My Account: Log In | Join | Renew
Search
Author
Title
Vol.
Issue
Year
1st Page

Agronomy Journal - Article

 

 

This article in AJ

  1. Vol. 105 No. 6, p. 1707-1720
    OPEN ACCESS
     
    Received: Apr 15, 2013
    Published: September 13, 2013September 13, 2013


    * Corresponding author(s): brian.beres@agr.gc.ca
 View
 Download
 Alerts
 Permissions
 Share

doi:10.2134/agronj2013.0193

A Canadian Ethanol Feedstock Study to Benchmark the Relative Performance of Triticale: II. Grain Quality and Ethanol Production

  1. Brian Beres *a,
  2. Curtis Pozniakb,
  3. David Bresslerc,
  4. Amera Gibreelc,
  5. Francois Eudesa,
  6. Robert Grafa,
  7. Harpinder Randhawaa,
  8. Don Salmond,
  9. Grant McLeoda,
  10. Yves Dionf,
  11. Byron Irvineg,
  12. Harvey Voldengh,
  13. Richard Martini,
  14. Denis Pageauj,
  15. Andre Comeauk,
  16. Ronald DePauwe,
  17. Sherrilyn Phelpsl and
  18. Dean Spanerc
  1. a Agriculture and Agri-Food Canada, Lethbridge Research Centre, 5403 1st Avenue South, Lethbridge, AB, Canada T1J 4B1
    b Crop Development Centre, Dep. of Plant Sciences, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK, Canada S7N 5A8
    c Dep. of Agricultural, Food, and Nutritional Sciences, Univ. of Alberta, 410 Ag/Forestry Building, Edmonton, AB, Canada T6G 2P5
    d Alberta Agriculture, Field Crop Development Centre, 5030 50th Street, Lacombe, AB, Canada T4L 1W8
    f CÉROM Agronome, 740 chemin Trudeau, Saint-Mathieu-de-Beloeil, QC, Canada J3G 4S5
    g Agriculture and Agri-Food Canada, Brandon Research Centre, Box 1000A, R.R. 3, Brandon, MB, Canada R7A 5Y3
    h Agriculture and Agri-Food Canada, Eastern Cereal and Oilseed Research Centre, 960 Carling Ave., Ottawa, ON, Canada K1A 0C6
    i Agriculture and Agri-Food Canada, Crops and Livestock Research Centre, P.O. Box 1210, Charlottetown, PEI, Canada C1A 7M8
    j Agriculture and Agri-Food Canada Research Farm, 1468 Saint-Cyrille St., Normandin, QC, Canada G8M 4K3
    k Agriculture and Agri-Food Canada Research Station, 2560 Hochelaga, Sainte-Foy, QC, Canada G1V 2J3
    e Agriculture and Agri-Food Canada, Semi-arid Prairie Agricultural Research Centre, P.O. Box 1030, Swift Current, SK, Canada S9H 3X2
    l Saskatchewan Ministry of Agriculture, 1192-102nd Street, North Battleford, SK, Canada S9A 1E9

Abstract

Cereal grain ethanol production may need to supplement biomass ethanol production to meet the increasing long-term demand for ethanol. A study was initiated to benchmark the relative performance of triticale (×Triticosecale ssp.) to wheat (Triticum aestivum L.) classes utilized for ethanol production. Ten cultivars: three triticale, two Canada prairie spring (CPS) wheat, three Canada western soft white spring (CWSWS) wheat, one Canada western red spring (CWRS) wheat, and one Canada western general purpose (CWGP) wheat cultivars were grown at 45 locations across Canada from 2006 to 2009. The locations were subgrouped by agroecological zone for western Canada, by province for Ontario and Quebec, and Charlottetown, PEI, for the Maritimes. The greatest grain yield was usually observed for Hoffman (red spring wheat) followed by triticale cultivars and CWSWS cultivars. Ethanol yield varied by region as a reflection of grain yield, and differences among cultivars generally were: triticale (excluding Tyndal) = Hoffman = CWSWS > CPS > CWRS. Ethanol concentration was least for Tyndal triticale and AC Superb CWRS. Stability assessments indicated that Pronghorn and AC Ultima triticales and Bhishaj CWSWS wheat provide consistent and high ethanol yields. The other CWSWS cultivars, AC Sadash and AC Andrew, had similarly high ethanol yields but were variable, indicating that utilization outside the Parkland and Western Prairies agroecological zones could pose greater risk for ethanol plants over Pronghorn and AC Ultima. Ethanol fermentation plants could therefore increase efficiency by replacing CPS wheat feedstocks with select triticales and potentially improve the consistency of production by using select triticales in regions where CWSWS wheats are less stable.


Abbreviations

    CPS, Canada prairie spring; CWGP, Canada western general purpose; CWRS, Canada western red spring; CWSWS, Canada western soft white spring

Globally, fuel ethanol production has now reached 75 billion L yr–1, and Canada’s contribution is approximately 2 billion L yr–1 (Canadian Renewable Fuels Association, 2010; Klein et al., 2004). Some economists have argued that ethanol fuel production from grain feedstocks relies too heavily on government programs to offset what is considered to be an inefficient system incapable of adequately reducing greenhouse gas emission targets (Freeze and Peters, 1999; Klein et al., 2004). Others have argued that ethanol should only be produced using crop biomass and residues (Canadian Renewable Fuels Association, 2010), considered by many as agricultural waste (Freeze and Peters, 1999). However, the dramatic rise in plant construction and output, including seven ethanol plants operating in western Canada with a collective annual output of 0.5 billion L (Canadian Renewable Fuels Association, 2010), suggests that the economics may not be as important as provincial and federal policies targeting energy diversity, agricultural benefits, and rural renewal (Coad and Bristow, 2011; Klein et al., 2004). Studies also report that >60% of crop residues must be retained to adequately maintain proper C cycling in the soil in wheat production systems (Freeze and Peters, 1999). Therefore, if crop residue exports are limited to 40% to maintain soil quality, ethanol production from grain is probably needed at some level to meet the increasing long-term demand for ethanol. Furthermore, grain-based ethanol production provides grain growers the opportunity to sell their grain into dual markets, which enhances marketing options for cereal production. Today, a producer of CWSWS and CPS wheat, which are preferred for wheat ethanol feedstocks, can choose to sell into a milling market or contract the production to an ethanol plant.

If cereal grains remain an important feedstock for ethanol production, does wheat possess superior attributes for fermentation over alternative cereals? Barley (Hordeum vulgare L.) was once considered inferior to wheat for fermentation efficiency and economic feasibility considerations (Klein et al., 2004). Newer studies using updated fermentation techniques report similar ethanol concentrations among some barley (Hordeum vulgare L.) and CPS wheat cultivars (Vigil et al., 2012). Triticale is a cereal crop first created in the late 19th century by crossing common wheat with rye (Secale cereale L.) (Oettler, 2005). It generally possesses a low grain protein concentration and high grain yield and biomass potential, which are more desirable traits in biorefinery processes than currently used wheat classes (Beres et al., 2010; Goyal et al., 2011). Moreover, the greater yield potential for triticale relative to Canadian wheat classes affords greater competitiveness with weeds (Beres et al., 2010; Oettler, 2005), and it also displays better tolerance to drought and pests than wheat (Darvey et al., 2000; Erekul and Köhn, 2006). Preliminary studies conducted in the western prairies of Canada indicated that triticale does have potential as an ethanol feedstock (McLeod et al., 2010).

