Comparative performance in Holstein-Friesian, Jersey and crossbred cows milked once daily under a pasture-based system in New Zealand

ABSTRACT

Production and efficiency of Holstein-Friesian (F), Jersey (J) and crossbred F × J milked once-a-day (OAD) were studied at Massey University dairy farm No. 1. Herd test records of milk yields, somatic cell score (SCS), live weight (LW) and body condition score (BCS) were used to model the lactation curves using a third-order orthogonal polynomial. Total lactation yields of milk (MY), fat (FY), protein (PY) and milk solids (MS = FY + PY) were calculated based on predicted daily yields. Predicted efficiencies were feed conversion efficiency (FCE; total lactation MS per kg of predicted total lactation dry matter intake [DMI]); biological efficiency (BE; total lactation MS per 100 kg of LW) and DMI capacity (DMIC; total lactation DMI per 100 kg of LW). Holstein-Friesian cows yielded 376 kg milk more than F × J cows and 1101 kg milk more than J cows per lactation. Holstein-Friesian and F × J produced similar total lactation MSY (366 and 369 kg, respectively). Jersey cows were more efficient than cows of the other breeds: FCE was 78.2, 82.3 and 86.3; BE was 69.7, 74.1 and 77.8; and DMIC was 887.9, 896.8 and 897.7 for F, F × J and J cows, respectively. There were significant differences in production and efficiency among the breeds, with J cows being more efficient per 100 kg of LW than F or F × J cows. Further research is required to compare breed profit per ha in order to conclude the best breed for OAD systems.

KEYWORDS:

• Breed efficiency
• Holstein-Friesian
• Jersey
• milk yield traits
• once-a-day milking
• stocking rate

Introduction

Dairy cows under seasonal pasture-based dairy farming in New Zealand traditionally have been milked twice-a-day (TAD). However, farmers have started to adopt once-a-day (OAD) milking for herd management and lifestyle benefits (Davis 2005). The economic viability of OAD systems, however, remains uncertain due to decreases in milk production (Clark et al. 2006; Lembeye et al. 2015).

Davis (2005) and Prendiville et al. (2009) indicated that traits related to efficiencies are critical to economic profitability in dairy systems. Potentially, the impact of efficiency on the profitability during the lactation period might be greater in OAD compared with TAD milking systems, due to the decrease in milk production. Production efficiency has previously been studied in dairy cattle (Mackle et al. 1996; González-Verdugo et al. 2005; López-Villalobos et al. 2008; Prendiville et al. 2009, 2010, 2011a, 2011b). Those studies, however, compared production and efficiency performances of Holstein-Friesian (F), Jersey (J) and their crosses (F × J) under pasture-based systems in cows milked TAD.

In July 2013, Massey University dairy farm No. 1 changed from a TAD split calving herd to a spring calving OAD herd with the feeding system based mainly on pasture of ryegrass and white clover. Stock numbers were reduced from 320 to 250 cows, resulting in a shift of stocking rate (SR) from 2.7 to 2.1 cows/ha. This was in line with farm policy to reduce nitrate leaching and achieve environmental guidelines. Switching from TAD to OAD milking was a strategy to reduce labour costs and remain profitable, while achieving good animal reproduction and health outcomes. This change represented an opportunity to evaluate the breed performance and production efficiencies of F, J and F × J cows milked OAD in a pasture-based system at Massey University’s dairy farm No. 1.

Different measurements of feed efficiency have been proposed in dairy cattle including feed conversion efficiency (Mackle et al. 1996), net efficiency (Schwager-Suter et al. 2001) and residual feed intake (López-Villalobos et al. 2008). In a comparative study among F, J and F × J grazing pasture, Prendiville et al. (2009) used alternative measurements of gross efficiencies defined as kg of milk solids (MS; fat + protein) per 100 kg of live weight (LW); MS per total dry matter intake (DMI) and total DMI per 100 kg of LW. These measurements can be referred to as biological efficiency, feed conversion efficiency and DMI capacity, respectively.

The objective of this study was to compare total lactation yields of milk production traits, somatic cell score (SCS), LW, body condition score (BCS), predicted total DMI, and efficiencies of three different breed groups milked OAD for the whole lactation.

