HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Zebarth, B.J. Articles by Telford, G.A. Search for Related Content PubMed Articles by Zebarth, B.J. Articles by Telford, G.A. Agricola Articles by Zebarth, B.J. Articles by Telford, G.A.

DIVISION S-8-NUTRIENT MANAGEMENT & SOIL & PLANT ANALYSIS

Fertilizer Banding Influence on Spatial and Temporal Distribution of Soil Inorganic Nitrogen in a Corn Field

^a Potato Research Centre, Agriculture and Agri-Food Canada, P.O. Box 20280, Fredericton, NB, Canada, E3B 4Z7
^b BC Environment, P.O. Box 159, 9880 South McGrath Rd., Rosedale, BC, Canada, V0X 1X0
^c Transform Compost Systems, 34642 Mierau St., Abbotsford, BC, Canada V2S 4W8
^d PFRA, Rm. 203 Federal Bldg., 704-4th Ave. S., Lethbridge, AB, Canada T1J 0N8
^e Pacific Agri-Food Research Centre, Agriculture and Agri-Food Canada, Summerland, BC, Canada, V0H 1Z0

zebarthb{at}em.agr.ca

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
Fertilizer N is commonly applied as a sidedress band application in corn (Zea mays L.) fields. The objectives of this study were to determine the implications of fertilizer band application on spatial and temporal variations in soil inorganic N and to evaluate different sampling strategies for their suitability in estimating the average soil inorganic N concentration of a field. Treatments were 0, 60, and 120 kg N ha^-1 in 1994, and 0, 120, and 240 kg N ha^-1 in 1995, replicated three times at each of two sites. Fertilizer was applied as a band of NH₄NO₃ 15 cm on each side of each corn row at the corn six-leaf stage. Soil samples taken at eight interrow locations 0, 5, 10, 15, 20, 25, 30, and 35 cm from the corn row and at depth increments of 0 to 15, 15 to 30, and 30 to 60 cm and on five sampling dates were analyzed for soil NO₃ and NH₄ concentrations. Random sampling and five systematic sampling strategies were evaluated with respect to bias in estimation of field soil NO₃ and NH₄ concentration for three depth increments, and with respect to the number of samples required to achieve a given precision and probability level combination for soil NO₃ concentration for the 0- to 15-cm depth. All systematic sampling strategies provided adequate estimates of the true soil inorganic N concentration as estimated based on uniform sampling across all interrow locations. There was no consistent benefit to using any one systematic strategy with respect to the number of cores required to obtain a given level of precision in soil NO₃ concentration at a given level of probability. Systematic sampling strategies were at least as good as, or superior to, random sampling, particularly after fertilizer band application and at high N rates. The apparent insensitivity of estimated field soil NO₃ concentration to sampling strategy was attributed to the relatively high mobility of NO₃ in soil and variation in placement of the fertilizer band relative to the corn row during sidedressing.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
THE STUDY OF N CYCLING in corn fields frequently requires estimation of the quantity of inorganic N in soil. Inorganic N fertilizer is commonly applied to corn fields in a band, either with the planter or sidedressed when the corn is at approximately the six-leaf stage. This type of concentrated application in a band increases the difficulty in obtaining a soil sample that is representative of the overall field. Knowledge of the spatial and temporal distribution of soil inorganic N concentration associated with fertilizer bands can be used to select a sampling strategy that results in collection of a soil sample with an inorganic N concentration representative of the overall field.

While much research has been carried out on soil N processes, relatively little research has been completed on soil sampling protocols to adequately account for the variability of inorganic N, particularly that associated with banding of N fertilizer. Clay et al. (1995) studied the influence of banded N fertilizer application with corn on soil sampling requirements using random and systematic sampling approaches. Restricting soil sampling to an area 8 to 22 cm from the band location reduced the number of samples required to accurately estimate inorganic N concentration by more than 60% compared with random sampling. Conversely, no difference in the estimate of soil NO₃ concentration after wheat (Triticum aestivum L.) harvest was measured using a systematic or random sampling pattern when N fertilizer was applied as a band in spring (Mahler, 1990).