A crop that does not occur naturally in the ecosystem, has low presence in human consumption markets, and is compatible with all on-farm and industrial equipment and infrastructure may be more attractive as a cereal platform technology and better received by society overall. The case for triticale may strengthen further if gene transformation is used to enhance a plant trait to be exploited in a bioindustrial process. Reports to date have not assessed triticale grain yield and ethanol production across an array of environments, nor have these reports assessed the crop using modern fermentation technologies. The Canadian Triticale Biorefinery Initiative (CTBI) is a consortium of stakeholders representing the triticale value chain with a focus on positioning triticale as a cereal platform technology for the bioindustry. One of the short-term goals for the CTBI is to benchmark the relative performance of triticale to Canadian wheat classes currently utilized for ethanol production. To address this goal, our objective was to compare grain ethanol fermentation and production results and other grain quality characteristics for 10 triticale and spring wheat cultivars.


MATERIALS AND METHODS

Experimental Design and Management

The experiment was conducted at 45 locations across Canada from 2006 to 2009 (Table 1; Fig. 1). Data generated from each location varied from a single year to 4 yr. Therefore, depending on the variable, data were collected at 71 to 94 sites (location ´ year combinations).


View Full Table | Close Full ViewTable 1.

Description of locations for ethanol feedstocks study.

 
Location Agroecological zone Soil zone Years Growing season precipitation
Latitude Longitude
2006 2007 2008 2009
mm
Western Canada
 Dawson Creek, BC Parklands Grey Wooded 2007, 2009 –† 438 240 55°48′ N 120°14′ W
 Fort St John, BC Parklands Grey Wooded 2007, 2009 562 222 56°17′ N 120°50′ W
 Donnelly, AB Parklands Grey Wooded 2007 270 55°43′ N 117°6′ W
 Edmonton, AB Parklands Black 2007, 2008, 2009 181 159 147 53°33′ N 113°29′ W
 Falher, AB Parklands Grey Wooded 2007, 2008, 2009 270 234 107 55°46′ N 117°10′ W
 Killam, AB Parklands Black 2006, 2007 416 370 52°47′ N 111°51′ W
 Kitscoty, AB Parklands Black 2006 469 53°20′ N 110°20′ W
 Lacombe, AB Parklands Black 2007, 2008, 2009 357 230 279 52°29′ N 113°43′ W
 Lethbridge, AB (dry) Western Prairies Dark Brown 2007, 2008, 2009 164 380 241 49°41′ N 112°50′ W
 Lethbridge, AB (irrigated) Western Prairies Dark Brown 2007, 2008, 2009 291 456 343 49°41′ N 112°50′ W
 Neapolis, AB Parklands Black 2007, 2009 472 114 51°40′ N 113°52′ W
 Sexsmith, AB Parklands Grey Wooded 2007 537 55°21′ N 118°46′ W
 Vermilion, AB Parklands Black 2007 397 53°21′ N 110°51′ W
 Westlock, AB Parklands Grey Wooded 2007 278 54°9′ N 113°51′ W
 Canora, SK Parklands Grey Wooded 2007 365 51°38′ N 102°26′ W
 Indian Head, SK Eastern Prairies Black 2007, 2008, 2009 275 217 210 50°32′ N 103°39′ W
 Lake Lenore, SK Parklands Black 2007, 2008 369 178 52°25′ N 104°58′ W
 Lashburn, SK Parklands Black 2006 339 53°7′ N 109°36′ W
 Melfort, SK Parklands Black 2007, 2008, 2009 351 190 243 52°52′ N 104°36′ W
 Outlook, SK Western Prairies Dark Brown 2007 291 51°29′ N 107°3′ W
 Redvers, SK Eastern Prairies Black 2007 283 49°34′ N 101°41′ W
 Regina, SK Western Prairies Dark Brown 2007, 2008 267 228 50°26′ N 104°35′ W
 Saskatoon, SK Western Prairies Dark Brown 2006, 2007, 2008, 2009 489 278 180 215 52°8′ N 106°38′ W
 Scott, SK Western. Prairies Dark Brown 2007, 2008, 2009 313 207 173 52°21′ N 108°49′ W
 Swift Current, SK Western Prairies Brown 2007, 2008, 2009 152 337 199 50°18′ N 107°46′ W
 Valparaiso, SK Parklands Grey Wooded 2006 489 52°51′ N 104°10′ W
 Watrous, SK W. Prairies Dark Brown 2007, 2008, 2009 210 238 256 51°40′ N 105°27′ W
 Arborg, MB Parklands Grey Wooded 2008 466 387 50°54′ N 97°13′ W
 Brandon, MB Eastern Prairies Black 2007, 2008, 2009 257 367 241 49°50′ N 99°56′ W
 Carberry, MB Eastern Prairies Black 2007, 2009 389 235 49°51′ N 99°21′ W
 Melita, MB Eastern Prairies Black 2006, 2007, 2008, 2009 378 283 258 213 49°16′ N 100°59′ W
 Minto, MB Eastern Prairies Black 2006 462 49°24′ N 100°1′ W
 Neepawa, MB Parklands Grey Wooded 2006, 2007 462 477 49°24′ N 100°1′ W
 Portage, MB Eastern Prairies Black 2007, 2009 395 209 49°58′ N 98°17′ W
 Roblin, MB Parklands Grey Wooded 2007, 2008, 2009 445 308 287 51°13′ N 101°21′ W
 Rosebank, MB Eastern Prairies Black 2007, 2008, 2009 388 315 274 49°22′ N 98°6′ W
Eastern Canada
 Elora, ON 2007 252 43°41′ N 80°25′ W
 Ottawa, ON Dark Grey Gleisolic 2007, 2008, 2009 346 337 488 45°24′ N 75°41′ W
 Renfrew County, ON Dark Grey Gleisolic 2007, 2008, 2009 442 354 339 45°39′ N 77°11′ W
 St. Isidore, ON Dark Grey Gleisolic 2008 354 351 45°23′ N 74°54′ W
 Normandin, QC 2007, 2008, 2009 253 339 322 48°50′ N 72°31′ W
 St. Foy, QC Dark Grey Gleisolic 2007, 2008 499 559 465 46°47′ N 71°14′ W
 St. Hyancinthe, QC Dark Grey Gleisolic 2007, 2008, 2009 351 487 409 45°37′ N 72°56′ W
 Charlottetown, PEI Orthic Humo-Ferric Podzol 2007, 2008, 2009 388 464 531 46°15′ N 63°7′ W
Precipitation data not collected.
Fig. 1.

Geographical distribution of study sites at locations in agroecological zones and provinces used to assess grain quality and ethanol production assessment from 2006 to 2009.

