Optimizing Yield and Quality of Cereal Silage

Download 86K pdf file ("optimizing.pdf")PDF
 Take Home Messages | Introduction | Species and Varieties | Management | Harvest Schedule and Strategy | Impact of Choices on Production and Economics | Conclusions | References

Proceedings of the 2000 Western Canadian Dairy Seminar
Vern S. Baron, Erasmus Okine and A. Campbell Dick

Western Forage-Beef Group, Agriculture and Agri-Food Canada,
6000 C&E Trail, Lacombe, AB,
Canada, T4L 1W1
E-mail: baronv@em.agr.ca

Take Home Messages

  • Semi-dwarf barley is the most nutritious, adapted crop for silage production.
  • Spring triticale is higher yielding than barley and has a nutritional value similar to or greater than some barley varieties, so has a place in dairies.
  • Corn has a very high nutritive value and while expensive to grow should be considered in some instances.
  • Mixtures can improve forage nutritional value over cereals pure stands, but yield loss and expense may negate gains in nutritive value, when compared to the best of the cereals grown alone.
  • The economic value of all options must be placed in the context of the entire milk production system and not evaluated as a single factor.

Cereals are harvested as silage or green feed from about 800 thousand ha annually in Alberta. Barley silage is used on virtually all dairy operations which grow feed on the farm, and is the staple feedstuff of the beef feedlot industry in Alberta. Cereal silage is neither as nutritious nor as inexpensive for milk production as alfalfa (Khorasani and Kennelly, 1997). Producers use cereal forages because they are flexible for land and time use when combined in a system with perennial forages. They not only spread out workload at harvest, but also reduce the risk of high feed costs as a result of perennial forage crop failure when untimely dry weather conditions and winterkill occur.

Currently there are more alternatives for cereal silage than ever before. The advances in crop production are:
  • Development of adapted semi-dwarf barley cultivars that can be used for silage.
  • Development of adapted triticale cultivars that have adequate nutritive value for dairy cattle.
  • Development of viable mixtures to improve forage quality over pure forage cereal stands.
  • Shorter maturing corn varieties, so that corn may be grown economically in short season areas.
The relevance of these advances have to be evaluated on the basis of how to optimize production of a feedstuff and how cost effective each advance is within the production system.

Species and Varieties

Over the last decade plant breeding programs have become more diverse in the genetic material they use and produce. This, in turn, provides variability among species and among varieties within species for their forage potential. Few cereals are bred specifically for forage production, but introductions of new genetic material to improve agronomic traits have brought improvements in forage yield and quality as well.

Three types of barley have been tested for forage potential. They are semi-dwarf, standard six-row and standard two-row types. Semi-dwarf types such as Duke, CDC Earl and Tukwa are adapted to irrigated and highly fertile black soils of western Alberta, where lodging is a problem. Under these conditions they are generally similar in yield to standard types like Virden, and AC Lacombe, even though they are shorter (McKenzie et al., 1997; Baron at al, 1999; Table 1). On dry land in southern Alberta, the standard six-row, AC Lacombe, and two-row, Seebe, were identified as high yielding varieties (McKenzie et al, 1997). The taller standard types have a tendency to yield more dry matter under dry land conditions than semi dwarfs, but the differences between the types are not always large (McKenzie et al., 1997; Baron et al., 1999).

Table 1. Forage yield, date of harvest and percent dry matter of spring and winter cereals at early dough stage averaged over 3 years (1993–1995) at Lacombe, AB.
Harvest date
Dry matter
Winter triticalePika
July 28
Fall ryePrima
July 14
Spring triticaleWapiti
Aug. 20
Spring barleyTukwa (semi-dwarf)
Aug. 2
Spring barleyVirden (standard)
Aug. 6
Spring oatCascade
Aug. 13

Spring triticale is an alternative to barley under some circumstances. It provides greater forage yield stability than barley under a variety of moist and dry conditions. Spring triticale forage yields were the same or greater than oat or barley at five locations across western Canada (McLeod et al, 1998). Averaged over 3 years at Lacombe, Alberta, Wapiti spring triticale yielded 20% more dry matter than a standard and semi dwarf spring barley (Table 1), but was not significantly greater than the oat variety. Several spring triticale cultivars are available and have been tested for forage potential in central and southern Alberta and southern Saskatchewan (Baron et al., 1999).

