Forage Yield and Quality Distribution of Diverse Pastures

J.R. Gerrish


  1. Determine total and seasonal forage yield and quality distribution patterns for monoculture and mixed cool-season grass-legume pastures in rotationally grazed swards.
  2. Evaluate persistence of forage species in rotationally grazed mixed swards.
  3. Characterize botanical composition of mixed swards with rotational grazing.

Research Procedures: Sixteen grass or grass-legume mixtures were established in a completely randomized design with four replications. The grass components were seeded on September 5, 1994 on a prepared seed bed. Grass seed was broadcast and rolled in with a cultipacker. The legume components were frost seeded on March 10, 1995. Red clover and birdsfoot trefoil establishment with frost seeding was very good while alfalfa establishment was very poor. Due to the near absence of alfalfa in the designated plots, the plots identified as TF or SB + alfalfa are being used as 0 N treatments for comparative purposes. Individual plots are 50 X 50 ft with a 10 ft bluegrass alleyway surrounding each plot. Swards were uniformly managed to encourage establishment during the 1995 growing season. Tall fescue and smooth bromegrass monocultures receive 120 lb N/acre annually as three 40 lb applications applied in March, June, and September.

Each mixture was sown with a target seeding rate of 80 seeds/sqft. Mixtures were constructed in the following manner. First order pastures were monocultures of either tall fescue or smooth bromegrass. Second order pastures were binary mixtures of each base grass with either red clover, birdsfoot trefoil, or alfalfa added. Binary mixtures were seeded at 40 seeds-m/sqft for each principal component. Third order mixtures contained the base grass (40 seeds/sqft) in combination with all three legumes (13 seeds/sqft for each component). Fourth order mixtures contained both base grasses (20 seeds/sqft for each component) with all three legumes. Fifth order mixtures contained both base grasses, all three legumes, plus orchardgrass and timothy all in equal proportions. Sixth order mixtures contained all of the above plus big bluestem.

Tall Fescue Base

1) Tall fescue + 120 lb N/acre
2) Tall fescue + alfalfa
3) Tall fescue + birdsfoot trefoil
4) Tall fescue + red clover
5) Tall fescue + alfalfa + birdsfoot trefoil
6) Tall fescue + alfalfa + birdsfoot trefoil + red clover
7) Tall fescue + alfalfa + birdsfoot trefoil + red clover + orchardgrass + timothy
8) Tall fescue + smooth bromegrass + alfalfa + birdsfoot trefoil + red clover + orchardgrass + timothy

Smooth Bromegrass Base

9) Smooth bromegrass + 120 lb N/acre
10) Smooth bromegrass + alfalfa
11) Smooth bromegrass + birdsfoot trefoil
12) Smooth bromegrass + red clover
13) Smooth bromegrass + alfalfa + birdsfoot trefoil
14) Smooth bromegrass + alfalfa + birdsfoot trefoil + red clover
15) Smooth bromegrass + alfalfa + birdsfoot trefoil + red clover + orchardgrass + timothy
16) Smooth bromegrass + tall fescue + alfalfa + birdsfoot trefoil + red clover + orchardgrass + timothy + big bluestem

In 1996 and 1997, 3 X 20 ft strips in all plots were mechanically harvested with a Carter Harvester at two-week intervals beginning around May 10 and continuing until mid July to develop baseline production data. One half of each plot was mown in early June so that summer regrowth could also be evaluated. Data collection began on regrowth plots 21 days after clipping. A sub-sample was taken from each bulk harvest sample for moisture determination and forage quality analysis. At each harvest forage height was measured both with a rising plate meter and simple yardstick. Visual estimate of species canopy cover was made at each date. Quadrat samples were clipped for hand separation of component species in late May and late June.

Beginning in 1998, plots were grazed according to the following protocol. Each plot was sampled and grazed individually whenever the mean sward surface height reached 8 to 10 inches. Mean sward surface height is defined as the height below which an estimated 90% of the forage biomass occurs. Plots were grazed with six steers for four to seven hours to remove approximately 50 – 60 % of the forage biomass. During the first grazing cycle, all fescue-based plots were grazed first followed by smooth bromegrass- based plots. In subsequent grazing cycles, plots were grazed strictly on an as needed basis.

Prior to grazing the following measurements were made on the plots to be grazed the following day.:

1) Mean sward surface height. Measured at each of the six locations where quadrats were cut.

2) Estimate of species composition within a 2-yard square area surrounding the quadrat.

3) Six .25 m2 quadrats clipped to 1 inch height.

    A) Three samples were cut from the south ½ of the plot and three samples from the north ½. Each set of three quadrats were combined in a single sample bag. The sample bag was labeled with date, plot number, and north or south designation.

