.0) The site was planted to corn on 21 May Copyright © 2002 Soil and Water Conservation Society. All rights reserved. Total 176 (5.0) 190 (8.0) 151(8.0) 1995,14 May 1996, and 15 May 1997.After Journal of Soil and Water Conservation +Numbers in parentheses are effluent application rates In cm. The differences in the corn emergence, aluminum access tubes were amount of N for the same rate at different times of application were due to variability in installed in each plot to monitor soil moisture the NH4-N concentratlon at each time. down to 1.8 m (6 ft) using neutron attenua- Due to supplemental irrigation, ail treatments received the same amount of water during the growing season. tion. Experiment IZ: commercial-N -fertilizer srudy. One of the objectives of this study was to ing depth increments of 0 - 15 cm (0 - 6 in), using standard procedures of the Colorado establish the N response threshold using 15 - 30 cm (6 - 12 in), and thereafier every State University SoilTest Lab. commercial-N fertilizer on the same site. Due 30 cm down to 3.0 m (10 a). Each depth Effluent hmthe second cell of a two- to the constraints of the center-pivot system, increment was analyzed for NH4-N and stage lagoon was analyzed prior to each appli- and to minimize soil variability, commercial-

N03-N. Electric conductivity (EC),pH, soil cation to determine the nutrient content N fertilizer plots were located inside the con- 57(2):111-120 organic matter, and AB-DTPA extractable-P (Table 4). Swine effluent hmthe two-stage trol plots of the swine effluent experiment, were estimated for the top 90 cm (36 in) only, lagoon contained hm257 - 356 mg L-' and they were kept in the same locations every year (Figure l).The size of each of the replications of the pie-shaped control plots Table 4. Analysis of effluent from a twu-stage lagoon from finishing unlts in 1995 1996

- www.swcs.org and breeding units in 1997. was approximately 1.2 ha (3.0 ac). Four application rates of commercial-N were randomly Constituents Units 1995 1996 1997 located inside the pie-shaped wedges of the NH4-N 356.0 314.7 257.0 field that did not receive swine effluent.The

Nos-N 1.9 0.3 0.1 size of the commercial-N plots was 6.6 x 16.6 Total N 371.9 321.6 263.1 m (22 x 54 fi) (Figure 1). Total P 100.0 64.9 45.9 The commercial-N fertilizer requirements K 270.0 247.3 241.0 for low, agronomic, and high rates were estimated similar to those for the swine effluent 7.1 7.7 7.6 PH rates (Table 2). Nitrogen application rates 5.4 3.9 4.8 Electric conductivity were determined by including N credlts from Dry matter 0.2 0.1 0.1 residual soil-N for the top 1.2 m (4 ft), min- Total organic carbon 0.05 0.06 0.04 eralized soil organic matter for the top 30 cm S 1.0 0.1 0.1 (12 in), and N credit from irrigation water Ca 130.0 91.6 72.5 [2 - 3 mg L-' (2-3 ppm) of N03-N], as in the Mg 60.0 50.8 42.5 swine effluent experiment. A starter fertilizer Na 70.0 73.5 52.0 containing 17 kg P ha-' (15 lb P ac-') was applied at planting.The N source used for Fe 1.0 2.0 0.9 the commercial-N fertllizer treatments was Mn 1.0 0.5 0.4 NH4-NO3 (34-0-0). The fertilizer was hand- cu 1.0 0.3 0.2 applied with a hand spreader at the same time Zn 1.0 0.3 0.1 (3-leaf, 6-leaf, 12-leaf, and tassel growth

