WICST Management-Intensive Rotational Grazing of Dairy Heifers

Janet Hedtcke [1] and Jim Stute [2]

Introduction

Management-intensive rotational grazing (MIRG) is a low input form of dairy herd management that is increasing throughout the United States. In Wisconsin, about 23 percent of dairy farmers used MIRG last year, more than triple that since 1993 (Jackson-Smith and Powell, 1999). Of the three dairy forage systems represented in the Wisconsin Integrated Cropping Systems Trial (WICST), rotational grazing is the lowest input system which measures productivity in terms of animal performance and forage production. Long-term effects of animals on soil and vegetation are also being monitored.

Materials and Methods

Pastures were established at the Lakeland Agricultural Complex (LAC) and the Arlington Agricultural Research Station (ARS) in 1990. Fencing and water lanes for pastures at both locations were built during the spring of 1992. Fig. 1 shows an example of paddock layout at ARS which was 0.70 acres; layout was similar at LAC but paddock size was 300’ x 120’ or 0.83 acres. Initially, species included red clover, smooth bromegrass and timothy; later, orchardgrass, perennial ryegrass and reed canarygrass were also included (see Table 1 below for seeding dates and rates). At LAC, the forage was mechanically harvested until the spring of 1992 (see Table 5 for yield) when grazing began. At ARS, severe winterkill of the grasses required reseeding in 1992 and grazing began in the spring of 1993. Red clover was seeded into the pastures on alternative years through 1995. Thereafter, red clover was seeded at a reduced rate when deemed necessary. Two of the paddocks at LAC (reps 1 and 4) were tilled and reseeded in August of 1996 to repair severe trampling damage which occurred during the early summer wet weather. The fall seeding of 1996 at LAC failed and there was no reseeding there until the fall of 1997.

Table 1. WICST rotational grazing seeding dates and rates.

* Seeded with no-till drill; otherwise broadcast seeded except for drilling with 1990 establishment.

* * reps 1 and 2
*** Heavy seeding due to poor pasture conditions

Results and discussion

Pasture Management

Arlington 1999. Grazing began May 4 in 1999. Continuing with the same management of the previous three years (put and take system), twelve dairy heifers flash grazed the paddocks the first week of the grazing season. We used the flash grazing practice to control some of the growth surge of the spring grass and avoid the need for mechanical harvest. No hay was mechanically harvested during the season. Beginning in mid-May, paddocks were subdivided with temporary electric wire into third or quarter sections and a back fence was put up to prevent animals from grazing new regrowth. An alleyway was set up in each paddock to allow animals access to water at all times. Approximately 1500 square feet of the paddock was used as this alleyway. Six heifers were removed on June 1 to accommodate pasture supply. Throughout the grazing season of 1999, the heifers were cycled through the four paddocks 5 times. Moisture was at or above the 30-year average through July. However, it was extremely dry from late July to October which severely slowed the rate of pasture growth. Grazing ended on September 14, about 50 days short of our goal of 180 grazing days.

Lakeland 1999. Seven dairy heifers were put on pasture trial on April 28. Three heifers were removed from the paddocks on June 30 to accommodate pasture growth. Since regrowth was slow, no back fence was set up; therefore, no alleyway was necessary. No hay was mechanically harvested during the season. Heifers were cycled through the paddocks five times during the season. Similar to ARS, dry weather in late summer reduced forage growth and grazing season ended September 23, for a total of 149 grazing days.

Arlington 2000. On April 21, paddocks were evaluated using a pasture condition score sheet compiled by Cosgrove et al. (1996). All plots were in the ‘good’ category. Late growth of legumes and low plant density reduced the score from the ‘ very good’ category. After several discussions with researchers and dairymen practicing MIRG, we decided to change our stocking strategy slightly to improve the forage stands. Instead of stocking the pastures heavily in the early spring to control spring growth, we decided to keep the stocking rate constant over the season and harvest excess forage as hay. Grazing began on April 25th. Rainfall was average in April but heavy in May and June. Pasture growth was good all season. Hay was harvested from plot 207 in mid-May yielding 1.15 tons dm/a and from plot 302 in mid-June yielding 1.68 tons dm/a. It was not necessary to feed this hay back to these heifers in mid-summer. All plots but 207 was fertilized with 40 lb N/a in mid-June. The four heifers cycled through each paddock approximately six times during the season for a total of 162 grazing days.