Studies have been completed on the spatial variation in P soil test values resulting from banded P fertilizer applications and the implications for soil sampling strategies (Mahler, 1990; Kitchen et al., 1990; Tyler and Howard, 1991; Rehm et al., 1995). The spatial distribution of P soil test is sensitive to tillage system and the method of P fertilizer band application (Rehm et al., 1995). Kitchen et al. (1990) proposed a method whereby the relative proportion of soil cores in a composite sample taken from the area of the fertilizer band could be calculated based on the distance between fertilizer bands. However, random sampling was often preferred when the accuracy of the P soil test value obtained, the number of cores required per composite sample, and the cost were considered (Mahler, 1990; Tyler and Howard, 1991; James and Hurst, 1995; Rehm et al., 1995). Random sampling resulted in lower and more variable P soil test values than systematic sampling after 1 yr of P fertilizer banding, but little difference between the P soil test values was measured when P fertilizer had been banded for two or more years (Zerkoune et al., 1994). This effect was attributed to fertilizer bands not being in the same locations in each year of application, and to the limited mobility of P in soil (Eghball et al., 1990). James and Hurst (1995) concluded that the number of individual cores required to estimate the P soil test value of a field following band application of P fertilizer was prohibitive. They recommended that a system be developed in which a slice of soil was collected perpendicular to the fertilizer band. A method for collecting soil samples perpendicularly to fertilizer bands with a modified chain saw to reduce the variability of P soil test values was described by Ashworth et al. (1994).

The first objective of this study was to determine the spatial and temporal distribution of soil inorganic N in a silage corn field receiving a band application of N fertilizer as sidedress. The second was to use the data to determine the most appropriate sampling strategies for estimating the quantity of soil inorganic N in the root zone at various times during and after the growing season.

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
Experimental trials were conducted in 1994 and 1995 at two sites in the Lower Fraser Valley, BC, Canada. The sites had either medium- (Site A) or coarse-textured (Site B) soil, and both were cropped to corn in the previous year. The soils at the experimental sites were stone-free, moderately well-drained floodplain soils, and classified as Eutrochrepts (Luttmerding, 1980) (Table 1) .

Corn (cv. Pioneer 3901) was planted on 28 Apr. 1994 and 26 Apr. 1995 at Site A and (cv. Pioneer 3845) on 3 May 1994 and 29 Apr. 1995 at Site B at 70000 plants ha^-1 with a 76-cm row spacing. Starter fertilizer applied with the planter included 9 kg N ha^-1 and 18 kg P ha^-1 as 11-52-0 in both years at Site A and 48 kg N ha^-1, 35 kg P ha^-1, and 6 kg K ha^-1 as a custom blend of 19-32-3 at Site B in both years. Starter fertilizer was placed 5 cm below and 5 cm away from the corn seed.

Each trial was a randomized complete block design with three blocks and three treatments. The experimental unit was a plot 3 m (four corn rows) wide by 12 m long. Treatments were 0, 60, and 120 kg N ha^-1 in 1994 and 0, 120, and 240 kg N ha^-1 in 1995 applied as NH₄NO₃ (34-0-0) in a sidedress application 15 cm away from each side of each corn row and at 5-cm depth at the six-leaf stage. The fertilizer was applied using a precision fertilizer applicator mounted on the back of a tractor. Cultivator tines mounted on the sidedress applicator provided tillage for weed control. The control plots received no fertilizer, but a sidedresser pass was applied to avoid a tillage effect. Some lateral movement (several centimeters) of the sidedresser unit was observed during fertilizer application. This was similar to what would be expected from a commercial application unit and resulted in the fertilizer band being applied at varying distances from the crop row.

