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 Google Scholar Google Scholar Articles by Kladivko, E. J. Articles by Santini, J. B. Search for Related Content PubMed Articles by Kladivko, E. J. Articles by Santini, J. B. Agricola Articles by Kladivko, E. J. Articles by Santini, J. B. Related Collections Water Management Crop Growth and Development Soil Hydrology Maize Production Agriculture Maize Management

Production Papers

Corn Growth and Yield Response to Subsurface Drain Spacing on Clermont Silt Loam Soil

Dep. of Agron., Purdue Univ., West Lafayette, IN 47907

^* Corresponding author (kladivko{at}purdue.edu)

Received for publication March 28, 2005.

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Subsurface drainage is an important water management practice on naturally poorly drained soils, and recommendations for appropriate drain spacings for particular soils continue to evolve. The objective of this long-term study was to measure corn (Zea mays L.) growth and yield as affected by subsurface drain spacing on a soil that was traditionally not tile drained. Three drain spacings (5, 10, and 20 m) were compared with a nondrained "control" (40 m) for plant population, grain yield, and moisture content over a 10-yr period on a low-organic-matter silt loam soil. In addition, corn populations, heights at 4 and 8 wk, yield, and moisture were measured with distance from the drain for the 5-, 10-, and 20-m spacings. Significant distance effects occurred more frequently for the 20-m spacing than for the 10- and 5-m spacings, especially for grain yield and moisture. The 10-yr average corn yields were 9.8, 9.7, 9.5, and 9.2 Mg ha^–1 for the 5-, 10-, and 20-m plots and the nondrained control plots, respectively. Grain yield was 1.3 to 1.7 Mg ha^–1 lower in the nondrained control than in the 5-m spacing in 3 of the 10 yr and was likely due to both planting date delays and wetter soil conditions after planting. The smaller-than-expected yield differences among treatments may reflect the excellent surface drainage in this field as well as optimal planting dates in 7 of the 10 years. The results demonstrate that drainage improvements are a long-term investment and may not provide yield benefit in every year.

Abbreviations: SEPAC, Southeast Purdue Agricultural Center

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
SUBSURFACE DRAINAGE is a common agricultural water management practice in the North Central USA. In soils with a seasonal perched water table, subsurface drainage "tile" has been utilized to lower the water table in spring, permitting more timely fieldwork and better crop establishment and early growth. The necessity of drainage for agricultural production has been discussed in numerous articles and textbooks over the years, and a recent extension publication (Zucker and Brown, 1998) and monograph (Skaggs and Van Schilfgaarde, 1999) present current thought on the practices. Although public concern about the loss of wetlands has prompted re-evaluation of agricultural drainage, Zucker and Brown (1998) point out that the current emphasis is to improve and optimize drainage on croplands already in production, rather than to drain remaining wetlands. Some current research evaluates technologies for drainage water management and subirrigation in conjunction with subsurface drainage, with goals to improve water use efficiency, water quality, and overall farm profitability on land already in agriculture.

The recent popularity of yield monitors and global positioning systems (GPS) among producers has had the effect of reinforcing some of the basic principles of crop production. Producers have seen the yield declines in wet spots or where tiles may be broken or plugged, as well as the better crop growth in areas near properly designed and functioning tile drains (Patrico, 1999). Consequently, the interest in subsurface drainage and the demand for information has increased.

Most states have published drainage recommendations for their agricultural soils, which include comments about surface and subsurface systems, recommended spacings, considerations about drainage outlets, etc. For the Clermont silt loam soil in southeastern Indiana, tile drainage was not generally recommended because of the low-porosity subsoil and the hazard of siltation to tile lines (Purdue Univ. Agric. Ext. Serv., 1958). Instead, surface drainage systems such as land smoothing and bedding were advocated (Sisson and Galloway, 1964). An updated Indiana Farm Drainage Guide (Purdue Univ. Coop. Ext. Serv., 1966) advised the use of tiles only to supplement a surface drainage system and suggested that surface drainage should be installed before complete tile systems were considered. In more recent years, however, subsurface drainage has been successfully performed on Clermont soil in Ohio (Fausey, 1983, 1984). Fausey and Brehm (1976) suggested that corrugated plastic draintubes and draintube plows made shallow placement of drains more practical, which may improve the effectiveness of drains on these soils.

