# Crop Science - Article

1. Vol. 52 No. 3, p. 1316-1329
OPEN ACCESS

Published: May, 2012

* Corresponding author(s):
View
Permissions
Share

doi:10.2135/cropsci2011.05.0239

# Influence of Seeding Rate, Nitrogen Management, and Micronutrient Blend Applications on Pith Expression in Solid-Stemmed Spring Wheat

1. Brian L. Beres *ab,
2. Ross H. McKenziec,
3. Héctor A. Cárcamob,
4. Lloyd M. Dosdalla,
5. Maya L. Evendend,
6. Rong-Cai Yanga and
7. Dean M. Spanera
1. a University of Alberta Department of Agricultural, Food, and Nutritional Science, 410 Ag/Forestry Building, Edmonton, Alberta, Canada T6G 2P5
b Agriculture and Agri-Food Canada, Lethbridge Research Centre, 5403 1st Avenue South, Lethbridge, Alberta, Canada, T1J 4B1
c Alberta Agriculture and Rural Development, Lethbridge Research Centre, 100, 5401 1st Avenue South, Lethbridge, Alberta, T1J 4V6
d University of Alberta Department of Biological Sciences, CW405, Biological Sciences Building, Edmonton, Alberta, T6G 2E9

## Abstract

The wheat stem sawfly (Cephus cinctus Norton [Hymenoptera: Cephidae]) is a serious threat to wheat (Triticum aestivum L.) and other cereal grains in the northern Great Plains. Wheat cultivars with high expression of pith in the culm of the stem (stem solidity) can minimize losses associated with sawfly infestations and subsequent stem boring of the larva. Based on the widespread area now sown to solid-stemmed wheat, our objective was to develop an integrated nutrient and planting strategy specific to solid-stemmed spring wheat using modern farming techniques. Five levels of banded N fertilizer (0, 30, 60, 90, and 120 kg N ha−1) were arranged in a factorial combination with three levels of sowing density (100, 300, and 500 seeds m−2) and grown at three sites in southern Alberta, Canada, from 2007 to 2009. Increased planting densities optimized yield, but an inverse relationship with pith expression (stem solidness) was observed. Low plant populations (100 seeds m−2) were often most effective at maximizing pith expression in solid-stemmed wheat and reducing sawfly cutting damage. However, this usually required the highest rates of N fertilizer, so a system of low seeding rates and high N may not be economical based on fertilizer input costs and the generally lower grain yield response (−9%). An integrated planting and nutrient strategy for solid-stemmed spring wheat cultivars consists of seeding rates no greater than 300 seeds m−2 and basal N applications in the range of 30 to 60 kg N ha−1.

### Abbreviations

EDTA, ethylenediamine tetra-acetate; NDVI, normalized difference vegetation index; WSS, wheat stem sawfly

The wheat stem sawfly (WSS) (Cephus cinctus Norton [Hymenoptera: Cephidae]), is one of the most economically important insect pests of wheat (Triticum aestivum L.) in the northern Great Plains (Beres et al., 2007, 2011c; Weiss and Morrill, 1992). A comprehensive review of WSS biology and management can be found in Beres et al. (2011c); a brief overview is provided here.

Adults emerge from the previous year's crop stubble in late spring to early summer and, following mating, the adult female seeks out a suitable host plant to oviposit, usually an adjacent wheat field (Criddle, 1922). A healthy female can successfully lay up to 50 eggs; thus, the population and subsequent damage to wheat can increase exponentially in a single generation (Ainslie, 1920). Shortly after an egg is deposited into a stem of wheat, a larva will hatch and begin boring the stem (Criddle, 1923). This activity continues throughout the growing season until the host plant reaches physiological maturity. Chlorosis associated with plant ripening and the reduction of whole plant moisture cues the larva to begin preparation to overwinter (Holmes, 1979). The larva moves to the base of the stem, notches a v-shaped groove around the stem, fills the region with frass, and encases itself in a cocoon below the groove. The groove weakens the stem and causes it to easily lodge or topple over, which proves difficult to recover at harvest (Ainslie, 1929).