In 1995, crop yield and N uptake were determined on each experimental plot. Two 6-m-long rows of corn were hand harvested from each plot at the end of the growing season to determine total wet silage yield. Corn yield and N uptake were not measured in the trials in 1994; however, corn yield and N uptake data were obtained in the same manner from adjacent experiments in the same fields with sidedress N rate treatments of 0, 30, 60, 90, 120, and 150 kg N ha^-1 as NH₄NO₃ (34-0-0) replicated twice. Ten representative corn plants from each plot in each year were chopped, and a 700-g subsample dried at 60°C for determination of dry matter yield. The subsample was ground to pass a 2-mm sieve and total N content of a 0.25-g subsample was determined using a Leco FP-428 N Determinator (Leco Corp., St. Joseph, MI).

Soil samples were collected at each site in each year by block after planting, and by plot after sidedress application in late August after the period of rapid crop N uptake, after harvest, and in late fall after the onset of heavy fall rainfall events (Table 2) . On each sampling date, two soil faces were exposed per block or plot to the 30-cm depth. A series of eight adjacent 5 cm wide by 10 cm long rectangular areas were sampled from each of the 0- to 15- and 15- to 30-cm depth increments using a custom sampler made from sheet metal. Thus, all soil in a trench 40 cm long (perpendicular to the corn row), 10 cm wide, and 30 cm deep was collected. The interrow location was designated by the distance from the corn row to the center of a soil sample in centimeters (Fig. 1) . Soil was sampled at the eight interrow locations of 0, 5, 10, 15, 20, 25, 30, and 35 cm. Five soil cores were then collected from each of the eight interrow locations for the 30- to 60-cm depth increment using a standard 2.5-cm-diameter soil probe. Soil from the two exposed soil faces was combined to give one composite soil sample for each interrow location by depth combination within each block or plot. Sampling was limited to the inside two corn rows and the middle 10 m of each plot.

View larger version (37K): [in this window] [in a new window]   Fig. 1 Soil sampling locations relative to the crop row and location of fertilizer applied with the planter (A) and as a band at sidedress (B)

All soil samples were frozen immediately after sampling and kept frozen until analyzed. A 15-g subsample of moist soil was oven dried at 105°C to determine gravimetric soil water content. A 20-g subsample of moist soil was extracted using 100 mL of 2 M KCl and a 1-h shaking time (Keeney and Nelson, 1982). Concentrations of NO₃ and NH₄ in the extract were determined spectrophotometrically by flow injection analysis (Zebarth et al., 1996).

Analysis of variance was conducted on the concentrations of soil NO₃ and NH₄ using the general linear model procedure of SAS (SAS Institute, 1990). The analysis assumed a nested arrangement of treatments, where N rate was considered to apply to the main plot, time of sampling to the subplot, interrow location to the sub-subplot, and depth to the sub-subsubplot.

Six sampling strategies were evaluated: (i) a random sampling strategy (RND) where soil cores were selected randomly from the eight interrow locations; (ii) a systematic sampling pattern where an equal number of soil cores were selected from each interrow location (SYS-ALL); (iii) a systematic sampling pattern where an equal number of soil cores were selected only within the crop row (interrow location 0), at the approximate location of the fertilizer band (interrow location 15), and midway between crop rows (interrow location 35) (SYS-0/15/35); (iv) a systematic sampling pattern where an equal number of soil cores were selected only at the approximate location of the fertilizer band and midway between crop rows (SYS-15/35); (v) a systematic sampling pattern where an equal number of soil cores were selected only within the crop row and at the approximate location of the fertilizer band (SYS-0/15); and (vi) a systematic sampling pattern where 25, 50, and 25% of the soil cores were selected within the crop row, at the approximate location of the fertilizer band, and midway between crop rows, respectively (SYS-0/15/15/35).

The experimental data, using combinations of year, site, N rate, and sampling date, were used as populations of NO₃ and NH₄ concentrations for evaluating the sampling strategies for three depth increments (0–15, 0–30, and 0–60 cm). When sampling from a population, each strategy generates a sampling distribution with its own mean (i.e., expected value) and standard deviation. The difference between the expected value in the sampling distribution and in the original population is the bias. The sampling strategies RND and SYS-ALL are unbiased because all eight interrow locations were sampled equally. Expected values and standard deviations were calculated for each sampling strategy using the methods for stratified random sampling (Snedecor and Cochran, 1967). Analysis of variance was performed on soil NO₃ concentration for each systematic sampling strategy with SYS-ALL as a covariate to test for the effects of the factors experimental site, N rate, and sampling date on the bias. There was no significant effect of any factor for depth increments of 0 to 15, 0 to 30, and 0 to 60 cm. Therefore, a single estimate of bias could be determined for each sampling strategy at each depth. Expected values were expressed as a percentage of the expected value for SYS-ALL. The same approach was used for soil NH₄ concentration.

