Top Navigation Bar

Article Request Page ASABE Journal Article

Variability in Harvest Moisture and Dry-Down in Multi-Hybrid Planting Systems

J. K. Ward, W. B. Henry, M. W. Hock


Published in Transactions of the ASABE 59(5): 1111-1115 (doi: 10.13031/trans.59.11572). Copyright 2016 American Society of Agricultural and Biological Engineers.


Submitted for review in September 2015 as manuscript number MS 11572; approved for publication by the Machinery Systems Community of ASABE in July 2016.

The authors are Jason K. Ward, ASABE Member, Assistant Extension Professor, Department of Agricultural and Biological Engineering, W.Brien Henry, Associate Professor, Department of Plant and Soil Sciences, and Matthew W. Hock, ASABE Member, Graduate Research Associate, Department of Plant and Soil Sciences, Mississippi State University, Mississippi State, Mississippi. Corresponding author: Jason K. Ward, Box 9632, Mississippi State University, Mississippi State, MS 39762; phone: 662-325-3280; e-mail: jason.ward@msstate.edu.

Abstract.  Significant differences among corn hybrid dry-down rates have been well documented. With the development of multi-hybrid planting systems, these differences become more important because crop moisture content has direct influence on harvest performance, yield estimation, and postharvest management. Therefore, the objectives of this study were to assess field drying performance of commercial corn hybrids across a range a relative maturities and to estimate additional variability that could be added to the production system due to differences in dry-down in a multi-hybrid system. A subset of two adjacent studies in which corn hybrids were planted on the same day, at the same population, and managed with the same practices were selected. Samples were collected weekly at 93, 100, 107, 114, 122, and 129 days after planting. Four ears were randomly collected from each plot, manually shelled, and oven-dried according to standard methods. Calculated wet-basis moisture contents were analyzed using repeated measures in a mixed model. There were significant differences in grain moisture content by hybrid and days after planting. The interaction effects were also significant. As crops approached harvest readiness, there were up to six percentage points of moisture content difference among hybrids. Differences of this magnitude influence yield monitor accuracy and can drastically change harvest and postharvest performance. Regression analysis indicated that the mean drying rate for all hybrids was one percentage point per day. This rate was higher than expected and did not capture differences in instantaneous drying rate that could exacerbate differences in field drying. The magnitude of the differences in moisture content indicates that care must be taken when choosing hybrids to pair in a single field when using multi-hybrid planting systems. Additional measures of drying performance may be needed to minimize unintended downstream effects of harvest variability.

Keywords.Corn, Dry-down, Field drying, Multi-hybrid, Planting.

Significant differences in field dry-down rates among corn (Zea mays) hybrids have been documented for over 50 years (Hillson and Penny, 1965). However, only within recent years has planting technology advanced to such a place to allow prescription-based multi-hybrid planting within the same field. The goal of these technologies is to place aggressive or strong-performing hybrids in field locations that can support high productivity demands. Likewise, a defensive or low-input hybrid can be selected for field locations that consistently underperform. Different hybrids are chosen to optimize yield potential, but additional variability is being added to the production system. Notably, variability in harvest readiness, as determined by crop moisture content, will need to be managed.

Corn moisture content exhibits moderate spatial dependence within a field (Miao et al., 2006). Miao et al. (2006) applied a split-planter approach to compare hybrids. Moisture content was significantly different by hybrid and was more variable than yield. If weather conditions are conducive to field drying, then the effects of hybrid differences are not as pronounced (Troyer and Ambrose, 1971). However, each hybrid has physiological differences that can directly influence grain dry-down rates, such as husk dry weight (Kang et al., 1986). Husk coverage of the ear and tightness of the husk can also influence the rate of in-field grain dry-down (Nielson, 2012). The differences among hybrids become more critical when the crop directly interfaces with equipment during the harvest process.

