|
|
||||||||
a Dep. of Soil, Water, and Climate, Univ. of Minnesota, St. Paul, MN 55108 USA
b Dep. of Agric. Engineering, North Dakota State Univ., Fargo, ND 58105 USA
waddeje{at}tetratech-ffx.com
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: AW, available soil water BD, buried drip irrigation BDF, buried drip irrigation with fertilizer injected through tape ET, evapotranspiration SCU, sulfur-coated urea SD, surface drip irrigation
| INTRODUCTION |
|---|
|
|
|---|
is moving from heavy clay soils of the Red River Valley (63% in 1985 and 43% in 1996) to glacial sandy outwash regions. The reason for this change is an ample irrigation water supply in sandy outwash regions that results in higher and more consistent potato yield. With this change, there is concern about the effects of current irrigation and N management practices on ground water quality. Myette (1984) showed that N concentrations in ground water have been steadily increasing under intensively farmed sandy outwash soils of central Minnesota.
Maximum productivity for potato occurs when the soil is kept consistently moist (Ojala et al., 1990; Saffigna et al., 1977) and with N available during periods of high demand (Sanderson and White, 1987; Stark et al., 1993). Sprinkler irrigation is the most common form of irrigation for potato production across the USA. However, alternative irrigation methods, such as drip irrigation, can save water and also decrease nutrient leaching (Phene et al., 1994; Phene and Sanders, 1976). Drip irrigation has not been extensively tested in humid regions, because of abundant precipitation and the high cost of drip line installation.
In central Minnesota, typical irrigation practice consists of an overhead sprinkler system with water applications when 70% or more of the available water remains in soil (Wright and Bergsrud, 1991). Studies have shown that irrigation applications that allow soil to become drier during noncritical growth periods can minimize deep percolation and N leaching without significantly decreasing potato yield (Roberts et al., 1982; Saffigna et al., 1977). Stegman (1983) showed no reduction in corn (Zea mays L.) yield when irrigation was applied at between 70 to 40% of the available water capacity. An unanswered question is: what is the impact of a 40% available water deficit on potato production and tuber quality in outwash soils of central Minnesota?
Common N fertilizers for potato production include either NH+4 or NO-3 derivatives. A management option to increase N use efficiency is to split N applications over the growing season. Recently, split applications have been common in potato production (Westermann et al., 1988; Sexton, 1993; Errebhi et al., 1998). Westermann et al. (1988) showed that post-hilling N application increased yield by potentially limiting N leaching beyond the root zone.
Nitrogen is often overapplied in potato production, to ensure against loss of yield and tuber quality. Liegel and Walsh (1976) showed frequent potato yield depressions with fast-release N fertilizers, mainly due to N leaching. Another way to combat N leaching is through the use of slow-release N fertilizers such as S-coated urea. Studies with S-coated urea in potato production have shown somewhat lower yields if N release is not synchronized with plant N uptake (Liegel and Walsh, 1976; Lorenz et al., 1974).
Another slow-release N source in central Minnesota is turkey manure. Minnesota is the second leading producer of turkey (Meleagris gallopavo) in the USA (Hunst, 1998), and a majority of the turkey operations are located on sandy outwash soils of central Minnesota. Therefore, the use of turkey manure in potato production provides opportunities for a beneficial means of its disposal. Currently, little information exists on the use of slow-release N fertilizers under humid conditions of central Minnesota.
Our objective was to evaluate the use of alternative management practices such as drip irrigation, various fertilization schemes, and slow-release inorganic and organic fertilizer on potato production and N leaching from Minnesota outwash soils. In this paper, we report the effect of these practices, compared with conventional practices, on potato yield, tuber quality, and N uptake. We intend to address the effect of these practices on N leaching in a subsequent paper.
| Materials and methods |
|---|
|
|
|---|
The previous crops were barley (Hordeum vulgare L.) in 1994 and corn in 1995. Russet Burbank potato was planted in 91-cm-wide rows with 25-cm spacing within rows. Each plot consisted of four rows (912 m long) of potato, with vine and tuber sampling restricted to the two inner rows. The experimental setup was a version of a split-plot design with whole-plot randomization restricted due to constraints associated with drip tape installation. All plots were disked twice prior to planting on 25 April each year. Seed tuber pieces were planted in both years on 26 April with a six-row planter that also formed a small hill.