Corn has been grown in Alberta for many years. A combination of earlier varieties and warmer than normal temperatures has spurred interest in silage corn. However, only the areas southeast of Brooks have air temperatures warm enough (greater than 2200 corn heat units) to allow traditional corn varieties to mature for grain. Corn development and yield are very heat-dependent compared to small grains. For example dry matter yield of one good variety (Pioneer 39K72) yielded slightly over 14 t/ha at Lacombe in 1998 (2,600 corn heat units) compared to about 8 t/ha in 1999 (1,700 corn heat units). The same variety yielded 17.8 t/ha in Alberta Corn Committee trials in southern Alberta averaged over 4 location-years. In recent years corn varieties with earlier maturity have been developed so that corn may be grown for grain in areas as low as 2000 corn heat units. Generally, corn varieties grown for silage need less heat units. They may be rated 200 corn heat units later than those grown for grain. Thus, very early corn hybrids may mature for silage in central Alberta. From 1993 until 1999 accumulated corn heat units have exceeded 2000 in five of seven years at Lacombe, where the long term average is 1800 corn heat units.

Forage quality
Among small grain species barley has had superior forage quality. However, there now appears to be more variability among varieties within barley for forage quality and spring triticale has made significant advances (Table 2).

In the absence of animal trials in vitro digestibility (in vitro dry matter digestibility, in vitro true (cell wall) digestibility and in vitro digestible organic matter (IVDOM)) probably best determines forage potential on the basis of nutritive value. Khorasani and Kennelly (1997) indicated that acid detergent fiber (ADF) may not be the best indicator of whole plant digestibility during the period between heading and maturity, but that neutral detergent fiber (NDF) could explain differences in intake between cereal species. Generally both NDF and ADF have been negatively associated with IVDOM in agronomic studies at Lacombe.

Table 2. Forage quality of spring and winter cereal species and varieties at early dough stage averaged over three years (1993-1995), grown at Lacombe AB.
VarietyCrude protein (%)IVDOM1NDF1ADF1
Winter triticalePika9.061
Fall ryePrima8.263
Spring triticaleWapiti9.766
Spring barleyTukwa11.068
Spring barleyVirden9.560
Spring oatCascade9.062
1IVDOM is in vitro digestible organic matter; NDF and ADF are neutral and acid detergent fiber.

When harvested for silage at the dough stage semi-dwarf barley (e.g. Tukwa, Duke and CDC Earl) has shown higher IVDOM and lower fiber levels than standard barley types (e.g. Virden). This is in agreement with Khorasani and Kennelly (1997) who also found the two-row feed variety Seebe to have greater nutritive value than AC Lacombe, a standard six-row type. Although there is some variability among spring triticale varieties (Baron et al 1999) spring triticale has higher IVDOM and lower fiber than some standard barley varieties (Table 2). However, winter triticale should not be used in milking rations as it has lower nutritive values than spring types and may not be greater than some oat varieties (Table 2 and Baron et al, 1999). Baron et al (1999) observed some oat varieties having as high IVDOM as some standard barley and spring triticale varieties, although it is likely more the exception than the rule. Mature well-eared corn silage has superior feeding value to standard barley. In vitro dry matter digestibility of whole-plant corn silage should range from 68 to 73 % (Daynard and Hunter, 1975). Relative maturity and grain content should not affect digestibility of whole plant corn (Daynard and Hunter, 1975; Fairey, 1983). Neutral and acid detergent fiber of well-eared corn was 48.6 and 29.8% compared to poorly eared corn at 57.2 and 29.8% averaged over three hybrids (Coors et al., 1997).

Year to year variability does exist for forage quality parameters. For example, NDF content for Tukwa barley (44.5, 52.5, and 48.3 % in 1993, 1994 and 1995) and Wapiti triticale (41.4, 52.5 and 51 in 1993, 1994 and 1995) ranged 8 and 11 percentage units, respectively, over 3 years. In cereal silage, harvest index or grain content may vary substantially from year to year (Baron et al., 1992a). This may have a dilution effect on whole plant fiber content as indicated by Coors et al (1997) for corn. Because NDF levels are negatively related to intake there may be an impact on digestible dry matter intake (Khorasani and Kennelly, 1997), so diets may have to be adjusted with supplemental grain from year to year.

