| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a USDA-ARS, Crop Genet. and Prod. Res. Unit, P.O. Box 343, Stoneville, MS 38776
b USDA-ARS Appl. and Prod. Technol. Res. Unit, P.O. Box 36, Stoneville, MS 38776
c Dep. of Agric. Econ., P.O. Box 9755, Mississippi State, MS 39762
* Corresponding author (lheatherly{at}ars.usda.gov)
Received for publication September 21, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: MG, maturity group • NR, narrow row • POST, postemergence • PRE, pre-emergence • RW, row width • WR, wide row • WTRT, weed management treatment
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Soybean planted in May and later in the midsouthern USA, especially nonirrigated soybean, provides relatively low gross returns with a small margin for profit (Heatherly et al., 1994; Heatherly and Spurlock, 1999; Williams, 1999). Soybean grown without irrigation has only a small profit potential. Thus, inputs used for producing a crop, especially direct costs for pest management, must be considered economically. Previous research at Stoneville, where drought is common during the reproductive period of soybean, demonstrated that weed management in addition to preplant foliar-applied glyphosate [N-(phosphonomethyl)glycine] and postemergence (POST) cultivation in nonirrigated conventional plantings of WRs did not increase seed yield and resulted in lower net returns (Heatherly et al., 1994). However, weed management expenditures are almost always made before the onset of drought and without knowledge of ensuing moisture status for subsequent crop and weed development.
At northern latitudes of the soybean-growing region of North America, soybean grown in NRs generally outyields soybean grown in WRs (Devlin et al., 1995; Elmore, 1998; Mickelson and Renner, 1997; Nelson and Renner, 1998; Swanton et al., 1998). Reasons for this NR advantage may be related to better weed control (Mickelson and Renner, 1997; Nelson and Renner, 1998), drought-free growing seasons (Devlin et al., 1995), and less weed resurgence following early season weed management in NRs (Yelverton and Coble, 1991). Results from Kansas reported by Devlin et al. (1995) imply that NR yield advantage may occur only in high-yielding environments (>3350 kg ha-1) such as those supported by irrigation. However, Oriade et al. (1997) reported that both yields and net returns obtained from NR systems consistently exceeded those from WR systems under both irrigated and nonirrigated conditions in northern Arkansas. In the Arkansas research, intensive weed management with broadcast herbicides was practiced in both RWs. Thus, a potential advantage of using less herbicides in WRs, with resulting lower weed management costs, was not realized.
In the southern USA, conventional plantings (May and later) of soybean grown in NRs (
50 cm wide) generally produce higher yield than soybean grown in WRs (Boquet, 1990; Ethredge et al., 1989; Heatherly, 1988; Oriade et al., 1997). However, the yield advantage of NRs may be inconsistent over years and relatively small without irrigation (Heatherly, 1988). Thus, choice of row spacing should not be based solely on the presumption that NR soybean systems will yield more than WR systems. A yield advantage for NRs must be measured against the economics of each system.
Wide-row soybean production systems are used because they match the row-spacing requirements of other row crops in a producer's crop mix. Banded herbicide application combined with interrow cultivation can be used to effectively manage weeds (Buhler et al., 1992; Poston et al., 1992), reduce weed control costs (Buhler et al., 1997), and reduce amount of herbicide introduced into the environment (Poston et al., 1992; Swanton et al., 1998). Narrow-row systems preclude POST cultivation normally used in WRs (Buhler et al., 1997; Hooker et al., 1997; Newsom and Shaw, 1996; Swanton et al., 1998). In NR soybean plantings made in a stale seedbed, effective weed management systems will almost exclusively involve herbicides (Johnson et al., 1997; Johnson et al., 1998; Oliver et al., 1993). This can lead to improved weed control in NR systems that will result in greater yield and net returns compared with WR systems (Mickelson and Renner, 1997; Swanton et al., 1998). However, increased net returns depend on both economical weed management for and increased yield from NR systems. Both of these requirements may not occur and if not, can lead to lower net returns.
