Managing for high yield � What worked in 2005?

A review of intensive wheat production research in North Dakota in 2005

March 2006

Compiled by Joel Ransom (joel.ransom@ndsu.nodak.edu)

This article summarizes research conducted during the 2005 growing season with the objective of identifying practices that could be used to improve the productivity of spring wheat.  This research replicates or builds on research conducted in 2003 and 2004.  Summaries of previous research are available at: http://www.ag.ndsu.nodak.edu/aginfo/smgrains/IntensW04.htm (for the 2003 growing season) and http://www.ag.ndsu.nodak.edu/aginfo/smgrains/InntensW05.htm (for the 2004 growing season).  Experiments conducted in 2005 as well as those conducted previous years studied the effect of one or more of the following factors and their interaction on wheat productivity: seeding rate, N rate, N timing, variety, and fungicide. 

Table 1 lists the experiments that were included in this review and the contact details of the principal investigators from whom additional information about the research can be obtained.

Table 1. Wheat management experiments conducted in North Dakota in 2005.

The experiments in most cases were designed so that the effects of each management factor studied could be evaluated as well as the interactions between the various factors.  Positive interactions between management practices are considered desirable, as it means that the output of two or more inputs is greater than the output of each input alone.  Unlike 2003 and 2004 a number of interactions were found to be significant in 2005.  The dominant factor affecting all of the experiments in 2005 was the prevalence of foliar diseases and Fusarium Head Blight (Scab).  Yields generally were much less than those obtained in the previous two years. 

Seeding rate

Increasing the seeding rate from 1.0 million (M) seeds/A to 1.75 million seeds/A increased yield significantly (about 3 bu/A) in the Carrington1 experiment, but not in Carrington2 (Table 2).  The positive effect of increasing the seeding rate at Carrington1 and not Carrington2, may be related to the varieties included (Briggs and Granite in Carrington1 and Glenn, Reeder, and SteeleND in Carrington2).  Varieties, however, did not significantly interact with seeding rates in either of the experiments.

Table 2.  Effect of seeding rate (averaged over other factors such as variety, fungicide and N timing) on wheat yield in ND, 2005

⁺ N.S. means that there is no statistically significant difference between the values within an experiment,
*, ** means there are statistically significance differences at the 5% and 1% level, respectively.

Nitrogen timing

Splitting or delaying the application of N until the 4 leaf stage did not significantly increase yield or grain protein content in the three experiments summarized (Table 3). 

Table 3. Effect of N timing on wheat yield in various experiments in ND, 2005

⁺ N.S. means that there is no statistically significant difference between the values within an experiment.

Fungicide

Applying fungicide (Folicur�, 4 fl oz/A) at flowering (Feekes 10.51 stage) dramatically increased yield in all experiments (Tables 4 and 5).  Moreover, at the Carrington site fungicide significantly reduced DON levels and scab severity (when averaged over varieties) (Table 4).  There was a significant variety by fungicide interaction.  Treating Reeder, the most disease susceptible of the three varieties, with fungicide resulted in a greater yield increase than in the other two varieties.  Nevertheless, even with a fungicide application, Reeder was not able to achieve the yield level of the other varieties.  There was no significant interaction between seeding rate and fungicide, indicating that plots seeded at rates of 1.00 M seed/A and 1.75 M seeds/A responded similarly to fungicide applications (data not shown). 

Table 4. Effect of genotype and fungicide at flowering on yield and DON levels, Carrington, 2005.

⁺ N.S. means that there is no statistical significant difference between the values within an experiment.

In the Prosper experiments (Table 5), yield increases due to fungicide were greater in durum than in HRSW.  Fungicide improved test weights in HRSW but not in durum.  Experiments at this location were designed to evaluate the response of varieties (27 were included in this trial) to more intensive management.  Therefore, plots receiving fungicide were also planted at a higher seeding rate than those not treated with fungicide.  In the HRSW trial, the response in yield of the 5 highest yielding varieties (as a group) to fungicide was much greater than that of the 5 lowest yielding varieties (Table 5).  In the durum experiment, however, the response to fungicides was similar in both groups (approximately 14 bu/A increase).  In both cases, achieving the highest yield was dependant on applying fungicide to genotypes that were high yielding and that possessed a good basic level of disease resistance. 

Conclusions and recommendations

The research in 2005 confirms the results obtained in the previous two years of research with additional observations regarding disease suppression/control: 

• There seems to be only marginal value in increasing seeding rates beyond 1.2 M seeds/acre (perhaps somewhat higher for durum). Avoid over-seeding and use the money saved for other inputs that will likely produce greater returns (i.e. nitrogen or fungicide).
• Based on a reasonable yield goal, apply all nitrogen before planting.  If during the early development of the crop, however, it appears that the yield goal will be exceeded, apply extra nitrogen using streamer bars.  Applications before the six-leaf stage are recommended in order to have the greatest chance of improving yield as well as protein.
• Fungicides have the potential of dramatically increasing yields in years like 2005 that are conducive to disease development (both foliar and scab).  The results from 2005 also demonstrate the importance of using varieties with good genetic disease resistance, even when applying fungicide.  The best yields were obtained when fungicide was applied to varieties that had the best level of disease resistance or tolerance. 

Table 5.  Effect of fungicide and variety on yield and test weight of HRSW and durum,
Prosper, 2005.

⁺N.S. indicates that there is no statistical significance between the values within an experiment. 
 *, ** indicate statistical significance at the 5% and 1% level, respectively.