Milk production
Khorasani and Kennelly (1997) reviewed relationships among forage quality, rumen dynamics, and milk production from small grain species as silage. While differences among barley types were not significant there was a trend for greater milk production from a semi-dwarf (37 kg/d) compared to a standard (32.8 kg/d) barley variety (Khorasani and Kennelly, 1997). Larger differences were detected in early vs. mid lactation animals. Intakes of semi-dwarf barley were greater than other barley types. This was related to a lower NDF level for the semi-dwarf barley. Also residence time of oat fiber in the rumen was longer than barley or triticale. These factors ultimately have an impact on digestible dry matter intake. Generally milk production from cereal silage was not superior to alfalfa. No comparisons were made among small grains and corn silage for milk production. However, considerable research has been carried out concerning grain content in corn and milk production. The consensus is that milk production responds positively to increasing proportions of grain in corn silage (Fisher and Fairey, 1979).

Awns and leaf diseases
Most semi dwarf barley varieties and triticale have rough awns, which are known to cause mouth and tongue lesions in young animals fed large amounts of silage. Chopping silage at moister stages of maturity tends to reduce these effects. Triticale varieties to be released in future will have a reduced-awn trait.
Most feed barley types (semi-dwarf and standard) have good levels of leaf disease resistance when released, but eventually the resistance breaks down making them susceptible to scald and net blotch. Some varieties are more susceptible than others, but when serious leaf damage occurs substantial yield loss will occur also. This can be minimized by varying the source (variety and where barley comes from) of barley grown on the same land or by growing a non-susceptible forage in rotation with barley. Spring triticale and corn are not susceptible to these same diseases. This is not to say that triticale and corn are disease-free.


Planting date
A rule of thumb is that planting early allows crops to take advantage of the best of soil and climatic resources, reducing risk of low or uneconomic yields. Research in southern Alberta had higher whole plant yields when planting occurred from early May to early June. Much lower forage yields resulted when planting occurred after mid-June (Riemer and Gaudiel, 1983). On going research at Lacombe showed that late-planted (mid-June) barley and oats yielded 60% and 75%, respectively of early planting (before mid-May) at several sites in central and Northern Alberta (Kibite, unpublished). Corn should be planted as early as soil conditions will allow. In the lower mainland of British Columbia each day of delayed planting resulted in 1 % less whole-plant corn yield (Fairey, 1983).

Soil fertility
Nitrogen and phosphorus response curves for barley silage have been developed in Alberta for each soil zone with different levels of stored soil moisture (McKenzie et al., 1997). Barley varieties had unique response relationships to N application rate (McKenzie et al., 1997), but in general, nitrogen recommendations for silage are 10-20 kg/ha higher than for grain production and phosphorus recommendations are 5-10 kg/ha higher for silage than for grain. Infrequent responses to K and S fertilizers are also seen. Guidelines developed for fertilization of grain crops are the best available for cereal silage crops at present. However, in silage operations the whole plant and not just the grain is removed. Unless manure is returned to the land, soil nutrient deficiencies are more likely to happen. Application of manure and fertilizer should be based on soil test recommendations and predicted crop use. While crude protein levels are affected by fertilizer nitrogen application, fiber and digestibility show only minor effects of fertilizer application.

Seeding rate
In southern Alberta maximum yield for barley and oat silage occurred at seeding rates from 125 to 150 kg/ha (Riemer and Gaudeil, 1983), which is in general agreement with Walton (1975) for the same species in central Alberta. Triticale tillers less profusely than oats and barley so seeding rates at the upper end of the range may be required. Seeding rates required to maximize yield on individual farms may vary greatly, as seed bed preparation, implement used, soil type and moisture and weather conditions, affect emergence, tiller production and death. Further, tiller size can compensate greatly for increases or decreases in tiller or plant density in small grains. Short-season corn for silage can be planted at higher densities than for grain. Corn silage yield increased with increasing planting densities up to 100, 000 plants per ha in studies conducted in short-season areas across Canada (Baron et al., 1987). However, maturity may be delayed and grain content in the silage may be reduced at populations much higher than this.

