Yield Loss of Corn Hybrids to Incremental Defoliation

Search PMN

Eric A. Adee, Principal Research Specialist, Northwestern Illinois Agricultural Research and Demonstration Center, 321 210th Ave. Monmouth 61462; Lyle E. Paul, Agronomist, Northern Illinois Agronomy Research Center, 14509 University Road, Shabbona 60550; Emerson D. Nafziger, Professor, and German Bollero, Associate Professor, Department of Crop Sciences, Turner Hall, MC-046, University of Illinois-Urbana/Champaign, 1102 S. Goodwin Ave., Urbana 61801

Abstract

A yield loss model is needed to describe field corn’s (Zea mays L.) response to progressive defoliation caused by defoliating agents such as plant pathogens or insects. Defoliation was achieved by physical removal of leaves at different growth stages starting two weeks prior to tasseling (V14) and continuing through physiological maturity. These treatments created different corn defoliation progress curves (CDPC). Three hybrids responded similarly to defoliation over the five environments tested. The general yield loss model for the regression of percent yield and CDPC was:

y (% yield) = -0.0005858x + 98.7, r² = 0.68.

The regressions for the individual environments were very similar to the general model. Loss of leaves below the ear leaf at silking resulted in 11% yield loss. When defoliation was initiated at dent stage yields were not reduced. This model improves the accuracy of predicting yield loss in field corn to defoliation that becomes progressively severe over time such as foliar diseases.

Introduction

The importance of leaf area of corn for grain production, especially after silking, has been established for many years (2,5,13). Yield loss models for corn in response to defoliation at single events, such as hail, have been available for many years (13,14). However, these models which assess yield loss to defoliation at a single growth stage are not suited for predicting the yield loss of corn from progressive defoliation, such as those caused by plant diseases. Area under disease progress curves (AUDPC) have been used to represent the changes in disease severity over time with plant pathogens (1,9) and would likely be a better predictor for yield losses caused by progressive defoliation.

The AUDPC, a unitless number describing development of defoliation effects over time, is derived by plotting periodic measurements of disease or defoliation severity over time and integrating the area under the curve (7). Such accumulation typically starts from the first rating (appearance of symptoms) and ends at an arbitrary point, often the time of the last rating. Instead of accumulating defoliation severity over calendar days as the AUDPC does, it is possible with corn to express defoliation as it accumulates over growing degree days (GDD) remaining until the crop reaches physiological maturity. Using GDD instead of calendar days more accurately relates leaf loss to the development of the corn crop under variable growing conditions and maturities of hybrids (8,11). Thus, calculating a corn defoliation progress curve (CDPC) using GDD remaining until maturity more accurately reflects the proportion of the growing season over which the ability of the plant to fill grain is decreased by defoliation.

Predicting yield loss in field corn to defoliators, such as foliar plant pathogens and insects, that progressively increase in severity over time, has been difficult because of the challenges in establishing the epidemics needed for field testing models. Several researchers have concluded that artificial inoculation to initiate epidemics of gray leaf spot (GLS), caused by the fungus Cercospora zeae-maydis (13), is not realistic due to the highly variable nature of inoculum efficiency and epidemic development in field conditions (10; P. E. Lipps, personal communication, 1997; D. G. White, personal communication, 1997). Also, it is extremely difficult to achieve the same degree of defoliation on hybrids differing in disease resistance with artificial inoculations of a pathogen. For that reason, physical removal of the leaves or portions of leaves to mimic the progressive defoliation over time on different corn hybrids may be preferable.

We recognize the limitations to using the progressive physical removal of leaves to mimic progressive defoliation patterns of foliar pathogens. One limitation is that shading of lower leaves by a diseased leaf or portion of a leaf that has ceased producing photosynthate; a factor that is eliminated by leaf removal. Physical removal of leaves also eliminates toxins produced by an infecting agent (4) or metabilites as a factor from the plant in response to the infection (3). Harvest losses due to increased stalk rot and lodging, which can be indirectly related to defoliation by foliar pathogens were not factored into this model. While these models may underestimate yield loss due to infectious agents, we consider this was the best method to derive a yield loss model for progressive defoliation of corn in the field.

