|
|
|
© 2003 Plant Management Network. Yield Penalty Due to Delayed Weed Control in Corn and Soybean Stevan Z. Knezevic, Assistant Professor, Sean P. Evans, Former Graduate Student, and Mike Mainz, Research Technician, Haskell Agricultural Laboratory, University of Nebraska, 57905 866 Rd., Concord 68728-2828 Corresponding author: Stevan Knezevic. sknezevic2@unl.edu Knezevic, S. Z., Evans, S. P., Mainz, M. 2003. Yield penalty due to delayed weed control in corn and soybean. Online. Crop Management doi:10.1094/CM-2003-0219-01-RS. Abstract With the widespread use of glyphosate-resistant crops, two common questions related to a weed control program have emerged: when is the most appropriate time for a postemergence weed control operation; and what is the cost of a delayed weed control operation? The first question was addressed using the concept of critical period for weed control. Our field studies determined the effects of three N rates on the critical period for weed control in dry land corn and of three row spacings on the critical time for weed removal in dry land soybean. When data were averaged over years and locations, the study in corn concluded that critical period for weed control ranged from V1 to V11, V3 to V10, V4 to V9 and V6 to V9 for N rates of 0, 55, 110, and 210 lb/acre, respectively. The study in soybean suggested that critical time for weed removal coincided with V3, V2, and V1 for soybean row spacing of 7.5, 15, and 30 inches, respectively. The second question was addressed by pooling yield loss data across locations-years from both studies and related to the extrapolated crop growth stage at the time of weed removal for both corn and soybean. The 2% yield loss per every leaf stage of delay past the critical time for weed removal was determined as the cost of delaying weed control in both corn and soybean. This recommendation is applicable from the critical time for weed removal up to canopy closure in corn (about 10 fully developed leaves) and the R3 stage in soybean (beginning pod) and for the fields with moderate to high weed densities (e.g., 30 to 100 plants per square yard). If weed control is delayed further than these indicated stages, the yield losses would be much higher than suggested, especially under drought conditions or higher weed densities. Low commodity prices and high input costs force producers to explore ways to optimize their crop production systems. With the widespread use of glyphosate-resistant crops, especially soybean, the two common questions that can help optimize the weed control program are: (i) when is the most appropriate time for a post emergence weed control operation; and (ii) what is the cost of a delayed weed control operation. One of the first steps in designing an optimized weed control system is to identify the critical period for weed control, the period in the crop growth cycle during which weeds must be controlled to prevent yield losses (8). In general, the critical period for weed control has its beginning and end, and weeds that emerge before or after critical period for weed control may not represent a threat to crop yield. In addition, the beginning of critical period for weed control is also commonly called the critical time for weed removal. In essence, it is a period in the crop growth cycle when weed control must be initialized to improve yield potential. The length of critical period for weed control is influenced by the cropping practices, such as nitrogen (N) fertilizer (1,2) and crop row spacing (9,10). We used this concept to develop a simple tool that can help practitioners make decisions on the need for and timing of weed control (8,4). Currently, no such tool is available for weed control in general. Therefore, the objectives of this research were: (a) to develop a simple decision aid concept for determining the best time for weed control; and (b) to provide information that can guide practitioners to determine the yield loss caused by delayed weed control in corn as affected by N level and in soybean as influenced by row spacing. The data for this paper were from field studies conducted in eastern Nebraska on the effect of three N rates on the critical period for weed control in dry land in corn (1) and of three row spacings on the critical time for weed removal in dry land soybean (9). Due to budget and labor constrains the soybean study was designed to determine only the beginning stage of the critical period for weed control in soybean (e.g., critical time for weed removal). The corn study (1) was conducted in 1999 and 2000 at the Haskell Agricultural Laboratory near Concord, Nebraska; and at the University of Nebraska Agricultural Research and Development Center near Mead, Nebraska. The experiment was a split-plot design with the N level as the main plot factor and the timing of weed removal as subplots. There were four replications. The three N rates included 0, 55, and 110 lb N per acre. An additional study was conducted at Mead in 1999, using a high N rate of 210 lb N per acre (unpublished data). There were also weed-free and weedy control treatments. The soybean study (9) was conducted in 2000 and 2001 at the Haskell Agricultural Laboratory near Concord, Nebraska; and in 1999, 2000, and 2001 at the University of Nebraska Agricultural Research and Development Center near Mead, Nebraska. The experiment