| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
DairyNZ (formerly Dexcel), Private Bag 3221, Hamilton, New Zealand
2 Corresponding author: john.roche{at}dairynz.co.nz
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: milk production • pasture • stocking rate • comparative stocking rate
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The profitability of such dairy systems depends on the efficiency of pasture use coupled with reasonable per cow production (Dillon et al., 2005). Dry matter intake, the main determinant of milk production in such systems, is primarily controlled by pasture allowance (Maher et al., 2003). However, this relationship is curvilinear, with DMI increasing at a declining rate with increasing pasture allowance (Poppi et al., 1987). The curvilinear nature of this relationship points to a potential optimum pasture allowance for both biological and economic efficiency. Furthermore, pasture quality, another essential factor in milk production, tends to be positively associated with grazing severity in the previous defoliation (Lee et al., 2008), decreasing with higher grazing residuals (reduced grazing severity), and representing a potential reduction in future milk production with increasing pasture allowances. Therefore, although short-term studies may point to an optimum pasture allowance and hence stocking intensity for milk production in the short term, they provide little information on the optimum stocking density for the farm as a whole.
Stocking rate (SR) is defined as the number of animals allocated to an area of land (i.e., cows/ha), and it has been recognized as the primary driver of milk production per hectare and profitability in grazing systems for over half a century (McMeekan, 1956; Castle et al., 1972). Extension of short-term studies investigating the effect of pasture allowance on DMI and pasture utilization indicates an increase in pasture utilization and a decrease in per cow production as SR increases (Castle et al., 1968; Gordon, 1973; Baker and Leaver, 1986; Kennedy et al., 2006), and this is consistent with full lactation studies (Dillon et al., 1995). Increasing SR therefore negatively affects the amount of pasture available to each animal, but increases the amount of pasture harvested per hectare.
The potential to increase per hectare profitability by increasing SR when farms were under stocked has been reported (Fales et al., 1995). This is consistent with the positive association between milk production per hectare and SR in many studies (Gordon, 1973; Baker and Leaver, 1986; Dillon et al., 1995). However, many of these experiments were short term (Castle et al., 1968; Gordon, 1973; Baker and Leaver, 1986; Kennedy et al., 2006), compared insufficient SR treatments (<3) to determine an optimum SR (Foot and Line, 1960; Castle et al., 1968; Gordon, 1973; Dillon et al., 1995), or included the use of purchased supplementary feeds (Fales et al., 1995), in effect modifying the true effect of SR on the biological (pasture utilization, milk production/cow, milk production/ha) and economic variables measured. Furthermore, much of this research is more than 30 yr old and may no longer be applicable to the modern dairy cow breed and strains being used globally.
An additional problem with utilizing the information from previous studies is that the reported experiments involve cows differing in genetic merit for milk production and farm systems, in which pasture production per hectare and supplementary feed amounts can be greatly different. Both of these factors are likely to modify the effect of SR. Therefore, we propose that accounting for these confounding factors would result in a much more robust definition of SR than cows per hectare, as has been the definition in previous studies.
Of the 2 confounding variables mentioned, pasture grown per hectare each year can be reasonably accurately estimated by regular (e.g., weekly) visual assessments of each paddock on the farm. O’Donovan et al. (2002b) concluded that a linear regression of actual pasture yield (i.e., a cut sample) on a visual estimate of pasture yield explained 95% of the variation in the data, and the residual SD was <12% of the pasture yield. In addition, there was a strong correlation between devices that record sward height and measured pasture yield (r2 > 92%; O’Donovan et al., 2002b), and there is continuing technology development in this area.
However, an accurate measure of cow genetic merit for milk production is more difficult. The currently accepted standard would be the EBV for milk production traits. However, the EBV is not routinely available, is dependent on lactation length, which is not consistent internationally in genetic evaluation systems, and has poor reliability (
35%; Mrode, 1996). Cow BW may offer an alternative indicator of cow genetic merit for the purpose of standardizing SR. Although the genetic correlation between BW and milk production is close to zero, Berry et al. (2003) reported moderate correlations between BW and milk production following adjustment for the confounding effect of BCS.