This study was begun in 1983 to provide a modern evaluation of the benefits of subsurface drainage on the Clermont soil in southeastern Indiana. Results on water flow and soil water content have been published in Huffman et al. (1986), Larney et al. (1988), and Kladivko et al. (1991)(1999), and the first 4 yr of corn growth and yield data were reported by Larney et al. (1989). Determining the optimal drain spacing for a soil is critical when evaluating the economics of a drainage system. Although narrower drain spacings remove more water from the field, they do not always produce the best crop growth and yield (Schwab et al., 1957). Obviously, the beneficial effects of drainage will be more pronounced in wet years while less-intensive drainage may be better in dryer years. To help make decisions about the optimal spacing for a given field, programs such as DRAINMOD (Skaggs and Nassehzadeh-Tabrizi, 1983) consider crop production costs and yields, risks and probabilities of weather in the region, investment costs for the drainage system, etc. But the basis for any economic decision on drainage system intensity is good long-term data on crop yields on different drain spacings over a range of weather conditions. The objectives of our study were to determine corn growth and grain yield on three tile drain spacings (5, 10, and 20 m) compared with a nondrained control and to evaluate crop growth, grain moisture, and yield with distance from the drain tile for each spacing.

	MATERIALS AND METHODS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
A subsurface drainage research study was established in 1983 on a naturally poorly drained Clermont silt loam soil (fine silty, mixed, superactive, mesic Typic Glossaqualfs) at the Southeast Purdue Agricultural Center (SEPAC) (39°01'33'' N, 85°32'24'' W) near Butlerville, IN. This soil is typical of similar soils extending from southwestern Ohio to eastern Kansas, formed in 50 to 120 cm of loess over Illinoian glacial till and with slowly permeable shallow layers. The surface soil is a light gray, low organic C (0.7%) silt loam containing 66% silt, 22% sand, and 12% clay. King and Franzmeier (1981) identified a borderline fragipan in its silty clay loam B horizon and a distinct paleosol surface at 2- to 3-m depth.

The experimental area was about 5 ha with less than 1% slope. Subsurface plastic drains (10 cm diam.) were installed in spring 1983 at 5-, 10-, and 20-m spacings at an average depth of 75 cm and a slope of 0.4%. Three drain lines (225 m long) were installed at each spacing (Fig. 1) with the outside drainlines on each spacing acting as common drains between treatments. Spacing treatments were randomized in the first block and not re-randomized in the second block. The 40-m spacing was the nondrained control and will be referred to as the nondrained control throughout the paper.

View larger version (38K): [in this window] [in a new window]   Fig. 1. Layout of subsurface drain-spacing experiment on the Southeast Purdue Agricultural Center (SEPAC) experimental drainage field. Surface elevation contours are shown in meters.

Corn was planted in 0.76-m rows parallel to the drain lines. Rows were positioned so that the center drain on each spacing lay directly beneath the middle of an interrow area. This ensured that the closest rows to the drain were 0.38 m (rounded to 0.4 m) on each side and that successively further rows on each side were also equidistant from the drain by increments of 0.76 m (rounded to 0.8 m). Corn was planted over the entire width of each plot, but measurements were made from midplane to midplane around the center drain only. On the nondrained control, the measurement area consisted of 12 rows in the middle of the plots (i.e., 15 to 20 m from the nearest drain).

Starting in 1985, crop measurements (plant height and yield) were made in eight harvest areas in each plot. The first harvest area was 38 m from the drain outlet for each plot and the remaining areas at successive 23-m intervals. On the control plots, harvest areas followed the same spatial pattern even though there was no drain outlet. This design resulted in 64 harvest areas (4 drain spacings x 8 harvest areas x 2 replications), which varied in width depending on drain spacing. In 1984, only four harvest areas per plot were used, the first one being about 50 m from the outlet and the remainder at 50-m intervals from there.