was a split-plot treatment design with four replications where row spacing was the main plot factor and the timing of weed control was the subplots. Three row spacings of 7.5, 15, and 30 inches were used. Soybeans were each planted at the density of 175,000 seeds per acre regardless of the row spacing. In both corn and soybean studies, a naturally occurring population of mixed weed species was utilized to obtain appropriate duration of weedy periods with most weeds emerging within a couple of days of crop emergence (± 2 days). Weed removal in both studies was conducted by application of commercially formulated glyphosate at the labeled rate of 32 oz per acre with ammonium sulfate (2 lb/acre) as an adjuvant. If it was necessary an additional weeding was done by hoeing. The timing of weed removal in both studies (1,9) was according to the crop developmental stages in the season-long weed-free treatments. To characterize the early-season competitive environment at each site, total mean density for each weed species was measured in all plots at the V3 stage in both crops. In addition, weed density and species composition were also assessed in all plots prior to every timing of weed removal (not shown). Weed counts were conducted at two locations within each subplot, using a 0.25 yard2 quadrant. The species composition in the corn study included velvetleaf (Abutilon theophrasti Medicus), pigweed species (Amaranthus spp.) and foxtails (Setaria spp.). The total combined density of weed species, ranged from 185 to 208 seedlings per square yard. The species composition in soybean study included velvetleaf, pigweed species, common lambs quarters (Chenopodium album), and foxtails under total combined densities ranging from 85 to 93 shoots per square yard. For additional information about each of the two studies see references 1 and 9. In addition, a series of data on ancillary variables was collected from each experimental plot to help determine potential sources of variation for each site before data were pooled. The ancillary variables such as weed densities, species composition, relative emergence times, and environmental conditions provided background information that quantified the competitive environment under which the experiments were conducted. In order to determine a generalized critical period for weed control in corn and critical time for weed removal in soybean, as well as the yield penalty due to delayed weed control, the yield loss data from both studies (1,9) were pooled over years-locations and related to the extrapolated crop growth stage at the time of weed removal for both corn and soybean. Crop yields and relative yields were subjected to an overall ANOVA using PROC MIXED (12) in SAS (15). Relative yield of each experimental plot was calculated as a percent of the corresponding weed-free yield for each level of main factor (e.g., row spacing in soybean and N level in corn). The significance of the main factor levels was evaluated at the P = 0.05 level. Air growing degree days (GDD) were related to the relative yields of both crops. The GDD corresponding to the beginning and the end of critical period for weed control in corn and critical time for weed removal in soybean was then related to crop growth stage. The potential yield losses (relative yields) for each location-year combination were then calculated and extrapolated for each corn (V1 to V10) and soybean (V1 to R3) stages using regression models presented in elsewhere (1,9). Yield loss data were then combined from all locations-years into a single data set for each crop. A four-parameter asymmetric sigmoidal regression was used to describe the effect of increasing duration of weed interference on the above mentioned yield losses, and to determine the yield penalty of delayed weed control for each row spacing in soybean and N level in corn: Y = A + (B-A) / ( 1 + ((CGS /C)) - P [1] where Y is crop yield loss (percent of season-long weed-free yield); CGS is the duration of weed interference measured by crop growth stage; B is the point of inflection in unit for yields; A, C and P are constants. From a practical standpoint, an arbitrary level of either 2, 5, or 10% yield loss can be used to signify the critical time for weed removal. This range can allow a practitioner to adjust timing of weed control depending on the risk one is willing to take. In this paper, an arbitrary level of five percent yield loss was used to determine the beginning of critical period for weed control in both crops (see the 5% yield-loss-line at Figs. 1 and 2). A five percent acceptable yield loss level was selected assuming: (i) statistical significance at the 5% level; (ii) reliable detection of the treatment effects on crop yields due natural variability in field; and (iii) acceptable loss with producers. Parameter estimates for asymmetric sigmoidal regressions were determined using FIG-P computer statistical software. Significant differences between the regression lines describing row spacings in soybean and N level in corn were determined by comparing parameter estimates and respective standard errors (± S.E.) using t-tests at the 5% significance level. If all regression coefficients of a model were not significantly different based on standard errors, the data could have been combined among row spacings in soybean or N levels in corn.