A comparative SR (CSR) can therefore be defined as the kilograms of BW (at a standard BCS; e.g., at mo 6) per tonne of feed available. Such a definition would allow producers to better calculate the optimum SR for any grazing dairy farm, accounting for known or modeled variations in pasture production, the inclusion of supplementary feeds, and cow BW as a proxy for genetic merit for milk production. The objective of the current study was to evaluate 5 SR over 3 lactations; measure the effect on pasture production and quality, milk production, reproduction, and herd health; and provide the biological information required to calculate the optimum CSR for production and profitability.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Experimental Design and Treatments
Ninety-four New Zealand Holstein-Friesian cows were randomly allocated to 1 of 5 SR (2.2, 2.7, 3.1, 3.7, and 4.3 cows/ha) in a completely randomized design. Treatment farmlets were 8.1, 7.3, 6.1, 4.9, and 4.5 ha in size, with 18, 20, 19, 18, and 19 cows, respectively. The SR were chosen to give a wide range either side of what would be considered average for the farming location, which is relative to the amount of pasture it can produce. Cows were allocated ensuring treatment groups were balanced for cow age (4.4 ± 2.20 yr), BW (515 ± 64.2 kg), BCS (4.4 ± 0.61; 1 to 10 scale; Roche et al., 2004), milk yield from the previous lactation (for multiparous cows; 4,128 ± 1,036.1 kg), calving date (July 28 ± 14 d), and EBV for milk production (677 ± 262.9 kg), milk fat (23.0 ± 8.69 kg), protein (20.3 ± 6.93 kg), and Breeding Worth (a measure of genetic merit accounting for the economic value of the trait; 70 ± 29.4). Assuming an estimated annual pasture production of 18 t of DM/ha, as measured in a previous study on the same farm area (McGrath et al., 1998), the applied SR resulted in an approximate annual feed allowance of 8.1, 6.8, 5.8, 4.9, and 4.2 t of DM/cow per year.
Thirty-one hectares subdivided into 0.4-ha paddocks (defined grazing area) were randomly allocated to SR treatment. Five farmlets were established, balanced for geographic location, soil type, distance from the milking parlor, and previous experimental treatments, such that farmlets were evenly spread over the farm in a checkerboard fashion. Each farmlet was then randomly allocated to 1 of the 5 SR treatments. Once farmlets were established they were unchanged throughout the trial.
Trial Management
Grazing and Fertilizer Management.
Although treatments compared a range of SR, grazing management decision rules were the same across farmlets, with the exception of intergrazing interval, which was managed to optimize each individual treatment, and are outlined in detail by Macdonald and Penno (1998). Cows were rotationally grazed similar to the method described by Roche et al. (2007a). Briefly, cows had access to a predetermined allocation of 0.4-ha paddocks (20, 18, 15, 12, and 11 paddocks for 2.2, 2.7, 3.1, 3.7, and 4.3 cows/ha treatments, respectively) and these paddocks were grazed in a rotational order. As a result, cows had access to a fresh allocation of pasture once daily and only returned to the same area when a minimum of 2 leaves had appeared on the majority (>66%) of perennial ryegrass (L. perenne L.) tillers (approximately 2,500 kg of DM/ha in spring, 4,000 kg of DM/ha in summer, and 3,000 kg of DM/ha in autumn and winter; all measurements to ground level). Desired postgrazing residual in all treatments was 40 mm.
Grazing management was determined by weekly monitoring of farm pasture cover. The total farmlet area was available for grazing for the entire year. However, pasture surplus to requirements was conserved as silage when growth rates exceeded cow requirements (primarily October and November). All pasture was conserved in bales (220 kg of DM ± 49.6), with silage from each farmlet kept separately for use on that farmlet. A sample of the pasture was taken before baling for DM determination, and bales were weighed from each paddock to give an estimate of the amount of feed conserved from each paddock. Mechanical cutting (topping) of pasture residuals in midseason occurred as deemed necessary to maintain pasture quality.
The farmlets received annual "maintenance" dressings of 54 kg of P/ha and 55 kg of S/ha as single superphosphate in March and 50 kg of K/ha as muriate of potash in November. If conserved silage was made from an area, an additional 50 kg of K/ha was applied. In addition, 200 kg of N/ha was applied (in 4 to 5 applications) annually to each farmlet.
Animal Management.