The standard deviation of NO₃ concentration increased with increasing mean concentration. For the data from the 0- to 15-cm depth in each year the relationship between the SD and the mean was expressed as:

(1)

Coefficients a and b were estimated from the data by regression for each combination of site, N rate, and sampling date. For each sampling strategy the effects of the factors site, N rate, and sampling date on the coefficients a and b were investigated using analysis of variance. The coefficient b was independent of the factors and was in most cases not significantly different from one. This implied that the relationship between the standard deviation and the mean could be represented by the CV. The coefficient a and consequently the CV were usually dependent on N rate and sampling date. Therefore, the CVs for each sampling strategy were estimated for each N rate by sampling date combination in each year.

The number of soil cores (n) required to estimate the mean of the sampling distribution within a certain percentage (100 D) of its expected value with a given probability was determined from the formula (adapted from Snedecor and Cochran, 1967):

(2)

where z is the normal deviate for a given probability (i.e., 0.01, 0.05, 0.10, and 0.20), and values considered for the precision, D, were 0.05, 0.10, and 0.20. For NO₃ concentration in the 0- to 15-cm depth increment, this calculation was done for each combination of year, N rate, and sampling date using the estimated CVs. For each sampling strategy, the number of cores required in a sample is an integer multiple of a minimum number, for example of 8 for SYS-ALL and of 4 for SYS-0/15/15/35, but this is not accounted for in the formula.

	Results

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
Growing season (April–October) air temperatures averaged 1.1 and 1.2°C above the long-term (1961–1990) average in 1994 and 1995, respectively (Fig. 2) . Total rainfall during the same period was 90 and 104% of the long-term average. Warm temperatures and near average growing season rainfall resulted in above-average yields in both years. Rainfall was below average in April and May of both years, and above average in July, August, and October of 1995.

View larger version (29K): [in this window] [in a new window]   Fig. 2 Growing season air temperature and rainfall (bars) at Agassiz in 1994 and 1995 in comparison with the long-term (1961–1990) average

Soil NH₄ concentrations at 0- to 15-cm depth measured 8 d after planting (12 May 1995) at Site B were higher at the 0- and 5-cm locations (21 and 13 mg NH₄–N kg^-1 soil, respectively) than at other interrow locations (average of 3 mg NH₄–N kg^-1 soil) in response to N applied with the seed at planting (Fig. 3) . Soil NH₄ concentrations at 15- to 30-cm depth were similar at all interrow locations and averaged 4 mg NH₄–N kg^-1 soil. A similar but smaller response was observed at Site A because less N had been applied.

View larger version (17K): [in this window] [in a new window]   Fig. 3 Soil NH4–N concentrations at the 0- to 15-cm depth for different interrow locations in May after planting and in June as influenced by rate of banded fertilizer N application at two sites in 1995

Similar to NH₄, slightly elevated NO₃ concentrations were observed in May at the 0-cm location at 0- to 15-cm depth at Site A because N had been applied with the seed (Fig. 4a) . A similar, but larger response was observed at Site B because of the higher N rate at planting (Fig. 5a) .