It is not known if hybrids respond the same to progressive defoliation, especially if there are differences among the hybrids in their response to a defoliating disease. As an example, GLS disease begins on the lower leaves and progresses to the upper leaves if the environmental conditions are favorable. Yield loss to GLS depends on the amount of defoliation in relation to crop development; however, some hybrids can produce greater yields than others with similar levels of disease severity, thus appear to be tolerant to GLS (16). It is not known if the relationship between defoliation and crop development is consistent between corn hybrids.

The objectives of this study were to: (i) determine the effect of leaf loss at different growth stages on yield, (ii) determine if yield losses from defoliation are consistent among hybrids, and (iii) determine if leaves below the ear leaf are important to yield.

Defoliation Experiments

This study was conducted from 1997 through 1999 at Northern Illinois Agronomy Research Center, Shabbona, IL (DeKalb), and in 1997 and 1999 at Northwestern Illinois Agricultural Research and Demonstration Center, Monmouth, IL (Monmouth); the 1998 study at Monmouth was lost due to greensnap. Greensnap can occur when corn has been growing very rapidly, becoming brittle, and breaks over near the crown in windy conditions as experienced at Monmouth in 1998. The previous crop was corn in all environments except DeKalb in 1997, when it was soybean. The studies were planted no-till in 1997 at Monmouth and in 1998 and 1999 at DeKalb, but followed tillage in the other two environments.

Defoliation treatments involved the removal of 20% of the leaf area at one or more of the following corn growth stages: V14, R1 (silking), R2 (blister), R3 (milk), and R5 (dent stage) in 1997, and V14 through milk stage (11) in 1998 and 1999 (Table 1). Other treatments included continued removal of 20% of the leaf area at subsequent growth stages resulting in defoliation ranging from 20 and 100% in 1997 and 20 and 80% in 1998 and 1999.

Table 1. Description of timing and severity of corn defoliation treatments.

Defoliation treatment^a	Crop growth stage when 20% of leaves removed					Total % defoliation
Defoliation treatment^a	V14	R1 (silk)	R2 (blister)	R3 (milk)	R5 (dent)	Total % defoliation
V	X					20
VS	X	X				40
VSB	X	X	X			60
VSBM	X	X	X	X		80
VSBMD	X	X	X	X	X	100
S		X				20
SB		X	X			40
SBM		X	X	X		60
SBMD		X	X	X	X	80
B			X			20
BM			X	X		40
BMD			X	X	X	60
M				X		20
MD				X	X	40
D					X	20

^a Defoliation at growth stages V = V14, S = Silking, B = Blister, M = Milk, D = Dent.

At each defoliation, between two and three leaves were removed to achieve 20% leaf area removal. Defoliation was accomplished by cutting perpendicularly across the leaf blade at the leaf collar, leaving the leaf sheath intact. Defoliation treatments began at the ear leaf and proceeded up the plant to simulate the progression of foliar diseases through the corn canopy. All leaves below the ear leaf were removed at silking from all plots being defoliated. The exception was Monmouth in 1997, whereby defoliation treatments started at the lowest viable leaf, generally the fifth or sixth leaf. Defoliation was timed to the growth stage of each hybrid and varied as much as seven days between hybrids. Two additional treatments were also included: plants with no defoliation (i.e., representing the untreated control) and plants defoliated below the ear leaf at silking. The decision to include removal of leaves below the ear leaf was based on current recommendations to apply fungicides when foliar diseases such as GLS were at the ear leaf or above two weeks prior and two to three weeks after silking (6,12). The removal of leaves below the ear would help determine the contribution of the lower leaves to yield, and reduce the variability of lower leaves that may or may not be healthy and contributing to yields of different hybrids. The treatment with plants defoliated below the ear leaf at silking was used as the control for the defoliation treatment starting at the ear leaf and above each year for the DeKalb studies and the Monmouth 1999 study. These two treatments were also imposed to determine the contribution of the leaves below the ear to grain yield. Defoliated plants were located in the center 10 ft segments of the middle two rows of each subplot. The border rows and plants on each end of the middle two rows were not defoliated in order to reduce the influence of neighboring plots, such as increased light infiltration. Yields were normalized as percent of appropriate controls for regression analysis.