1 114 1 JOURNALOF SOIL AND WATER CONSERVATION MIA 2002 I Table 5. Grain yield of irrigated corn under swine effluent treatments. Experiment I: Yield effluent rates Total N available+ 1995 1996 1997 stages) as when swine effluent was applied. I Irrigation of commercial-N plots followed kg ha-l the same irrigation-scheduling program for Control 84 2195 2634 2759 the entire study during the growing season. Low 151 5707 4264 8215 Measurements and data collection. Similar Agronomic 207 7212 7400 11288 data were collected for both swine effluent High 263 7713 8529 12229 and commercial-N fertilizer experiments. LSD(0.05) 3382 2695 2875 On a weekly basis, data were collected for P>F(0.051 0.026 0.006 0.058 soil moisture to 1.8 m (6.0 fi) depth using +Total amount N available for effluent includes credits from soiCN, organic matter, starter- neutron attenuation, leaf area index, and total N, and Irrigation water-N. Differences between treatments greater than Least significant plant dry matter accumulation. In addition, difference value are significant at = 0.05. soil samples were taken prior to planting and afier harvest each year and were analyzed for NH4-N, NO3-N, pH, EC, organic matter, and AB-DTPA extractable-P using Colorado Figure 2 State University Soil Test Lab standard proce- Seasonal accumulated growing degree days (GDD) for 1995,1996,and 1997.GDD was calculated dures. Plant and grain samples for total N and for corn by subtracting base temperature of io C from average daily max. and min. temperatures. P plant removal were collected at harvest fiom an area of 2 rows x 6.6 m (21.66 6)long Copyright © 2002 Soil and Water Conservation Society. All rights reserved. of both effluent and commercial-N treat- 1,946 Journal of Soil and Water Conservation ments.The final grain yield for both experiments (effluent and commercial-N) was determined by hand picking an area of 4 rows x 13.3 m (44fi) fi-om each application rate for 8 both swine effluent and commercial-N ferd- 1,668 izer experiments. Soil and plant analyses for both experiments were performed at the Colorado State University Soil Test Lab. Weather data was collected throughout the season for irrigation scheduling. 1990

Data analysis. Data generated by this study 57(2):111-120 and the results presented were statistically analyzed using the StatisticalAnalysis System (SAS Inst. 1988). The general linear model (GLM) procedure was used to perform the www.swcs.org analyses of variance. Swine effluent and a commercial-N fertilizer experiments were 8 analyzed separately, and were quahtatively compared to observe differences in plant 834

1 performance and soil nitrogen accumulation. U Results And Discussion Experiment I: swine fluent study. Swine efluent nutrient analysis. Ammonium-N 556 represented on average approximately 90% of the total N for the two-stage lagoon, where total dry matter was only 0.1 - 0.2% by volume. Emuent analysis from a two-stage lagoon containing waste fiom swine finishing 278 units in 1995 and 1996 and swine breeding units in 1997 revealed large hfferences, especially for NH4-N and P concentrations (Table 4). Another hfference observed was that the concentration of micronutrients fiom the 0 finishing units was greater than the concen- Growing Season tration of the breedmg units' effluent. The analysis of swine effluent at hfferent

I MIA2002 VOLUME 57 NUMBER 2 I 115 I Table 6. Total plant removal and recovery percentage of N and P for swine effluent treatments. 1995 1996 1997 Totalt Totalt Totalt Experiment I: amount Total amount Total amount Total swine effluent available removal Recovery' available removal Recovery' available removal Recovery' -kg ha-l- % kg ha-l- % kg ha-l- %

Nitrogen: Control 84 37 - 87 43 - 81 61 - Low 151 85 32 151 76 22 151 132 47 Agronomic 207 124 42 207 140 47 207 172 54 High 263 153 44 263 172 49 263 239 68 LSD(O.05) 64 56 116 P>F(0.05) 0.002 0.004 0.046

Phosphorous: Control 30 15 - 29 16 - 48 22 - Low 39 27 31 51 22 12 53 38 30 Agronomic 51 30 29 88 37 24 81 39 21 High 65 30 23 81 36 25 103 48 25 LSD(0.05) 13 9 20 P>N0.05) 0.093 0.002 0.083 ?Total available N includes applied effluent, irrigation water-N, soil organlc matter, and residual solCN. Available P sources are from effluent Copyright © 2002 Soil and Water Conservation Society. All rights reserved. plus AB-DTPA extractable soiEP in the top 15 cm. Journal of Soil and Water Conservation 'Recovery % = (N or P treatment total removakontroi total removal)/total amount available x 100. Differences between treatments greater than Least significant dlfference value are significant at = 0.05.

times during the crop growing season showed atures and three to four events of high rainfall concentration increased significantly under considerable temporal variabdity in N con- [>4.0 cm (1.6 in)] occurred between early the high effluent application rate over three centration, generally decreasing late in the May and mid-August in 1995 and 1996, years, while no significant change was growing season (Table 3). This decrease was while only one high rainfall event occurred observed under the lower rates of effluent due to the high use of fi-esh water in the in early August of 1997. These con&tions application (Figure 3). This accumulation flushing system late in the growing season, as contributed to relatively low yield perform- indicates that applied P was in excess of crop compared with early in the season. Also, ance in 1995 and 1996. needs in the high rate plots.