4) In the lab, each sample bag was individually weighed and a 100 to 150 g subsample was collected to determine moisture content of the forage.

5) The subsample was weighed wet and then oven dried at 155o F for 72 hours.

6) The subsample was retained for potential forage quality analysis. When the samples were ground, the two subsample bags were combined for a single quality sample and were labeled with the date and plot number.

After grazing the following measurements were made:

1) Estimate of percent bare ground.

2) Residual sward height

3) Residual forage on selected plots was also measured by clipping six quadrats as described above.

4) Processing for determination of dry matter content was as described above but no residual samples were processed for quality analysis.

Results and discussion

ResultsThe mean sward height at initiation of grazing was approximately 10.5 inches which was slightly taller than our target of 8 to 10 inches (Figure 1). Rapid growth early in the season allowed some plots to exceed target height before they could be grazed the second time. The mean dry matter availability at initiation was slightly less than the expected target level of 2700 lb/acre (Figure 2). The expected forage availability level was based on height:yield relationships previously developed at FSRC. In this study, quadrats were clipped above the thatch layer rather than at ground level as was done in the calibration study.

Figure 1. The mean sward height at grazing initiation was slightly above the target level of 8 to 10 inches.
Figure 2 Forage availablity at turn in was less than what was predicted by mean sward height.

Mean rest period required to reach the target grazing height was greater for smooth brome based pastures than for tall fescue based pastures (Figure 3). Mean rest period was 28 days for SB pastures while only 22 days for TF pastures. Nitrogen fertilized TF had the shortest mean rest period at 19 days. The range in length of rest period for TF + N pastures was 15 days in May to 28 days in July-August. The longest mean rest period was required by smooth brome + red clover at 31 days with a range from 25 days in May to 41 day in July-August. Excluding the N-fertilized TF and SB treatments, SB-based pastures usually required about five days additional rest through the summer months compared to TF-based pastures (Figure 4). The shorter rest periods in August are a reflection of the summer annual grass component, primarily crabgrass, in the swards.

While it has been fairly easy to maintain orchardgrass as a strong component in tall fescue based pastures with rotational grazing, maintaining smooth bromegrass in a tall fescue mixture has been much more difficult. If a pasture is grazed using tall fescue condition as the guide for initiating grazing, the smooth bromegrass will likely not be adequately rested. Basing turn-in on smooth bromegrass condition may reduce palatability and quality of the fescue component which may result in selective grazing of the brome component. Based on these observations, pasture mixtures should not include both smooth bromegrass and tall fescue.

Figure 3. Mean length required to reach target grazing height of 8 to 10 inches for sixteen pasture mixtures.
Figure 4. Smooth brome based pastures typically require a longer rest period than tall fescue based pastures.

As would be expected based on relative length of rest period, mean daily growth rate for TF-based pastures was significantly greater than SB-based pastures (Figure 5). Mean daily growth rate was similar for TF + N, TF + BFT, and TF + BFT + RC, however, there were very notable monthly differences among the three treatments (Figure 6). Daily growth rate of TF + N increased very markedly following N applications in early April and mid-June and was significantly greater than the TF + legume mixtures during April, May, and July. During June and August, the TF-legume mixtures exhibited significantly greater daily growth. Daily growth rate of TF + N pastures fluctuated from 33 to 89 pounds/acre/day while the range for TF-legume mixtures was from 47 to 70 pounds/acre/day. The range for TF + N is 270% compared to only 50% for TF + legumes. Greater flexibility in management is required to accommodate the larger changes in daily growth rate associated with N fertilization. Either stocking rate, size of paddock, or amount of forage harvested as hay or silage should be adjusted more often and to a greater extent for TF + N compared to TF + legumes. There was less overall variance in daily growth rate among SB-based pastures than among TF-based pastures (Figure 7).

Figure 5. Mean daily growth rate of sixteen pasture mixtures during April through August growing period.
Figure 6. Monthly variance in daily pasture growth rate between tall fescue + 120 lb N and tall fescue + legumes.
Figure 7. Monthly variance in pasture daily growth rate for tall fescue (TF) or smooth bromegrass (SB) based pastures.

The low relative variance in monthly mean daily growth rate is in stark contrast to the growth curves for cool season forages that are often presented to producers. The typical growth curves represent either unmanaged growth or what might be expected in a hay management system. The data presented here indicate that seasonal variance in growth distribution in managed pastures is much less than many graziers believe. The variance occurring in mixed grass-legume pastures is much less than that occurring in grass monocultures. The inclusion of warm season annuals such as crabgrass and lespedeza in the mixture are also very likely contributing to the higher level of growth observed during July and August. The summer annual species present in these plots were not sown as part of the mixtures but are volunteer components. Many pastures in this region contain these or other summer annual species which may contribute significantly to summer carrying capacity.