A total of 11 treatments were evaluated (Table 1) . The treatments were three N sources (urea, turkey manure, and S-coated urea) and two irrigation techniques (sprinkler and drip). Two levels of deficit irrigation (wet and dry) were tested in the sprinkler-irrigated plots. The three drip irrigation treatments were surface, subsurface, and subsurface with fertigation. Crop factors quantified were potato yield, tuber size distribution, specific gravity, hollow heart, scab [caused by Streptomyces scabies (Thaxter) Waksman & Henrici], and total N uptake. All treatments were replicated four times.
|
|
![]() | (1) |
Drip irrigation treatments included surface drip (SD), buried drip (BD), and buried drip with fertigation (BDF). Buried drip tape was installed at the 25-cm depth just above the argillic horizon with a knife followed by a press wheel to close the slot. After insertion, the drip tape was inflated with water to expand the tape and check for leaks. For the surface drip treatment, the drip tape was laid on top of the hill after potato planting. According to the manufacturer (Chapin Watermatics,1 Watertown, NY), this would result in a water flow rate of 11.2 L h-1 m-1 at an operating pressure of 70 kPa. Water through drip irrigation (10 mm per event in 1994 and 1 mm per event in 1995) was applied when soil matric potential dropped below -30 kPa at the 25-cm depth next to the buried tape in the BD and BDF treatments, and at the 15-cm depth below the drip tape in the SD treatment. Each of the three drip irrigation treatments were irrigated independently.
Soil matric potential was measured using tensiometers that were fitted with pressure transducers (Soil Moisture Equipment Corp., Santa Barbara, CA). A datalogger was used to monitor the hourly values of soil matric potential in various treatments. In addition to the tensiometers, digital flow meters, solenoids to control the flow, and a chemical injection pump were also connected to the datalogger (Model 21X, Campbell Scientific, Logan, UT). This arrangement allowed automated data collection on irrigation and fertigation. A direct phone line was also installed at the site to remotely monitor and control irrigation.
Fertilizer Treatments
The fertilizers used in this study can be categorized into three groups: urea, fast release; turkey manure, fast and slow release; and S-coated urea, slow release. Except for the turkey manure treatment (250 and 217 kg N ha-1 in 1994 and 1995, respectively) and the control, N was applied to all treatments at a rate of 224 kg ha-1 (Table 1). There was no application of N fertilizer in the control treatment. Diammonium phosphate (180 kg P2O5 ha-1) and KCl (224 kg K2O ha-1) were applied at planting to all treatments except the turkey manure treatment. Starter N applications were made at planting (Table 1). For the BDF treatment, no starter N fertilizer was applied in 1994, but 45 kg N ha-1 was applied at planting in 1995. Cultural practices related to pesticide application followed guidelines of Hutchison (1994).
In both years, manure was surface applied with a manure spreader (beater disengaged) and then incorporated by disking. In 1994, manure was applied on 25 April at a rate of 31.4 Mg ha-1 (wet wt.). Moisture content (by wt.) of the manure was 36.3%. Manure was spread on each plot by making two side-by-side passes. Rakes were used to evenly spread the manure across each plot. Mineral and organic N in manure (wet wt. basis) were 2.3 and 18.7 kg Mg-1, respectively. Using guidelines for N availability from organic sources of Sutton et al. (1985), estimated available N from this manure application was 250 kg ha-1. In 1995, manure was applied on 25 April at a rate of 20.8 Mg ha-1 (wet wt.). The manure moisture content (by wt.) was 45.3% and the mineral and the organic N (wet wt. basis) were 6.7 kg Mg-1 and 12.2 kg Mg-1, respectively. Estimated available N from this application was 217 kg ha-1.