Cows on all farmlets were managed in a similar manner. The system of milk production was seasonal, with approximately 50% of cows calving in 2 wk, 40% calving in the next 4 wk, and the remaining cows calving during wk 7 and 8. Cows with a due calving date later than wk 8 of the seasonal calving period were hormonally induced to calve during wk 7 and 8 using a 2-step combination of dexamethasone (Opticortenol S, Novartis Animal Health, Switzerland; Voren, Boehringer-Ingelheim, Berkshire, UK) and prostaglandin (Estrumate, Schering-Plough Coopers, New Zealand), provided they had low SCC before dry-off (<200,000 cells/mL), were in a BCS of 5.0 or greater (on a 10-point scale), and blood Mg (>0.8 mmol/L) and 
-glutamyl transferase (15 to 22 U/L plasma) measured the week before planned induction did not indicate health concerns.
The routine mating management policy at No. 2 Dairy was to record any cows exhibiting signs of estrus before the planned start of mating (PSM). Estrous detection was performed by twice-daily visual observation of estrous behavior with the aid of paint applied to the tail-head of the cow. Cows not detected in estrus by PSM were presented for veterinary examination. Those without a palpable corpus luteum were treated with an intravaginal controlled internal drug releasing insert (InterAg, Hamilton, New Zealand) according to the Genermate program (Cliff et al., 1995). Artificial insemination was performed for the first 6 wk from PSM, followed by a further 6 wk of natural breeding. Pregnancy diagnosis was performed by manual palpation of uterine contents at least 5 wk after the end of the 12-wk mating period.
At the end of each lactation, approximately 20% of the cows from each farmlet was culled on the basis of reproductive failure, health, age, and genetic merit, and were replaced with primiparous cows 1 mo before the planned start of calving. Age structure did not differ across farmlets for the duration of the study. Lactation was curtailed for individual cows within each season, based on BCS, level of production, and number of days from calving, as described by Macdonald and Penno (1998).
Animal Health.
Because of low pasture Mg and relatively high K and N concentrations in pasture during winter and spring, Mg supplementation was routinely practiced to prevent hypomagnesemia and associated hypocalcemia (Roche and Berry, 2006). Magnesium oxide was top dressed on pastures grazed by the nonlactating cows from 3 wk before planned start of calving until calving was completed. After calving, all lactating cows received 20 g Mg daily as an oral supplement of magnesium chloride (MgCl2·6H2O) at the a.m. milking until late November when Mg supplementation ceased. During periods of increased risk of bloat, an antibloating solution (Pluronic, Ecolab, Hamilton, New Zealand) was also provided at the morning milking. Zinc sulfate (3.6 g of ZnSO4·7H2O/100 kg of BW) was given orally to the cows during periods of increased vulnerability to facial eczema, as determined by fungal (Pithomyces chartarum) spore counts. Water troughs were located in each paddock so that cows had continual access to water.
Measurements
Pasture Measurement.
Pasture herbage mass was estimated by 2 people by calibrated visual assessment of each paddock from a weekly farm walk similar to the method described by O’Donovan et al. (2002a). On each occasion, 11 calibration quadrats (each 0.3 m2) covering the range of herbage mass present were assessed before and after the farm walk. After the final assessment, the pasture within the quadrats was cut to ground level, washed, and dried in a forced-draft oven at 100°C until dry (approximately 48 h). The herbage mass estimate for each paddock (for that week) was then adjusted using a regression of quadrat visual assessment on quadrat herbage mass. The net herbage accumulation was calculated weekly from the increase in herbage mass on ungrazed paddocks. Throughout the trial, one assessor, calibrated in the manner outlined previously, visually assessed pasture mass pre- and postgrazing as outlined by O’Donovan et al. (2002a). Those data were used to estimate total pasture consumed per hectare per year.
Representative samples of pasture were hand-clipped to grazing height (to represent what the cows would be consuming) from paddocks due to be grazed on each farmlet on one day each month. Representative samples of pasture silage were taken just before feeding. Duplicate samples of all feeds were dried at either 100°C for DM analysis, or at 60°C for analysis of nutrient composition. All samples dried at 60°C were dried for 72 h, ground to pass through a 1.0-mm sieve (Christy Lab Mill, Suffolk, UK) and analyzed for CP, NDF, ADF, NSC, lipid, and ash by near infrared spectroscopy (Corson et al., 1999). The ME content was derived directly from predicted OM digestibility on the basis of an in vitro cellulase digestibility assay, which had been calibrated against in vivo standards (Corson et al., 1999). Mineral concentrations were determined by inductively coupled plasma emission spectroscopy.