## RESULTS

Average annual precipitation at Lethbridge and Bow Island during the study period was close to normal but below average rainfall occurred in 2007 at both sites and in 2009 at Bow Island (Table 1). Growing season precipitation was 150% of normal at Lethbridge in 2008. Mean temperature and light intensity were also unique in 2007. Trends in light intensity and temperature were similar at all sites with the notable exception of Bow Island fallow in 2007, where higher light intensity was recorded in August (Fig. 1). Light intensity levels were similar for June, July, and August at the other sites in 2007; however, light intensity peaked in June for 2008 and 2009 at all sites. Temperature also peaked in 2008 and 2009 at all sites but was highest in July in 2007, which was the hottest month of the entire study period (Fig. 1). Background levels of available soil NO3N were generally low at all sites and lowest in the continuously cropped system at Bow Island (Table 1).

Figure 1.

Monthly mean temperature and light intensity received at each study site from June to August in 2007 to 2009 at Lethbridge (continuously cropped) and Bow Island (BI) (wheat-fallow and continuously cropped regimes), AB, Canada. lm, lumens.

Reduced pith expression in solid-stemmed wheat concomitant with increased stem cutting caused by WSS was observed as seeding rates increased over the low rate of 100 seeds m−2 (Table 2; Fig. 2). Single degree of freedom contrast results indicate that the best fit of the response was linear but the significant quadratic response for pith expression may indicate that the downward trend for pith expression was not strictly linear. The average pith rating of main stems was reduced by 13% when planting density was increased to 300 seeds m−2 and an additional 5% when increased to 500 seeds m−2. Similar results were observed for stem diameter. The magnitude of change was greater in stem cutting as visual estimates increased by 38 and 50%, respectively, for moderate and high seeding rates (Table 2). Across all site years, N management practices had no effect on stem solidness but did influence stem diameter and visual estimates for stem cutting. Higher N rates resulted in greater stem cutting and increased stem diameter from 2.17 to 2.44 mm at the highest rate of N (Table 2; Fig. 2). In site–years where N did influence pith expression when averaged over seeding rate (Bow Island stubble 2008 and Bow Island fallow 2007), the response generally resulted in greater pith at 60 to 90 kg N ha−1 vs. the control or 30 kg N ha−1 (data not shown).

View Full Table | Close Full ViewTable 2.

Influence of sowing density and N rate on solid stem spring wheat (cv. AC Lillian) tolerance to wheat stem sawfly in rainfed environments in the Brown and Dark Brown soil zones of southern Alberta, Canada, in 2007 to 2009.

 Factor Treatment Stem solidness (pith expression)† Stem diameter (mm) Sawfly damage (% stems cut) Seed rate, seeds m−2 100 3.03 2.53 24 300 2.66 2.30 33 500 2.50 2.18 36 SED‡ 0.05 0.03 3.52 Error df 16 16 16 p > F <0.0001 <0.0001 0.0111 Contrasts Linear <0.0001 <0.0001 0.0044 Quadratic 0.0353 0.0374 0.3016 Nitrogen rate, kg N ha−1 0 2.66 2.17 22 30 2.69 2.31 28 60 2.78 2.37 31 90 2.77 2.40 37 120 2.76 2.44 39 SED 0.08 0.06 4.60 Error df 32 32 32 p > F 0.4513 0.0035 0.0061 Contrasts Linear 0.1843 0.0113 0.0314 p > F Quadratic 0.8323 0.0923 0.4759 Cubic 0.6480 0.0184 0.0052 Quartic 0.2054 0.0409 0.0813 Seeding rate × fertilizer p > F 0.9408 0.2236 0.7325
Pith rating based on average of stem internodes (1, hollow; 5, solid).
SED, standard error of the difference between two means.
Figure 2.

Summary of main effect means for sawfly resistance parameters.