Mixtures have become novel systems in the past few years, but they are really a reinvention from the past. Producers of the past used mixtures to attain yield stability or reduce year to year variability in yield due to unpredictable weather patterns. Yield and quality of forage or grain mixtures will almost always be an average of their components as pure stands. Rarely will a mixture yield more than the better of the pure stand components. When cereals are used as a forage mixture, the mixture yield is usually less than the cereal component. Therefore any gain in quality must be enough to overcome the value-loss in yield from the cereal. It is important to recognize this because mixtures often cost more to grow than the cereal pure stand. Mixtures require more seed; usually 60 to 75 % of either pure stand component is required. When legumes are used as part of the mixture, some saving in nitrogen fertilizer occurs. Depending on the weed problem, herbicide requirements for these mixtures may be more expensive and complicated than for the cereal grown in a pure stand.

The strategy in choosing mixture components is to find a lower degree of competition between components. While one component will dominate, successful partners in a mixture usually have similar degrees of aggressiveness. Generally from a cereal perspective oats and barley perform better in mixtures than wheat and triticale, with oats better than barley (Berkenkamp and Meeres, 1992). Thompson et al. (1992) observed lower yield using Samson, a semi-dwarf barley, than with two standard barley types in mixtures with Italian ryegrass.

Table 3. Yield, yield loss, percent pea component and change in crude protein (CP) content for cereal-pea combinations harvested for silage at Lacombe (1981-1983)1
loss (%)
Change CP (%)
1Modified from Berkenkamp and Meeres (1992). Pea cultivar was Trapper. All cereals were spring cereals. Mixtures were sown at 90 kg peas:20 kg cereal per ha.

Table 4. Forage yield and quality of pure stands and mixtures from previous studies carried out at Lacombe, AB.
Crude protein
Pea-barley (Aasen, unpublished) grown 1992 and 19931
Virden barley
Radley pea
Winter triticale-barley, (Baron et al., 1992a) grown 1987 and 19881
Leduc barley
Pika triticale
1Pea-barley mixture was planted at peas 160 kg:25 kg/ha barley; tritcale-barley was planted 1:1 by seed number or triticale 68 kg:50 kg/ha barley.

Two types of mixtures are being used: cereals with peas and cereals planted with winter cereals in the spring or with Italian ryegrass. All will improve forage quality, but the cereal-pea and winter cereal-spring cereal mixtures are perhaps better adapted to the prairie climate. Yield loss occurs in all mixtures with lowest losses in mixtures with oats (Table 3). Yield losses of peas and barley can be as high as 19% (Table 3) and as low as zero (Aasen, unpublished). Work by Baron et al. (1992b) compared yields of winter cereal-spring cereal mixtures with yields of spring cereal monocrops and showed responses ranging from gains in silage yields up to 15% to losses of 19% at Lacombe and Brooks over two years. Peas and winter cereals are relatively high in forage quality (Table 4), so gain in quality for the mixture may be expected. Digestibility of the mixture should increase over the pure stand cereal, but usually the biggest difference between pure cereal and mixtures is the reduction in fiber levels. Table 4 indicates substantial reduction in NDF levels for both types of mixtures and Khorasani and Kennelly (1997) observed a 4 to 5 percentage unit decrease in a barley-pea mixture compared to a pure barley stand. However, Khorasani and Kennelly (1997) did not realize an increase in milk production from the mixture compared to a pure stand of barley.

Harvest Schedule and Strategy

Maturity and species
The first step to optimizing yield and quality over large acreage is to spread out the time required to harvest. Using only barley, this period may well be limited. To optimize fermentation and to reduce dry matter losses due to oxidation, silage should reach the silo above 40% dry matter (Daynard, 1978). Heading is the first signal that whole plant dry down has been initiated and is a good indication of relative maturity among cereal species. Averaged over 3 years at Lacombe heading date among several species started as early as June 8 for fall rye and ended July 17 for oats. Tukwa and Virden barley headed on July 15. Dough stage (which is about as late as one could harvest silage) ranged from July 14 for fall rye to August 20 for spring triticale. Dough stage for barley ranged from Aug 2 to 6 (Table 1). Using both barley and triticale could spread harvest out over 3 weeks. Corn would be available for harvest in mid-September at earliest.