Milk, BCS, and BW.
Individual cow milk yields were recorded weekly (Tru-Test milk meter system, Palmerston North, New Zealand). Milk fat, CP, true protein, casein, and lactose concentrations determined on composite p.m. and a.m. aliquot samples by Fossomatic FT120 (Foss Electric, Hillerød, Denmark). Milk component data were verified by reference techniques for a subset of milk samples [milk fat: Röse-Gottlieb (IDF, 1987); CP, true protein, and casein: macro-Kjeldahl techniques (Barbano et al., 1991)].
Body weight and BCS were determined every other week following the a.m. milking or at approximately 0900 h during the nonlactating period. Body condition score was assessed pre- and postcalving on a 10-point scale, where 1 is emaciated and 10 is obese (Roche et al., 2004). Those scores can be converted to the 5-point scale using the regression equation generated by Roche et al. (2004; 5-point scale = 1.5 + 0.32 x 10-point scale). Calving BW and calf birth weight were recorded within 18 h of calving.
Lactation Profiles.
Preliminary graphical analysis of the mean milk yield lactation curve and individual cow lactation curves revealed a profile similar to that described by the Wilmink exponential function (Wilmink, 1987) and recently presented by Roche et al. (2006) for grazing dairy cows. Although a similar shaped profile was observed for FCM, fat, protein, and lactose yield, it was not as consistent across individual lactations. Therefore, the Wilmink exponential function was fitted to only milk test-day yields within lactation and is described as follows:
 |
In this model, yt represents yield (kg) at day t of lactation, and a, b, and c are estimated parameters relating to the height of the curve, the initial phase of postcalving incline to peak, and the subsequent post-peak decline phase, respectively. The regression parameters were estimated for each cow-lactation separately using PROC NLIN (SAS Institute, 2006). The first derivate of the Wilmink function with respect to time (dY/dt) for each cow-lactation was set equal to zero and solved for DIM to determine DIM at peak yield; DIM was rounded to the nearest whole integer. Peak yield was the yield corresponding to DIM at peak, and the final milk yield measure was regarded as the final DIM for that lactation. Total lactation milk yield was derived as the definite integral of the function for each lactation, and 305-d lactation yield was estimated by setting last DIM to 305 for all cows.
DMI.
Individual animal intake estimates were obtained at pasture using the n-alkane technique as modified by Dillon and Stakelum (1989). Briefly, each cow was dosed twice daily (at milking) for a 10-d period with a pellet containing 356 mg of n-dotriacontane (C32; i.e., 712 mg of C32/cow per d) on 4 occasions in each lactation. Fecal grab samples were collected twice daily from each cow (after milking) during the last 5 d of the 10-d period. The fecal samples from each cow for the 5-d period were then bulked and stored at –17°C until alkane analysis. During the same 5-d period, pasture samples were "plucked" to grazing height, following close observation of the grazing animal, to represent pasture grazed. The n-alkane contents (C25 to C36) of the pasture and feces were analyzed by gas chromatography.
The ratio of herbage C33 (tritriacontane) to dosed C32 (n-dotriacontane) was used to estimate intake. Estimates of daily herbage intake were calculated as follows:
 |
where Fi, Si, and Pi are the concentrations (mg/kg of DM) of the natural odd-chain n-alkane (C33) in feces, supplement, and pasture, respectively; Fj, Sj, and Pj are the concentrations (mg/kg of DM) of the dosed even-chain n-alkane (C32) in feces, supplements, and pasture, respectively; IS is the daily supplement intake, and Dj is the dose rate (mg/d) of the even-chain n-alkane (C32).
Statistical Analyses
For each year, cumulative milk and FCM yield and yield of milk components were calculated for each cow as the sum of yields from calving to the end of lactation. Average lactation fat, protein, and lactose percentages were also calculated as the average of all test-day records during lactation. Body weight and BCS were averaged across the intercalving period. One mean was calculated for each variable for each farmlet. Average DMI was calculated for each farmlet for the 4 periods of the year during which DMI was estimated. Reproduction variables were as defined by Roche et al. (2007a) and averages calculated for each treatment.