 

Three triticale and two CPS, three CWSWS, one CWRS, and one CWGP wheat candidate cultivars (10 in total) were grown at the study sites (Table 2). Cultivars were chosen using the following criteria: (i) used as mid- to long-term checks used in either cultivar trials or cooperative registration trials or both; and (ii) provide a spectrum of yield and quality characteristics and disease resistance. Superb, for example, represented the CWRS class for most regional cultivar trials as well as cooperative registration trials and was popular with western Canadian producers. Hoffman, for example, was chosen for its high yield potential and its agronomic and disease attributes, particularly in the context of cultivar suitability to general purpose end uses such as feed or ethanol. Cultivars were randomly assigned to each of three replicates at each location × year combination according to a randomized complete block design.


View Full Table | Close Full ViewTable 2.

Summary of cultivars evaluated and corresponding market class description.

 
Cultivar Classification Reference
AC Ultima spring triticale McLeod et al. (2001)
Pronghorn spring triticale Salmon et al. (1997)
Tyndal spring triticale Salmon et al. (2007)
AC Andrew Canada western soft white spring wheat Sadasivaiah et al. (2004)
AC Sadash Canada western soft white spring wheat Sadasivaiah et al. (2009)
Bhishaj Canada western soft white spring wheat Randhawa et al. (2011)
AC Superb Canada western red spring wheat Townley-Smith et al. (2010)
5700PR Canada prairie spring red wheat AgriPro/Syngenta (unpublished data, 2000)
AC Crystal Canada prairie spring red wheat Fernandez et al. (1998)
Hoffman Canada eastern red spring wheat; Canada western general purpose candidate H. Voldeng (unpublished data, 2004)

The locations in Ontario, Quebec, and Prince Edward Island were grouped together to represent eastern Canada (Table 1). All other locations represented western Canada. Locations in western Canada were subdivided into three agroecological zones of the Canadian prairie: Western Prairies, Eastern Prairies, and the Parkland (Table 1). The Western Prairies region has soil types that are generally Orthic Dark Brown Chernozem clay loam soils (Typic Borolls), with approximately 30 g kg–1 organic matter content, or Brown Chernozem loam soils (Aridic Borolls), with approximately 20 g kg–1 organic matter. The Western Prairies region is considered semiarid, with growing season precipitation ranging from 152 mm (Swift Current, SK) to 380 mm (Lethbridge, AB).

The Eastern Prairies region site soils were predominantly Orthic Black Chernozem clay loam soils (Udic Borolls), with 60 to 80 g kg–1 organic matter, but also included Dark Grey Gleisolic soils. From an area standpoint, this region is smallest relative to other regions in the Western Prairies. Growing season precipitation ranges from 210 mm (Watrous, SK) to 462 mm (Minto, MB), and has relatively high yield and disease potential.

The Parkland region soils were typically Grey Wooded Luvisols in northern regions (Boralfs and Udalfs) and Orthic Black Chernozem clay loam soils in southern and transitional regions. The Parkland region is the largest, extending from Neepawa, MB (49°16′ N) northwest to Fort St. John, BC (56°17′ N, 120°50′ W), and east to Arborg, MB (97°13′ W) (Table 1). A shorter, humid growing season is typical for this region, but severe drought can occur; e.g., growing season rainfall was as low as 107 mm reported in Falher, AB, in 2009, but as high as 562 mm in Fort St. John, BC, in 2007.

The eastern Canadian sites in Ontario and Quebec were represented by Dark Grey Gleisolic (aquic suborders) or Podzolic (Spodosols) soils. The sites in Ontario were similar in terms of precipitation accumulation and did not vary greatly along latitudinal lines; these sites generally received approximately 350 mm of precipitation with extremes as high as 488 mm in Ottawa in 2009. The Ontario region included an area extending from the southwestern point of Elora (43°41′ N, 80°25′ W), north to Renfrew County (45°39′ N, 77°11′ W), and as far east as St. Isidore (45°23′ N, 74°54′ W) (Table 1). The Quebec sites were as far south and west as St. Hyacinthe (45°37′ N, 72°56′ W) and as far north as Normandin (48°50′ N, 72°31′ W) (Table 1). Normandin is in a region similar to the Parkland of western Canada and is typified by a short growing season and cooler climate. The Normandin area was relatively dry and received the least amount of precipitation of all locations in Quebec during the study period (average 305 mm). St. Foy was the wettest site during the study period and frequently experienced severe fusarium head blight (FHB) outbreaks. The Maritimes site of Charlottetown, PEI, had an Orthic Humo-Ferric Podzol soil characterized by fine sandy loam texture. The coastal climate of the Maritimes region is characterized by relatively high precipitation as well as high disease pressure from FHB (caused by Fusarium graminearum Schwabe [teleomorph Gibberella zeae (Schwein.) Petch]) (Table 1).

The plots were seeded at a rate of 300 seeds m–2 using a plot seeder equipped with a cone splitter and zero-tillage double disk openers. Seeding dates were typical for the respective regions within western and eastern Canada. Soil macronutrients were amended to levels that optimized wheat production for the region based on soil test recommendations. Plots were scouted for incidence of weed, disease, and insect pressure, and pesticides were applied as needed based on product labels and field crop protection guides.

Experimental Measurements

Yield and Grain Quality

Each plot was harvested using a Wintersteiger plot combine (Wintersteiger AG) equipped with a straight-cut header, pickup reel, and crop lifters. Grain yield was calculated from the entire plot area, and a subsample was retained to characterize test weight and whole grain protein, starch, and pentosan concentrations. Grain yield and protein concentration were calculated at a moisture content of 135 g kg–1. Protein concentration was estimated by means of near-infrared reflectance spectroscopy (Foss Decater GrainSpec, Foss Food Technology). Composite samples of grain for each cultivar at each location were analyzed for starch, pentosan, and residual starch content. Starch concentration was measured using AACC International (1976). Total pentosan concentration was estimated from flour by the orcinol-HCl method (Hashimoto et al., 1987).

Ethanol Fermentation

Grain from the composite samples was milled with a Perten 3100 Laboratory Mill (Perten Instruments) until the flour could pass through a 0.5-mm mesh screen and then mixed with water to obtain a mash with 32% (w/w) solids. The pH of the mash was adjusted to 4.0 using 12 mol L–1 HCl for viscosity and raw starch hydrolysis analyses. An enzymatic treatment with Optimash TGB (317 μL kg–1 of grain) and Fermgen (952 mL kg–1 of grain) (Genencor International) was performed at 52 to 55°C in a water bath for 1 h with frequent stirring. At the end of the enzymatic treatment, the pH and total solids were adjusted to original conditions to compensate for any change. The mash was transferred aseptically to a sterile 500-mL flask. Diethyl pyrocarbonate (Sigma-Aldrich, 97%) was added (779 mL kg–1 of mash) to each flask as a sterilizing agent and kept at 4°C for 72 h.

Before starting fermentation, the mash was warmed to 53to 55°C in a shaking incubator (Model Innova 44, New Brunswick Scientific). Stargen 002 (1.071 mL kg–1 of mash) (Genencor International) was then added to each flask for a 1-h incubation period. At this point, the temperature of the mash was reduced to 30°C and the stirrer speed was maintained at 200 rpm for the rest of the fermentation.