Yield components and grain quality parameter results are summarized in Table 3 and Fig. 3 and 4. Seeding and N rates affected all traits except grain bulk density (Table 3). With the exception of seed mass and grain bulk density, single degree of freedom contrasts indicate that the response to varying rates of both main effects was primarily linear (Table 3; Fig. 3 and 4). Grain yield and spike density responses were greater from low to moderate seeding rates but diminished from moderate to high seeding rates. Plant establishment decreased from 87% at the lowest plant density to 62% at 300 seeds m−2; however, plant stands increased proportionally at each seed rate level (Table 3). All yield components were optimized at the 60 to 90 kg N ha−1 N rate, but grain protein concentration continued to respond through the 120 kg N ha−1 rate. However, grain protein diminished with increased plant density (Table 3; Fig. 4). According to the biplot summaries, grain yield and protein associated with the lower rates of N were either too low or too variable. Increasing plant density to 300 seeds m−2 produced high and stable grain yield with 60 to 120 kg N ha−1. The same plant density combined with 90 or 120 kg N ha−1 also produced high protein. The low plant density produced high and stable protein at the higher N rates, and the same combinations also produced consistent grain yield, but the results were average or just below average (Fig. 5).

View Full Table | Close Full ViewTable 3.

Response of solid stem spring wheat (cv. AC Lillian) yield components and grain protein concentration to sowing density and N rate in rainfed environments located in the Brown and Dark Brown soil zones of southern Alberta, Canada, in 2007 to 2009.

 Factor Treatment Yield at 14% m.c.† Stand establishment Spike density Seed mass Grain bulk density Grain protein Mg ha−1 plants m−2 heads m−2 g 1000−1 kg hL−1 % Seed rate, seeds m−2 100 2.74 86 293 31.2 70.3 14.2 300 3.01 212 449 29.8 70.6 13.9 500 3.00 310 502 29.1 71.1 13.7 SED‡ 0.11 8.7 18.5 0.35 1.0 0.12 Error df 16 16 16 16 16 16 p > F 0.0425 <0.0001 <0.0001 <0.0001 0.6089 0.0046 Contrasts, p > F Linear 0.0323 <0.0001 <0.0001 <0.0001 0.3268 0.0013 Quadratic 0.1528 0.0743 0.0057 0.3380 09820 0.5019 Nitrogen rate, kg N ha−1 0 2.47 194 345 30.8 71.0 12.6 30 2.82 199 401 30.2 70.2 12.8 60 3.01 208 431 29.8 70.6 13.8 90 3.16 209 455 29.8 71.3 14.9 120 3.14 204 443 29.4 70.1 15.4 SED 0.16 5.4 15.8 0.43 1.1 0.30 Error df 32 32 32 32 32 32 p > F 0.0006 0.0551 <0.0001 0.0329 0.7606 <0.0001 Contrasts, p > F Linear 0.0013 0.0085 <0.0001 0.1030 0.6801 0.0002 Quadratic 0.2160 0.7330 0.0804 0.1899 0.2271 0.9989 Cubic 0.0115 0.6209 0.0007 0.0635 0.7912 <0.0001 Quartic 0.0352 0.1504 0.0044 0.0635 0.9491 <0.0001 Seeding rate × fertilizer p > F 0.3900 0.1940 0.6579 0.7431 0.6773 0.9237
m.c., moisture content.
SED, standard error of the difference between two means.
Figure 3.

Summary of main effect means for yield and agronomic components.

Figure 4.

Summary of main effect means for grain quality parameters.

Figure 5.

Biplot (mean vs. CV) of seed rate by banded N fertilizer rate combinations for data collected at Lethbridge and Bow Island, AB, Canada, in 2007 to 2009. The first letter of the labels, “S,” indicates the seed rate factor followed by planting density (100, 300, or 500 seeds m−2) and the letter “N” indicates the N rate effect followed by the rate of the treatment (0, 30, 60, 90, or 120 kg N ha−1). Grouping categories: group I, high mean and low variability; group II, high mean and high variability; group III, low mean and high variability; group IV, low mean and low variability.