Determination of economically feasible weed management systems in nonirrigated and irrigated environments using broadcast-applied pre-emergence (PRE) and POST herbicides without POST cultivation in NRs and band-applied PRE and POST herbicides with POST cultivation in WRs is necessary. Comparisons of strategies based purely on yield results may be biased in favor of production practices that result in higher yields at greater cost. Soybean management systems must provide options for economic control of both summer broadleaf weeds and grasses to attain maximum economic yield. Inputs used for weed management in soybean represent a significant cost (Buhler et al., 1997; Heatherly et al., 1994; Johnson et al., 1997) and must be managed early (PRE) or on an as-needed basis (POST).
Use of combinations of PRE and POST herbicides with POST cultivation is common in WR production systems in the midsouthern USA (Askew et al., 1998; Heatherly and Elmore, 1991; Heatherly et al., 1993; Heatherly et al., 1994; Hydrick and Shaw, 1995; Oliver et al., 1993; Poston et al., 1992). Herbicides banded over the crop row and cultivation of interrow areas can provide complementary weed control (Newson and Shaw, 1996; Griffin et al., 1993) and may result in lower weed management costs than broadcast applications of herbicides (Krausz et al., 1995) in any row spacing. Interrow cultivation alone will not control weeds over time and will result in lower yield and net returns (Buhler et al., 1997) than when supplemented with herbicide weed control. On the other hand, soybean culture in NRs will not allow the option of POST cultivation, and thus involves the exclusive use of and reliance on PRE- and/or POST-broadcasting of herbicides.
Many weed management systems provide similar control of weeds, but cost differences can be large (Buhler et al., 1997; Heatherly et al., 1993, 1994). Cost difference, coupled with yield differences among weed management systems, can mean significant differences in net return among weed control systems (Buhler et al., 1997; Heatherly et al., 1993, 1994; Johnson et al., 1997; Poston et al., 1992). The best weed management systems must be determined to maximize profits.
The economics of RW–weed management–cultivar systems for continuous cropping of conventional soybean grown with and without irrigation in the midsouthern USA have not been evaluated. The objectives of this study were to (i) determine the effect of PRE and POST broadleaf and grass herbicides, used alone or in combination, on weed cover in and seed yield from continuous nonirrigated and irrigated plantings of conventional soybean cultivars grown in NR (50 cm wide) and WR (100 cm wide); and (ii) use estimated costs and returns to evaluate the economic impact of treatment effects.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Separate nonirrigated and irrigated experiments were conducted using a randomized complete block design with four replicates each year. Treatments were in a split split-plot factorial arrangement, with RW as the main plot, cultivar as the subplot, and weed management treatment (WTRT) as the sub-subplot. Treatments were randomly assigned to plots in 1994 and remained in the same location thereafter to assess the effect of continued use of a system.
Plantings were made on 12 May 1994, 10 May 1995, and 16 May 1996. Maturity Group V ‘P 9592’ (classified as medium tall) and ‘DP 3589’ (classified as tall) were chosen to provide a difference in stature. Their selection was based on previous experience, regional variety trial results, and use patterns by producers. Seed were treated with metalaxyl [N-(2,6-dimethylphenyl)-N-(methoxyacetyl)-DL-alanine methyl ester] fungicide at 0.3 g a.i. kg-1 before seeding.
Row widths were 0.5 (NR) and 1 m (WR). Seeding rates were 16 and 33 seed m-1 NR and WR, respectively, or about 50 kg ha-1 seed. Plots were 4 m (eight NR or four WR) wide and 21.5 m long. All experiments were seeded into a stale or untilled seedbed (Heatherly and Elmore, 1983; Heatherly et al., 1993; Heatherly, 1999b) that had been tilled with a disk harrow and/or a spring-tooth field cultivator the previous fall. Glyphosate at either 840 or 1120 g a.i. ha-1 in 94 L ha-1 water was preplant foliar-applied to each experimental site in April and/or May of each year to kill weed vegetation.