during the first year of the study, the rate of Total plant N and P removal and recovery rate. Impact of swine egluent on selected soil proper- 57(2):111-120 application was designed to be 1.25,2.5, and The amounts of available N and extractable- ties. Swine effluent can alter soil properties, 5.0 cm (0.5, 0.75, and 2.0 in) of effluent P, plant N and P removal, and N and P recov- such as extractable soil-P, pH, and EC. Figure (Table 3) due to high initial residual soil-N. In ery rate for effluent treatments are summa- 3 summarizes the changes measured in soil 1996 and 1997 the rates were increased to rized in Table 6.The total plant N and P properties afier crop harvest over three years meet the designed N rates to 2.5,5.0, and 8.0 removal (grain, leaf, and stalks) include of swine effluent application at different rates. www.swcs.org cm (1.0,2.0,and 3.0 in) of effluent based on removal &om both applied and soil-available The extractable-P increased in the top 15 cni the residual soil-N and effluent nitrogen con- N and I? As the application rate increased, (6.0 in) as the swine effluent application rate tent. No foliar burn was observed at any rate the total plant removal of N and P and the increased. However, during the first year of

or application time over the three years of this percent of N recovery generally increased swine effluent application (1995) no signifi- study. (Table 6).The increase of P removal with the cant increase in extractable-P value at the top Corn yield response to swine efluent. Yield increase of the swine effluent application rate 15 cm (6.0 in) for all rates was observed, as performance of irrigated corn tended to may have contributed to better plant growth, compared to the initial level prior to swine increase with the increase of effluent applica- even though soil tests indicate that P was not effluent application. The increase in P value tion rates across all three years. However, the needed. Also, the high K content of swine was significant afier three consecutive years of highest rate of effluent did not sipficantly effluent (Table 4) along with other micro- swine effluent application (1997), especidly increase grain yield over the recommended and macronutrients may have contributed to under the high application rate. agronomic rate (Table 5). Poor weather con- hgher biomass and grain yields and greater N Soil pH did not change in the top 15 cni ditions, such as cool temperatures and hail and P removal, in spite of soil test recom- (6.0 in) during the three years of effluent damage, affected plant growth and grain mendation to the contrary. The change in application. On the other hand, soil EC yields in 1995 and 1996, as compared with recovery rate was not significantly dfferent increased slightly in the top 15 cm (6.0 in) the 1997 growing season, where growing between agronomic and high application afier the second and third year of swine efflu- conditions were better for corn (Figure 2). rates.The amount of P that was received by ent application.The average EC value was 0.3 The total amount of rainfall received between the plants under the swine effluent treatments dS m-l (0.3 mmho cm-'), whch is not consid- May and August during 1995,1996, and 1997 was highly variable during different applica- ered high enough to impact corn grain yield. was 46, 48, and 37 cm (18.1, 18.9, and 14.6 tion times and rates. Soil analysis for the top Soil-N profile aJer three years (rf swine efluent in), respectively. Cooler than normal temper- 15 cm (6.0 in) showed that extractable-P field application. Residual soil-N distribution

1 116 I JOURNALOF SOIL AND WATER CONSERVATION MIA 2002 I Figure 3 Soil EC, pH, and AB-DTPA extractable Pin the top 15 cm of soil under different swine effluent application rates after harvest in 1995-1997. Error bars represent standard error of mean. potential for N leaching below the root zone, where N is beyond the typical plant rooting depth. 20 Experiment If: commerciul-N fertilizer Prior to Effluent Appl. in 1995 study. Corn yield rerponre to N rates. Corn grain 1 Control ylelds increased with the addition of com- l5 rnLow mercial-N fertilizer up to the agronomic rate. - 69 Medium The increase in N rate by 56 kg N ha-' (50 lb N ac-') above the agronomic rate did not 10 produce a significant increase in grain yields. Yield differences between the years was primarily a fbnction of weather conditions 5 (Table 7). In 1995 and 1996 grain yields were lower than those obtained in 1997, when the growing conditions were better for corn 0 production. The trend of yield response to 10 I 1 commercial-N fertilizer rates was similar to those of swine effluent.