Critical Period for Weed Control in Corn as Affected by N We reported that the critical period for weed control in corn was influenced by the level of N fertilizer (1) and that the critical period for weed control varied among years and locations; however the trends were identical. When data were combined, the critical period for weed control ranged from V1 to V11, V3 to V10, V4 to V9 and V6 to V9 for N-rates of 0, 55, 110, and 210 lb/acre, respectively (Table 1). These results indicate that an increase in the supply of N increased the tolerance of corn to the early season weed presence. The mechanisms by which the addition of N increased corn tolerance to weeds are not completely understood. We reported that the addition of N increased early season corn growth rates and leaf area expansion, suggesting that the supply of N available to a crop significantly improved corn competitiveness against weeds (3). Table 1. Critical period for weed control in corn expressed as crop leaf stage and days after crop emergence (DAE) as affected by the level of nitrogen (N) fertilizer.
Critical Time for Weed Removal in Soybean as Affected by Row Spacing We reported that the critical time for weed removal in soybean was influenced by the row spacing, the critical time for weed removal was very similar among years and locations, and that the trends were identical (9). When data were combined for this paper, the critical time for weed removal coincided with the V1 (1st trifoliate), V2, and V3 stages of soybean in 7.5, 15, and 30 inches of row spacing, respectively (Table 2). These results indicate that the soybean crop planted in 30-inch-wide rows was least tolerant to early season weed presence compared to 15- and 7.5-inch rows. The mechanisms by which the wider rows reduced soybean tolerance to weed interference are not completely understood. We also suggested that perhaps wider rows provided additional space for early-season weed growth and later canopy closure, which influenced change in the quality and quantity of light available for weed growth due to crop shading effects (9). Consequently, weeds were better competitors earlier in wide than in narrow rows, resulting in an earlier critical time for weed removal due to weed competitive advantage for soil and light resources. It was also observed that even though weeds were present in the narrow-row plots at similar densities as in other row spacing, they were not growing as vigorously, and thus consequently not being as competitive against the crop in 7.5 inches as in 15- and 30-inch-wide rows. This is an area that needs further investigation. Table 2. The critical time for weed removal in soybean expressed as crop leaf stage and days after crop emergence (DAE) as affected by the row spacing.
Yield Cost Due to Delayed Weed Control The general trends of yield reduction with delayed weed control were similar both in corn and soybean (Figs. 1 and 2). The regression lines were significantly different from each other based on the difference in the regression coefficients (Tables 3 and 4). An average of 2% yield loss per every leaf stage of delay past the critical time for weed removal (5% yield loss) was determined as the cost of delaying weed control in both corn and soybean (Figs. 1 and 2). For example, the critical time for weed removal in 7.5 in. rows soybean was the V3 (third trifoliate) stage. If weed control was delayed to the V4 (fourth trifoliate) the yield loss was about 7%, costing about 2% in yield losses due to prolonged competition from weeds (Fig. 2). The same was true if weed control were delayed past the recommended critical time in other row spacings in soybean (Fig. 2). The yield losses in corn ranged from 1.5 to 3% depending on the N levels (Fig. 1). Higher yield penalty occurred at zero lbs of N (~3% per every leaf stage) than at 210 lb of N (~1.5%); however, the 2% yield penalty remained relatively stable across commonly used N rates (e.g., 55 to 110 lb/acre). Table 3. Parameter estimates (standard error) by N level (lb/acre) for the four-parameter regression model characterizing the influence of the duration of weed interference on the yield loss in corn (Fig. 1).