All harvest areas were based on 5.3-m row lengths during 1984 to 1986; harvest row lengths were increased to 9.1-m lengths starting in 1987 due to a switch from hand-harvesting to combining the plots. Plant population was measured on each row of each harvest area at 4 wk from planting. Plant height was measured on three plants from each row in each harvest area at 4 and 8 wk from planting in 1985 to 1993, excluding 1989 (in 1984, 4-wk heights were taken at midplane positions on the spacings and in middle of control plots only). The same plants were measured at 8 wk as at 4 wk, allowing an average daily growth rate (change in height over change in time) for the intervening 4-wk period to be calculated. In 1984 to 1986, each row in each harvest area was hand-harvested individually. In 1984, the combined weight of equidistant rows on each side of the drain was used for yield calculation on each spacing (i.e., equidistant rows were not treated as subsamples). Starting in 1987, corn rows in each harvest area were combine-harvested in pairs (the 5-m plots therefore consisted of four rows on each side of the center drain, rather than three rows as previously). All corn yields were calculated on a 15.5% moisture basis.

Statistical analysis was performed using a Generalized Linear Models procedure (SAS Inst., 1999). Data were analyzed in two different ways. First, each spacing (except the control) was analyzed individually as a split block experimental design, with distance from the drain as a main treatment [each equidistant row, or pair of rows for yields in 1987 to 1993, on either side (east, west) of the drain being deemed as subsamples] and harvest area (i.e., distance from tile outlet) also as a main treatment, with both factors as fixed effects. Since distance from the drain represented a linear progression, the distance variance in the analysis was separated into its linear and quadratic components (trend analysis). In this analysis, linear and quadratic F values of distance may be significant even though the distance effect itself may be nonsignificant. The LSD values were computed for the P 0.05 level while the linear and quadratic contrasts were considered significant up to the P 0.10 level.

Second, an unweighted whole-plot means analysis of variance was performed as a split-block experimental design, with subsurface drainage tile spacing and years both as whole-unit treatments and two blocked replicates, again using a Generalized Linear Models procedure (SAS Inst., 1999). The four subsurface drainage tile-spacing treatments were randomized in the first block, but the treatments were not re-randomized in the second block; thus the spacing treatments were confounded with blocks. Confounding is assumed to have no effect, so the experiment was analyzed as if the tile-spacing treatments were re-randomized in each block. Two models were considered, one with years random and the other with years fixed. Averages of single-row yields and location (east or west of tile) within plot, year, drainage tile distance, and harvest area were calculated first and then averages across the distance and finally averages across harvest areas within each plot and year. This method of successive averaging weighted each area appropriately and minimized any possible distortion due to the occasional missing data. Thus, whole-plot averages consisted of the mean of all rows measured even though this number varied by spacing (6 or 8, 12, 24, and 12 rows on the 5-, 10-, and 20-m plots and nondrained control plots, respectively).

	RESULTS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Effects of Distance from Drain within 5-, 10-, and 20-m Spacings
Corn Establishment and Early Growth
Corn populations at 4 wk (Table 2), height at 4 and 8 wk (Table 2), and growth rate between 4 and 8 wk from planting (data not shown) were all significantly affected by distance from the drain in the 5-, 10-, and 20-m tile-spacing treatments in some years. Corn growth rate results were similar to the 8-wk height results and are not discussed further. Detailed data for the years 1988 to 1993 are presented in Table 2 while data from 1985 to 1987 are given in Table 2 of Larney et al. (1989). Growing season rainfall data for each of the 10 yr are presented in Fig. 2 .

View larger version (44K): [in this window] [in a new window]   Fig. 2. Monthly rainfall during April through August for 30-yr normal and all 10 yr of the drainage study at the Southeast Purdue Agricultural Center (SEPAC).

There were more years with significant distance effects for 8-wk heights compared with 4-wk corn heights and populations in the 10-m spacing. In contrast to the 5-m spacing, a linear response with distance from the tile was more common (4 of 8 yr for 8-wk height) than a quadratic response (1 of 8 yr). The linear response generally showed a decrease in height as distance from the drain increased (Table 2).

The 20-m spacing response was somewhat different than the 5- and 10-m spacings in the occurrence of early- vs. later-season effects. The 4-wk populations and heights had about the same number of significant distance effects as did 8-wk heights. This may reflect that with a wider spacing and the wetter conditions that exist further from the drain, both early- and later-season growth can be affected, depending on the weather that year. Linear responses with distance were more common than quadratic responses, with the most common response being a linear decrease in height with distance from the drain.