Moisture content has a direct influence on the harvest readiness and harvest quality of crops. Crops that are too wet do not separate from unwanted plant material easily. The harvested crop costs more to dry and will degrade faster than a drier crop. Crops that are too dry may result in increased harvest losses attributable to shattering and ear drop; overly dry grain can be damaged during harvest and has greater exposure to field risks such as weather, ear rot, and animal predation prior to harvest. Changes in harvest moisture can directly influence the accuracy of yield monitors for harvest yield estimation. Proper calibration includes harvesting across a range of mass flow rates, which is usually achieved by altering the combine speed. Manufacturers suggest that yield monitors should be calibrated multiple times in the harvest season as the crop dries. Moisture content changes of greater that 5% require recalibration of the yield monitors; a 1% error in moisture resulted in a yield error of 2.5 bu acre-1 in on-farm tests (Taylor et al., 2011). The ASABE Standard for yield monitor performance testing suggests using a range of moisture con-tents if a moisture sensor is used to calculate mass flow and that crop moisture should be reported for all tests (ASABE, 2012). Therefore, it is important that variability introduced to harvest readiness and moisture content be examined in the context of multi-hybrid planting systems.

Objectives

The objectives of this study were to (1) evaluate the variability of in-field dry-down across a range of modern commercial corn hybrids and (2) estimate the additional variability that these differences could contribute to harvest readiness in a multi-hybrid planting system.

Materials and Methods

Experimental Design

Samples were collected from a subset of two adjacent corn studies evaluating the effects of planting date and seeding density on a range of commercial corn hybrids with varying maturity characteristics (table 1). The studies were located at the R. R. Foil Plant Science Research Center on the campus of Mississippi State University near Starkville, Mississippi. The selected plots were all planted on 21 April 2014. Plots were established with corn planted 6.3 cm (2.5 in.) deep using a four-row John Deere 7100 MaxEmerge vacuum planter (Deere and Co., Moline, Ill.) in slight excess of the target density at 25,076 plants acre-1 and hand-thinned to the desired population of 25,000 plants acre-1. The plots consisted of four rows, each 97 cm (38 in.) wide by 9.1 m (30 ft) long. Field experiments were planted in a Marietta fine sandy loam (fine-loamy, siliceous, active, Thermic Fluvaquentic) following a previous crop of cotton in 2014 (NRCS, 2015).

Table 1. Description of hybrids.
BrandHybridTechnology
Traits[a]
Relative Maturity
(days)
Mid-Silk
(GDU)[b]
Black Layer
(GDU)[b]
DekalbDKC 67-57GENVT3P11713502925
DekalbDKC 69-29GENVT3P11913592975
PioneerP-1498HX1/LL/RR211413702700
PioneerP-1319HX1/LL/RR211314002730
SyngentaAGR-N68B-31131111114052580

    [a]    GENVT3P = Genuity VT Triple PRO, HX1 = Herculex, LL = Liberty Link, RR2 = Roundup Ready Corn 2, and 311 = Agrisure Viptera 311.

    [b]    GDU = growing degree units for corn at base temperature of 10°C (50°F).

All of the plots received the same agronomic and pest management treatments. Pre-plant soil samples were taken for both studies, and soil test results indicated that all extractable nutrients other than nitrogen (N) were within recommended levels. N was applied with a four-row liquid fertilizer applicator equipped with coulter-knives approximately 20 cm from the center row in a split application of 224 kg ha-1 using 32% urea ammonium nitrate (UAN) solution. The first application of N was applied post-emergence to plants at the 3 to 4 leaf stage. The second N application was applied at the 6 to 8 leaf stage.

Figure 1. Plot and sample area layout (not to scale).

Weed management for both experiments consisted of a pre-emergence application of Roundup PowerMax and Halex GT at recommended label rates. Post-emergence weed control entailed an additional Roundup PowerMax application as needed per label recommendations. Field preparation for each location consisted of using a chisel plow to break the soil to a depth of 20 cm in the fall. Fall bed preparation was accomplished using a packer/roller to flatten the tops of the rows in order to have a wider surface for spring planting.

The studies were designed as a randomized complete block with four replications. Each experimental plot consisted of four rows; the middle two rows were designated for yield measurement, and the outer rows were used for sample collection (fig. 1). For each sampling date, four ears were selected from each plot, two randomly from each of the two outer rows. Samples were collected at least five feet into row. Border plots were placed around the study location so that edge effects were minimized.