Urea (Fisher Scientific, 98%), Superstar yeast (Lallemand Ethanol Technology), and water were added to the mash in each flask to adjust to 30% (w/w) solids. Urea was used as an external N source at an initial concentration of 16 mmol L–1. The yeast was prepared by rehydration with 0.2 g yeast mL–1 water in a 250-mL flask followed by incubation at 30°C for 30 min with shaking at 200 rpm. Each flask was individually inoculated with yeast to have an approximate concentration of 2 × 10 7 colony-forming units mL–1.

After inoculation, each flask was capped with a rubber stopper containing a gas trap (American Brewmaster) to allow CO2 venting. To prevent ethanol evaporation from the flasks during fermentation, 2 mL of sterile water was added to each gas trap before starting fermentation. The inoculated flask was then incubated at 30°C in a shaking incubator at 200 rpm for 72 h. At the end of the fermentation, a sample of the mash was subjected to gas chromatography analysis to determine the ethanol concentration. All remaining mash was freeze-dried for residual starch concentration determination.

Statistical Analysis

Data were analyzed with the PROC GLIMMIX procedure of SAS version 9.2 (SAS Institute). The effects of replicate and site (location × year combinations) were considered random, and the cultivar effect was considered fixed. A Gaussian error distribution was used for the analysis. Pairwise comparisons were assessed using the SAS pdmix800 macro, developed by Saxton (1998), which accounts for pairwise probabilities and converts them into letter groupings, and a Bonferroni adjustment was used to provide some protection against Type I errors. Variability for the cultivar effect among sites was assessed with a statistical test to determine if the variance estimates were significantly different from zero and also by comparing the relative size of the site ´ cultivar variance estimate to the total variance associated with the site (main effect of site plus site × cultivar interaction). Treatment effects were declared significant at P < 0.05.

The genotype × environment interaction was assessed with a grouping methodology biplot, as described by Francis and Kannenberg (1978). The mean and CV were estimated across sites and replicates for each cultivar. These means were plotted against CV to explore average responses relative to variability for all cultivars. The mean of the cultivar values and CVs was used in the plot to divide the ordination space into four quadrants or categories: Group I—high mean, low variability (optimal); Group II—high mean, high variability; Group III—low mean, high variability (poor); and Group IV—low mean, low variability.

A general form of principal component analysis, otherwise known as multidimensional preference analysis, was performed to further explore the relationships among mean responses for the different crop traits (multivariate analysis of means). The data matrix for the analysis included cultivar means as rows and means for selected response variables as columns. The analysis was conducted with the PRINQUAL procedure of SAS version 9.2 (SAS Institute). using an identity transformation. The results were summarized in a biplot, which is a plot of the mean principal component scores for treatments for the first two principal components. Eigenvectors (correlation between the transformed and original data) for the crop responses were plotted as points at the end of vectors projecting from the origin into various positions in the ordination space. The coincidence of response variable vectors and cultivars across the ordination space suggested crop response variable associations with the cultivars. The lack of coincidence for response variable vectors and cultivars indicates cultivars for which the associated responses were lesser than other cultivars. The relative lengths of the vectors indicated the strength of these associations.


RESULTS

Entry and Class Differences

The analysis of variance indicated that the overall effect of cultivar always was highly significant (P < 0.001). The statistical test for the interaction effect of cultivar with agroecological zone or province was often important (P < 0.041), with a few exceptions. The cultivar × agroecological zone interaction effect for starch and ethanol concentration was not statistically significant.

Hoffman usually was the highest yielding cultivar, and the triticale and CWSWS cultivars almost always were part of the highest yielding group with Hoffman in and across all Canadian agroecological zones and provinces (Table 3). The CPS and CWRS cultivars consistently yielded least across most agroecological zones and provinces. The average yields of Hoffman, Pronghorn, and AC Ultima exceeded those for the CWRS cultivar AC Superb by an average of 24% and CPS red cultivars 5700PR and AC Crystal by 19% across western Canada. The overall yield potential was reduced in eastern Canada, but the magnitude of the difference over the CPS and CWRS cultivars was usually greater (the difference was 46% greater relative to AC Superb and 52% greater relative to CPS cultivars). Responses for Ontario were unique relative to the other eastern Canada zones because of increased variability and lower yield potential, which made statistically significant differences difficult to detect. The average yield for the triticale class was similar to that for CWSWS wheat in western Canada, whereas the triticales yielded 23% more than CWSWS cultivars in eastern Canada; however, specific within- and across-class comparisons for the triticale and CWSWS cultivars by agroecological zone in western Canada deviated from the preceding overall differences.


View Full Table | Close Full ViewTable 3.

Grain yield means for data collected in six agroecological zones and provinces across Canada from 2006 to 2009.

 
Cultivar Class† Western Prairies Eastern Prairies Parkland Western Canada Ontario Quebec Maritimes Eastern Canada Overall
Mg ha–1
AC Ultima TRIT 5.48 a‡ 5.12 a 5.71 ab 5.44 a 2.95 a 3.56 bc 3.51 ab 3.34 ab 4.39 ab
Pronghorn TRIT 5.66 a 5.24 a 5.84 ab 5.58 a 2.97 a 4.24 ab 3.74 a 3.65 a 4.61 a
Tyndal TRIT 5.10 abc 4.66 abc 5.39 bc 5.05 b 2.89 a 3.59 bc 2.99 ab 3.16 abc 4.10 bc
AC Andrew CWSWS 5.49 a 4.78 abc 5.78 ab 5.35 ab 2.75 a 3.87 abc 1.63 b 2.75 bc 4.05 bc
AC Sadash CWSWS 5.51 a 4.74 abc 6.14 a 5.46 a 2.70 a 3.82 abc 2.24 ab 2.92 bc 4.19 bc
Bhishaj CWSWS 5.46 a 4.84 abc 5.81 ab 5.37 ab 2.61 a 3.37 bc 1.64 b 2.54 c 3.95 cd
AC Superb CWRS 4.39 c 4.19 c 4.75 c 4.44 c 2.49 a 3.52 bc 1.94 ab 2.65 c 3.55 e
Hoffman CWGP 5.27 ab 5.05 ab 6.06 a 5.46 a 3.31 a 4.87 a 2.89 ab 3.69 a 4.58 a
5700PR CPS-R 4.51 c 4.35 bc 4.89 c 4.58 c 2.60 a 3.06 c 1.96 ab 2.54 c 3.56 e
AC Crystal CPS-R 4.66bc 4.28 c 5.10 c 4.68 c 2.34 a 3.51 bc 1.72 b 2.52 c 3.60 de
LSD(0.05)§ 0.55 0.63 0.46 0.32 0.92 0.92 1.50 0.66 0.37
Red¶ 4.71 4.47 5.20 4.79 2.68 3.74 2.13 2.85 3.82
CWSWS 5.49 4.79 5.91 5.39 2.69 3.69 1.84 2.74 4.07
Triticale 5.41 5.01 5.65 5.35 2.93 3.80 3.41 3.38 4.37
P value <0.001 <0.001 <0.001 <0.001 0.041 <0.001 <0.001 <0.001 <0.001
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

The superior milling and baking qualities of a CWRS class cultivar were evident in the study: AC Superb was always in a group of cultivars with the highest test weights and protein concentration in or across all agroecological zones and provinces (Tables 4 and 5). Yield components other than test weight must be responsible for the high grain yield potential of the triticales because they all displayed the lowest test weights (Table 4). Lower protein concentration also occurred for the triticales relative to the other cultivars (Table 5). The other hard red spring wheat cultivars often had moderate to high test weights, which indicates other yield components were unable to maintain the grain yield (Table 4). Also, moderate to high protein concentration was observed for this same group of cultivars (Table 5). The CWSWS class generally had intermediate test weight and protein concentration values (Tables 4 and 5). These trends were similar for all regions, although test weights were generally higher in western Canada. Differences in pentosan concentration were sometimes detected (Table 6); however, the most notable and consistent difference was the elevated pentosan levels for Tyndal, especially in the Western Prairies and Quebec.