Weed management treatments were selected along the following premises. First, uncontrolled weeds will reduce soybean yield. Therefore, no weedy check was included. Second, the inclusion of economic analyses in this study dictated that all WTRTs be practical and realistic. Also, there was no intent to determine how WTRT related to an economically unattainable or unfeasible weed-free environment. Therefore, a weed-free check was not included. Third, cultivation in WRs was assumed to be a POST weed-control measure and could be used in lieu of POST herbicides in a treatment with a POST component. Based on these premises, WTRTs each year were (i) control of only broadleaf weeds using PRE and POST broadleaf herbicides in NRs and PRE and POST broadleaf herbicides and/or POST cultivation in WRs (WTRT 1); (ii) control of both broadleaf and grass weeds using PRE broadleaf and grass herbicides and POST broadleaf herbicides in NRs and PRE broadleaf and grass herbicides plus POST broadleaf herbicides and/or POST cultivation in WRs (WTRT 2); (iii) control of only broadleaf weeds using POST broadleaf herbicides in NRs and POST broadleaf herbicides and/or POST cultivation in WRs (WTRT 3); and (iv) control of both broadleaf and grass weeds using POST broadleaf and grass herbicides in NRs and POST herbicides and/or POST cultivation in WRs (WTRT 4). Postemergent cultivation was conducted in WRs three times in 1994 and 1995 and two times in 1996. Herbicides were broadcast-applied to NRs and band-applied (0.5-m-wide band centered over each row) to WRs each year at labeled rates with recommended adjuvants and in recommended tank mixes. Pre-emergent herbicides were applied immediately after planting each year. Rainfall amounts received within 10 d after PRE applications were 15, 8, and 28 mm in 1994, 1995, and 1996, respectively.
Application of PRE herbicides and their combinations within each WTRT was dictated by expected weed populations. Expert opinion during the early growing season was used to determine when populations of weeds within each WTRT were sufficient to justify application of POST herbicides and what herbicides to use. The POST weed management inputs that were applied were determined to be necessary based on weed presence; i.e., they were not frivolous but were applied to control specific weed problems. Pre-emergent herbicides and POST broadleaf herbicides were applied in 187 L ha-1 water, whereas POST grass herbicides were applied in 94 L ha-1 water. Herbicides used each year and their application rates are shown in Table 1. All PRE and POST broadcast herbicides were applied using a canopied sprayer (Ginn et al., 1998a) for over-the-top applications (to prevent drift to adjacent plots of different treatments) or a directed sprayer (Ginn et al., 1998b) for applications underneath the developing soybean canopy.
|
|
Just before harvest each year, mature plant height (length from the soil surface to the tip of stem) was measured in NRs and WRs of WTRT-4 plots to verify height differences between cultivars each year. No visually detectable differences in height occurred among the WTRTs. This is supported by results from studies with similar WTRTs at this location (Heatherly et al., 1992). Lodging ratings were recorded each year, but none exceeded a score of 2 (either all plants leaning slightly or a few plants down). This minor leaning did not affect harvestability; therefore, lodging data are not presented.
A field combine modified for small plots was used to harvest the two (WR) or four (NR) center rows of each plot on 27 and 28 Sept. 1994, 29 Sept. and 2 Oct. 1995, and 9 Oct. 1996. Seed from all plots were cleaned by the harvesting machine; thus, correction for foreign matter content in seed of any treatment combination was not necessary in any year. Harvested seed were weighed and adjusted to 130 g moisture kg-1 seed.
Estimates of total costs and returns (U.S. dollars) were developed for each annual cycle of each experimental unit using the Mississippi State Budget Generator (Spurlock and Laughlin, 1992). Total specified expenses were calculated using actual inputs for each treatment in each year of the experiment and included all direct and fixed costs but excluded costs for land, management, and general farm overhead, which were assumed to be the same for all treatment combinations. Direct expenses included costs for herbicides and adjuvants, seed, rollout vinyl pipe used in irrigation, and labor; costs for fuel, repair, and maintenance of machinery and irrigation systems; cost of hauling harvested seed; and interest on operating capital. Weed management costs after planting were calculated for each treatment and are shown in Table 3. These costs included charges for herbicides, surfactants, and their application and POST cultivation in WRs. All application costs included both variable and fixed expenses associated with tractors and sprayers. Planting costs associated with NRs were $4 to $6 ha-1 greater than those for WRs. Fixed expenses were ownership costs for tractors, self-propelled harvesters, tillage implements, sprayers, and the irrigation system. Costs of variable inputs and machinery were based on prices paid by Mississippi farmers each year. Irrigation costs were based on a 65-ha furrow irrigation setup and included an annualized cost for the engine, well, pump, gearhead, generator, fuel tank and lines, and land leveling. Total fixed costs of the irrigation system consisted of annual depreciation, interest on investment, and insurance. Machinery ownership cost was estimated by computing the annual capital recovery charge for each machine and applying its per-hectare rate to each field operation. Insurance was estimated at 1% of the original investment. Within the nonirrigated and irrigated environments, expenses for both cultivars within a WTRT and year were essentially the same.