Spring weather conditions at the study site were generally considered normal (table 2). Early spring conditions were wet and cold; however, late-season conditions were less representative of what has been seen in recent years. Rainfall was more frequent in the latter part of the growing season, especially June. The overall temperatures in 2014 were lower, and nighttime temperatures appeared especially lower, than what is considered normal. Rainfall was below the 30-year average for March, May, July, and August. However, April was 45% and June was 49% above the 30-year average. Monthly temperatures averaged below the 30-year average during March, April, and July. Temperatures during May and June were slightly higher, averaging between 0.2°C and 0.5°C above the 30-year average. August mean temperatures were normal.

Sample Collection and Processing

Samples collection started on 23 July 2014, which was 93 days after planting (DAP). Sampling continued every subsequent week thereafter on 30 July, 6 August, 13 August, 21 August, and 28 August 2014. Including the first date, sampling events occurred on 93, 100, 107, 114, 122, and 129 DAP, respectively. Harvested ears for most of the plots were immature at the initial sampling dates, which was not conducive for shelling. Therefore, ears were frozen to allow shelling. All collected samples from all events were thereafter frozen to prevent bias. Ears were manually shelled using hand corn shellers (Item No. 115, Seedburo Equipment Co., Des Planes, Ill.). Kernels from all four ears from each plot were comingled. In duplicate, approximately 100 g of kernels from each plot were weighed and placed in a drying oven at 103°C for 72 h according to standard methods (ASABE, 2008).

Analysis

Wet-basis moisture content (MCwb) was calculated from measured wet and dry weights. The data were arcsine transformed because they were proportional and limited between 0 and 0.1 (0% to 100%). Data were analyzed as a mixed model with random replication effects in SAS (ver. 9.3, SAS Institute, Inc., Cary, N.C.). Repeated measures were based on sampling event, and the subject was each individual plot. The autocovariance model was selected to minimize Akaike’s corrected information criterion (AICC) using a method similar to that of Littell et al. (2000). A macro was used to divide treatment means (Saxton, 1998). All determinations of significant difference occurred at the 0.05 significance level.

A linear regression was fitted to each hybrid’s weekly mean MCwb. The regression was intended to asses mean dry-down rate (MCwb percentage points per day) and to calculate estimated days for each hybrid to reach 25%, 20%, and 15% MCwb (D25, D20, and D15, respectively). Each regression’s suitability was assessed using R2 and standard errors.

Results and Discussion

There were significant differences (p < 0.0001) in moisture content over time among the assessed hybrids (fig. 2). Expectedly, DAP and the interaction between DAP and hybrid were also significant (both p < 0.0001). Generally, MCwb descended linearly for each hybrid over time. For each sampling event other than the initial sampling, there were significant differences among hybrids. The relative maturity of the selected hybrid does not provide any additional insight into harvest readiness. At the last sampling event, two hybrids with 111 and 114 day relative maturity were at the same MCwb, while three hybrids with 113, 117, and 119 day relative maturity were at a higher MCwb. The lack of information regarding field drying as related to relative maturity was expected due to the lack of standardization of maturity determination within the seed industry. The best comparisons of maturity occur with a particular genetics provider’s offerings, rather than comparing across providers. Additionally, the unit of days attached to the numerical relative maturity value is not completely accurate and is arbitrary (Nielsen, 2012).

Figure 2. Wet-basis moisture content by days after planting. Error bars indicate standard errors. Means are divided within sampling events.