Dry down
In small grains silage, dry down begins about 7 days after heading when grain filling begins in earnest. At first the whole-plant dries slowly, but as the kernels fill, the insoluble starch in the kernels dilutes the moisture in the vegetative part of the plant. As the vegetative material dies or senesces dry down rate accelerates. For barley, given average conditions at Lacombe, 30% dry matter will be attained about 17 days after heading (Baron et al, 1992a). Whole-plant barley will dry at about a percentage unit per day under average temperature conditions in Central Alberta. If temperatures are above average and the crop is under stress, drying can occur at a faster rate. Thus barley may move through the 30 to 40% range in less than 10 days. Some drying will occur between cutting and when it enters the silo, so producers should allow more time than indicated by the dry down rate or cut earlier to avoid over-dry silage.

Less information exists on whole-plant dry-down in oats and triticale. Work at Lacombe indicates that oat and triticale dry at the same rate as barley, initially. But, drying does not accelerate in the later stages of grain filling. It appears that whole plant drying of triticale actually slows down as the kernels enter the later dough stage. While Daynard (1978) concluded that corn silage should not be harvested before 30% dry matter, Fairey and Fisher (1979) argued that corn could be stored in bunker silos at 25% dry matter without risk of seepage losses. Fairey (1980) observed that 40 corn heat units were required to increase percent dry matter by 1 unit after silking. To move from silking to 25% dry matter would require 400 corn heat units and from silking to 30% dry matter, 600 corn heat units. Given that it would take about 1400 corn heat units to reach silking with short season corn hybrids, at least 1800 and 2000 corn heat units would be required to reach the silage stage (i.e. 25 and 30% dry matter, respectively).

Silage yield implications
Depending on the year and species about 30 to 60% of final whole plant yield accumulates in small grains during the approximately 30-day period between heading and grain maturity (grain filling period). At the point where silage harvest should occur in barley whole-plant yield is about 85% of maximum. By the time 40% dry matter is attained (perhaps 10 days later) whole-plant yield is at about 95% of maximum. In achieving an optimum percent dry matter, yield loss will range from 5 to 15% (Baron et al., 1992a). Some wilting will occur between cutting and silo filling; therefore it may be necessary to initiate harvest earlier, with greater initial losses of yield so that average dry matter percentages are within the 30 to 40% range. Since vegetative die-off does not move as quickly for triticale and oats, it may be possible to delay initial harvesting while attaining acceptable moisture content for harvest and silo filling. This would result in less yield loss. Daynard and Hunter (1975) observed that maximum whole-plant yield for corn occurred at 30 to 34% dry matter.

Forage quality implications
In both corn and barley digestibility tends to remain constant, or decreases very slightly (Daynard and Hunter, 1975; Baron et al, 1992a). Oat decreases in digestibility during the grain filling period. Triticale decreases less than oats but more than barley. During grain filling corn stover declines in digestibility. The whole-plant remains constant for digestibility, because the declining stover is offset by highly digestible corn grain, which is accumulating to a high proportion of the whole-plant (Daynard and Hunter, 1975). It can be assumed that the same mechanism occurs in barley. In oats the effect of accumulating grain is not sufficient to overcome the declining straw quality. Whole-plant NDF and ADF contents, tend to decrease slightly for small grains during the filling period (Cherney and Marten, 1982). Harvesting at this point should be advantageous for intake, but likely occurs in the 30 to 40% dry matter zone, since this is in the mid to latter portion of the grain filling period. Coors et al (1997) observed that whole-plant corn had about five percentage units lower NDF when ears were normally filled compared to when ears were restricted to half of their kernel capacity.

Table 5. Relationships among whole-plant dry matter percentage, relative silage yield, relative kernel fill, in vitro digestible organic matter (IVDOM) and dough stage in barley1.
Whole-plant dry matter303540
Relative silage yield1859295
Relative kernel fill1607787
Dough stageEarlyMidHard
1Baron et al. (1992a). Relative silage yield is a percent of maximum yield. Relative kernel fill is percent of final or mature kernel weight.