View Full Table | Close Full ViewTable 4.

Test weight means for data collected in six agroecological zones and provinces across Canada from 2006 to 2009.

 
Cultivar Class† Western Prairies Eastern Prairies Parkland Western Canada Ontario Quebec Maritimes Eastern Canada Overall
g L–1
AC Ultima TRIT 712 b‡ 693 d 699 d 701 f 642 e 642 d 611 d 632 f 666 d
Pronghorn TRIT 705 b 683 d 689 d 692 f 639 e 678 cd 646 cd 655 e 673 d
Tyndal TRIT 708 b 689 d 694 d 697 f 651 de 672 cd 613 d 645 ef 671 d
AC Andrew CWSWS 770 a 722 c 757 c 749 e 679 cd 697 bc 652 cd 676 d 713 c
AC Sadash CWSWS 779 a 733 bc 770 abc 760 cd 691 bc 704 bc 693 abc 696 cd 728 b
Bhishaj CWSWS 770 a 727 c 759 bc 752 de 687 cd 697 bc 656 bcd 680 d 716 c
AC Superb CWRS 786 a 752 ab 778 a 772 ab 727 ab 755 a 695 abc 726 ab 749 a
Hoffman CWGP 785 a 762 a 781 a 776 a 750 a 769 a 713 ab 744 a 760 a
5700PR CPS-R 786 a 755 ab 776 ab 772 ab 739 a 745 a 726 a 736 a 754 a
AC Crystal CPS-R 780 a 744 abc 769 abc 764 bc 699 bc 732 ab 691 abc 708 bc 736 b
LSD(0.05)§ 18 2.0 1.5 10 29 29 47 21 12
Red¶ 784 75.3 77.6 771 729 750 706 728 750
CWSWS 773 72.7 76.2 754 686 700 667 684 719
Triticale 708 68.8 69.4 697 644 664 623 644 670
P value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

View Full Table | Close Full ViewTable 5.

Protein concentration means for data collected in six agroecological zones and provinces across Canada from 2006 to 2009.

 
Cultivar Class† Western Prairies Eastern Prairies Parkland Western Canada Ontario Quebec Maritimes Eastern Canada Overall
g kg–1
AC Ultima TRIT 116 f‡ 106 e 110 ef 111 f 109 d 105 d 116 b 110 e 111 f
Pronghorn TRIT 112 f 105 e 101 f 106 f 112 cd 105 d 119 b 112 e 109 f
Tyndal TRIT 115 f 112 de 107 ef 111 f 116 cd 108 cd 122 b 115 de 113 f
AC Andrew CWSWS 131 cde 123 bcd 117 de 124 d 126 bcd 125 bc 134 ab 128 bc 126 de
AC Sadash CWSWS 124 ef 119 cde 112 ef 118 e 124 bcd 123 bcd 127 b 125 cd 121 e
Bhishaj CWSWS 129 de 123 bcd 117 de 123 de 126 bcd 127 b 130 ab 128 bc 126 de
AC Superb CWRS 160 a 145 a 146 a 150 a 148 a 147 a 157 a 150 a 150 a
Hoffman CWGP 140 bcd 126 bcd 123 cd 130 c 122 bcd 124 bc 129 ab 125 cd 127 cd
5700PR CPS-R 145 b 136 ab 134 b 138 b 137 ab 140 ab 137 ab 138 b 138 b
AC Crystal CPS-R 141 bc 132 abc 127 bc 133 bc 129 bc 132 ab 136 ab 132 bc 133 bc
LSD(0.05)§ 9 11 8 5 14 14 23 10 6
Red¶ 146 135 132 138 134 136 140 137 137
CWSWS 128 122 115 122 125 125 130 127 124
Triticale 114 108 106 109 112 106 119 112 111
P value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.013
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

View Full Table | Close Full ViewTable 6.

Pentosan concentration means for data collected in six agroecological zones and provinces across Canada from 2006 to 2009.

 
Cultivar Class† Western Prairies Eastern Prairies Parkland Western Canada Ontario Quebec Maritimes Eastern Canada Overall
g kg–1
AC Ultima TRIT 58.7 ab‡ 68.1 a 61.0 a 62.6 ab 71.3 a 74.5 ab 58.7 a 68.2 a 65.4 ab
Pronghorn TRIT 51.8 b 65.6 a 56.6 a 58.0 b 73.2 a 72.6 ab 54.7 a 66.8 a 62.4 b
Tyndal TRIT 71.9 a 68.6 a 62.5 a 67.7 a 71.8 a 80.2 a 65.0 a 72.3 a 70.0 a
AC Andrew CWSWS 56.0 b 62.2 a 62.4 a 60.2 b 68.7 a 69.3 ab 70.0 a 69.3 a 64.8 ab
AC Sadash CWSWS 55.1 b 66.7 a 58.2 a 60.0 b 72.4 a 69.2 ab 66.7 a 69.4 a 64.7 ab
Bhishaj CWSWS 55.6 b 63.2 a 55.0 a 57.9 b 72.4 a 58.2 b 66.4 a 65.6 a 61.8 b
AC Superb CWRS 58.6 b 63.6 a 56.7 a 59.7 b 75.5 a 68.7 ab 66.9 a 70.4 a 65.0 ab
Hoffman CWGP 57.3 b 68.4 a 58.6 a 61.4 ab 67.0 a 71.0 ab 69.0 a 69.0 a 65.2 ab
5700PR CPS-R 58.7 ab 69.5 a 60.0 a 62.7 ab 65.5 a 67.3 ab 64.8 a 65.9 a 64.3 ab
AC Crystal CPS-R 57.2 b 61.9 a 58.0 a 59.1 b 68.5 a 65.2 ab 53.6 a 62.4 a 60.7 b
LSD(0.05)§ 9.8 12.3 8.9 6.0 14.9 14.1 24.4 10.6 6.1
Red¶ 58.0 65.9 58.3 60.7 69.1 68.0 63.6 66.9 63.8
CWSWS 55.6 64.0 58.5 59.4 71.2 65.5 67.7 68.1 63.8
Triticale 60.8 67.4 60.0 62.8 72.1 75.7 59.5 69.1 65.9
P value <0.001 0.302 0.080 <0.001 0.514 <0.001 0.275 0.138 <0.001
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