The naturally occurring variability in corn MCwb across a field could exacerbate or mitigate hybrid differences, as documented above. At 122 DAP without additional variability, a field planted in hybrids with similar dry-down characteristics as DKC69-29 (26% MCwb) and AGR-N68B (20% MCwb) resulted in a six percentage point difference in MCwb. Differences of this magnitude could directly influence harvest performance, drying management, storage, and market-ing strategy. The estimated allowable storage time to minimize dry matter loss, assuming standard conditions at 15.6°C (60°F) air temperature and no additional damage, ranges between 8 and 29 days at the previously described moisture levels (ASABE, 2010). Stored grain quality and value could be negatively influenced by MCwb differences this large. Additionally, error could be introduced in calculating harvested bushels from yield monitor data, which could create discrepancies between estimated harvest yield data and marketed yield records depending on the sampling rate and calculation method used to estimate dry or standard yield. Capacitance sensors have proven sensitive to mass density or crop maturity (Reyns et al., 2002). Therefore, when choosing multiple hybrids to plant in the same field, attention must be paid to the dry-down qualities of the selected hybrids.

Regressions constructed from the mean weekly MCwb over DAP were highly linear (table 3). Regression slope and intercept p-values for all hybrids indicated significance at or below a = 0.05. Standard errors of the regressions ranged from 1.3 to 1.7 percentage points, which is comparable to the standard error of 1.88 percentage points for the isotherm equation for corn used to estimate equilibrium moisture content (ASABE, 2007). The regressions indicated that on average all of the hybrids lost one percentage point of MCwb per day during in-field drying. This drying rate was above the accepted rule-of-thumb in the southern U.S., which is 0.6 to 0.7 percentage points per day (Larson, 2015). Therefore, an average seven-day period should result in seven percentage points of MCwb loss. The average value, however, did not take into account additional variables, such as weather conditions or disease incidence. The range of MCwb lost over a normalized seven-day period during field drying was between 5 and 13 percentage points of moisture. An already heterogeneous MCwb was subject to additional variability by inconsistent instantaneous drying rates as opposed to average rates. Regressions were only considered valid during corn stages in which kernel MCwb was decreasing predominately by moisture loss from the kernel. The regression was considered trustworthy after late dough stage (R4); previous to R4, regression outputs are likely not valid. The regression intercepts ranged from 1.47 to 1.55 decimal MCwb, which equates to 147% to 155% MCwb, which is not physiologically feasible if the regressions is considered valid to the first day of planting.

Table 3. Mean MCwb linear regression by DAP.
HybridSlope
(MCwb d-1)
Intercept
(MCwb)
R2SE
(MCwb)
D25
(d)
D20
(d)
D15
(d)
DKC67-57-0.0101.410.990.014121126132
DKC69-29-0.0101.440.990.014123128133
P-1498-0.0101.490.990.013119124129
P-1319-0.0101.470.980.016123128133
AGR-N68B-0.0111.550.990.017119124129

The regressions were algebraically transformed to output the estimated days to reach a target MCwb (%). Inverting the regression reduces systematic error; however, the standard errors of the regressions can still provide error bounds to estimate the target moisture range. For example, for hybrid P-1319, the regression predicts that 123 days are required to reach 25% MCwb. The standard error for this hybrid’s regres-sion was 1.6% MCwb; therefore, at a predicted 123 days, the grain moisture could range from 26.6% to 23.4% MCwb. The range of times to meet a selected MCwb, as discussed earlier, does not appear to be informed by the published relative maturities of the selected hybrids. Differences in geographic location between where relative maturities were estimated and where the study was established could lead to differences. To minimize variability in harvest readiness, hybrids with similar dry-down characteristics should be chosen for use in the same field.

Choosing hybrids with overlapping moisture ranges at a particular DAP or desired target harvest MCwb will be an important component of hybrid selection in multi-hybrid planting systems. Ideally, these data could be used to minimize in-field variability of MCwb and still allow for a well-staged harvest across fields. In the age of big data and rapid hybrid development, additional indicators of relative harvest readiness may need to be developed and should be readily available to better manage a newly developing planting system through harvest and into successful delivery of a high-quality product.