|
Analysis of variance [PROC MIXED (SAS Inst., 1996)] was used to evaluate the significance of treatment effects on weed cover, plant height, seed yield, and net returns within the separate nonirrigated and irrigated experiments. Analyses across years treated year as a fixed effect to determine interactions involving year. Analyses for individual years treated RW, cultivar, and WTRT as fixed effects. Mean separation was achieved with an LSD0.05. Data tables have footnotes that denote significant main effects and interactions with an LSD. If no LSD is given, then the effect or interaction did not have a significant effect at P 0.05.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Overall Analyses
Analysis of all variables across years showed that year significantly interacted with one or more of the other factors. Therefore, results from statistical analyses are presented by year. Because nonirrigated and irrigated production systems are separate, results were not analyzed across the two. Cost for weed management in each WTRT within each RW was the same for each cultivar within each year. Therefore, expense associated with each WTRT is shown only for each RW (Table 3). Because each sub-subunit received the same weed management across replicates, all differences in weed management expense are significant.
Nonirrigated
Plant Height
Row width significantly affected plant height only in 1994 when plants in WRs averaged 94 cm and those in NRs averaged 88 cm. In 1995 and 1996, plants in NRs averaged 81 and 82 cm, respectively, while those in WRs averaged 79 and 84 cm. Plants of DP 3589 averaged 19, 28, and 9 cm taller than plants of P 9592 in 1994, 1995, and 1996, respectively. Thus, the objective of having two cultivars different in stature was realized.
Weed Management Expense and Weed Cover
Weed management in NRs costed more than that in WRs in 2 of the 3 yr (Table 3). Thus, the banding of herbicides plus POST cultivation resulted in lower weed management cost, which concurs with results from previous research (Buhler et al., 1997; Krausz et al., 1995). In the first year of the study, weed management in WTRT 4 (POST broadleaf and POST grass herbicides) cost the most, and weed management in WTRT 1 (PRE and POST broadleaf herbicides) cost the least. Thereafter, WTRT 2 (PRE broadleaf, PRE grass, and POST broadleaf herbicides) cost the most while WTRT 3 (POST broadleaf herbicides) cost the least.
Weed cover at harvest was significantly affected by RW and cultivar each year and by WTRT in 1994 and 1995 (Table 4). In 1994, average weed cover was 6% in WRs and 1% in NRs. Barnyardgrass [Echinochloa crus-galli (L.) Beauv.] was the dominant grass while prickly sida (Sida spinosa L.) was the dominant broadleaf. Differences between cultivars and among WTRTs, though significant, were small in this nonirrigated environment.
|
5%, and all differences were small. In 1996, WRs and P 9592 had the greater weed cover values. Average cover was 7% in WRs and 4% in NRs and 8% for P 9592 and 3% for DP 3589. Prickly sida was the dominant broadleaf weed, and red sprangletop [Leptochloa filiformis (L.) Beauv.] was the dominant grass weed.
Seed Yield and Net Return
In 1994, the WTRT main effect and the RW x cultivar interaction significantly affected yield (Table 5). The 2430 kg ha-1 average yield from WTRT 1 (PRE and POST broadleaf herbicides) was greater than that from the other WTRTs. Average yield from P 9592 in NRs (2375 kg ha-1) exceeded that from P 9592 in WRs (2130 kg ha-1) while average yields from DP 3589 in NRs (2495 kg ha-1) and WRs (2400 kg ha-1) were not significantly different. Average yield from DP 3589 (2445 kg ha-1) was greater than that from P 9592 (2255 kg ha-1).
|
|
|
Net return in 1995 was affected by the RW x WTRT interaction (Tables 6 and 7), and differences between or among treatment combinations were the same when using either the annual price or the average price. Average net returns from WTRTs 1, 2, and 4 in WRs exceeded those from WTRT counterparts in NRs while average net returns from WTRT 3 in NRs and WRs (POST broadleaf herbicides) were not different. In essence, net returns from WTRTs in WRs exceeded those from WTRTs in NRs in all but one case. This resulted from the higher expense for weed management in NRs than in WRs (Table 3), with insufficient yield increase to cover the additional cost. Within RW, average net return from WTRT 3 in NRs was greater than that from the other WTRTs in NRs while in WR, net return from WTRT 3 was greater than that from WTRTs 1 and 2. Expense associated with weed management in WTRT 3 (Table 3) was the least in both RWs. Thus, use of only POST broadleaf herbicides resulted in greatest net return, and the use of grass herbicides was not necessary for highest net returns.