Cumulative pasture production was estimated for each paddock annually and the treatment average calculated. The average of all of the quality variables measured was calculated as the average of monthly measurements.
All data were averaged across years to provide one value per farmlet and analyzed by ANOVA using the statistical procedures of Genstat, with SR as the fixed effect. Linear and quadratic contrasts of SR were included in the model.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
CSR
An additional limitation of previous data is that the reported experiments involve cows differing in genetic merit for milk production and farm systems in which pasture production per hectare and supplementary feed amounts can be greatly different. Both of these factors are likely to modify the effect of SR. The current study evaluated 5 different SR over 3 years, providing data on the effect of SR on pasture production and quality variables, enabling the provision of a more robust and extendable definition of SR, termed CSR. We defined CSR as kilograms of BW (at a standard BCS; e.g., at mo 6) per tonne of feed available. Considering that all treatment herds in the current study were balanced for cow BW at the beginning of the study, the CSR equivalent to the SR imposed would be 62, 76, 90, 103, and 120 kg of BW/t of DM per year for 2.2, 2.7, 3.1, 3.7, and 4.3 cows/ha, respectively, assuming no feeds were acquired externally to the grazing platform and 18.0 t of DM pasture was grown/ha per year (McGrath et al., 1998).
Pasture Production and Quality
Table 1
highlights some of the changes in farm management that occur as SR changes and the effect of SR on the amount of pasture grown and consumed. The effect of SR on key pasture quality characteristics are presented in Table 2. Pasture grown tended (P = 0.11) to increase linearly with increasing SR, partly offsetting the decline in available pasture per cow. This resulted in a smaller increase in CSR than was originally expected. Actual CSR were 60, 70, 76, 89, and 91 kg of BW/t of DM for 2.2, 2.7, 3.1, 3.7, and 4.3 cows/ha, respectively, using the BW of the cows in mo 6. This is equivalent to an annual allowance of 8.2, 6.7, 6.3, 5.1, and 4.9 t of DM/cow, respectively. In addition to the tendency for an increase in pasture grown per hectare with increasing SR, postgrazing residual mass also declined linearly, reflecting an increase in pasture utilization rates with increasing SR that is consistent with the linear increase in pasture consumed per hectare. These data indicate less of a difference in annual pasture allocation than would be reflected in either SR or CSR. There was a 4 and 3 t of DM/cow difference in the amount of pasture available to the highest and lowest stocked treatments if compared on a SR and CSR basis, respectively. However, measured pasture disappearance (the accumulated difference between pre- and postgrazing pasture masses across the year) and the amount of pasture conserved for silage suggests that the actual difference in pasture harvested at the highest and lowest SR was 2.1 t of DM/cow, and the difference in actual pasture consumed was approximately 1.5 t of DM/cow. Therefore, the decline in utilizable pasture with increasing SR was not as dramatic as would have been predicted. There were insufficient detailed measurements of pasture production in the previous studies investigating SR to have been able to determine this effect on the availability of utilizable pasture.
|
|
depicts the average pasture accumulation and pasture DMI per hectare of the 5 farmlets over the 3 years. During spring (mo 3, 4, and 5), pasture production vastly exceeded requirements and so there was little need for longer intergrazing intervals (i.e., are not affected by SR). In comparison, during late summer, autumn, and winter there was a deficit of pasture grown relative to requirements. This deficit increased with increasing SR and was partially alleviated by longer rotations. Therefore, increasing the intergrazing interval and capturing the advantage of additional pasture grown is an essential element of managing increasing SR.
|
The effect of lower grazing residuals with increasing SR is also manifested in the average pasture quality characteristics presented in Table 2
. The increasing SR and the associated decline in postgrazing pasture residual mass resulted in a quadratic decline in NDF and ADF concentrations and a quadratic increase in OM digestibility, NSC, and ME. So, in general, it can be said that management changes associated with increased SR result in an improvement in pasture quality. This is despite the longer grazing interval and associated advancement in pasture maturity, and the greater amount of mechanical cutting that was undertaken in the lower stocked farmlets (Table 1). Fales et al. (1995) also reported increased pasture quality with higher SR, and Lee et al. (2007) reported a lower NDF and ADF and higher OM digestibility, NSC, and ME in swards that had been grazed more severely in the previous grazing; the result is an effect of a greater proportion of live leaf and less senescent material.