Corn Grain Yield and Moisture Content
Corn grain yield and moisture content with distance from the drain for the 5-, 10-, and 20-m spacings for all 10 yr are presented in Tables 3 through 5. The 5-m spacing showed a distance effect for yield in 1984 when both a linear and quadratic decrease with distance occurred (Table 3). Yield in 1984 was highest near the drain, decreased at the 1.2-m distance, and then increased again at the 2.0-m distance, which was similar to the quadratic responses of 4- and 8-wk heights in some years on that treatment. Grain moisture content showed a significant linear decrease with distance from the drain for 1984 and 1986. Starting in the fourth year (1987) of the 10-yr study, the 5-m spacing had only two harvest distances (since pairs of rows were combine-harvested together); thus linear and quadratic responses were not applicable, and little effect of distance was observed.

Grain yield had a linear response to distance in 5 and quadratic response in 4 of the 10 yr in the 20-m spacing (Tables 3–4), with yields generally higher near the drain and declining with distance. A linear response with distance occurred about as frequently for grain yield and moisture as for plant growth parameters while quadratic responses with distance occurred more frequently for yield than for plant growth. Four of the 5 yr with linear responses of grain moisture in the 20-m plots showed higher moisture contents at greater distances from the drain. The occasional higher moisture contents for the first row from the drain observed in the 5- and 10-m spacings (1984, 1986) were not expected but may reflect soil disturbance from drain installation in 1983 or the slightly higher soil moisture contents sometimes found immediately adjacent to the drain (Larney et al., 1988).

The 20-m spacing plots had more years with significant distance effects on yield (5 yr) compared with 10-m (3 yr) and 5-m (1 yr) plots. There was generally a yield decline as distance from the drain increased in the 10- and 20-m spacings. This would indicate to farmers that less corn grain yield variability may occur with the 5-m spacing compared with the 10- and 20-m spacings. Significant yield declines with distance from the drain occurred in several dry years as well as wet years, which is consistent with Bolton et al. (1982) and Fausey (1984). In general, across all three spacings, some years showed similar responses to distance from the drain for plant growth parameters and yield while other years showed distance effects on some parameters and not others. Grain moisture differences with distance from the drain were generally less apparent in years when harvest occurred after grain had dried down to the 15 to 18% range than in years with overall wetter grain.

Effect of Tile Spacing across Ten Years: Whole-Plot Analysis
Producers are generally more interested in yields from the entire spacing treatment rather than individual rows with distance from a drain. Decisions about the economics of a drainage system are based on the difference in yields among the various spacing treatments across many years compared with the initial investment for each system. Whole-plot corn yields from the 5-, 10-, and 20-m spacings as well as the nondrained control are shown in Table 6. The 10-yr average corn yields were not significantly different among treatments and were 9.8, 9.7, 9.5, and 9.2 Mg ha^–1 (156, 154, 151, and 146 bushels acre^–1) for the 5-, 10-, and 20-m plots and nondrained control plots, respectively. Year and the spacing x year interaction were highly significant. The 5-m drain spacing had significantly higher yield than all the other treatments in 1984. The 5-m drain spacing increased grain yield 1.3 to 1.7 Mg ha^–1 in 3 of the 10 yr when compared with the nondrained control and had the highest grain yield in 5 of the 10 yr. The 10-m spacing increased grain yield 1.3 to 1.4 Mg ha^–1 in 1989 and 1992 compared with the nondrained control and had the highest grain yield in 2 of the 10 yr. Yield differences in 1984 and 1989 were likely due in part to planting date differences (Table 1), as discussed later. The nondrained plot had the lowest yield of all plots in 7 of the 10 yr (significant in 3 yr).