Two urea N fertilization treatments were compared, termed conventional and alternative, both at a seasonal application of 224 kg N ha-1. The conventional N treatment had urea application of 45 kg N ha-1 at planting followed by 90 kg N ha-1 at emergence (3 June 1994; 30 May 1995) and 90 kg N ha-1 at hilling (14 June 1994; 15 June 1995). The alternative N treatment had slightly lower urea applications (73 kg N ha-1 each) at emergence and at hilling compared with the conventional N treatment. The dates of urea application were the same for both alternative and conventional N treatments. For the alternative N treatment, the post-hilling N applications (26 June and 12 July 1994; 18 July and 31 July 1995) were applied based on an N sufficiency index as indicated by the sap NO3 levels (Errebhi et al., 1998). In both conventional and alternative N treatments, urea was broadcast applied at the soil surface and then incorporated. The S-coated urea was banded 10 to 15 cm on either side of the seed piece at planting and broadcast at emergence.
The SD and BD followed the alternative N fertilization scheme (i.e., fertilization at planting, emergence, hilling, and two post-hilling applications). For the BDF treatment, a 5 M urea solution was applied according to plant growth stage and N uptake patterns: a lag phase beginning at emergence (3 wk) in which less soil N is assimilated (at 8 kg ha-1 every third day); a rapid growth period (3 wk) requiring high soil N for optimal yield (at 16 kg ha-1 every third day); and a final lag phase (3 wk) in which senescence controls plant N uptake (at 8 kg ha-1 every third day). These N amounts were applied through irrigation regardless of the soil moisture status.
Yield and Tuber Quality
One factor considered for tuber quality was tuber size. The following size categories were considered marketable: >400 g (Jumbo), 200400 g (No. 1), and 85200 g (No. 2). Culls (<85 g) and misshapen tubers (knobs) represent the difference between total and marketable yields. After harvest, tubers from each plot were graded into various size classes and weighed. From these, 1 Jumbo, 12 No. 1, and 12 No. 2 tubers were then randomly selected for determination of other quality attributes such as specific gravity, hollow heart, and scab.
Laboratory Measurements
Before vine kill, samples of vines were taken to determine total aboveground biomass and vine N uptake. However, in 1995, many of the vines had already senesced. Of the tubers selected for yield and quality, six tubers were used to determine tuber N uptake. Tubers and vines were dried and ground to pass through a 2-mm sieve. Plant N was determined by digesting in 3 mL H2SO4, 3.5 g K2SO4, and 10 mg CuSO4 at 437°C. Ammonia in the digested sample was then analyzed conductimetrically (Carlson, 1978). Tissue N concentration was multiplied by dry matter yield to calculate tuber and vine N uptake.
Subsamples of manure were taken to determine its moisture and N content. In preparation for inorganic and total N determination, manure subsamples were mixed with distilled water at a 1:3 (by wt.) ratio, and then homogenized in a blender for 5 min. Extraction and digestion procedures for N determination in the manure samples were similar to those for soil and tissue samples. The organic N fraction was determined from the difference between total N and inorganic N. Ammonium in soil water samples was analyzed conductimetrically. This was followed by reduction of NO3 in solution to NH4, which was again measured conductimetrically to give total N in solution (Carlson, 1978). Nitrate concentration was calculated by difference in total N and NH4N (Carlson, 1986).
Statistical Analysis
Statistical analysis was conducted using SYSTAT (Wilkinson, 1996). The incomplete nature of N treatments within irrigation schemes prevented the use of a split-plot technique for analysis of variance. Instead, the treatments were run as a single-factor ANOVA. Where appropriate, single degree of freedom orthogonal polynomial contrasts were used to determine whether differences existed between certain comparisons. The probability level for determination of significance was 0.05.
| Results and discussion |
|---|
|
|
|---|
Irrigation
Amount and distribution of supplemental irrigation varied with the method of application. For the wet and the dry sprinkler irrigation treatments, respectively, approximately 20 and 40 mm of water was applied per event to bring the soil to field capacity. For the drip irrigation, the volume of water was converted to a depth using the entire plot area, even though only about half to a third of the area received irrigation.