There were notable interactions for starch and ethanol yield, which are proxies for energy potential per unit area, although there was no interaction between cultivar and region for starch and ethanol concentration. The overall cultivar differences for starch and ethanol yields matched closely those for grain yield; starch and ethanol yields often were greater for the triticales Pronghorn and AC Ultima, the CWSWS class, and Hoffman than for the CPS and CWRS cultivars (Tables 7 and 8); however, the residual starch concentration, which is a reflection of resistance to starch digestibility, did reveal notable cultivar differences, with one exception (Table 9). The class mean for triticale displayed lower resistant starch values than the mean for the CWSWS class in two of three western Canada agroecological zones and Ontario; the CWSWS class produced 37% more residual starch than the triticales in these agroecological zones. These differences were not related to ethanol yield in western Canada and Ontario because the triticales (AC Ultima and Pronghorn) and the CWSWS class produced similar ethanol yield in most regions of western Canada and Ontario. The triticale class, however, mainly Pronghorn and AC Ultima, did produce more ethanol at the Maritimes site (78% more) and when averaged across all of Canada (10% more; the percentage increase did not include Tyndal) (Table 8). When assessing starch and ethanol concentrations averaged across all regions, the values were relatively similar among cultivars with the exception of Tyndal and AC Superb, which were among a group of cultivars that tended to produce lower concentrations of starch and ethanol (Table 10).


View Full Table | Close Full ViewTable 7.

Cultivar starch yield means for data collected in six agroecological zones across Canada from 2006–2009.

 
Cultivar Class† Western Prairies Eastern Prairies Parkland Western Canada Ontario Quebec Maritimes Eastern Canada Overall
Mg ha–1
AC Ultima TRIT 3.37 a‡ 3.05 a 4.11 abc 3.51 a 2.02 a 2.36 ab 1.94 a 2.11 abc 2.81 bc
Pronghorn TRIT 3.41 a 3.06 a 4.17 abc 3.54 a 2.16 a 2.88 ab 2.14 a 2.39 a 2.97 a
Tyndal TRIT 2.92 ab 2.54 ab 3.75 cd 3.07 b 1.98 a 2.25 ab 1.51 a 1.91 abc 2.49 cde
AC Andrew CWSWS 3.29 ab 2.80 ab 4.23 abc 3.44 a 2.05 a 2.63 ab 0.71 a 1.80 bc 2.62 cd
AC Sadash CWSWS 3.34 a 2.80 ab 4.49 a 3.54 a 2.11 a 2.55 ab 0.99 a 1.88 abc 2.71 abcd
Bhishaj CWSWS 3.29 ab 2.79 ab 4.16 abc 3.41 a 1.91 a 2.09 b 0.74 a 1.58 c 2.50 cde
AC Superb CWRS 2.75 b 2.31 b 3.37 d 2.81 b 1.80 a 2.23 ab 0.94 a 1.66 c 2.23 e
Hoffman CWGP 3.25 ab 3.00 a 4.40 ab 3.55 a 2.25 a 3.30 a 1.41 a 2.32 ab 2.94 ab
5700PR CPS-R 2.74 b 2.43 ab 3.61 cd 2.93 b 1.84 a 2.02 b 0.96 a 1.61 c 2.27 e
AC Crystal CPS-R 2.86 ab 2.52 ab 3.80 bcd 3.06 b 1.84 a 2.36 ab 0.92 a 1.71 c 2.38 de
LSD(0.05)§ 0.57 0.67 0.50 0.34 0.84 0.83 1.33 0.60 0.34
Red¶ 2.90 2.57 2.80 3.09 1.93 2.48 1.06 1.82 2.45
CWSWS 3.30 2.80 4.29 3.47 2.02 2.43 0.81 1.75 2.61
Triticale 3.24 2.88 4.01 3.38 2.05 2.50 1.86 2.14 2.76
P value <0.001 <0.001 <0.001 <0.001 0.716 <0.001 0.001 <0.001 <0.001
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

View Full Table | Close Full ViewTable 8.

Ethanol yield means for data collected in six agroecological zones and provinces across Canada from 2007 to 2009.

 
Cultivar Class† Western Prairies Eastern Prairies Parkland Western Canada Ontario Quebec Maritimes Eastern Canada Overall
L ha–1
AC Ultima TRIT 2279 a‡ 2025 abc 2518 abcd 2274 a 1074 a 1326 b 1283 a 1228 abc 1751 abc
Pronghorn TRIT 2289 a 2122 a 2659 abc 2356 a 1094 a 1596 ab 1344 a 1345 ab 1851 a
Tyndal TRIT 2011 abcd 1717 abc 2361 cde 2030 b 1008 a 1260 b 1073 ac 1113 abcd 1571 def
AC Andrew CWSWS 2209 ab 1926 abc 2716 ab 2284 a 937 a 1458 ab 628 d 1008 cd 1646 cd
AC Sadash CWSWS 2139 abc 1950 abc 2752 a 2280 a 920 a 1461 ab 846 bcd 1075 bcd 1678 bcd
Bhishaj CWSWS 2215 ab 1873 abc 2620 abc 2236 a 976 a 1274 b 603 cd 951 cd 1594 cde
AC Superb CWRS 1681 d 1597 c 2123 e 1800 c 857 a 1267 b 716 cd 947 cd 1374 g
Hoffman CWGP 2030 abcd 2054 ab 2780 a 2288 a 1192 a 1863 a 1119 ab 1391 a 1840 ab
5700PR CPS-R 1819 cd 1629 bc 2276 de 1908 bc 930 a 1098 b 768 bcd 932 d 1420 fg
AC Crystal CPS-R 1870 bcd 1673 bc 2396 bcde 1980 b 823 a 1295 b 619 d 912 d 1446 efg
LSD(0.05)§ 287 346 263 173 405 404 395 291 169
Red¶ 1850 1739 2394 1994 951 1381 805 1046 1520
CWSWS 2188 1917 2696 2267 944 1398 692 1011 1639
Triticale 2193 1955 2512 2220 1059 1394 1233 1229 1724
P value <0.001 <0.001 <0.001 <0.001 0.090 <0.001 <0.001 <0.001 <0.001
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

View Full Table | Close Full ViewTable 9.

Residual starch concentration means for data collected in six agroecological zones and provinces across Canada from 2007 to 2009.