*Within a column the same letter indicates that the parameter values did not differ significantly according to a t-test (P = 0.05). Table 4. Parameter estimates (standard error) by row spacing (inches) for the four-parameter regression model characterizing the influence of the duration of weed interference on the yield loss in soybean (Fig. 2).
*Within a column, the same letter indicates that the parameter values did not differ significantly according to a t-test (P = 0.05). The determined 2% yield penalty is suggested for use as a general rule-of-thumb, in order to help producers make decisions on timing post emergence weed control to verify their own intuition. This recommendation is applicable from the critical time for weed removal up to canopy closure in corn (about 10 fully developed leaves) and the R3 stage in soybean (beginning pod) and for the fields with moderate to high weed densities (e.g., 30 to 100 plants per square yard). If the weed control is delayed further than the indicated stages, the yield losses would be much higher than suggested, especially under drought conditions and higher weed densities. If the weed densities were lower than indicated, then the anticipated yield losses would be smaller. In addition, the actual economic losses per every leaf stage of delayed weed control can be easily calculated using anticipated crop price and potential yield. For example, economic losses in corn could be about $4/acre for every corn leaf stage of delay, assuming a price of $2/bu and a yield of 100 bu. In soybean, it could be about $5/acre for every soybean leaf stage of delay, assuming a price of $5/bu and a yield of 40 bu/acre. We believe that the 2% yield penalty per every crop leaf stage of delayed weed control past the critical time for weed removal is a reasonable prediction rule with moderate to heavy weed infestations. It is well known, for example in corn, that yield potential can be significantly reduced during three major phases of crop growth and development: growing point determination; pollination; and grain filling period. Any significant stress, including weed competition, can reduce crop yield during those stages (14). The following example illustrates potential corn yield loss in a hypothetical scenario utilizing a 2% yield loss rule from weeds not controlled: (a) by the time of canopy closure (e.g., 10 leaves), (b) by the tasseling period, or (c) during the end of grain fill period. If the weeds are not controlled for the first 10 leaves of corn growth, (e.g., canopy closure in 30-inch-wide rows, scenario a) the potential for yield reduction is about 20% (e.g., 2% per every leaf stage). Most corn hybrid in major corn growing areas develop 18 to 21 leaves, thus if the weeds are not controlled for another 8 to 11 leaves it is possible to have another 20-25% yield loss. Therefore, if the weeds are not controlled by tasseling time (scenario b), potential yield losses can be 40 to 50%. Furthermore, if the weeds are not controlled during the grain-fill period it is possible to have another 30 to 40% yield loss (scenario c). In summary, combining the hypothetical yield loss values from all three scenarios suggests possible yield losses ranging from 70 to 90% , which had been observed in many studies of crop-weed competition (7,11). Therefore, we suggest that the 2% yield penalty per every leaf stage of delay is a reasonable guideline for predicting yield penalty due to delayed weed control past the critical time for weed removal in both corn and soybean. It is a simple tool that can help in making the decision on the need for and timing of POST herbicide. Movement towards the use of POST herbicides with little or no residual activity has resulted in renewed interest in determining the most appropriate timing and periodicity for weed control, especially in systems involving the use of herbicide tolerant crops. The popularity of crops tolerant to glyphosate, especially soybean, has generated many studies across the United States (2,5,13). Some reported the timing of weed removal based on weed height (5), while others used weeks or days after crop emergence (6), and crop growth stage (2,13). It was reported (4) that weed heights varied considerably among years and locations due to variable mixture of weed species, differences in the relative time of weed emergence and environmental and soil variables. We suggest that weed height does not provide sufficient information for timing weed control efforts unless it is coupled with crop growth