The superior pith expression observed in the low planting density is apparent in the biplot (Fig. 5). The lowest planting population produced higher pith and low variability when combined with all fertilized treatments. The moderate planting density also produced above average pith expression and low variability when combined with the 60 kg N ha−1 fertilizer rate. However, the visual stem cutting estimate for this treatment (33%) was just above the average of 31%. Lower and stable stem cutting was observed when planting at 100 seeds m−2 with 0, 60, or 90 kg N ha−1 (Fig. 5). Adding a micronutrient blend to a basal rate of 90 kg N ha−1 as previously described had no effect on pith expression or cutting susceptibility compared to the same basal rate of N without added micronutrients (Table 4).

View Full Table | Close Full ViewTable 4.

Response of solid stem spring wheat (cv. AC Lillian) to supplemental micronutrient blend applied in-crop at five times the recommended rate at the three-leaf stage and repeated at the six-leaf stage in southern Alberta, Canada, in 2008 to 2009.

 Factor Treatment Yield at 14% m.c.†(Mg ha−1) Stem solidness(pith expression)‡ Sawfly damage(% stems cut) Grain protein (%) Basal fertilizer, 90 kg N ha−1 Without micronutrient 3.66 2.85 36 13.6 Basal fertilizer, 90 kg N ha−1 With micronutrient 3.45 2.78 35 13.9 SED§ 0.14 0.07 5.59 0.32 Error df 25 25 25 25 p > F 0.1542 0.4464 0.8536 0.3787
m.c., moisture content.
Pith rating based on average of stem internodes (1, hollow; 5, solid).
§SED, standard error of the difference between two means.

Correlation coefficients were generated to further explore relationships between sawfly related parameters and crop production parameters (Table 5). Positive relationships exist between stem diameter and grain yield, protein, NDVI, and plant chlorophyll, but an inverse relationship exists with plant stands. There was also a positive relationship with stem solidness and stem diameter indicating larger stems generally produce more pith than smaller diameter stems. Negative correlations were observed between stem solidness and stem cutting, stand establishment. and spike density.

View Full Table | Close Full ViewTable 5.

Correlation matrix relating stem diameter, stem solidness, and wheat stem sawfly damage to yield components, protein, and canopy reflectance data.

 Pearson correlation coefficients Stem solidness Sawfly damage Stand establishment Spike density Grain yield Grain protein NDVI† Plant chlorophyll Stem diameter 0.49 –‡ −0.41 – 0.43 0.39 0.25 0.43 Stem solidness 1 −0.30 −0.51 −0.44 – – – – Sawfly damage 1 – – −0.30 0.40 – −0.56 Stand establishment 1 0.72 0.29 – 0.48 0.56 Spike density 1 0.50 0.29 0.82 0.94 Grain yield 1 – 0.60 0.88 Grain protein 1 0.33 – NDVI 1 0.96 Plant chlorophyll 1
NDVI, normalized difference vegetation index. NDVI and plant chlorophyll readings taken at Zadoks 31to 37 (Zadoks et al., 1974) and based on six and three site–years, respectively. All other correlation coefficients are based on means averaged over nine site–years.
“–” represents p > 0.05; all other r values presented at p ≤ 0.05.

A similar pattern for leaf area index was observed for N applications and sowing density (Fig. 6). Leaf area index peaked at Zadoks 58 (Zadoks et al., 1974) with the greatest accumulation observed using the higher rates of N or sowing density. There was no apparent benefit to leaf area index when in-crop applications of the micronutrient blend was applied to the 90 kg N ha−1 basal rate of N (Fig. 6).

Figure 6.

Leaf area index readings first recorded at Zadoks 30 growth stage (Zadoks et al., 1974) for plant density and N main effects averaged over data collected from Lethbridge and Bow Island, AB, Canada, in 2007 to 2009. Micro, micronutrients; 90N, 90 kg N ha−1.