Lakeland 2000. Similar to ARS, the stocking density of four heifers was held constant over the grazing season. Grazing began on May 2 and ended on October 13 for a total of 164 grazing days. Hay was harvested once in late June from plot 408 yielding 1.59 tons DM/a. No N fertilizer was applied. Although heavy rains in May and June (200% above long-term average) hampered field operations in other systems, grazing continued without problems. Grasses, especially reed canarygrass are extremely productive in wet conditions. Heifers cycled through the plots four times during the grazing season.

Animal Details

Arlington 1999. Eighteen dairy heifers were chosen for the study. All 18 animals were weighed on a shrunk-weight basis at the initiation of the grazing season. The control group, which would remain in confinement, weighed an average of 499 lbs. The trial group, which would be on pasture, weighed an average of 439 lbs.

Heifers were supplemented with four lbs of grain per head per day from May 4–June 23. After this time, the grass was essentially all vegetative, leafy material and grain supplement was decreased to two lbs per head per day for the remainder of the grazing season. In total, 364 lbs grain/animal was supplemented for the season.

Midseason weights were recorded on the six controls and the six animals remaining in the trial group (see Table 2). The six heifers that were taken off the trial early were weighed the day after they were removed from the trial (group A, Table 2). This group’s minimal weight gain may be due to stress and poor adjustment to high fiber diet as well as poor forage quality during rapid grass maturity.

The remaining six heifers (group B, Table 2) adapted to the pasture system and the frequent movement with final gains of 1.82 lb/head/day. Confinement heifers gained an average of 2.08 lb/head/day. The range in gain was quite similar for both groups: 1.59 to 2.16 and 1.82 to 2.20 lb/head/day for pasture and confinement groups, respectively (data not shown). Final body weights of pastured heifers averaged 668 lbs/head, slightly below breeding size.

Lakeland 1999. Seven heifers were weighed before being put on pasture with an average initial weight of at 609 lbs. Supplement included barley grain at 2.5 lbs/hd/day from April 28 to June 30. Grain supplementation increased to 3.7 lbs/hd/day for the remainder of the season. A total of 475 lbs grain/animal was supplemented during the season. During the last week on pasture (September 19-23), heifers were given 40 lbs corn silage, 8 lb steer mix, and 35 lbs alfalfa hay to compensate for low forage availability.

Midseason weights averaged 735 lbs on June 30. Three heifers were removed from the trial at this time with an average daily gain of 1.41 lb/day/head (Table 2, LAC Group A). Final gains averaged 1.84 lb/day/head (Table 2, LAC Group B). Final body weights averaged 886 lbs/head, ideal size for breeding. No data was collected for the control group.

Arlington 2000. Ten heifers were selected for the study and initial weights were recorded on a shrunk-weight basis. Six of the ten heifers were controls and remained in confinement with initial average weight of 499 lbs. The initial average weight of the four heifers in the pastured group was 476 lbs. They were given the entire paddocks for the first ten days on the trial to control early forage growth. Thereafter, paddocks were divided into quarters and heifers were moved about every 2 to 3 days to new area.

Heifers were supplemented four lbs of corn grain per head per day for the first month; grain was later reduced to two lbs per head per day for the remainder of the grazing season. In total, 366 lbs grain/animal was supplemented for the season.

Midseason weights averaged 628 lbs for pastured heifers and 713 lbs for confinement heifers. At this time, heifers were dewormed and vaccinated if not already done. Heifers were removed from the trial on October 5 due to cold weather conditions. Average final weight was 759 lbs from the pastured heifers and 856 for confinement heifers. Average daily gain was 1.75 lbs/hd (ranging from 1.65 to 1.86 lbs/day) for pastured animals and 2.20 lb/hd (ranging from 1.89 to 2.27 lb/day).

Lakeland 2000. Initial average shrunk weight of the four heifers was 529 lbs. There was no data collected on a confinement group. Paddocks were quartered and heifers were given new sections every two days.

Heifers received two lbs corn grain per day per head for the entire season. Total grain supplemented for the season was 328 lbs/hd.