Total plant N and P removal and recovery rate. Copyright © 2002 Soil and Water Conservation Society. All rights reserved. Total plant N and P removal for commercial- Journal of Soil and Water Conservation N fertihzer treatments showed no significant increase with the increase in the N fertilizer application rate (Table 8).The N use ratio of total N removal to yield was 1.2 for the high application rate and less than 1.0 for the control, low, and agronomic rates. These results indicate that yield response to the increase in N application above the agronomic rate under irrigated sandy soil conditions was not 0.5 I 1 significant. The resulting accumulation of N

in the soil profile increases the potential for N 57(2):111-120 leaching. Soil-N proJle ajer three years of commercial- N 0.3 applications. The commercial-N fertilizer treatments showed a significant N buildup below the root zone, 1.5 - 3.0 m www.swcs.org 0.2 (5 - 10 fi), where an average of approximately 50 kg N ha-' (45 lb N ac-') accumulated at 0.1 depths 1.5 - 3.0 m (5 - 10 fi) (Figure 5B).This

buildup was also reflected in the residual soil- n N dlstribution through the soil profile for all 1995 1996 1997 rates, inhcating a potential N leaching below the root zone (Figure 5A). Lower grain yield Year resulted in low plant N removal and high N leaching below the root zone. Kesidual soil- in the soil profile after three years of swine Ths was evident where continued increase in N decreased significantly in the soil profile effluent application at dfferent rates is sum- plant-N removal under swine effluent treat- between 1995 and 1997. marized in Figure 4A. In general, different ments was observed hm1995 to 1997. effluent application rates show no sipficant Soil-N buildup was greater for the high Summary And Conclusion differences in residual soil-N at all depths.The rate of effluent application in the top of Several important aspects of utilizing swine residual soil-N afier harvest was significantly the soil profile (Figure 4B).The cumulative effluent became apparent from this study.The lower for all swine effluent treatments, as residual soil-N in the 1.5 - 3.0 m (5 - 10 fi) N in swine effluent fkom two-stage lagoons compared to the initial residual soil-N prior zone for all swine effluent application rates was almost entirely (90%) in the ammoniacal to the first effluent application in 1995.The was greater than in the root zone [0 - 1.2 m form (NH4-N). This is important because decrease in residual soil-N content afier three (0 - 4 fi)].An average of approximately 20 kg ammonium-N is immediately available to the consecutive years of swine effluent applica- N ha-' (18 lb N ac-') accumulated below the crop, unlike organic forms of N found in tion was due to soil-N removal by the crop. root zone. Ths accumulation indlcates a many other waste products. Therefore, man-

I MIA2002 VOLUME57 NUMBER2 I 117 I Figure 4 (A) Residual soil-N (N03-N + NH,-N) distribution in the profiie and (B) Total amount of residual soil-N in and below the root zone after 3 years of swine effluent application on irrigated cornfield. Error bars represent standard error of mean.

2 4 6 8 lo O0 224

1% 0.5

168 Experiment I: Swine Effluent 1.0

Y-140 n T El 1.5 k W a .se 8112 a 5 2.0 rl

m 884 Journal of Soil and Water Conservation

2.5 56

3.0 28 --*- Agro.after3-yr Experiment I: Effluent -.-* - High,after%yr 0 0.12 1.5-3.0 Root Zone Below Root Zone Soil Depth (m) 57(2):111-120

Table 7. Grain yield of Irrigated corn under commercial-N fertilizer treatments. rate increased.The increase in total N and P removal with the increase of the effluent www.swcs.org Yield application rate contributed to the increase of Experiment 11: commercial-N rates Total N availablet 1995 1996 1997 grain yrelds. This increase in N removal resulted in lower residual soil-N after three kg ha-1