Grain moisture content at harvest was significantly higher (P 0.01) for the nondrained control than for the 5-, 10-, and 20-m spacings when averaged over the 10-yr period (Table 7). Significant differences (P 0.05) among spacings within years occurred in 1989, 1990, and 1993, which were 3 of the 4 yr with the latest planting dates. The years with significant grain moisture differences were also years with overall wetter grain at harvest (>22% moisture), with no significant differences occurring in years when grain had already dried down to the 15 to 18% range. In 1989 and 1990, the control was different from the three drained treatments, with no differences among the three drained spacings. In 1993, the control was different from the 5- and 10-m spacings but not the 20-m spacing. Of those 3 yr, only 1989 had a substantial planting date delay (15 d) for the control compared with the three drained treatments.

Although there was near-normal precipitation during April–May in 1989, rainfall timing caused planting to be delayed until mid-May for the three drain spacings and an additional 15 d (63 mm of rain) for the nondrained control. There were significant differences in yields but no significant differences in plant populations among treatments. This year had the largest drop in yield in the nondrained control compared with the 5-m spacing (17% decrease or 1.66 Mg ha^–1 lower). Although the 15-d planting date delay may have contributed to the lower yields for the nondrained control compared with the 5-m spacing, the 20-m spacing had only a 1-d planting delay but yet a significantly decreased yield (1.04 Mg ha^–1 lower) compared with the 5-m spacing. The wetter-than-normal June (130% of normal) likely contributed to the lower yields in the 20-m spacing and nondrained control compared with the narrowest spacing.

May 1990 was extraordinarily wet (210% of normal), leading to the latest overall planting dates of the 10-yr study (29 May–1 June). The absence of any wet period after planting (Fig. 2, Table 9), however, and the uniformly late planting dates of all treatments led to similar yields for the 5-m spacing and the nondrained control. This was the only year where the two intermediate spacings (10 and 20 m) tended to have lower yields than both the 5-m spacing and the nondrained control although the differences were not significant.

Several years had normal to dry weather in spring, allowing timely planting of all plots, but then had a wetter-than-normal period in the summer that led to lower yields for the nondrained control (Table 9). Extremely high precipitation in July 1992 (about 300% of normal) led to the second-largest yield drop for the nondrained control compared with the 5-m spacing (13% decrease or 1.42 Mg ha^–1 lower). The 10-m spacing also had 1.42 Mg ha^–1 greater yield than the nondrained control while the 20-m spacing was intermediate in yield. In 1993, a wet August (about 160% of normal) may have contributed to the 8% yield drop (0.82 Mg ha^–1 lower) in the nondrained control compared with the 5-m spacing.

The other 5 yr of the 10-yr study were marked by drier-than-normal weather for substantial periods of the growing season (Table 9, Fig. 2). In 1985 through 1987, precipitation in April was 38 to 62% of normal, leading to early planting dates (22 to 30 April) in all 3 yr. May–June precipitation was lower than normal in 1985 and 1986 and much lower than normal in 1987, and the intermediate spacings (10 and 20 m) tended to have the highest yields, as discussed by Larney et al. (1989). Although April 1988 had near-normal precipitation, leading to early May planting dates, May–June precipitation during this drought year was only 23% of normal. Overall yields were low, but the 10-m spacing tended to have the highest yield. The 1991 season was another drought year, with only 10 mm of precipitation (11% of normal) in June. Although total July rainfall was 80% of normal, there was little rain the first 3 wk in July except for an intense, 50 mm in 1-h event on 2 July that likely suffered large runoff losses. Highest yields were obtained in the widest drain spacing (20 m).

The results from the 10-yr period are consistent with the initial 4-yr period of our study. Years dominated by substantial dry periods in the growing season tended to have higher yields in the intermediate (10- and 20-m) spacings compared with the 5-m spacing and nondrained control although differences were not statistically significant. Years with substantial wet periods either before or after planting showed yield declines with decreasing drainage intensity, and these differences were significant in three of five wet years.