Grain content implications
Over the range of 30 to 40% whole-plant dry matter, Baron et al. (1992a) found that barley grain ranged from 60 to 87% of fullness (a measure of grain maturity). Daynard and Hunter (1975) estimated that grain corn was approximately 80 to 92 % of fullness. These relationships are about grain maturity during silage harvest, but the effect of grain content on milk production is likely a function of grain as a percentage of whole plant (harvest index) which is highly variable from year to year. For example in barley, Baron et al. (1992a) found that harvest index could range from 30 to 43% at 30% silage dry matter and from 37 to 52 % grain content at 35% silage dry matter. Fisher and Fairey (1979) reviewed corn silage literature, which indicated that milk production responded positively to increases in grain content within narrow ranges of whole-plant percent dry matter. This may have been due to effects of starch on rumen dynamics or simply a lower whole-plant NDF as a result of dilution of cell walls by highly digestible grain as suggested by Khorasani and Kennelly (1997) for barley. Other cereal species did not show the same effect. These effects are hard to manipulate by management, because of the impact of the environment on grain to straw or stover ratio.

Harvest time indicators
Traditionally dough stage has been used as an indicator of whole-plant yield or moisture level. The doughiness of kernels is a function of kernel starch and moisture content. Thus there is room for a lot of error. Initiating harvest at milky to early dough is likely a good policy for most small grains. Checking whole-plant moisture content is much more reliable, when we know how fast dry down may be and how many days after heading the linear rate of drying may occur. In corn the kernel milk line has been used as an indicator for harvest (Crookston and Kurle, 1988). This is likely valid where corn is well adapted and in years when there is an abundance of corn heat units. In short-season areas like Alberta, in some years corn will only reach slightly beyond the blister stage. Waiting for milk stage may invite yield loss due to frost damage. Having a good knowledge of the hybrid grown (maturity), and following corn heat unit accumulation after silking as suggested by Fairey (1980), is likely a good approach to decide when to begin corn silage harvest.

Impacts of Choices on Production and Economics

Optimizing implies compromising factors to achieve an objective or overall gain. Professionals often present results of individual factors, indicating percentage improvement in production. However, the true value of individual factors can only be seen in the context of the larger production system (milk production in this case). Decision support systems help integrate several factors, in this case cereal silage yield and forage quality, predicted milk response, costs and projected revenues. In these examples we will use the Speadsheet Milk95 (Undersander et al., 1993), produced by the University of Wisconsin and modified for use in Alberta.

Assumptions for milk production are listed in Table 6a. Equations utilized may be seen in Undersander et al. (1993) and on the University of Wisconsin Web Site. It is important to note that responses here will not be identical to your farm and to recognize that year to year and location to location differences among species and varieties occurs. The University of Wisconsin web site indicates that only treatments with known significant differences should be compared. In some examples to follow we have adjusted representative data to observe economic implications, because all treatments are not found within the same or identical experiments.

Semi-dwarf vs. standard barley (different forage quality - same yield)
The biggest advance for barley silage is the use of semi-dwarf types. Improved digestibility and lower NDF and in this case higher crude protein, means that less barley grain and protein supplement need to be added to the ration. Potentially, more milk is produced for roughly the same production cost. The standard type compared is a variety, which has very low quality so the difference between semi-dwarf and standard type is overstated. Here, the semi-dwarf produces 2.2 times the milk per tonne (Table 6b) of silage and $200 more milk per ha (Table 6c). While difference between all standards and semi dwarfs may not always be this large, this example shows the wide variation for forage quality within barley and the economic impact it has.

Table 6a. Assumptions used in spreadsheet calculations1.
Cow Weight
Stage of lactation
Mid lactation
Milk yield
Milk price
Feeding losses
% of harvested yield
Barley price
44% CP soybean supplement
Daily overhead
Estimated DMI
1Equations may be found in Undersander et al. (1993) and University of Wisconsin Web Site

Spring triticale vs. barley (similar forage quality – different yield)
Triticale has emerged as a high yielding silage crop in central Alberta. Generally, triticale nutritive value is not as high as semi-dwarf barley, but it usually yields more. The higher cost of production for triticale is due to a higher seed cost and harvesting cost (Table 6c). Similar milk production per tonne of triticale compared to semi-dwarf barley is due to a lower NDF in this example. (Table 6b). The higher yield for triticale results in a 30% increase in milk value per hectare over semi-dwarf barley.

Corn silage vs. small grain silage (high cost, high quality vs. moderate cost, lower quality)
Corn silage produced the highest milk production per tonne (Table 6b). This is mainly because of a high digestibility, requiring less barley added to the ration than small grains to achieve the milk production target (Table 6a). The high milk production per tonne for corn is offset by a high cost of crop production (Table 6c). For corn, seed and irrigation costs were $98.80 and $56.56/ha, respectively. Thus while milk production per tonne was relatively high, it’s net value of milk production was less than the semi-dwarf barley and triticale (Table 6c). As corn production becomes more reliable, it could become a major player in the dairy industry, at least in southern Alberta.