The experimental design was a randomized complete block using three replicates in a split-plot design, with defoliation treatment as the main plot and corn hybrids as the subplots. The subplot size was 15 ft long by 10 ft wide, four 30-inch-wide rows per plot. Three corn hybrids differing in relative maturity and resistance to GLS -- Pioneer hybrids 3394 (2660 GDD, susceptible); 3489 (2630 GDD , moderately resistant); and 33Y18 (2710 GDD, moderately resistant) -- were over planted and hand-thinned to 26,000 plants per acre (relative maturity and GLS resistance information from Pioneer Hi-Bred International, Inc. product guides and representatives). Planting dates were 23 May 1997, 21 May 1998 and 22 May 1999 at DeKalb, and 29 April 1997 and 11 May 1999 at Monmouth. Weeds and insects were chemically controlled with recommended herbicides and insecticides, and hand-weeded as necessary.

All treatments were protected from foliar fungi with one or two applications of propiconazole (Tilt) at 4 oz/acre, at the onset of disease and until 50% brown silks, if necessary. Disease ratings were made of the leaf area infected on the ear leaf and leaves above the ear at dent stage in the control plots. Grain yields were determined by hand harvesting and shelling grain from 18 ears from the middle 6 ft of both treated rows, and expressed in bu/acre at 15% moisture grain.

Equivalent CDPCs for each defoliation treatment were derived by plotting the severity of defoliation against the GDD remaining, based on seed company information, from each defoliation event until physiological maturity for each hybrid. The GDD for a single day is calculated by subtracting 50 from the average daily temperature (°F). The maximum average daily temperature used is 86 and the minimum used is 50. The GDDs necessary for different hybrids to reach physiological maturity are readily available and could be used in deriving CDPCs throughout the season.

Statistical Analysis

Analysis of variance and regression analyses were performed on grain yield and yield loss using the PROC GLM and REG procedures of SAS 8.1 (SAS Institute, Cary, NC). The analysis for yield loss was conducted on the combined data from five environments.

Modeling Yield Loss of Corn to Defoliation

The model predicting the relationship between incremental defoliation, as represented by CDPC, and percent of control treatment yield over three corn hybrids in five environments was:

y (% yield) = -0.0005858X + 98.7, r² = 0.68,

where X = CDPC (a unitless number describing defoliation severity over potential grain fill period remaining) (Fig. 1). This model allows for an estimate of yield loss any time during the season based on the severity of defoliation and growth stage or percent of grain fill remaining. For example, the expected percent yield loss of a hybrid requiring 2700 GDD (50 to 86°F) to reach maturity has 25% defoliation with 1440 GDD needed to reach maturity can be derived as follows:

77.6% yield = ((1440 × 25) × -0.0005858) + 98.7;

thus expected percent yield loss is 22.4%; (100 - 77.6 = 22.4% yield loss). This yield loss model, if used in conjunction with other factors such as the cost of disease control and grain price, could provide the economic thresholds to ensure that control measures are profitably employed. Yield declined linearly with defoliation initiated at V14 through milk stage (Table 2). The yield loss associated with any level of defoliation did not differ between V14 and milk growth stages. Defoliation after the milk stage resulted in less yield loss than at the earlier growth stages. But if leaf area reductions were delayed until the dent stage, then grain yields were similar to the control treatments at Monmouth and DeKalb in 1997 (Table 2). Grain fill had apparently progressed sufficiently by dent stage that additional defoliation did not limit yield further. It had previously been reported that dry matter accumulation begins to slow about 45 days after silking (5). Dent stage averaged 39 days after silking in these studies in 1997. Therefore, defoliations at dent stage were not included in subsequent years.

Fig. 1. Regression of field corn yield as percent of controls of three hybrids against Corn Defoliation Progress Curve (CDPC) at five environments. Severity of defoliation (percent leaf area) was plotted against Growing Degree Days (GDD) remaining to maturity and integrated to calculate CDPC for defoliation treatments for each hybrid.

Table 2. Yield (bu/acre) of corn in response to progressive defoliation initiated at different growth stages in five environments.

Defoliation treatments^b	Mon 97^a	DK 97	DK 98	Mon 99	DK 99	Aver.
No leaves removed^c	109	192	209	188	180	180
Leaves below ear removed^d	100	160	193	193	151	159
V	110	124	170	169	126	140
VS	96	107	143	136	110	118
VSB	76	89	110	82	91	90
VSBM	57	68	84	71	79	72
VSBMD	57	69
S	119	123	151	170	133	140
SB	93	99	120	133	117	112
SBM	64	94	118	104	100	98
SBMD	70	84
B	98	130	142	171	138	136
BM	114	109	135	147	117	124
BMD	95	109
M	105	135	170	175	127	145
MD	106	133
D	120	181
LSD (P=0.05)	25	10	27	20	20	11

^a Mon = Monmouth, DK = DeKalb.