stage. The practical value of weed height information is primarily for adjusting the dose of herbicide (e.g., smaller weeds may require lower than the label rate). Therefore, we recommend that, from the practical standpoint, the timing of weed control should be based primarily on the crop growth stage. Weeds are controlled because we try to protect the crop, therefore the crop should be the focus of the program (8). Summary and Conclusions Results presented confirm that the crop row spacing and nutrient management can significantly influence crop-weed interference relationships. Obviously, the timing for weed removal were affected by N rate in corn and row spacing in soybean, while the 2% yield penalty remained relatively stable thereafter. The documented differences in the critical period for weed control in corn and in the critical time for weed removal in soybean highlight the importance of integrating decisions regarding cropping practices and the timing of weed control into integrated weed management (IWM) programs. With the growing popularity of herbicide-tolerant crops (HTC), dependence upon total POST programs will likely become more prevalent in future cropping systems. Such a shift in cropping practices highlights the importance of appropriately timed weed control. Therefore from a practical standpoint, reducing N rate in corn and planting wider soybean rows may warrant more intensive weed management (e.g., weed control measures applied several times). Since glyphosate-resistant crops, especially soybeans, have received high levels of acceptance, the concepts of critical period for weed control and critical time for weed removal are an important part of integrated weed management in answering a fundamental question of if and when to apply POST herbicide. A generally sound strategy in glyphosate-resistant soybean would be to apply glyphosate tank-mixed with a residual herbicide at the critical time for weed removal, which should provide adequate weed control the entire critical period. Acknowledgments Published as University of Nebraska Agricultural Research Division Journal Series No. 13887. Many thanks to Ray Brentlinger for assisting with plot maintenance and treatment implementation. Literature Cited 1. Evans, S. P., Knezevic, S. Z., Shapiro, C., and Lindquist, J. L. 2003. Nitrogen level affects critical period for weed control in corn. Weed Sci., in press. 2. Evans, S. P., and Knezevic, S. Z. 2000. Critical period of weed control in corn as affected by nitrogen supply. Proc. North Cent. Weed Sci. Soc. 55:151. 3. Evans, S. P., Knezevic, S. Z., Shapiro, C., and Lindquist, J. L. 2002. Influence of nitrogen level and duration of weed interference on corn growth and development. Weed Sci., in press. 4. Evans, S. P. 2001. Effects of varying N supply on the critical period for weed control in corn (Zea mays L.). M.S. thesis, University of Nebraska, Lincoln. 5. Gower, A., Loux, M. M., and Cardina, J. 1999. Determining the critical period of weed management in glyphosate tolerant corn. Proc. North Cent. Weed Sci. Soc. 54:66. 6. Halford, C., Hamill, A. S., Zhang, J., and Doucet, C. 2001. Critical period of weed control in no-till soybean and corn. Weed Tech. 15:737-744. 7. Hall, M. R., Swanton, C. J., and Anderson, G. W. 1992. The critical period of weed control in grain corn (Zea mays). Weed Sci. 40:441-447. 8. Knezevic, S. Z., Evans, S. P., Blankenship, E. E., Van Acker, R. C., and Lindquist, J. L. 2002. Critical period for weed control: The concept and data analysis. Weed Sci. 50:773-786. 9. Knezevic, S. Z., Evans, S., and Mainz, M. 2002. Row spacing influences critical time of weed removal in soybean. Weed Tech., in press. 10. Knezevic, Z. S., Evans, S., and Mainz, M. 2001. Yield penalty due to delayed weed control in corn and soybean. Proc. North Cent. Weed Sci. Soc. 56:106 11. Knezevic, Z. S., Weise, S. F., and Swanton, C. J. 1994. Interference of redroot pigweed in corn. Weed Sci. 42:568. 12. Littell, R. C., Milliken, G. A., Stroup, W. W., and Wolfinger, R. D. 1996. SAS System for Mixed Models. Cary, NC: SAS Institute, Inc. 13. Mulugeta, D., and Boerboom, C. M. 2000. Critical time of weed removal in glyphosate-resistant Glycine max. Weed Sci. 48:35-42. 14. Ritchie, W. S., Hanway, J. J., and Benson, G. O. 1997. How a corn plant develops. Special Report No. 48. (Revised). Iowa State University of Sciences and Technology, Cooperative Extension Service, Ames, Iowa. 15. Statistical Analysis Systems Institute, Inc. 1999. SAS Online, Version 8. Cary, NC. |