View larger version (37K): [in this window] [in a new window]   Fig. 4 Soil NO3–N concentrations for the 0- to 15- and 15- to 30-cm depths at different interrow locations and on different sampling dates at Site A in 1995 for three N rates

View larger version (33K): [in this window] [in a new window]   Fig. 5 Soil NO3–N concentrations for 0- to 15- and 15- to 30-cm depths at different interrow locations and on different sampling dates at Site B in 1995 for three N rates

Soil NO₃ concentrations in the 0- to 15-cm depth increment at Site A generally increased from the mid May sampling date to a maximum value on the June sampling date because of fertilizer application, net soil N mineralization, and nitrification of NH₄ (Fig. 4). Subsequently, soil NO₃ concentrations decreased throughout the remainder of the growing season for 240 kg N ha^-1, whereas a small increase in soil NO₃ concentrations occurred between August and September for the control and 120 kg N ha^-1. A similar response was observed at Site B, except that soil NO₃ concentrations did not increase between the May and June sampling dates in the control (Fig. 5).

Soil NO₃ concentrations increased in the 15- to 30-cm depth increment at both sites for 240 kg N ha^-1 only (Fig. 4 and 5), and did not change during the growing season for any treatment in the 30- to 60-cm depth increment (data not presented). Soil NO₃ contents were uniformly low at all depths for all treatments in November (Fig. 4 and 5). Soil NO₃ concentrations to 60-cm depth in the location of the fertilizer band were low for all treatments measured in January or February.

When sampling strategies were evaluated, expected values for NO₃ concentrations were generally within 5%, and almost always within 10%, of the true value for both years and for the 0- to 15-, 0- to 30-, and 0- to 60-cm depths (Table 4) . In most cases, the expected value was closer to the true value when a greater depth of soil was considered (i.e., 0–60 vs. 0–15 cm). Overall, the SYS-0/15/35 sampling strategy appeared to give somewhat better results than other sampling strategies, having generally low differences in the expected value from the true value and a low standard deviation in both years. The difference between the expected and the true values for the 0- to 15-cm depth increment was somewhat higher for NH₄ concentration than for NO₃, especially in 1995 (Table 4).

For a given sampling strategy, the number of cores required to estimate soil NO₃ concentration in the 0- to 15-cm depth varied by as much as two orders of magnitude depending on the choice of precision and probability levels (Table 6) . Precision generally had a greater effect on the number of cores than the probability level.

	Discussion

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

The observed changes in soil NO₃ concentrations with time are consistent with the results of Zebarth and Paul (1997), where soil NO₃ contents generally increased until June due to high net N mineralization and nitrification of NH₄, decreased after June due to rapid crop N uptake, and increased again in late summer due to significant net N mineralization.

The increase in soil NO₃ concentrations in the 15- to 30-cm depth increment below the location of the fertilizer band suggests that some downward movement of NO₃ occurred during the growing season, possibly in response to the above-average precipitation in July and August of 1995. This downward movement was observed only for 240 kg N ha^-1, well above the rate required for optimal corn growth, and resulted in elevated soil NO₃ concentrations throughout the growing season. The lack of any increase in the 30- to 60-cm depth increment suggests that loss of NO₃ from the root zone was limited during the growing season. This is consistent with the minimal downward movement of NO₃ during the growing season for medium-textured, nonirrigated soils, as reported previously for this region (Kowalenko, 1987, 1989; Zebarth and Paul, 1997), although downward movement of NO₃ was measured by Zebarth and Paul (1997) in a coarse-textured corn field under unusually wet spring conditions.

There was an almost complete loss of soil NO₃ from the top 30 cm of the soil by the November sampling date. This is consistent with the essentially complete leaching of NO₃ from the root zone of south coastal British Columbia soils that occurs in response to heavy fall and winter rainfall events (Kowalenko, 1987, 1989; Zebarth et al., 1996).

The various soil N processes resulted in varying distributions of NO₃ and NH₄ concentrations with interrow location, N rate, and depth on different sampling dates. These variations have implications for the collection of soil samples to estimate average inorganic N concentrations on a field basis.

Using current fertilizer N recommendations for silage corn in south coastal British Columbia based on the pre-sidedress soil NO₃ test (Zebarth et al., 1994), it can be assumed that a precision level of 0.20 with a probability level of 0.10 is adequate for on-farm testing. Somewhat higher levels of precision may be required for other purposes. Using this criterion, all systematic sampling strategies appeared to be adequate in terms of the estimated value that would be obtained.