Two economic net return scenarios were considered, based exclusively on N input costs (Fig. 7a and 7b) or the combined input costs of seed and fertilizers (Fig. 7c and 7d). In the first scenario, pricing was based on the 2009 input costs per tonne of urea N. In the second scenario, the input cost for N was increased by 50%, which is similar to a pricing scenario for fertilizer in 2008. In both cases, the seed input cost was based on the 2009 price of approximately C$13 per bushel, which was a typical price for AC Lillian in the first 3 yr of commercial production (2007 to 2009). In both cost scenarios, net returns were maximized at the 90 kg N ha−1 rate with either the 300 or 500 seeds m−2 sowing density (Fig. 7a and 7b). However, when seed input costs were added to the model, the highest and lowest sowing densities provided similarly low net returns compared to 300 seeds m−2 (Fig. 7c and 7d). Figure 7. Summary of net returns based on 2009 N fertilizer input costs (a) and for a scenario of when the N cost is increased by 50% (b). Summary of net returns based on 2009 N fertilizer and seed input costs (c) and for a scenario of when the N cost is increased by 50% (d). ## DISCUSSION Our findings do not correspond with a previous study on planting density that reported an increase in stem cutting of Rescue solid-stemmed wheat with lower plant populations (Luginbill and McNeal, 1958). Rescue planted at a rate of approximately 100 seeds m−2 sustained 57% cutting whereas plots sown at 600 seeds m−2 had 50% less cutting. The N results of our study agree with Luginbill and McNeal (1954) as the authors reported no effect when N was applied alone at rates of 60 and 120 kg N ha−1. A study comparing unfertilized spring wheat plots to those with basal rates of approximately 40 kg N ha−1 also report little effect from fertilization (O'Keeffe et al., 1960), and DePauw and Read (1982) reported that environmental factors were more important for pith expression than fertility management practices. Environmental factors that would affect pith expression are those related to precipitation. Cloudy conditions and reduced photoperiod negatively effect pith expression (Holmes, 1984). Shading effects within the canopy could have a similar effect. Leaf area index was greatest for the highest levels of N and seeding rates (Fig. 6), which corresponds to the responses observed in the same treatments for pith expression and sawfly damage. Therefore, the management of crop canopy architecture through appropriate seeding and N rates will probably influence plant tolerance or cutting susceptibility. There are two important considerations for plant density and N management recommendations for solid-stemmed wheat. First, any recommendation based on this study or previous findings must have agronomic merit. The second consideration is for a system that enhances the plant's ability to produce maximum pith in the culm of the stem. Wallace et al. (1973) reported that a mean pith expression of 3.75 would be required to achieve consistent high tolerance to WSS infestations. We did not observe this degree of stem solidness in AC Lillian in any of the treatments, which indicates that the weather parameters or the genetic potential of AC Lillian prevented maximum resistance to WSS infestation. Low plant populations maximized pith expression in our study and all combinations with fertilized treatments at this level displayed acceptable stability. However, low plant populations may not always optimize grain yield. A number of studies report that higher plant populations enhance weed competitive ability and herbicide efficacy and produce higher grain yield (Beres et al., 2010b; O'Donovan et al., 2006). A study of variety selection and seeding rates ranging from 150 to 450 seeds m−2 reports that the ideal sowing density for solid-stemmed wheat is in the range of 250 to 350 seeds m−2 (Beres et al., 2011b). Therefore, the compromise may be to combine moderate seed rates with moderate levels of N to maximize pith expression and grain yield and to meet the minimum protein standards (13.5%) required by the Canadian Wheat Board for marketing purposes. This strategy may improve net economic returns as herbicide management and weed competitive ability is enhanced over the low seeding rate. Luginbill and McNeal (1954) reported increased susceptibility to cutting in Rescue wheat when N was combined with P or P alone. Other studies report a reduction in yield and cutting damage when plants are P deficient (Delaney et al., 2010). Our N treatments generally did not affect pith expression but we did observe reduced pith in two site–years for the control and 30 kg N ha−1 N treatments (data not shown). This suggests that once N or P sufficiency