Forage dry matter production is presented for the April through mid-September period Figure 8) and also for the total growing season (Figure 9). All pastures were rested from their final grazing in mid-August or early September and allowed to stockpile growth for winter grazing. Stockpile forage yield was measured after the end of the growing season in early November. During the spring-summer period, forage yield was similar for most of the pastures, although yield distribution varied. Only six mixtures were significantly lower yielding than the highest yielding mixture. When the stockpile phase is included, TF + 120 N was the highest yielding pasture and it was significantly greater than ten other treatments. Previous studies at FSRC have shown TF + RC to produce fall stockpile yields comparable to TF + 60 lb N/acre with N applied in mid-August. Because there was over 30 days variance, both within and among treatments, in the beginning of the stockpiling phase in this study, that data is not presented as a single component.

Figure 8. Forage dry matter yield produced from April to late August-early September.
Figure 9. Total season forage dry matter yield produced from April through October.

Visual estimates of percent canopy cover by each individual species were made prior to grazing each plot. All legumes, cool season grasses, warm season perennial grasses, and summer annual grasses were combined to give total forage cover. Total forage cover did not vary greatly through the season for most pasture mixtures (Figure 10). Comparison of the TF and SB monocultures to mixtures of each base grass with five other species shows little difference in total cover. There was a trend towards lower total forage cover in SB-based pastures compared to TF-based pastures during the fall period. An average forage cover of 75 to 90 % is considered to be in the good category.

When individual components or classes of components are compared, the pastures appear much more dynamic in composition. Even the monoculture pastures exhibit a degree of diversity and composition changes through the season (Figures 11 & 12). In plots seeded with only tall fescue, fescue accounted for less than 50% canopy cover on a seasonal basis. Other cool season grasses, primarily bluegrass, and summer annual grasses made up an additional 25 to 35% of the canopy. Legumes were less than 10%. Smooth bromegrass presence declined rapidly through the season, accounting for only 10% of canopy cover in August and September, even in the smooth bromegrass monoculture pasture. Because of the open sod nature of bromegrass and its lack of competitive ability in summer, summer annual grasses reached 50% canopy cover in late summer. Another factor contributing to the high presence of summer annual grasses was the extremely wet June with 10.85 inches of rain. Bromegrass is very susceptible to trampling damage during wet periods while fescue is much more resilient. As with fescue monoculture, legumes contributed less than 10% of the seasonal yield, illustrating the detrimental effect of N-fertilization on legume persistence.

Figure 10. Seasonal changes in total forage canopy cover for monoculture and mixed pastures.
Figure 11. Seasonal changes in component composition of tall fescue + 120 lb N/acre pastures.
Figure 12. Seasonal changes in component composition of smooth bromegrass + 120 lb N/acre pastures
Figure 13. Seasonal changes in component composition of tall fescue based pasture with five other sown components.
Figure 14. Seasonal changes in component composition of smooth bromegrass based pasture with five other sown components.

Trends in seasonal composition changes of mixed pastures was similar between TF and SB-based pastures (Figures 13 & 14). As with the monoculture pasture, smooth bromegrass as a component in the mixture declined to less than 10% of canopy cover during late summer. Legume content was greatest in spring for both pastures and declined in late summer. Brome based pasture maintained a higher level of legume presence season long compared to tall fescue based pasture. This is probably a reflection of the greater competitive ability of tall fescue through the summer months. Summer annual grasses were the major component of total forage yield in August and September for both pasture types. Crabgrass was the major summer annual constituent with lesser amounts of foxtail and barnyardgrass. While foxtail and barnyardgrass declined in palatability later in the season as evidenced by observed animal selection, crabgrass was readily grazed through the last grazing in September. Surprisingly, observed dead material did not significantly increase late in the season in response to the high presence of summer annual grasses. This observation also provides evidence that most summer annuals were being grazed even late in the season.


This study is evaluating the relative merits of pastures varying in degree of species complexity. Initial findings indicate that within context of similar base grasses, even simple mixtures are as productive as pastures of single sown grasses receiving 120 lb N/acre. Complex mixtures do not necessarily provide for greater forage production or uniformity of seasonal yield distribution than do simple mixtures. One very revealing result was that just three years after seeding of a pure stand of a single grass species, that sown specie may provide less than 50% of the total forage yield. The very significant contribution of weedy summer annual grasses to total annual forage production in cool season pastures has also been demonstrated. We hope to continue this study for several years to learn more about seasonal changes in composition and quality of mixed pastures.

May 17, 1999