The number of cores required for different sampling strategies varied for different precision and probability combinations, and for different sampling dates and N rates within a given precision and probability combination. Overall, there was no consistent benefit to any systematic sampling strategy. However in certain cases, a higher number of cores were required for the SYS-ALL and SYS-0/15 sampling strategies. In almost all cases, the number of cores required to obtain 0.20 precision and 0.10 probability was sufficiently low (i.e., 40 or less) to be practical. A smaller number of cores was required for systematic than for random sampling strategies in cases where the spatial variation in soil NO₃ was most pronounced, after sidedress application and where the N application rate was high.

Therefore, all systematic sampling strategies appeared to be acceptable, and no single sampling strategy appeared to be consistently better than any other. Systematic sampling was at least as good as or superior to random sampling, particularly after fertilizer band application and for high N application rates.

The results of this study suggest that measured soil NO₃ and NH₄ concentrations are less sensitive to the sampling strategy than might be expected. This lack of sensitivity is probably due to several factors. First, this study looked at variation in N as opposed to P concentration. Greater mobility of N in soil should allow greater redistribution of N in the soil than P. In addition, the essentially complete leaching of NO₃ that occurs in south coastal soils each fall and winter (Kowalenko, 1987, 1989; Zebarth et al., 1996) should prevent patterns associated with band application in one growing season from persisting until the next growing season, unlike P. Secondly, the fertilizer in this study was applied with a sidedresser similar to a commercial unit. This resulted in several centimeters of variation in the location of the fertilizer band relative to the corn row. In most other studies, the N or P was applied with the planter, resulting in precise application of the fertilizer relative to the corn row. More precise placement of the fertilizer would be expected to increase the sensitivity of measured soil inorganic N concentrations to sampling strategy. Thirdly, soils in south coastal BC have relatively high soil N mineralization rates (Zebarth et al., 1996), resulting in significant quantities of inorganic N being present in the soil regardless of the fertilizer placement. This is consistent with the high corn yield and N uptake obtained with 0 kg N ha^-1. This pattern of mineralization is probably also variable in space because of variable patterns of manure application in previous years. Both the magnitude and variation in mineralization would be expected to reduce sensitivity to the sampling strategy. Finally, soil NO₃ and NH₄ concentrations were measured in discrete soil volumes that were larger in size than a typical soil sample. This should reduce the apparent variability in soil inorganic N across interrow locations. However, it is unclear how important this factor is given the apparent variability in fertilizer placement relative to the corn row.

The systematic sampling patterns chosen in this study, with the exception of the SYS-ALL, were chosen to be practical for use in field sampling. The results of this study suggest that these sampling patterns are practical to use and may be advantageous relative to random sampling for two reasons. First, it is difficult to choose a truly random sampling pattern in a corn field because of the obvious pattern provided by the corn rows. Thus, it is likely that some sort of systematic sampling will be done regardless, in which case, a predefined sampling strategy would be preferable. Secondly, the systematic sampling strategy required fewer cores than random sampling in some cases (i.e., after sidedress application and at high N application rates) to meet a given precision and probability combination. For ease of field sampling, the simpler systematic sampling strategies such as SYS-15/35 and SYS-0/15/35 may be preferable.

	ACKNOWLEDGMENTS

 
This research was supported by Agriculture and Agri-Food Canada, Canada's Green Plan for Agriculture, Fraser River Action Plan of Environment Canada, and Coast Agri Ltd. The assistance of Mr. John Dieleman, the owner of one of the sites, is gratefully acknowledged. Field and laboratory assistance were provided by Brian Harding, Lisa Birston, David Chapple, Cynthia Watson, Jennifer Sadlish, and Chad Cowie.