is achieved, additional application of macronutrients elicits a physiological response (vegetative growth leads to shading) in spring wheat that inhibits pith expression. Therefore, N rates beyond 60 kg N ha−1 may begin to reduce the efficacy of a solid-stemmed wheat variety due to shading effects unless a very low sowing density is used (Fig. 5). In our study, we were also interested in expanding these findings to test if a blend of both macronutrients and micronutrients would create a similar result. Adding a micronutrient blend to a basal rate of 90 kg N ha−1 as previously described had no effect on pith expression or cutting susceptibility compared to the same basal rate of N without added micronutrients (Table 4). Furthermore, there were no differences observed in grain yield or grain protein concentration. Therefore, the hypothesis that micronutrient blends would have a positive effect on pith expression, grain yield, or quality parameters in solid-stemmed wheat was not supported. Nitrogen management and the use of micronutrient blends will alter canopy architecture in a similar fashion as seeding rates. However, there was no direct effect on pith expression observed in solid-stemmed wheat that was attributed to anything other than shading effects; fertilization did not influence pith expression but N did influence cutting susceptibility. Nutrient management should focus on plant health and therefore standard amendments are recommended: that is, 60 kg N ha−1. Most agronomists would recommend N amendments in accordance to soil available N results from soil sampling and lab analysis to confirm N requirements. The inverse relation observed between chlorophyll and stem cutting indicates that prolonged “greenness” or delayed crop maturity could reduce stem cutting susceptibility. This could simply be a case of asynchrony during the WSS flight period as later maturing treatments were at a crop phenology stage unattractive to an ovipositing female. Ovipositing females require a host that has entered the reproductive phase and has reached at least Zadoks 31 growth stage (Holmes, 1979; Zadoks et al., 1974). The positive relation between chlorophyll content and yield components suggests that the chlorophyll meter could be a very effective tool for predicting both yield potential and the threat to WSS cutting in solid-stemmed spring wheat. Further research into the relationship between chlorophyll, sawfly infestation, and subsequent damage from stem cutting is warranted. The results of this study indicate that low plant populations were often most effective at maximizing pith expression in solid-stemmed wheat and reducing sawfly cutting damage. However, this required one of the three the highest rates of N fertilizer (Fig. 5), and a system of low seeding rates and high N may not be economical based on current fertilizer input costs and the generally lower grain yield response. In our economic analysis, the lowest seeding rate produced the lowest economic returns at every rate of applied N (Fig. 7). Therefore, a decision to adopt a system for the exclusive purpose of optimizing pith would be cost prohibitive. The biplot for pith indicates that moderate seeding and N rates produced high and stable pith in AC Lillian. The economic returns for this system appear greatly improved over a system with a lower density. However, the economic returns may improve further with N rates over 60 kg N ha−1 at the moderate sowing density (+C$80 to C\$100 ha−1).

A producer is therefore faced with a decision to risk less than desirable pith and possible greater stem cutting vs. potentially higher net returns. The problem with the former scenario is that the higher net returns would probably erode in an environment of heavy WSS pressure due to greater harvest losses (Beres et al., 2007). A recommendation to hold N applications to moderate levels may appear conservative but is likely to be more stable over time, particularly in environments with heavy WSS pressure. In summary, an integrated planting and nutrient management plan when using a solid-stemmed spring wheat cultivar consists of seeding rates no greater than 300 seeds m−2 and 30 to 60 kg N ha−1.

## Acknowledgments

Special thanks to T. Entz for statistical guidance and R. Beck and S. Simmill for graphics and formatting support. The laborious nature of this work could not have been completed without the technical expertise provided by R. Dyck, S. Simmill, S. Kendrick, C. Vucuverich, A. Middleton, T. Larson, C. Herle, D. Friesen, R. Nielson, C. Hietamma, J. Michaelis, R. Diakow, J. Hudak, M. Williams, and K. Ziegler. This study was funded through Agriculture and Agri-Food Canada's A-Base program. This article is LRC contribution no. 387-11041.