Milk Production, BW, and BCS
The effect of SR on DMI and milk production per cow and per hectare, and the estimated parameters of the lactation profile are presented in Tables 3 and 4, respectively; Figure 2 depicts the effect of SR on the shape of the lactation profile. The median R2 obtained from fitting the Wilmink function to the milk yield data was 0.84. All of the per-cow milk production variables, with the exception of fat and lactose percentage, declined linearly with increasing SR. The quadratic tendency for a decline in milk fat percentage can probably be attributed to the lower BCS at calving (Roche et al., 2007b) and the decline in protein percentage a result of lower energy intakes in early lactation (Spörndly, 1989). The linear decline in the per-cow yield of milk, FCM, and milk components was caused by both the shortening lactation length (Table 3) and the lower (P = 0.06) peak and poorer persistency (c parameter; P < 0.05) with increasing SR (Table 4 and Figure 2). The published effect of SR on milk yield is not consistent. A recent study (Kennedy et al., 2006) compared a high and medium SR and reported a negative effect of SR on milk and milk component yield and milk protein percentage, consistent with the findings reported here. Similarly, earlier studies (Castle et al., 1968; Gordon, 1973) reported a negative effect of SR on milk yield per cow, but a positive effect on milk yield per hectare. However, the 2 most recent comprehensive studies evaluating the effect of SR showed only a very small (100 kg of milk; Dillon et al. 1995) or no effect (Fales et al., 1995) of SR on milk yield. However, in the study of Fales et al. (1995), concentrates were fed in proportion to milk production (0.25 kg of grain/kg of milk, with a minimum of 4 kg of DM/d), effectively removing the negative effect of the SR-imposed feed restriction in early lactation. Similarly, the comparison published by Dillon et al. (1995) delayed the calving date by 7 wk and then compared high vs. low SR; the later calving date in effect removed the early-lactation feed deficit associated with the higher SR (Figure 1). In addition, neither system was purely grazing, with both requiring confinement during early and late lactation and including a strategy of supplementing cows with purchased supplements when sufficient pasture was not available. Similarly, the short-term SR comparison of Baker and Leaver (1986) included a supplementation strategy that was altered weekly to maintain desired grazing residuals.
|
|
|
The shortening of the lactation in autumn (late lactation) with increasing SR is necessary in a "closed system" in which supplementary feed is unavailable (for reasons such as availability, price, organic certification, and so on), to ensure that there is sufficient pasture grown and stored before the beginning of the new milk production season. The milk yields of cows stocked at 2.2 and 4.3 cows/ha predicted from the Wilmink function were 5,109 and 3,401 kg, respectively, and cows on these SR treatments had lactation lengths of 291 and 221 d, respectively. If the lactation length of all herds was standardized to 305 d (Table 4), 24% of the reduced milk production resulting from increasing SR could be attributed to the shorter lactation lengths, whereas 76% was a result of the lower peak and less persistent lactation.
Although there was a linear decline in calving BCS with increasing SR (Table 3), a factor known to negatively affect milk production, the difference between the highest and lowest SR (0.5 BCS units; 10-point system) would only have been expected to reduce milk yield by 73 kg milk (Roche et al., 2007b) or 4%. Therefore, the reason for the lower yield per cow as SR increased was most likely a result of the lower intakes of fresh pasture during winter (early lactation). Intakes estimated using a dual indigestible marker system (n-alkanes; Dillon and Stakelum, 1989) showed a quadratic decline in pasture consumed during winter. This is consistent with the tendency for milk and FCM yield (P < 0.1) to decline quadratically with increasing SR, and is supported by the increasing deficit in pasture availability in early lactation with increasing SR (Figure 1). Cows on the high stocked treatments received pasture silage in lieu of available fresh pasture, but even high quality pasture silage does not support milk production as effectively as fresh pasture (Dillon et al., 2002). This is an important observation because it infers that if feed of equivalent quality to the pasture being consumed were purchased for the more highly stocked herds in early lactation to elevate their lactation profile, and in particular their peak milk yield, to that of the less severely stocked herds, it may be possible to capture more than 70% of the difference in milk production per cow between SR with only a modest investment in supplementary feeds in early lactation. This is consistent with the lack of effect of SR on milk yield per cow in the study of Fales et al. (1995) and the recent findings of Roche (2007), who reported responses to supplementing restricted cows with additional pasture during the first 5 wk of lactation that were 2 times greater than the immediate response measured and approximately 2.5 times greater than the average published response (Bargo et al., 2003). Further research is required to determine the benefit of strategic use of supplementary feeds to dairy cows on high SR systems in early lactation.