No mid-season weights were measured. Final weight averaged 762 lbs/hd. Average daily gain averaged 1.42 lb/head (ranging from 1.13 to 1.77 lbs/hd). Cold temperatures and cold rains in October may have hindered animal productivity.

Table 2. Weight gain of pasture animals at Arlington Research Station (ARS) and Lakeland Agricultural Complex (LAC), WICST 1999 and 2000.

Table 3 shows the current summary of animal performance at both ARS and LAC from 1992-2000. It is apparent that the grazing days has been well below the expected 180 days at both sites over the years. Grazing usually begins in late April but rarely extends beyond mid-September due to either lack of forage or cold weather conditions.

Table 3. Summary of animal weight gain for WICST rotational grazing for 1992-2000.

* Animals at LAC were on pasture 86 and 122 days; only 53 and 56 of those days were on plots.
* * Days on pasture include 6 days in confinement for mid-season shrink wts & deworming.

Forage Quantity and Quality

At ARS, four random areas of the paddock was sampled to ground level using a hand held electric grass shears and a 0.5m² quadrant. At LAC, one sample was clipped to grazing height (approximately 10” for reed canarygrass) with a quadrant and sheers. The samples represented forage that would be immediately grazed by the heifers. Samples were dried and yields of dry matter per acre were calculated. Samples were sent to University of Wisconsin Soil and Plant Analysis Lab. Using NIRS, forage quality was estimated for crude protein (CP), neutral detergent fiber (NDF), and acid detergent fiber (ADF). Generally, fiber and protein levels were similar at each site for most of the season and at levels appropriate for this class of livestock.

1999. Throughout the growing seasons, weekly forage samples were collected at LAC and monthly samples were collected at ARS. Across sites and months, crude protein levels ranged from 13.0-21.0%; NDF levels ranged from 47.9-69.4%; and ADF levels ranged from 29.8-42.3% (Figs. 2-3). Relative feed values ranged from 108-161 at ARS and 76-123 at LAC (Fig. 4). Mid-season estimates of CP and ADF and NDF show high quality forage was available at ARS and marginal quality forage was available at LAC (Table 4). Differences in quality between the two sites was due to different plant management and diversity between sites. Several species made up the pasture at ARS while reed canarygrass was the main species present at Lakeland.

2000. Samples were collected weekly at ARS and monthly at LAC. Protein levels were lower than expected at both sites suggesting that red clover should be reseeded next year (Fig. 2) Typical of any cool-season grass pasture, our pastures started out with very high quality in late April/early May but declined until July when most grasses were finished heading out. Fiber levels increased to their highest levels in June but remained relatively constant over the remainder of the season (Fig. 3), when the pastures were mostly leafy, vegetative material. Crude protein levels ranged from 13.0 to 18.9 across sites, similar to 1999. Fiber levels were also in a similar range as 1999.

Table 4. Forage quality on WICST rotational grazing pastures for 2 periods each summer 1993-2000.

* Forage quality data is an average from the two closest sampling dates to the time period date.

Seasonal forage availability is shown in Figs. 5a and 5b. Forage availability at ARS was measured indirectly each week by the use of a plate meter in which height and density of the forage was measured and quantity was predicted using a regression equation (EQ 1). In addition, samples were clipped once a month to ground level for forage quality evaluation and dry weights were recorded which compared with the plate meter estimates. Each paddock was not sampled equally but each was measured throughout the season. As shown in Fig. 5a, approximately 1 ton DM/a was available to the animals at ARS in 1999. Between 1.5 and 2.0 ton DM/a was available in 2000 and possibly another harvest of hay could have been taken.

Forage availability at LAC was measured directly from the dry weight of the samples collected weekly in 1999. In Fig. 5b, over 2 ton DM/a was available for most of May and June at LAC. Early-season rapid forage growth can be controlled with mechanical defoliation or higher stocking rates. Lack of mechanical defoliation and understocking in the early months likely contributed to the reduced forage quality at LAC. No data was available on forage height/seasonal forage distribution for 2000 at LAC.