Conclusions

The five commercial corn hybrids with a range of relative maturities assessed in this study exhibited significantly different dry-down performances. The range of moisture contents was large enough that harvest performance, yield monitor accuracy, and postharvest management could all be negatively affected. Relative maturity did not provide adequate information to create compatible pairings in multi-hybrid planting systems. The mean field-drying rate for all hybrids was approximately one percentage point of MCwb per day. This drying rate was higher than the expected rate of 0.6 to 0.7 percentage points MCwb per day typical in the southern U.S. Despite the similarities of the mean drying rate, actual weekly moisture loss ranged from 5 to 13 percentage points. Natural variability in crop moisture content could be further compounded by the addition of multiple hybrids into the same field. Therefore, hybrids with similar dry-down performance should be paired when implementing multi-hybrid planting technology. Additional indicators of drying performance may be useful and could be integrated into algorithms to better pair hybrids.

Acknowledgements

Funding for this study was graciously provided by the Mississippi Corn Promotion Board. Thanks must be given to the staff of the R. R. Foil Plant Science Research Center for their time and effort. Additional thanks go to Ms. Christina Cooper, Mr. Caleb McKee, and the ABE shop for their support in collecting and processing samples.

References

ASABE. (2007). D245.6: Moisture relationships of plant-based agricultural products. St. Joseph, MI: ASABE.

ASABE. (2008). D241.4: Density, specific gravity, and mass-moisture relationships of grain for storage. St. Joseph, MI: ASABE.

ASABE. (2010). D271:2: Psychrometric data. St. Joseph, MI: ASABE.

ASABE. (2012). S579.1: Yield monitor field test engineering procedure. St. Joseph, MI: ASABE.

Hillson, M. T., & Penny, L. H. (1965). Dry matter accumulation and moisture loss during maturation of corn grain. Agron. J., 57(2), 150-153. http://dx.doi.org/10.2134/agronj1965.00021962005700020007x

Kang, M. S., Zuber, M. S., Colbert, T. R., & Horrocks, R. D. (1986). Effects of certain agronomic traits on and relationship between rates of grain-moisture reduction and grain fill during the filling period in maize. Field Crops Res., 14, 339-347. http://dx.doi.org/10.1016/0378-4290(86)90068-7

Larson, E. (2015). When should I start corn harvest? Mississippi State, MS: Mississippi State University Extension. Retrieved from http://www.mississippi-crops.com/2015/07/31/when-should-i-start-corn-harvest/

Littell, R. C., Pendergast, J., & Natarajan, R. (2000). Modelling covariance structure in the analysis of repeated measures data. Stat. Med., 19(13), 1793-1819. http://dx.doi.org/10.1002/1097-0258(20000715)19:13<1793::AID-SIM482>3.0.CO;2-Q

Miao, Y., Mulla, D. J., & Robert, P. C. (2006). Spatial variability of soil properties, corn quality, and yield in two Illinois, USA, fields: Implications for precision corn management. Prec. Agric., 7(1), 5-20. http://dx.doi.org/10.1007/s11119-005-6786-2

Nielsen, R. L. (2012). Interpreting corn hybrid maturidy ratings. West Lafayette, IN: Purdue University. Retrieved from https://www.agry.purdue.edu/ext/corn/news/timeless/HybridMaturity.html

NRCS. (2015). Web soil survey. Washington, DC: USDA Natural Resource Conservation Service. Retrieved from http://websoilsurvey.nrcs.usda.gov/

Reyns, P., Missotten, B., Ramon, H., & De Baerdemaeker, J. (2002). A review of combine sensors for precision farming. Prec. Agric., 3(2), 169-182. http://dx.doi.org/10.1023/a:1013823603735

Saxton, A. M. (1998). A macro for converting mean separation output to letter groupings in Proc Mixed. Proc. 23rd SAS Users Group Intl. Conf. Cary, NC: SAS Institute, Inc.

Taylor, R., Fulton, J., Mullenix, D., Darr, M., McNaull, R., Haag, L., & Staggenborg, S. (2011). Using yield monitors to assess on-farm test plots. ASABE Paper No. 1110690. St. Joseph, MI: ASABE.

Troyer, A. F., & Ambrose, W. B. (1971). Plant characteristics affecting field drying rate of ear corn. Crop Sci., 11(4), 529-531. http://dx.doi.org/10.2135/cropsci1971.0011183X001100040019x