In 1996, both RW and cultivar significantly affected yield, with average yield from NRs (2270 kg ha-1) exceeding that from WRs (1865 kg ha-1) and average yield from DP 3589 (2155 kg ha-1) exceeding that from P 9592 (1980 kg ha-1) (Table 5). When using the annual price, RW and cultivar significantly affected net return (Table 6), whereas all three main effects significantly affected net return when using the average price (Table 7). Average net return from NRs exceeded that from WRs. This resulted from greater yield from NRs because there was only a small difference in weed management expense between the two RWs (Table 3). Average net return from DP 3589 was greater than that from P 9592. Using the 3-yr average price, average net return of $236 ha-1 from WTRT 3 (POST broadleaf herbicides) was greater than that from WTRT 2 but not greater than returns from WTRTs 1 and 4 (Table 7). As in 1995, use of both PRE and POST broadleaf herbicides and grass herbicides was not necessary for highest net returns.
Irrigated
Plant Height
Plants in NRs averaged 93, 80, and 96 cm in height in 1994, 1995, and 1996, respectively, while plants in WRs averaged 91, 78, and 94 cm in height for the same years. The differences between RWs were not significant. Swanton et al. (1998) measured no significant difference in height of soybean grown in 19- and 76-cm WRs. Average height of DP 3589 plants was a significant 21, 30, and 10 cm greater than that of P 9592 plants in 1994, 1995, and 1996, respectively.
Weed Management Expense and Weed Cover
Cost for each WTRT within each RW was the same for both cultivars within each year. Therefore, expense associated with each WTRT is shown only for each RW (Table 3). Because each sub-subunit received the same weed management across replicates, all differences in weed management expense are significant.
Weed management in NRs cost more than that in WRs for all years (Table 3). Thus, banding herbicides plus POST cultivation resulted in lower cost for WRs. As shown earlier, this is common when economic comparisons are made between NR and WR systems. In the first year, weed management in WTRT 4 (POST broadleaf and POST grass herbicides) cost the most, and weed management in WTRT 1 (PRE broadleaf and POST broadleaf herbicides) cost the least. Thereafter, WTRT 2 (PRE broadleaf, PRE grass, and POST broadleaf herbicides) cost the most while WTRT 3 (POST broadleaf herbicides) cost the least.
Weed cover at harvest was significantly affected by RW and cultivar each year and by WTRT in 1994 and 1995 (Table 4). In 1994, weed cover averaged 11% in WRs and 2% in NRs. Weed cover averages in P 9592 and DP 3589 were 9 and 4%, respectively. The relatively high cover values in WRs of P 9592 resulted from infestations with barnyardgrass, browntop millet [Brachiaria ramosa (L.) Stapf], and prickly sida in WTRTs 1 and 3 and prickly sida and barnyardgrass in WTRTs 2 and 4.
In 1995, the RW x cultivar and RW x WTRT interactions were significant for weed cover. For P 9592, average cover was 18% in WRs compared with 9% in NRs, whereas average cover values in both NRs and WRs were 6% for DP 3589. Weed cover values for WTRT 3 in NRs of both cultivars were relatively high while cover values for all treatments in WRs of P 9592 were relatively high. These high values resulted from a predominant infestation with pitted morningglory (Ipomoea lacunosa L.). Barnyardgrass and browntop millet were again the dominant grasses.
In the 1996 studies, WRs and P 9592 had the greater weed cover values. Average cover was 11% in WRs and P 9592 and 5% in NRs and DP 3589. Barnyardgrass and browntop millet were the dominant grasses, and pitted morningglory and prickly sida were the dominant broadleaves.