	DISCUSSION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The average 6% lower yield in the nondrained control compared with the 5-m spacing was not as much lower as had been expected for this soil. In early drainage studies on a Clermont soil about 8 km from our SEPAC study, corn yields were about 24% lower in undrained land compared with all the drained treatments (Wiancko et al., 1939). The different cropping systems and overall lower yields of that time (3.4 Mg ha^–1 for drained plots) may have accentuated the importance of drainage more than in modern, high-yielding systems. On a Clermont soil in Ohio, Fausey (1983) reported that the average yield within 6 m of shallow drains (45- to 50-cm depth) was 9.69 Mg ha^–1 while average yield for all areas > 6 m from the drain was 8.84 Mg ha^–1 (9% lower). The smaller differences at our site may reflect both the excellent surface drainage on this field and the fact that the center of the 40-m spacing is not truly non-drained. In contrast, plots established in an adjacent field under drained (15 m) and undrained conditions had only fair surface drainage, with pockets of ponded water persisting in the undrained areas for several days after a rainfall. Corn yield differences between subsurface-drained and undrained areas of that field over an 8-yr period averaged between 0.9 and 1.4 Mg ha^–1, depending on the particular agronomic treatment (Kladivko, unpublished data, 1994). These larger yield differences suggest that in typical producers' fields with imperfect surface drainage, tile drains are likely to give a greater benefit than the 0.6 Mg ha^–1 yield difference between the 5-m spacing and nondrained control treatments in our study. In addition, producers are tending to plant corn earlier than previously, and yield differences may be greater with the cooler, wetter conditions common in early spring.

The smaller-than-expected differences in yield among drainage treatments also reflect the relatively drier weather cycle near planting time in this 10-yr study. As discussed previously, only 2 of 10 yr were wet enough at planting time to have substantial planting date delays between the 5-m spacing and the nondrained control (Table 1). The other years all had a "break in the weather" that was long enough to move sequentially with field operations from the narrow spacing to the wider spacings without having an intervening rainfall to prevent fieldwork. Thus the timeliness benefits that might be expected on larger acreage, having an extra day or two to plant before fieldwork is stopped again by another rain, could not be tested during this study period and on these plot sizes. One of the 10 yr had the same planting date for all four treatments, and another 3 yr had the same planting dates for the 5-, 10-, and 20-m spacings. The decision of when each treatment was "ready" for tillage and planting was made by the farm superintendent in consultation with the senior author, but the warm, dry weather most years at planting time made the decision relatively straightforward.

Previous studies in the Midwest have often found little additional benefit of the narrowest drain spacings (4 to 5 m) compared with intermediate spacings (10 to 20 m), depending on the weather during the study period. For example, corn yield from a 9-m drain spacing averaged 3% higher than that on a 4.5-m spacing and 13% higher than that on an 18-m spacing in a 6-yr study in southern Iowa (Schwab et al., 1957). The 4.5-m spacing had higher yields than the 18-m spacing in 4 of the 6 yr, the exceptions being the two driest years. The early study near SEPAC (Wiancko et al., 1939) found similar corn yields in the 5-, 10-, and 15-m spacings and yield decreases of 7 and 11% for the 20- and 25-m spacings, respectively. The authors commented that yield differences would not justify tiling any closer than 15 to 20 m. In a current study on a high-organic-matter soil in Indiana, Hofmann et al. (2004) reported 6-yr average annual corn yields were 0.3 Mg ha^–1 higher in the 20-m spacing than in either the narrower (10-m) or wider (30-m) spacings. The different results with different weather patterns and soil characteristics highlight the need for both long-term field research and accurate models to predict drainage response over the expected economic lifetime of a system.

Recent drainage research has focused on conserving water during dry periods while providing adequate drainage during wet periods by use of controlled drainage systems. These systems build on results such as those of a 6-yr study on seven soils in Indiana (Galloway and Sisson, 1969), which found greater yields in dry years at 30 m from the tile drain than at the tile drain on some of the soils. On a silt loam soil with 5-m drain spacings in Ohio, Fisher et al. (1999) reported 2-yr average corn yields were 19% greater with the controlled drainage/subirrigation system compared with conventional subsurface drainage. This again suggests that more water is removed by a 5-m drain spacing than is often optimal for corn growth later in the season. Galloway and Sisson (1969), for example, advocated blocking tile outflow if heavy rain caused discharge during July and August, arguing that corn would not be damaged by excess water at this time and yields would benefit, especially if a dry spell followed. New technologies have made drainage water management (controlled drainage) more feasible, and the practice is receiving new attention for both crop production and water quality purposes (Agricultural Drainage Management Systems Task Force, 2005).