years of effluent application and reduced the Control 84 2195 2634 2759 potential of N movement below the root Low 151 4954 5330 4766 zone. The N and P concentrations in the Agro nom ic 207 5895 7274 8215 swine effluent were highly variable over time as a fhnction of the waste flushing system, High 263 6835 7024 8654 time of the year, and swine type. Therefore, LSD(0.05) 1483 1599 2739 the rate of N and P loading is highly variable P>F(0.05) 0.001 0.002 0.005 in a commercial-scale system and must be +Total amount N available for commercial fertilizer includes credits from soil-N, organic considered in nutrient management plans. matter, starter-N, and irrigation water*. Differences between treatments greater than Significant changes in EC and extractable-P Least significant difference value are significant at = 0.05. were observed in the soil after three years of swine effluent application. Applications of aging swine effluent-N becomes very sdar Increased N application by 56 kg N ha-' swine effluent at different times during the to managing commercial-N fertilizer under (50 lb N ac-') above the recommended agro- growing season appears to be an effective way irrigated conditions. The main difficulty in nomic rate did not produce a significant yield of managing swine effluent under irrigated constructing a nutrient management plan is increase under swine effluent or the commer- conditions when it is applied through a in determining the appropriate rate of cial-N fertdizer.Total plant N and P removal sprinkler system, as no plant leaf burn was ammonia volatilization. increased as the swine effluent application observed, even at high application rates.

L18 I JOURNAL OF SOIL AND WATER CONSERVATION MIA 2002 I Table 8. Total plant removal and recovery percentage of N and P from commercial-N fertilizer treatments. 1995 1996 1997 Experiment I: Totart Totalt Total7 commercial-N amount Total amount Total amount Total fertilizer available removal Recovery' available removal Recoveryt available removal Recovery'

Nitrogen : Control 84 37 - 87 43 - 81 61 - Low 151 57 13 151 96 35 151 69 5 Agronomic 207 63 13 207 133 43 207 162 45 High 263 113 29 263 157 43 263 209 56 LSD(0.05) 31 19 48 P>F( 0.05) 0.005 0.001 0.001

Phosphorous: Control 26 15 - 27 16 - 27 13 - Low 26 20 19 27 23 26 27 15 7 Agronomic 26 20 19 27 23 26 27 22 33 High 26 23 31 27 21 19 27 24 41 LSD(0.05) 7 6 6 P>F(0.05) 0.151 0.081 0.135 ?Total available N includes applied fertilizer, irrigation water-N, soil organlc matter, and residual s01l-N. Available P sources are commercial fertilizer P that was applied as starter 17 kg ha, plus AEDTPA extractable soil-P In the top 15 em. Copyright © 2002 Soil and Water Conservation Society. All rights reserved.

'Recovery % = (N or P treatment total removal - control total removal)/total amount available x 100. Journal of Soil and Water Conservation Differences between treatments greater than Least signlflcant difference value are slgnlficant at = 0.05.

Figure 5 (A) Residual soil-N (NO,-N + NH,-N) distribution in the profile and (6) Total amount of residual soil-N in and below the root zone after 3 years of commercial-N fertilizer application on irrigated cornfield. Error bars represent standard error of mean.

Soil-N (mg kg-') 2 4 6 8 0' lo 224 Prior to Effluent Appl. m 95 57(2):111-120 Ll Control, after 3-yr tah,acter3yr (B) 0.5 1% rsAgro,ater3yr GI High, &er 3-yr

168 Experiment 11: Commercial-N Fertilizer www.swcs.org 1.o

F 140 n -I

1.5 W8 A .s 3 g 112 m 5 2.0 -a v) s84

2.5 56

3 .O 28

wriment 11: Commercial-N 0 0-12 1.5-3.0 Root Zone Below Root Zone Soil Depth (m)