EQ1: Y (tons DM/a) = 0.004073 x forage ht in mm\[3]

Although forage availability, determined by weekly measurements, is a useful estimate of pasture production, it is not what the animal actually consumes as dry matter intake (DMI). Actual pasture production is difficult to determine for grazing studies. We used animal energy consumption formulas, tables of feed composition, and animal nutrient requirements to estimate dry matter consumed by the grazing animals (Table 5). Daily net energy for maintenance (NEM) and net energy for growth (NEG) requirements for the animals were determined by the standard formulas from the 1989 NRC Publication on Nutrient Requirements for Dairy Cattle (shown below) using average weight of each animal during the grazing period and their average rate of gain.

NEM (in Mcal/day) = .086 * (LW).75

and

NEG (in Mcal/day) = .035 * (LW).75 * (LWG/1000)1.119 + 1.0 * LWG/1000

where LW is live weight in kg and LWG is live weight gain in g/day.

Daily dry matter consumption was calculated using the following equation:

DMI (kg/d) = (daily NEM requirement/ NEM per kg ration) + (daily NEG requirement/ NEG per kg ration)

The pasture NEM and NEG (in Mcal/kg DM) used for ARS were an average of the following three values taken from the NRC feed energy tables:

1) NEM = 1.52 and NEG = 0.93 for avg. of early and medium bloom red clover
2) NEM = 1.69 and NEG = 1.08 for young orchardgrass
3) NEM = 1.48 and NEG = 0.89 for avg. of early and medium maturity bromegrass

Avg NEM = 1.56 and NEG = 0.97 for grass/clover mix pasture

Energy value of supplemented grain mix is

NEM = 1.96 and NEG = 1.30

Pasture NEM and NEG (in Mcal/kg DM) values used for Lakeland were taken from the NRC feed energy tables and are shown below:

NEM = 1.31 and NEG = 0.74 for reed canarygrass
NEM = 2.06 and NEG = 1.40 for grain supplement

The actual NEM and NEG of the complete ration was calculated after determining the percent of daily dry matter intake from the grain. Forage consumption was then determined per animal per day by subtracting off the grain from the daily dry matter intake. Then, we multiplied the number of days the animals were on pasture by the daily dry matter consumed per animal. Total forage production per acre was determined by multiplying forage consumption per animal by the total number of animals on each paddock and adding any mechanically harvested forage. Table 5 shows resulting forage production using this method at both sites, as well as quality parameters from 1990 through 2000.

Table 5. Average Forage Quality, and Estimated Dry Matter Intake (DMI) Using Animal Energy Requirements for WICST Rotational Grazing System, 1990-2000*.

* Grazing began in 1993 at ARS and in 1992 at LAC, harvested mechanically in previous years. ** Quality determined from hand cut samples during each season; does not include mechanically harvested forage when cattle were grazing. *** Does not include growth before July 23.

Conclusion

Rotational grazing offers a low input system with very good forage quality and respectable animal performance. Close attention to forage growth and weather conditions is important to maintain high forage quality. Excess forage growth should be harvested as hay in May and June. Grain supplementation may not be necessary all year but it is a relatively inexpensive energy source that can improve the dry matter intake when needed. By managing both forage and animal needs, we can realize the potential of management-intensive rotational grazing as a viable alternative to conventional heifer rearing.

Figure 1: Pasture layout at Arlington (similar at Lakeland)

Figure 2

Figure 3

Figure 4

Figure 5a

Figure 5b

Literature Cited

Cosgrove, D., D.Undersander, and M. Davis. 1996. Determining pasture condition. UW extension bulletin A3667.

Jackson-Smith, D., and J. Mark Powell. 2000. How Wisconsin dairy farmers feed their cows: Results of the 1999 Wisconsin dairy herd feeding study. Program on Agricultural Technologies Studies. College of Agriculture and Life sciences. University of Wisconsin- Madison. No 5. 16 pages.

[1] Research Specialist, Agronomy Dept. UW-Madison. E-mail: jlrieste@facstaff.wisc.edu

[2] Agronomist, Michael Fields Agricultural Institute, East Troy, WI (262) 642-3303. Email: jstute@michaelfieldsaginst.org

[3] Equation from Undersander and Casler, 1999.

Home | Who we are | Research | Publications | Links | Site map | Contact us

University of Wisconsin-Madison
College of Agricultural and Life Sciences
If you have any difficulty accessing this page, please contact our web master.

Site hosted by UW-Madison Center for Integrated Agricultural Systems