	CONCLUSIONS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Corn growth, grain yield, and grain moisture content were significantly different at different distances from the tile drain in some years. Significant differences occurred more frequently for the wider spacing (20 m) than in the narrower spacings (10 and 5 m). When linear or quadratic effects occurred, they generally showed greater growth or yield values near the tile, decreasing values moving out from the tile, and then (for quadratic responses) an increase again at greater distance, but occasionally the lower values were near the tile.

The 10-yr whole-plot average corn yields were not significantly different among treatments and were 9.8, 9.7, 9.5, and 9.2 Mg ha^–1 for the 5-, 10-, and 20-m plots and nondrained control plots, respectively. The smaller-than-expected yield differences may reflect the excellent surface drainage in this field as well as optimal planting dates in 7 of the 10 yr. Only 2 of the 10 yr had substantial planting date delays between the 5-m spacing and the nondrained control, and both years had significantly different yields. For a producer with greater acreage than our study site, however, it is likely that the timeliness benefits of drainage would be greater than on our research farm plots, and this would give a larger yield advantage in some years.

The excellent surface drainage on our 5-ha site highlights a limitation of our research for evaluating drainage benefits at the larger scale of field production units. The Clermont soils occur in very flat landscapes where even small microtopography differences can result in ponded areas for prolonged time periods in the absence of subsurface drainage. On large production fields, these microtopography differences are more common than on our field site and result in planting delays or yield losses in some years. On producers' fields with less than ideal surface drainage, the overall yield benefit of tile drainage is probably greater than what we found on our 5-ha research field.

The optimal drain spacing for crop production depends on the number of years with yield benefits, crop prices, and the cost of the drainage system and would likely be between the 10- and 20-m spacings for this soil. An economic analysis of this system is the subject of future work. However, the optimal spacing for water quality and nitrate leaching is generally the wider spacing, as less nitrate was lost through the tile drains in the 20-m spacing than in the 10- and 5-m plots (Kladivko et al., 2004). Technologies such as drainage water management, whereby drainage and nitrate leaching are minimized during the fallow season, may be one way to better meet both crop production and water quality goals.

	ACKNOWLEDGMENTS

 
The authors thank the many people who have contributed to this work over the years, including field assistance by D. Biehle, W. Maschino, L. Sherwood, R. Martin, M. Hatton, D. Bauerle, C. Parker, D. Taylor, T. Reutebuch, E. Stath, and C. Kiefer, and data analysis assistance by F. Larney, J. Dickey, and B. Worstell. Financial support from the Purdue Office of Agricultural Research Programs is gratefully acknowledged.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Contribution of the Indiana Agric. Res. Progr., Purdue Journal Paper 2005-17688.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 