1 MIA2002 VOLUME57 NUMBER2 I 119 I Sprinkler-applied swine effluent at the rec- Harvey, R.W., J.P. Mueller, J.A. Barker, M.H. Poore, and J.P. SAS Institute. 1988. SAS Users Guide: Statistics (Version ommended agronomic rate resulted in maxi- Zublena. 1996. Forage Characteristics, Steer 6.03). Cary, NC: SAS Institute. Inc.. Performance, and Water Quality hmBermudagrass Sutton,A.L.,D.W. Nelson,J.I). Hoff,andV.B. Mayrose. 1983. mum yields and minimal N accumulation Pastures Fertilized with Two Levels of Nitrogen from Effects of Injection and Surface Applications of Liquid below the crop root zone. Swine Lagoon Effluent. Journal of Animal Science Swine Manure on Corn Yield and Soil Composition. 74~457-464. Journal of Environriiental Quality. 1 1 468-472. Acknowledgements Hegg, R.O., A.T. Shearn, D.L. Handlin, and L.W. Grimes. Sweeten, J.M., M.L. Wolfe, E.S. Chasteen, M.A. Sanderson, 1984. Irrigation of Swine Lagoon Effluent onto Pine B.A. Auvermann, and G.D. Alston. 1994. Dairy Lagoon The authors express sincere appreciation to the and Hardwood Forests. Transaction of the American Effluent Irrigation: Effects on Runoff Quality, Soil funding partners, the advisory committee, Ken Society ofAgricultural Engineers. 83:1411-1418. Chemistry, and Forage Yield. In: Proceedings of the Goeglein, farmer, and Bonnie Fisher for their help King, L.D., J.C. Burns, and PW. Westerman. 1990. Long- Great Plains Animal Waste Conference on Confined during the course of the study. Term Swine Lagoon Effluent Applications on Coastal Animal Production and Water Quality, October 19-2 1, Bermudagrass: 11. Effect on Nutrient Accumulation in 1994 at Denver, CO. Great Plains Agricultural Concil Soil. Journal of Environmental Quality 19:756-760. Publication No. Great Plains Agricultural Council References Cited King, L.D., P.W. Westerman, G.A. Cummings, M.R. Publication No. 15199-104. Burns, J.C., L.D. King, and P.W. Westerman. 1990. Long- Overcash, and J.C. Burns. 1985. Swine Lagoon Effluent Waskom, R.M. and J.G. Davis. 2000. Best management prsc- Term Swine Lagoon Effluent Applications on Coastal Applied to Coastal Bermudagrass: 11. Effects on Soil. tices for manure utilization. Colorado State University Bermudagrass: I. Yield, Quality, and Element Removal. Journal of Environmental Quality 14:14-21. Cooperative Extension Bull. No. 568A. Journal of Environmental Quality 19:749-756. King, L.D. 1981. Effect of Swine Manure Lagoon Sludge and Westerman, PW., M.R. Overcash, R.O. Evans, L.1). King, Burns, J.C., PW. Westerman, L.D. King, G.A. Cummings, Municipal Sewage Sludge on Growth, Nitrogen J.C. Burns, and G.A. Cummings. 1985. Swine Lagoon M.R. Overcash, and L. Goode. 1985. Swine lagoon Recovery, and Heavy Metal Content of Fescuegrass. Effluent Applied to Coastal Bermudagrass: 111. Irrigation Effluent Applied to Coastal Bermudagrass: I. Forage Journal of Environmental Quality 10:465-472. and Rainfall Runoff.Journal of Environmental Quality Yield, Quality, and Element Removal. Journal of Liu, E, C.C. Mitchell, J.W. Odom, D.T. Hill, and E.W. 14:22-25. Environmental Quality 14:9-14. Rochester, 1997. Swine Lagoon Effluent Disposal by Evans, R.O., PW. Westerman, and M.R. Overcash. 1984. Overland Flow: Effects on Forage Production and Subsurface Drainage Water Quality from Land Uptake of Nitrogen and Phosphorus. Agronomy Application of Swine Effluent. Transaction of the Journal. 89:900-904. Copyright © 2002 Soil and Water Conservation Society. All rights reserved. American Society of Agricultural Engineers 81:473- Mortvedt, J.J., D.G. Westfall, and R.L. Croissant. 1995. Journal of Soil and Water Conservation 480. Fertilizer suggestions for corn. Colorado State Gangbazo, G., A.R. Pesant, G.M. Barnett, J.P Charuest, and University, Fact Sheet No. 0.538. D. Cluis. 1995. Water Contamination by Ammonium Newton, G.L., R.K. Hubbard, J.C. Johnson, J.G. Davis, G. Nitrogen Following the Spreading of Hog Manure and Vellidis, R. Lorance, A.W. Johnson, R.G. Williams, and Mineral Fertilizers. Journal of Environmental Quality C. Dove. 1994. Utilization and environmental conse- 241420-425. quences of land application of liquid manure in south- Gast, R.G., W.W. Nelson, and G.W. Randall. 1978. Nitrate eastern U.S. coastal plains. In: Proceedings of the Great Accumulation in Soils and Loss in Tile Drainage Plains Animal Waste Conference on Confined Animal Following Nitrogen Applications to Continuous Corn. Production and Water Quality, October 19-21,1994 at Journal of Environmental Quality 7:258-261. Denver, CO. Great Plains Agricultural Concil Publication No. 15 1:66-73. 57(2):111-120 www.swcs.org

lizo I JOURNAL OF SOILAND WATER CONSERVATION MIA 2002 I