Mixtures vs. small grain pure stand
There are several issues here. The first is that the increase in forage quality of mixtures is greater if the species grown with either peas, winter cereals or ryegrass is of low quality (Baron et al., 1992b). If we compare the forage quality of any of the mixtures (Table 6b) with the high quality semi-dwarf vs. the standard type, the improvements are not as great with the semi-dwarf as with the standard variety. Thus without any yield loss for mixtures (high yield examples), the milk yield per tonne may only be slightly larger than the semi-dwarf pure stand (Table 6b). The value of milk production per tonne of silage is slightly less for the mixture than the pure stand semi dwarf or triticale, because the cost of crop production is greater for the mixture (Table 6c). Seed cost for the pea mixture is especially high, even though fertilizer requirements are reduced.

Table 6b. Silage yield, in vitro organic matter digestibility (IVDOM), neutral detergent fiber (NDF) and estimated milk produced per tonne of forage.
Crude protein (%)IVDOM
Pure stands of cereal silage
Spring triticale1
Corn (irrigated)2
Barley based mixtures
Low yield
High yield
Barley-W. triticale4 Low yield
Barley-W. triticale4
High yield
1 Cereal silage yield and quality averaged over 3 years of data from Lacombe, AB.
2 Irrigated corn yield southern AB (Kotowich and Van Biert, 1998).
3 Yields set at standard barley (1) for High yield and 20% yield loss for low yield. Forage quality data a composite of Aasen (unpublished) and Khorasani and Kennelly (1997).
4 Yield set at standard barley (1) for High yield and 20% yield loss for low yield. W. stands for winter. Forage quality data (Baron et al., 1992a).
5 Milk yield calculated from Milk95 spreadsheet (Undersander et al., 1993); University of Wisconsin Web Site.

The second mixture issue is the cost of yield loss. Yield losses, relative to the pure stand of barley will happen. If they don’t happen this year they will next year. In our literature review the responses for forage quality of the mixtures were remarkably similar for either the pea or winter cereal type (Tables 3 and 4), so both mixture types will improve forage quality. The yield loss used here (Table 6b) may be at the extreme at 20%, although these levels were reported for pea-barley mixtures (Berkenkamp and Meeres, 1992). The yield loss of either mixture-type reduces the potential profit per acre to below all of the small grain types. (Table 6c). It costs the same to grow the mixture with a loss or no yield loss. Thus, while mixtures have the potential to improve forage quality considerably, the extra cost per ha and the possibility of yield loss may make mixtures no better than the highest quality cereal variety grown as a pure stand.

Table 6c. Crop production and harvest costs and estimated value of milk production above costs per tonne of forage and per ha.
Value of milk produced
Production ($/ha)
HarvestAbove costs ($/t)($/ha)
Pure stands of cereal silage
Spring triticale1
Corn (irrigated) 2
Barley based mixtures
Low yield
High yield
Barley-W. triticale4 Low yield
Barley-W. triticale4
High yield
1Cereal silage yield and quality averaged over 3 years of data from Lacombe, AB.
2Irrigated corn yield southern AB (Kotowich and Van Biert, 1998).
3Yields set at standard barley (1) for High yield and 20% yield loss for low yield. Forage quality data a composite of Aasen (unpublished) and Khorasani and Kennelly (1997).
4Yield set at standard barley (1) for High yield and 20% yield loss for low yield. W. stands for winter. Forage quality data (Baron et al., 1992a).


Optimizing cereal silage production practices for the dairy system involves give and take among yield, nutritive value and percent dry matter for the species of choice. In Alberta, this species is likely barley, especially semi-dwarf barley where it can be grown. Spring triticale and corn are also viable options in some scenarios. Harvesting cereal silage within the 30 to 40% dry matter range should optimize silage fermentation, yield and nutritive value, but silage dry matter content is the critical factor in deciding when to harvest. Many options are available in cereal silage production, but their adoption should be weighed against the profitability of barley grown in a pure stand on the basis of economic impact within the milk production system.