^b Defoliation at growth stages V = V14, S = Silking, B = Blister, M = Milk, D = Dent.

^c The treatment with no defoliation was used as the control for Monmouth in 1997.

^d The treatment with plants defoliated below the ear leaf at silking was the control for DeKalb 1997-1999 and Monmouth 1999.

Influence of Environment on Yield Loss to Defoliation

Environments did not have a large affect on the percent yield loss as only the regression slope for the Monmouth 1999 location was different from the slopes for the other environments (Table 3). While yields of the control treatments for each environment ranged from 109 to 193 bu/acre, the percent yield loss to defoliation relationship was consistent. For example, corn yields at Monmouth in 1997 were 30 to 40% less than the other location years, however, the regression for yield loss was not different from three of the four other environments. The lower yields at Monmouth in 1997 were somewhat influenced by growing season precipitation (Table 4), but they were also influenced by no-till planting following corn. Corn yields were higher if planted in tilled soil, or following soybeans, or with adequate moisture in July (7) with the other environments. However, environments should be considered random factors, hence the general model can be used to predict yield loss to progressive defoliation over a relatively wide range of conditions.

Table 3. Regression models for percent corn yield loss related to CDPC for individual environments

Environment	Slope estimate	Intercept	Model fit
Environment	CDPC	Yield (% of control)	R²
Monmouth 97	-0.000536	98.0	0.49
DeKalb 97	-0.000586	99.0	0.73
DeKalb 98	-0.000547	94.5	0.68
Monmouth 99	-0.000733	103.9	0.82
DeKalb 99	-0.000547	98.3	0.67
General model	-0.000586	98.7	0.68
LSD (P = 0.05)	0.000129	5.9

Table 4. Growing season precipitation (inches) by month for DeKalb and Monmouth.

Month	1997		1998	1999		1971-2000 average
Month	DeKalb	Mon	DeKalb	DeKalb	Mon	DeKalb	Mon
April	2.07	2.03	5.01	5.21	7.60	3.27	3.76
May	2.72	3.90	2.92	3.95	2.92	3.98	4.27
June	2.15	3.61	4.27	4.91	4.51	4.20	4.26
July	0.67	1.46	4.31	3.34	2.34	3.74	4.33
Aug.	4.49	9.53	4.36	2.28	3.14	4.31	4.02
Sept.	4.36	2.77	2.72	5.53	5.19	3.64	3.45
Total	16.46	23.3	23.59	25.22	25.7	23.14	24.11

Hybrids Response to Defoliation

All hybrids responded similarly to the defoliation treatments; there was no interaction (P = 0.29) between defoliation and hybrid for yield loss (data not shown). Averaged over all treatments, 3489 yielded more than 3394 and 33Y18 (134, 125, and 126 bu/acre, respectively). Previously it had been reported that corn hybrids that yield more than other hybrids, even under the same degree of GLS severity, can be classified as tolerant (16). However, this data suggests that the yield differences may in part be simply due to differences in yield potential of hybrids and not differences in their response to defoliation.

Timing to Control Defoliation

The hybrids lost 11.7% in yield when the leaves below the ear leaf were removed at silking (157 bu/acre) compared to untreated control (177 bu/acre) (Table 2). Allison and Watson (2) reported that the top eight or nine leaves contribute 75 to 90% of the grain fill. This agrees with the data from our study, in which the eight to ten leaves remaining at the ear and above contributed 88% of the grain yield. However, the importance of the lower leaves to grain fill, and to the epidemic development of some foliar diseases cannot be overlooked. Ward et al. (16) found that in environments conducive for disease development, a single application of fungicide is most effective in reducing yield loss to GLS when lesions are first observed on the five basal leaves of the plant. Depending on the yield potential, the price of grain and the cost of control measures an 11% increase in grain from protecting the lower leaves could be economically justified in the Corn Belt, where changing weather conditions can halt or slow disease from reaching the upper leaves. If the conditions are favorable and the disease is present on the lower leaves prior to the reproductive stages of the crop, our data and that from previous work (15) indicate that control measures may be economically feasible even if the environmental conditions are less favorable for foliar disease development later in the season.