All WTRTs resulted in excellent weed control at the end of the treatment period each year (before canopy closure in WRs and before irrigation). These results indicate that late-season weed infestations were more problematic with the shorter P 9592, especially in the WR environment where the final soybean canopy was incomplete and soil moisture from irrigation enhanced weed growth. Weed cover was greater in WRs regardless of weed management. Nelson and Renner (1998) and Swanton et al. (1998) reported that weed control by herbicide treatments in studies conducted at more northern latitudes (Michigan and Ontario) was also enhanced in 19- vs. 75-cm RWs.
Seed Yield and Net Return
In 1994, the RW and WTRT main effects significantly affected yield (Table 5). Average yield of 3365 kg ha-1 from NRs was greater than the 3075 kg ha-1 from WRs, presumably because of the better weed control in NRs (Table 4). Average yields from WTRT 1 (PRE and POST broadleaf herbicides) and WTRT 2 (PRE broadleaf, PRE grass, and POST broadleaf herbicides) exceeded average yields from WTRT 3 (POST broadleaf herbicides) and WTRT 4 (POST broadleaf and grass herbicides).
Net returns were significantly affected by the WTRT main effect (Tables 6 and 7). Average net return from WTRT 1 was the highest, and average net return from WTRT 4 was the lowest. Differences in net returns among WTRTs resulted from both yield differences (Table 5) and differences in weed management expenses (Table 3). Use of grass herbicides did not enhance either yield or net return.
In 1995, both yield and net return were significantly affected by the RW x WTRT interaction (Tables 5, 6, and 7). Across RWs, average yields from WTRTs 1, 2, and 4 in NRs exceeded those from WR counterparts (Table 5). Within NRs, average yield from WTRT 1 exceeded average yield from WTRTs 3 and 4. Across RWs, average net return from WTRT 1 in NRs exceeded that from WTRT 1 in WRs (Tables 6 and 7). Within NRs, average net returns from WTRTs 1, 3, and 4 exceeded average net return from WTRT 2. Within WRs, average net returns from WTRTs 1, 2, and 4 were less than average net return from WTRT 3. Thus, yield differences between RWs and among WTRTs did not translate into the same differences in net returns between RWs and among WTRTs because of the differing weed management expenses associated with the WTRTs (Table 3). As in 1994, use of grass herbicides did not enhance either yield or net return.
In 1996, all main effects significantly affected both yield and net return (Tables 5, 6, and 7). Average yield from NRs (3225 kg ha-1) exceeded that from WRs (2855 kg ha-1) (Table 5), and average net return from NRs was greater than that from WRs (Tables 6 and 7). Average yield and net return from DP 3589 exceeded those from P 9592. Yields and net returns from WTRTs 1 and 3 exceeded those from WTRT 4. As in the previous years, use of grass herbicides (WTRTs 2 and 4) provided no benefit for either yield or net return.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In both the nonirrigated and irrigated environments, use of NRs vs. WRs resulted in less weed cover at soybean maturity (Table 3). Others have reported varying degrees of enhanced weed control in NRs vs. WRs (Mickelson and Renner, 1997; Nelson and Renner, 1998). Thus, use of NR culture for soybean production under conditions represented by those at this site should enhance weed management efforts although costs associated with weed management will be greater in NRs than in WRs. This implies that a small yield advantage for NRs, such as that reported from Nebraska (Elmore, 1998), should be evaluated economically to determine the true worth of the yield increase. These results could temper the results from research conducted in northern Arkansas (Oriade et al., 1997) where net returns from NRs consistently exceeded those from WRs. In that study, weed management was the same across both NRs and WRs; that is, all herbicides were broadcast, and WRs were cultivated. Thus, the advantage of banding herbicides and the subsequent lower expense for weed control in WRs was not realized.
In both nonirrigated and irrigated plots, both cultivars grown in NRs yielded as much as or more than when both were grown in WRs. These results are somewhat different from those of Devlin et al. (1995), who determined that soybean yield response to RW depended on seasonal rainfall amount, with NRs producing higher yields with high rainfall and WRs producing higher yields with low rainfall.