 
Cultivar Class† Western Prairies Eastern Prairies Parkland Western Canada Ontario Quebec Maritimes Eastern Canada Overall
g kg–1
AC Ultima TRIT 61 b‡ 114 ab 39 a 71 abc 71 b 46 a 164 a 94 ab 83 abc
Pronghorn TRIT 74 ab 74 c 44 a 64 c 65 b 38 a 62 cd 55 b 59 c
Tyndal TRIT 94 ab 88 bc 51 a 78 abc 82 ab 53 a 98 bcd 77 ab 78 abc
AC Andrew CWSWS 108 ab 124 a 50 a 94 ab 122 a 58 a 79 bcd 87 ab 90 ab
AC Sadash CWSWS 122 a 117 ab 53 a 97 a 121 a 56 a 121 ac 99 ab 98 a
Bhishaj CWSWS 99 ab 101 ac 51 a 84 abc 81 ab 42 a 109 ac 78 ab 81 abc
AC Superb CWRS 100 ab 111 ab 58 a 90 abc 103 ab 63 a 40 d 69 ab 79 abc
Hoffman CWGP 96 ab 121 a 52 a 90 abc 92 ab 49 a 85 bcd 76 ab 83 abc
5700PR CPS-R 79 ab 74 c 47 a 67 bc 85 ab 59 a 70 cd 71 ab 69 bc
AC Crystal CPS-R 81 ab 73 c 49 a 68 bc 117 a 52 a 141 ab 103 a 86 abc
LSD(0.05)§ 47 32 43 28 41 63 65 48 28
Red¶ 89 95 52 78 99 56 84 80 79
CWSWS 109 114 52 92 108 52 103 88 90
Triticale 76 92 45 71 73 45 108 75 73
P value 0.001 0.001 0.967 <0.001 0.044 0.961 0.007 0.029 0.001
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

View Full Table | Close Full ViewTable 10.

Means for grain and ethanol characteristics averaged across six agroecological zones and provinces across Canada from 2006 to 2009.

 
Cultivar Class† Starch conc. Ethanol conc.
g kg–1 L Mg–1
AC Ultima TRIT 606 ab‡ 139 a 372 a
Pronghorn TRIT 609 ab 140 a 375 a
Tyndal TRIT 583 c 133 c 356 c
AC Andrew CWSWS 608 ab 139 a 373 a
AC Sadash CWSWS 616 a 139 a 372 a
Bhishaj CWSWS 607 ab 140 a 374 a
AC Superb CWRS 587 c 134 bc 358 bc
Hoffman CWGP 604 ab 139 a 373 a
5700PR CPS-R 597 bc 139 a 373 a
AC Crystal CPS-R 603 ab 138 ab 368 ab
LSD(0.05)§ 15 2 11
Red¶ 598 137 368
CWSWS 610 139 373
Triticale 599 137 368
P value <0.001 <0.001 <0.001
TRIT, spring triticale; CWSWS, Canada western soft white spring wheat; CWRS, Canada western red spring wheat; CWGP, Canada western general purpose candidate; CPS-W, Canada prairie spring white wheat; CPS-R, Canada prairie spring red wheat.
Means followed by the same letter in a column are not significantly different (P < 0.05; Bonferroni adjustment).
§Adjusted LSD(0.05) can be used only to compare cultivar means.
Red includes CPS-R, CWRS, and CWGP (Hoffman).

A multivariate representation of the means in a biplot was used to identify overall cultivar and class differences for all crop responses. Mean principal component scores for treatments for the first two principal components were plotted as points in the ordination space, and eigenvectors (correlation between the transformed and original data) for crop responses were plotted as points at the end of vectors projecting from the origin. The coincidence, or lack of coincidence, of points and vectors (the length of a vector indicates the strength of the relationship) on the ordination space suggested crop response variable associations with the cultivars. The Western Prairies biplot most closely corresponded with the representation across all agroecozones (Fig. 2). Triticale and the CWRS and CPS wheat cultivars tended to deviate farthest from the origin, and these two groups of cultivars were most polarized among all 10 cultivars, mainly along the first, horizontal principal component. This distinction among cultivars corresponded with protein-related and test-weight vectors often aligning with CWRS and CPS cultivars, while yield-related vectors coincided with triticale and Hoffman. In the eastern Canada regions, Hoffman was most polarized from the other cultivars. The CWSWS cultivars generally positioned closer to the origin than the other cultivars; this indicated that these cultivars were generally moderate with regard to the traits tested in this study. Moreover, starch-related and ethanol-concentration vectors tended to position closest to the CWSWS cultivars, especially in the western Canada agroecological zones. In eastern Canada, starch- and ethanol-related variables had greater coincidence with the triticales (Pronghorn and AC Ultima) and Hoffman. Pentosan concentration was associated with Tyndal in the western agroecological zones and with all triticale cultivars in the Parkland agroecological zone and Quebec. The Maritimes and Ontario tended to deviate from these trends because the CPS cultivar AC Crystal and the CWSWS cultivar Bhishaj were associated with the pentosan concentration vector.

Fig. 2.

Biplot of agronomic, grain quality, and ethanol responses (vectors) relative to cultivars (points) generated with multidimensional preference analysis (generalized form of principal components analysis) for data collected at locations across Canada from 2006 to 2009. The percentage of variance explained by the first two principal components (x axis = first component and y axis = second component) is indicated on the respective axes.

 

Performance Stability and Variance

The random effect of site captured the variability among location × year combinations not explained by the fixed effect of region. The site × cultivar variance estimate was always highly significant (P < 0.01) (Table 11). The percentage of the total site variance accounted for by this interaction often varied around 10 to 15% but was as high as 24 to 31% for protein, pentosan, and residual starch and starch concentrations.


View Full Table | Close Full ViewTable 11.

Variance estimates associated with the random effect of site (location ´ year combinations) for wheat and triticale cultivar traits collected across Canada from 2006 to 2009.

 
Effect Variance estimate
Yield Test wt. Protein conc. Pentosan conc. Starch yield Residual starch conc. Starch conc. Ethanol conc. (g kg–1) Ethanol conc. (L Mg–1) Ethanol yield
Site 2.60** 13.1** 161** 106** 0.811** 3594** 433** 167** 385519** 1202**
Site × cultivar 0.25** 2.3** 66** 41** 0.126** 1105** 199** 23** 47575** 189**
Site × cultivar, %† 9 15 29 28 13 24 31 14 11 14
**Significant at P < 0.01.
Percentage of the total variance associated with the effect of site.

Mean vs. CV biplots were used to further explore and understand the response variability relative to the mean responses (Fig. 3 and 4). Overall, Hoffman wheat and AC Ultima and Pronghorn triticale produced consistent, maximum yields in most areas, although Pronghorn was positioned closely only to the quadrant associated with high and consistent yields. Tyndal and the CWSWS cultivars were less consistent but comparatively high yielding. The CWRS cultivar AC Superb and the CPS cultivars produced lower grain yields, with CPS red wheat yields being variable and AC Superb yields being consistent. Hoffman clearly showed that it is highly adapted (Group I: maximum crop response with low CV across sites) in all regions except the Maritimes and Eastern Prairies. Similar to Hoffman, at least one triticale, usually Pronghorn or AC Ultima, displayed high adaptation to all regions except the Western Prairies. In this agroecological zone, high and variable grain yields were observed for the triticales. The CWSWS cultivars were not adapted (lower and/or more variable responses) to areas outside the Western Prairies and Parklands. The CPS and CWRS cultivars were frequently highly adapted in all regions in terms of protein concentration (Fig. 3). It was apparent that AC Superb and Tyndal often were not adapted in terms of starch and ethanol concentration (Fig. 3 and 4); however, AC Ultima, AC Sadash, and AC Crystal exhibited clear adaptation in most regions for starch concentration (Fig. 3). The other CSWSW cultivars, Pronghorn triticale, and Hoffman also produced high starch concentration but with greater variability (Fig. 3). Ethanol yield trends, not surprisingly, followed closely with those for grain yield (Fig. 4).