Baron, V.S., T.B. Daynard, G. Dupuis, N.A. Fairey, D.J. Major, J.F. Muldoon, and R.P. White. 1987. Evaluation of early-maturing European and Canadian corn hybrids for grain and forage production in Canada. Maydica XXXII: 33.

Baron, V.S., A.C. Dick, and M.S. Wolynetz. 1992a. Characterization of barley silage-maturity relationships for central Alberta. Can. J. Plant Sci. 72: 1009.

Baron, V.S., H.G. Najda, D..F. Salmon, and A.C. Dick. 1992b. Post flowering forage potential of spring and winter cereal mixtures. Can. J. Plant Sci. 72: 137.

Baron, V.S., D.F. Salmon, and J.G. McLeod. 1999. The evaluation of spring and winter triticale lines for forage quality. AARI Final Report # 95M788 29 pp.

Berkenkamp, B. and J. Meeres. 1992. Mixtures of annual crops for forage in central Alberta. Can. J. Plant Sci. 67: 175.

Cherney, J.H. and G.C. Marten. 1982. Small grain crop forage potential: I. Biological and chemical determinations of quality and yield. Crop Sci. 22: 227.

Coors, J.G., K.A. Albrecht, and E.T. Bures. 1997. Ear fill effects on yield and quality of silage corn. Crop Sci. 37: 243.

Crookston, R.K. and J.E. Kurle. 1988. Using the kernel line to determine when to harvest corn for silage. J. Prod. Agric. 1: 293.

Daynard, T.B. 1978. Practices affecting quality and preservation of whole-plant corn silage. Can. J. Plant Sci. 58: 651.

Daynard, T.B. and R.B. Hunter. 1975. Relationships among whole-plant corn silage. Can. J. Plant Sci. 55: 77.

Fairey, N.A. 1980. The effects of hybrid maturity, date of planting and date of harvesting on growth and development of forage maize. Can. J. Plant Sci. 60: 1367.

Fairey, N.A. 1983. Yield, quality and development of forage maize as influenced by dates of planting and harvesting. Can. J. Plant Sci. 63: 157-168.

Fisher, L.J. and N.A. Fairey. 1979. Factors influencing the utilization by ruminants of corn silage in marginal growing areas. Can. J. Anim. Sci. 59: 427.

Khorasani, G.R. and J.J. Kennelly. 1997. Optimizing cereal silage quality. Proc. of the 1997 Western Canadian Dairy Seminar. 9: 249.

Kotowich, J. and P. Van Biert. 1998. 1998 Forage and pasture enterprise cost & return analysis (by ecoregion). Alberta Agriculture Food and Rural Development, Markets & Economics Division) Edmonton, AB. 67pp.

McKenzie, R.H., A. Middleton, E. Solberg, J. DeMulder, and H. Najda. 1997. Optimizing barley silage production in Alberta. Alberta Agricultural Research Institute Project 94M608 Final Report. Edmonton, AB.

McLeod, J.G., Y.T. Gan, D.F. Salmon, and V.S. Baron. 1998. Triticale – biomass potential and quality on the Canadian prairies. P 264 – 267 in P.Juskiw (ed) Proc. 4th International Triticale Symposium, Vol. 2. July 26-31, 1998, Red Deer, AB.

Riemer, G.E. and R.G. Gaudiel. 1984. Annual forage crops for Alberta. Farming for the Future Project #78-0181 Final Report. Alberta Horticultural Research Center, Brooks, AB.

Thompson, D. J., D.G. Stout, T. Moore, and Z. Mir. 1992. Yield and quality of forage from intercrops of barley and annual ryegrass. Can. J. Plant Sci. 72: 163.

Undersander, D.J., W.T. Howard, and R.D. Shaver. 1993. Milk per acre spreadsheet for combining yield and quality into a single term. J. Prod. Agric. 6: 231.

Walton, P.D. 1975. Annual forage seeding rates and mixtures for central Alberta. Can. J. Plant Sci. 55: 987.
  For more information about the content of this document, contact Grant Lastiwka.
This document is maintained by Linda Hunt.
This information published to the web on June 17, 2004.
Last Reviewed/Revised on June 4, 2014.

Home | Contact Us | Privacy Statement
The user agrees to the terms and conditions set out in the Copyright and Disclaimer © 2004 - 2017 Her Majesty the Queen