In the nonirrigated environment, the shorter statured P 9592 provided greater net return when grown in NRs rather than in WRs in all WTRTs, whereas DP 3589, the taller cultivar, produced greater net returns from NRs in WTRTs 1 and 3. In the irrigated environment, net returns from both cultivars grown in NRs were greater for all WTRTs except DP 3589 in WTRT 2. Over the 3 yr, and using annual prices, average net returns from using NRs vs. WRs were $22 and $31 ha-1 greater in nonirrigated and irrigated plots, respectively. Using the 3-yr average price, differences were $20 and $30 ha-1, respectively.
Regardless of RW or cultivar, use of only broadleaf herbicides applied either PRE and POST (WTRT 1) or POST (WTRT 3) in both nonirrigated and irrigated plots resulted in greatest net return under the field conditions at this site. Use of grass herbicides (WTRTs 2 and 4) provided no net return benefit in any year, regardless of RW or cultivar. This is in spite of the relatively high infestations with annual grasses in some irrigated WTRT treatments, which indicates that these late infestations with annual grass species did not provide competition for soybean. Johnsongrass (Sorghum halepense L.) cover was relatively high (8%) only in irrigated WTRT 3 of P 9592 in NRs in 1996. Again, its late appearance evidently did not affect soybean yield. In previous research at this location, use of grass herbicides in irrigated WRs for 4 yr did not affect yields (Heatherly and Elmore, 1991). If johnsongrass had been present, or if population densities of annual grasses such as barnyardgrass had been greater, these results may have been different.
Johnson et al. (1997) inferred from results of research conducted on sites with annual grasses in southeast Missouri that both POST broadleaf and grass weed control is necessary to reduce variability in yield of no-till soybean grown in 19-cm RWs. Swanton et al. (1998) concluded from research conducted in southern Ontario that highest gross returns and the most risk-efficient weed management on clay soils resulted from application of PRE grass and broadleaf herbicides. However, the above studies did not include a treatment without grass herbicides. The conclusion drawn from the present study does not mean that grass weed control will not be necessary, but rather that it can be done on an as-needed basis with POST herbicides in both nonirrigated and irrigated systems. Thus, the POST-only weed management, represented by WTRT 3, offers the best opportunity for maximum profit because any grass weed management will be done by POST herbicides only when needed. As mentioned earlier, the POST weed management inputs used in this study were determined to be necessary based on weed presence; i.e., they were not frivolous. However, further research may identify POST weed management options that are more economically optimum than those used in this research.
Yields from nonirrigated environments in this study are common for the region (Heatherly, 1999a). They indicate that, regardless of RW, minimum inputs represented by those in the lowest-cost WTRTs offer the least risk for soybean production where seasonal drought is common, weeds are moderately competitive, and commodity prices are low. This agrees with earlier findings from a WR study at this location (Heatherly et al., 1994).
These results indicate that NRs should be used for soybean culture in the midsouthern USA even though weed management expense will be greater. A decrease in cost of POST herbicides and/or an increase in fuel prices could change the relationship between NR and WR weed management costs. Also, the advent of glyphosate-tolerant cultivars will add a new dimension to PRE vs. POST weed management strategies in soybean.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. J. Heitholt, J. B. Farr, and R. Eason Planting Configuration x Cultivar Effects on Soybean Production in Low-Yield Environments Crop Sci., August 1, 2005; 45(5): 1800 - 1808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. G. Heatherly, K. N. Reddy, and S. R. Spurlock Weed Management in Glyphosate-Resistant and Non-Glyphosate-Resistant Soybean Grown Continuously and in Rotation Agron. J., March 1, 2005; 97(2): 568 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. G. Heatherly, S. R. Spurlock, and K. N. Reddy Influence of Early-Season Nitrogen and Weed Management on Irrigated and Nonirrigated Glyphosate-Resistant and Susceptible Soybean Agron. J., March 1, 2003; 95(2): 446 - 453. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. G. Heatherly, C. D. Elmore, and S. R. Spurlock Weed Management Systems for Conventional and Glyphosate-Resistant Soybean with and without Irrigation Agron. J., November 1, 2002; 94(6): 1419 - 1428. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. G. Heatherly, S. R. Spurlock, and C. D. Elmore Row Width and Weed Management Systems for Early Soybean Production System Plantings in the Midsouthern USA Agron. J., September 1, 2002; 94(5): 1172 - 1180. [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 | |||