Fig. 3.

Biplot (mean on y axis vs. CV on x axis) summarized across and by zone or province for yield (Mg ha–1), protein concentration (g kg–1), and starch dry weight concentration (g kg–1) data in six agroecological zones and provinces collected at locations across Canada from 2006 to 2009. In addition to cultivar estimates, estimates are also provided for the Canada western soft white spring (CWSWS) and Canada prairie spring red (CPS Red) wheat classes and indicated with different symbols. Grouping categories: Group I—high mean, low variability; Group II—high mean, high variability; Group III—low mean, high variability; Group IV—low mean, low variability.

 
Fig. 4.

Biplot (mean on y axis vs. CV on x axis) summarized across and by agroecological zone for ethanol concentration (L Mg–1) and yield (L ha–1) in six agroecological zones and provinces collected at locations across Canada from 2006 to 2009. In addition to cultivar estimates, estimates are also provided for the Canada western soft white spring (CWSWS) and Canada prairie spring red (CPS Red) wheat classes and indicated with different symbols. Grouping categories: Group I—high mean, low variability; Group II—high mean, high variability; Group III—low mean, high variability; Group IV—low mean, low variability.

 


DISCUSSION

The ethanol production potential of triticale in this study can generally be summarized as: triticale (excluding Tyndal) = Hoffman red spring wheat = CWSWS > CPS > CWRS. Hoffman was not recommended for registration in western Canada in 2010 based on susceptibility to stem rust. For this reason, it was deemed that Hoffman was not a suitable ethanol feedstock alternative in western Canada. Although ethanol concentration differed only for the lesser performing Tyndal and AC Superb (Table 2), the mean vs. CV biplots (Fig. 4) suggest that the Pronghorn and AC Ultima triticales, Bhishaj CWSWS wheat, and the CPS class would provide consistently greater levels of ethanol concentration across most regions. The other CWSWS cultivars, AC Sadash and AC Andrew, were more variable in most regions, which indicates that utilization of this class outside the Parkland and Western Prairies could pose greater risk (i.e., variable supply) for ethanol plants, especially when compared with Pronghorn and AC Ultima.

With the exception of the Western Prairies, where greater variability was observed, Pronghorn consistently produced high grain yield, which means it was probably best adapted to produce stable, maximum ethanol yields across most of Canada. Hoffman and AC Ultima also displayed reasonable ethanol yield traits; however, Hoffman has stem rust susceptibility in western Canada and AC Ultima displayed a slightly narrower range of adaption across the three agroecological zones. The CWRS and CPS classes are not as well suited to ethanol production relative to select triticales, the CWSWS class, and Hoffman. It was not surprising that the CWRS class is not suitable as an ethanol feedstock because this has been established in other studies (McLeod et al., 2010). Our findings for CPS, however, differ from the conclusions of McLeod et al. (2010) that both CPS red and white classes were superior to triticale at study sites in 1993 to 1996.

There are a few possibilities suggesting that the ethanol yield potential of triticale has improved in recent years. Also, innovations to fermentation technologies have evolved, along with new-generation enzymes that seem to have improved starch digestibility to the point that most cereal grain feedstocks, including those in this study, do not differ in terms of ethanol concentration (Gibreel et al., 2011). The residual starch values presented in this study would seem to support this argument because the values were generally similar among most cultivars. When elevated residual starch values were observed, they occurred in the CWSWS class, the preferred feedstock for ethanol plants. Inferior residual starch responses for CWSWS further support the use of select triticale as a suitable ethanol feedstock in Canada.

It is thought that advancements in cultivar development have overcome concerns about growing degree day requirements, elevated concentrations of pentosans, and other major deficiencies responsible for a lack of utilization of triticale for ethanol plants. Our findings indicate that select triticales will be suitable for ethanol production, which corresponds with other studies conducted in the Parkland and the southern region of Alberta, Canada (Beres et al., 2010; Collier et al., 2013; Goyal et al., 2011). For example, Collier et al. (2013) reported that triticale could be successfully grown in regions that accumulate ≥1700 growing degree days and that modern cultivars appeared to have improved stress tolerance and grain quality. This provides evidence that these issues are no longer relevant if appropriate cultivars are utilized. Pronghorn and AC Ultima appear well suited for an ethanol end use because they were widely adapted and displayed pentosan levels similar to the preferred cultivars. Tyndal, however, would not be well suited due to elevated pentosan contents and longer growing degree day requirements—attributes that hindered the adoption of previous triticale cultivars (McLeod et al., 2010).

The generally small variance estimates relative to total variance for the random effect of site suggests that rank changes between cultivars, in exclusion of agroecological zone or region interactions, probably were not important. In the first study of this series (Beres et al., 2013), similar observations were reported for agronomic and disease traits with the exception of lodging, which had a site × cultivar variance estimate of 31%. Entry difference variability may increase when the area expands beyond the agroecological zones; however, the area required to observe a response can be large. For example, a study assessing genotype × region interactions for two-row barley cultivars also reported no notable adaptation to subregions; responses within western Canada were similar (Atlin et al., 2000). Regional adaptation occurred only when the area was expanded to all of western Canada vs. eastern Canada.

In conclusion, the high ethanol yield potential for triticale, along with relatively stable ethanol yields, reflects its potential as an ethanol feedstock in most regions of Canada. This broad range of adaptation and stability would also suggest that the potential extends into the northern and central Great Plains of the United States. There is greater perceived risk in growing triticale, however, due to diseases such as ergot, a lack of information regarding the fermentation efficacy of triticale, and the fact that crop insurance programs will not ensure triticale at a rate similar to CWSWS wheat (Keith Rueve, personal communication, 2011). There was greater potential for risk (in terms of additional variability) growing the preferred CWSWS cultivars AC Andrew and AC Sadash in all regions except the Parkland in western Canada. Compared with AC Andrew or AC Sadash, Pronghorn and AC Ultima appeared to have similar or greater ethanol production overall, in most regions, and with greater consistency. Therefore, the results of this study indicate that triticale offers advantages over CPS and CWSWS as an ethanol feedstock. Changing ethanol feedstock acreages from CPS and CWSWS wheats could occur when Canadian growers have a revenue insurance option to cover the price of risk of growing triticale.

Acknowledgments

Special thanks to Craig Stevenson for statistical analyses of data presented in this manuscript; and to Ryan Dyck, Shannon Chant (Saskatchewan Ministry of Agriculture), Sheree Daniels, Ryan Beck, Dan Yagos, and Steven Simmill for trial coordination. The laborious nature of this work could not have been completed without the expertise provided by the technical teams at each of the author and co-author research facilities as well as the many applied research associations that participated in this project. This study was funded through Agriculture and Agri-Food Canada’s Agricultural Bioproducts Innovation Program, the Saskatchewan Agricultural Food Council, and Saskatchewan Ministry of Agriculture. This article is LRC contribution no. 387-13027.

 

References

Footnotes


Comments
Be the first to comment.



Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.