• Agricultural Drainage Management Systems Task Force. 2005. Agricultural Drainage Management Systems Task Force [Online]. Available at www.ag.ohio-state.edu/~usdasdru/ADMS/ADMSindex.htm (verified 27 June 2005). The Ohio State Univ., Columbus.
• Bolton, E.F., V.A. Dirks, and M.M. McDonnell. 1982. The effect of drainage, rotation and fertilizer on corn yield, plant height, leaf nutrient composition and physical properties of Brookston clay soil in southwestern Ontario. Can. J. Soil Sci. 62:297–309.
• Fausey, N.R. 1983. Shallow subsurface drain performance in Clermont soil. Trans. ASAE 26:782–784.
• Fausey, N.R. 1984. Drainage-tillage interaction on Clermont silt loam. Trans. ASAE 27:403–406.
• Fausey, N.R., and R.D. Brehm. 1976. Shallow subsurface drainage-field performance. Trans. ASAE 19:1082–1084, 1088.
• Fisher, M.J., N.R. Fausey, S.E. Subler, L.C. Brown, and P.M. Bierman. 1999. Water table management, nitrogen dynamics, and yields of corn and soybeans. Soil Sci. Soc. Am. J. 63:1786–1795.[Abstract/Free Full Text]
• Galloway, H.M., and D.R. Sisson. 1969. Corn yields as affected by distance from drain tile. Res. Progress Rep. 356. Purdue Univ. Agric. Exp. Stn., Lafayette, IN.
• Hofmann, B.S., S.M. Brouder, and R.F. Turco. 2004. Tile spacing impacts on Zea mays L. yield and drainage water nitrate load. Ecol. Eng. 23:251–267.
• Huffman, R.L., E.J. Monke, and E.J. Kladivko. 1986. Drainage studies on Clermont Silt loam. Paper 86–2554. ASAE, St. Joseph, MI.
• King, J.J., and D.P. Franzmeier. 1981. Morphology, hydrology and management of Clermont soils. Proc. Indiana Acad. Sci. 90:416–422.
• Kladivko, E.J., J.R. Frankenberger, D.B. Jaynes, D.W. Meek, B.J. Jenkinson, and N.R. Fausey. 2004. Nitrate leaching to subsurface drains as affected by drain spacing and changes in crop production system. J. Environ. Qual. 33:1803–1813.[Abstract/Free Full Text]
• Kladivko, E.J., J. Grochulska, R.F. Turco, G.E. Van Scoyoc, and J.D. Eigel. 1999. Pesticide and nitrate transport into subsurface tile drains of different spacings. J. Environ. Qual. 28:997–1004.[Abstract/Free Full Text]
• Kladivko, E.J., G.E. Van Scoyoc, E.J. Monke, K.M. Oates, and W. Pask. 1991. Pesticide and nutrient movement into subsurface tile drains on a silt loam soil in Indiana. J. Environ. Qual. 20:264–270.[Abstract/Free Full Text]
• Larney, F.J., E.J. Kladivko, and E.J. Monke. 1988. Subsurface drain spacing effects on soil moisture regime of Clermont silt loam. Trans. ASAE 31:1128–1134.
• Larney, F.J., E.J. Kladivko, E.J. Monke, and J.V. Mannering. 1989. Corn growth and yield behavior with distance from subsurface drains in drier-than-normal growing seasons. Trans. ASAE 32:579–587.
• Patrico, J. 1999. Tiling for dollars. Progressive Farmer 114(10):18–20.
• Purdue University Agricultural Extension Service. 1958. Indiana farm drainage guide. Mimeo AE-48. Purdue Univ. Agric. Ext. Serv. and USDA-SCS, Lafayette, IN.
• Purdue University Cooperative Extension Service. 1966. Indiana farm drainage guide. Mimeo ID-55. Purdue Univ. Coop. Ext. Serv. and USDA-SCS, Lafayette, IN.
• SAS Institute. 1999. SAS/STAT user's guide. Version 8. SAS Inst., Inc., Cary, NC.
• Schwab, G.O., D. Kirkham, and H.P. Johnson. 1957. Effect of tile spacing on crop yield and water table level in a Planosol soil. Soil Sci. Soc. Am. Proc. 21:448–452.
• Sisson, D.R., and H.M. Galloway. 1964. Surface drainage on Clermont silt loam. Res. Progr. Rep. 159. Purdue Univ. Agric. Exp. Stn., Lafayette, IN.
• Skaggs, R.W., and A. Nassehzadeh-Tabrizi. 1983. Optimum drainage for corn production. Tech. Bull. 274. North Carolina Agric. Res. Serv., North Carolina State Univ., Raleigh.
• Skaggs, R.W., and J. Van Schilfgaarde (ed.) 1999. Agricultural drainage. Agron. Monogr. 38. ASA, Madison, WI.
• Wiancko, A.T., G.P. Walker, and C. Robbins. 1939. Jennings County Experiment Field Report of Progress: 1921–1938. Circ. 244. Purdue Univ. Agric. Exp. Stn., Lafayette, IN.
• Zucker, L.A., and L.C. Brown (ed.) 1998. Agricultural drainage: Water quality impacts and subsurface drainage studies in the Midwest. Bull. 871. The Ohio State Univ. Ext., Columbus.

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 Google Scholar Google Scholar Articles by Kladivko, E. J. Articles by Santini, J. B. Search for Related Content PubMed Articles by Kladivko, E. J. Articles by Santini, J. B. Agricola Articles by Kladivko, E. J. Articles by Santini, J. B. Related Collections Water Management Crop Growth and Development Soil Hydrology Maize Production Agriculture Maize Management