Cumulative irrigation and the sum of rain and irrigation are shown in Table 2 . In 1994, the wet sprinkler treatment received 230 mm of supplemental water in 11 irrigations during the growing season. In comparison, the dry sprinkler treatment received 220 mm of irrigation in five waterings during the growing season. Surface and buried drip irrigation applications were similar, each receiving 110 mm of water during the growing season. The surface drip treatment received this water in 27 applications, compared with 26 for the buried drip treatment. The least amount of water was applied to the buried drip with fertigation treatment: 95 mm of water in 25 irrigation events.
|
Tuber Quality and Yield
Yield and Quality in 1994
Significant differences were detected in total, marketable, and No. 1 grade (200 to 400 g) tuber yields (Table 3)
. Yields of No. 1 tubers from treatments amended with turkey manure were significantly greater than the control. The trends in marketable tuber yields were the same as the trends in No. 1 tuber yields, with the manure-amended treatments yielding significantly more tubers than the control. Statistically higher tuber yields for the manure treatment were possibly due to extra N and other nutrients present in manure but not in urea or S-coated urea. There were no other significant differences in No. 1 or marketable tuber yields. The total tuber yield was the highest for the 70% AW manure treatment and was significantly greater than the S-coated urea (SCU), BDF, and control treatments. Total tuber yields for these low-yielding treatments (SCU and BDF) were not statistically different than the control. In all categories, it appears that either manure or urea (both the conventional and alternative application schedules) produced the greatest tuber yields.
|
A significant difference between treatments occurred for tuber specific gravity and hollow heart (Table 3). The specific gravity of tubers in treatments amended with turkey manure was less than with other N treatments; however, tubers in turkey manure treatments also showed less incidence of hollow heart. No differences were observed in the infestation of scab on tubers for various N and irrigation treatments.
Yield and Quality in 1995
Except for the control and BDF treatments, total tuber yield was not affected by various irrigation and N treatments (Table 4) . The SD treatment had the highest total tuber yield, although it was not significantly different from the total tuber yield of the BD treatment. The yield of Grade 1 tubers in the SD was higher than in the BD treatment.
|
The specific gravity of tubers from the dry sprinkler treatments was significantly lower than the specific gravity of tubers from the wet sprinkler treatment. The drip treatments (SD and BD) had tubers with the highest specific gravities. The reasons for this variation in specific gravities among irrigation treatments are not clear. Specific gravity is an important quality parameter associated with the processing of tubers and needs further study with respect to the impact of drip irrigation.
Averaged over all treatments, total tuber yield in 1994 (49.6 Mg ha-1) exceeded the tuber yields in 1995 (32.0 Mg ha-1). A major difference in tuber yield between the two years was in the proportion of No. 1 grade (200400 g). In 1994, 45% of the yield consisted of tubers weighing 200 to 400 g and 33% in the 85- to 200-g category (No. 2). In 1995, by contrast, 26% of the yield consisted of tubers weighing 200 to 400 g and 45% were in the 85- to 200-g category. These differences were due to more favorable weather for tuber bulking in 1994 than in 1995. For example, the early part of the 1995 growing season was cool, cloudy, and wet and resulted in fewer growing degree days than in 1994 (Fig. 2) . Subsequently in the growing season, the temperatures were higher in 1995 than 1994, as indicated by the increase in growing degree days (Fig. 2). Ewing (1981) showed that tuber initiation is enhanced by cool soil conditions (<15°C), while warm soil temperatures (>25°C) inhibit or even reverse tuberization. Not counting the nonoptimum weather in 1995, potato yield may have been depressed for other unknown reasons, because the haulms died before harvest. Even though 1995 yields were low, the trends among various irrigation and N management strategies were similar to those in 1994.
|
|
| Conclusions |
|---|
|
|
|---|
In both years, drip irrigation significantly reduced the irrigation amount compared with sprinkler irrigation. Less N application at emergence and hilling (alternative treatment) supplemented with post-hilling N applications had little effect on tuber yield and quality. Turkey manure produced yields similar to the urea treatments. The SCU and BDF treatments resulted in lower tuber yields in both years. In general, the wet sprinkler treatment and the SD treatment produced more No. 1 grade (200400 g) tubers than either the dry sprinkler or the BD treatment, even though total yields were not significantly different. The trends in total N uptake by plants closely followed that of total yield with relatively few differences occurring in either year.
| ACKNOWLEDGMENTS |
|---|
| NOTES |
|---|
|
|
|---|
Received for publication August 17, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Zvomuya, C. J. Rosen, M. P. Russelle, and S. C. Gupta Nitrate Leaching and Nitrogen Recovery Following Application of Polyolefin-Coated Urea to Potato J. Environ. Qual., March 1, 2003; 32(2): 480 - 489. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||