Received for publication July 28, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 

• Ashworth J., Crépin J.M., Repski R. Soil sampling with a modified chainsaw. Agron. J. 1994;86:736-740.[Abstract/Free Full Text]
• Ashworth J., Flint R.C. Delayed nitrification and controlled recovery of aqueous ammonia injected under grass. J. Agric. Sci. (Cambridge) 1974;83:327-333.
• Clay D.E., Carlson C.C., Holman P.W., Schumacher T.E., Clay S.A. Banding nitrogen fertilizer influence on inorganic nitrogen distribution. J. Plant Nutr. 1995;18:331-341.
• Eghball B., Sander D.H., Skopp J. Diffusion, adsorption, and predicted longevity of banded phosphorus fertilizer in three soils. Soil Sci. Soc. Am. J. 1990;54:1161-1165.[Abstract/Free Full Text]
• James D.W., Hurst R.L. Soil sampling technique for band-fertilized, no-till fields with Monte Carlo simulations. Soil Sci. Soc. Am. J. 1995;59:1768-1772.[Abstract/Free Full Text]
• Keeney D.R., Nelson D.W. Nitrogen — Inorganic forms. In: Page A.L., ed. Methods of soil analysis. Part. 2, 2nd ed Madison, WI: Agron. Monogr. 9. ASA and SSSA, 1982:643-698.
• Kitchen N.R., Havlin J.L., Westfall D.G. Soil sampling under no-till banded phosphorus. Soil Sci. Soc. Am. J. 1990;54:1661-1665.[Abstract/Free Full Text]
• Kowalenko C.G. The dynamics of inorganic nitrogen in a Fraser Valley soil with and without spring or fall ammonium nitrate applications. Can. J. Soil Sci. 1987;67:367-382.
• Kowalenko C.G. The fate of applied nitrogen in a Fraser Valley soil using ¹⁵N in field microplots. Can. J. Soil Sci. 1989;69:825-833.
• Luttmerding, H.A. 1980. Soils of the Langley-Vancouver map area. Vol. 1. B.C. Soil Survey Rep. 15.
• Mahler R.L. Soil sampling fields that have received banded fertilizer applications. Commun. Soil Sci. Plant Anal. 1990;21:1793-1802.
• Rehm G.W., Randall G.W., Scobbie A.J., Vetsch J.A. Impact of fertilizer placement and tillage system on phosphorus distribution in soil. Soil Sci. Soc. Am. J. 1995;59:1661-1665.[Abstract/Free Full Text]
• SAS Institute. SAS/STAT user's guide: Statistics. Version 6, 4th ed Cary, NC: SAS Inst, 1990.
• Snedecor G.W., Cochran W.G. Statistical methods, 6th ed Ames: Iowa State University Press, 1967.
• Tyler D., Howard D.D. Soil sampling patterns for assessing no-tillage fertilization techniques. J. Fert. Issues 1991;8:52-56.
• Wetselaar R., Passioura J.B., Singh B.R. Consequences of banding nitrogen fertilisers in soil. I. Effects on nitrification. Plant Soil 1972;36:159-175.
• Zebarth B.J., Paul J.W. Growing season nitrogen dynamics in manured soils in south coastal British Columbia: Implications for a soil nitrate test for silage corn. Can. J. Soil Sci. 1997;77:67-76.
• Zebarth B.J., Paul J.W., Schmidt O., McDougall R. Influence of the time and rate of liquid manure application on yield and nitrogen utilization of silage corn in south coastal British Columbia. Can. J. Soil Sci. 1996;76:153-164.
• Zebarth B.J., Younie M., Paul J.W., Bittman S. The pre-sidedress soil nitrate test for silage corn: 1994 results. Agassiz, BC, Canada: Pacific Agriculture Research Centre Tech. Rept. 108. Agriculture and Agri-Food Canada, 1994.
• Zerkoune, M.A., D.H. Sander, and C.A. Shapiro. 1994. Soil sampling and soil test-evaluation of banded phosphorus under no-till. p. 105–118. In J.L. Havlin (ed.) Great Plains Soil Fert. Conf. Proc. Denver, CO. 7–9 Mar. 1994. Kansas State Univ., Manhattan, KS.

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Zebarth, B.J. Articles by Telford, G.A. Search for Related Content PubMed Articles by Zebarth, B.J. Articles by Telford, G.A. Agricola Articles by Zebarth, B.J. Articles by Telford, G.A.