The application of propiconazole, a systemic fungicide, was very effective in protecting the leaves throughout grain fill. At dent, over five weeks after the last fungicide application, there was an average of 6% leaf area infested above the ear of the most susceptible hybrid, 3394, across all environments (data not shown). The primary foliar disease in these environments was GLS, with some northern corn blight (Exserohilum turcicum (Pass.) Leonard & Suggs) and common rust (Puccinia sorghi Schw.) observed. The fungicide helped limit the confounding effects of foliar diseases on the different hybrids with the defoliation treatments.

Conclusions

A yield loss model for corn that is applicable to progressive defoliation was derived from the data gathered from these corn defoliation studies. Yield loss models are essential for developing economic thresholds, which are necessary to make environmentally and economically sound decisions regarding treatments. These data will assist in the development of treatment recommendations that include stage of crop and severity of defoliating agent. The consistency of the model with the three different hybrids and the different environments indicates that this model should be applicable for corn under a variety of conditions.

Acknowledgments

This work was funded in part by Pioneer Hi-Bred International, Inc. through the Pioneer Crop Management Research Awards program. We would like to thank Marty Johnson and Shelly Adee for their assistance in the preparation of this manuscript.

Literature Cited

1. Adee, E. A., and Pfender, W. F. 1989. The effect of primary inoculum level of Pyrenophora tritici-repentis on tan spot epidemic development in wheat. Phytopathology 79:873-877.

2. Allison, J. C. S., and Watson, D. J. 1966. The production and distribution of dry matter in maize after flowering. Ann. Bot. N.S. 30:365-381.

3. Birch, R. G. 2001. Disease resistance. Pages 356-363 in: Encyclopedia of Plant Pathology. Vol.1. O. C. Maloy and T. D. Murray, eds. John Wiley & Sons, Inc. New York.

4. Desjardins, A. E. 2001. Toxins. Pages 1040-1041 in: Encyclopedia of Plant Pathology. Vol. 1. O. C. Maloy and T. D. Murray, eds. John Wiley & Sons, Inc. New York.

5. Eik, K., and Hanway, J. J. 1966. Leaf area in relation to yield of corn grain. Agron. J. 58:16-18.

6. Kirby, H. W. 1997. Conditions favorable for Gray Leaf Spot. Coop. Ext. Serv., Univ. Ill., Urbana-Champaign. Pest Manage. Dis. Bull. No. 16, July 11, 1997.

7. Mengistu, A., and Grau, C. R. 1987. Seasonal progress of brown stem rot and its impact on soybean productivity. Phytopathology 77:1521-1529.

8. Nafziger, E. D. 2002. Corn. Illinois Agronomy Handbook 23rd Edition. Ext. Pub. 1372. Univ. of Illinois Extension Service. Univ. of Illinois at Urbana-Champaign.

9. Nutter, F. W., Jr., and Guan, J. 2001. Disease losses. Pages 340-351 in: Encyclopedia of plant pathology. Vol. 1. O. C. Maloy and T. D. Murray, eds. John Wiley & Sons, Inc. New York.

10. Ringer, E. E., and Grybauskas, A. P. 1995. Infection cycle components and disease progress of gray leaf spot on field corn. Plant Dis. 79:24-28.

11. Ritchie, S. W., Hanway, J. J., Thompson, H. E., and Benson, G. O. 1996. How a corn plant develops. Iowa State Univ. Coop. Ext. Serv. Spec. Rep. 48.

12. Stack, J. 1999. Gray Leaf Spot of corn. Neb. Coop. Ext., Publ. No. G99-1384-A.

13. Vasilas, B. L., and Seif, R. D. 1985. Defoliation effects on two corn inbreds and their single-cross hybrid. Agron. J. 77:816-820.

14. Vorst, J. V. 1993. Assessing hail damage to corn. National Corn Handbook. NCH-1. Univ. Ill., Urbana-Champaign. Coop. Ext. Serv.

15. Ward, J. M. J., Laing, M. D., and Rijkenberg, F. H. J. 1997. Frequency and timing of fungicide applications for the control of gray leaf spot in maize. Plant Dis. 81:41-48.

16. Ward, J. M. J., Stromberg, E. L., Nowell, D. C., and Nutter, F. W., Jr. 1999. Gray leaf spot: A disease of global importance in maize production. Plant Dis. 83:884-895.