Although there was a negative association between SR and milk yield per cow, the effect of SR on milk yield per hectare was positive (Table 3). This is consistent with earlier research findings (Castle et al., 1968, 1972; Gordon, 1973; Fales et al., 1995) in which milk production per hectare increased with increasing SR irrespective of the effect of SR on milk yield per cow. In fact, Castle et al. (1972) reported that SR explained 92% of the variation in milk production per hectare. Irrespective of SR, there was no significant (P = 0.25) effect on the amount of milk produced per kilogram of feed consumed (0.9 kg of milk/kg of pasture + silage consumed). However, the decline in protein percentage and the tendency for a decline in milk fat with increasing SR resulted in a decline in milk fat and protein yield, FCM, and NEL per kilogram of feed consumed (P < 0.05; adjusted R2 = 0.79, 0.74, and 0.73, respectively) with increasing SR, implying a reduced efficiency of converting available pasture into saleable product. This would be expected considering the linear increase in energy required per hectare for maintenance, activity, and pregnancy as SR increases. However, the effect is small, with a decline of 3 to 5% in feed DM conversion efficiency with each additional cow per hectare and is offset by the 5.5% increase in pasture DM availability and the increase in pasture ME with increasing SR.
Reproduction
Seasonal pasture-based systems require a compact calving season (8 to 10 wk) and an intercalving interval of 365 d to ensure that maximum animal demand coincides with peak pasture growth (Dillon et al., 1995). Consequently, the seasonal nature of milk production in such systems places additional requirements on reproductive performance in that cows need to initiate ovarian cyclicity early after calving and preferably conceive within 6 wk from a fixed calendar date (PSM; Roche et al., 2007a) to achieve the 365-d intercalving interval. The effect of SR on the measured reproductive variables and the pattern of calving is presented in Table 5. There was no effect of SR or CSR on the proportion of cows that calved within the first 4, 6, or 8 wk following the onset of calving. Between 75 and 80% of cows calved in the first 4 wk and between 95 and 100% calved within the desired 8-wk window. The number of cows induced to achieve this calving spread was not significantly affected by SR, but declined in a quadratic fashion with increasing CSR.
|
). This is consistent with the tendency for a linear decline in the number of cows submitted to AI in the first 3 wk of the breeding season as SR (P < 0.15) and CSR (P = 0.07) increased. Several authors have linked low calving BCS, as was evident with increasing SR and CSR, with longer periods of postpartum anestrus (Markusfeld et al., 1997; Titterton and Weaver, 1999; Roche et al., 2007a), but calving BCS has been reported to not affect pregnancy (Buckley et al., 2003; Roche et al., 2007a). This lack of effect of SR on pregnancy rate is probably because all cows had sufficient fresh pasture to meet demand for the month before PSM. In addition, the quality of pasture on offer increased with increasing SR. |
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Comparisons with other studies support the need for a more robust measure of SR than cows per hectare, because farms will differ in the genetic merit of their cows, in their potential to produce pasture, and in their ability to import suitable feed supplements. We propose the concept of a CSR, whereby SR is defined by the BW of the cows, the potential of the land to produce pasture, and the amount of supplement sourced externally to the allocated grazing area (kg of BW/t of DM). Although SR is sufficient to compare between very similar farming systems, the adoption of CSR would facilitate the extrapolation and transfer of research findings between different systems.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Received for publication August 22, 2007. Accepted for publication January 8, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. M. Lee, D. J. Donaghy, and J. R. Roche Short Communication: Effect of Postgrazing Residual Pasture Height on Milk Production J Dairy Sci, November 1, 2008; 91(11): 4307 - 4311. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), 2.7 (
), 3.1 (
), 3.7 (x), and 4.3 (•) cows/ha over a 3-yr period. Months 1 to 12 refer to the southern hemisphere pasture growth season of July to June. PSM = planned start of mating.