Materials and methods

The data analysis consisted of herd test records from 297 cows (77 F, 71 J and 149 F × J) that calved during spring of the dairy seasons 2013–2014 and 2014–2015 at dairy farm No. 1 of Massey University (40°22′S, 175°36′E). Those cows were daughters of 124 sires (39 F, 65 J and 20 F × J). The average breeding worth (BW) of F, J and F × J cows were 130 (range 17–221), 146 (range 30–229) and 123 (range 29–202), respectively. Reliabilities ranged from 25%–56%. Average production worth (PW) was 164 (range 1–409), 155 (range -33–314) and 180 (range -10–412) for F, J and F × J respectively, with reliabilities ranging from 35%–93%. The farm has a total effective area of 119 ha, consisting of 10 ha of lucerne crop, 10 ha of mixed herb chicory and plantain pasture, and the remainder ryegrass–white clover pasture, with the whole area stocked at an SR of 2.1 cow/ha.

As supplementary feeding, during calving cows had access to a pasture comprised of plantain, chicory and red clover, lucerne and maize silage (an average of 3.5 kg DM supplement/cow/day). During October and November, a mixed-herb pasture containing chicory and plantain (an average of 3.7 kg DM/cow/day) was rotationally grazed for 2 weeks followed by a 2 week rest period. A brassica crop (6.3 kg DM/cow/day on average during summer) and lucerne (4.6 kg DM/cow/day) were grazed during late lactation.

Mean calving dates for seasons 2013–2014 and 2014–2015 were 11 August and 8 August, respectively. Breed groups were determined using pedigree information. Pure-bred F and J cows were defined as those animals where the breed composition was ≥93.75% (15/16) from a particular breed (F or J). Proportion of F and J genes in the F × J cows were on average 8/16 and 7/16, respectively, while 1/16 corresponded to unknown breed composition. In total, data from 427 lactating cows were analysed, 104 of them corresponded to first-lactation cows (22 F [19.8%], 33 J [33.6%] and 49 F × J [11.5%]).

Herd test records (2562) of daily milk yield (MY), percentage of fat and protein, days in milk (DIM) and somatic cell count (SCC) were used. Daily fat (FY) and protein (PY) yield were calculated from daily MY by multiplying by their respective percentages. Herd test SCC was transformed to SCS calculated as SCS = log₂ (SCC). Lactation length was calculated as the difference between date at dry-off and date at calving. For cows that were culled before the end of the season, lactation length was calculated as the maximum DIM provided by the herd test records. Monthly (season 2013–2014) and bimonthly (season 2014–2015) average LW records (4945), along with individual BCS, on a 1–10 scale, six times during the lactation period (in both seasons) were recorded (1916).

Model

Lactation curves were modelled using a third-order orthogonal polynomial (Kirkpatrick et al. 1990) for each cow–lactation combination. Considering as the level of production of a trait ‘i’ measured on day (t) of the lactation from calving. The polynomial was defined as:

where are the regression coefficients to be estimated for each trait.

The Legendre polynomial’s functions of were calculated as (Spiegel 1971):where (Schaeffer 2004).

Total lactation yields for milk production traits were estimated for each cow–lactation combination using the polynomial equation, as the sum from day 1 to maximum DIM. Average SCS, LW and BCS were calculated as the mean of these traits using the polynomial equation. Lactation persistency was calculated as total lactation MY divided by peak of daily MY multiplied by lactation length (modified from López-Villalobos et al. 2005).

Metabolisable energy requirements and dry matter intake

It was assumed that pasture contained 18.4 MJ gross energy (GE) and 10.5 MJ ME kg⁻¹ DM; where , efficiencies of ME utilisation, and energy requirements for maintenance, milk production and body weight change, were calculated as follows (AFRC 1993):

ME for maintenance () was calculated from the following formula:where is the energy requirement before feeding and is the energy required for activity (cows walking 3 km under grazing conditions). and were calculated as follows:where corresponds to LW on day t after calving.

ME for lactation () was calculated with the following formula:where is daily milk yield, is energy value of 1 kg of milk, , where and are fat and protein contents .

ME required for live weight change () was calculated as:where is change in LW and is the energy value of a unit of change LW of the cow; this value was assumed constant (19.3 MJ NE).

The total requirements for ME () were adjusted for feeding level as:where FL was calculated as:Daily DMI per cow was estimated by dividing by the content of ME per kg pasture dry matter. It was assumed that cows could consume the pasture needed to achieve their specific energy demands (Brookes 2002). Total lactation of DMI (kg/cow) was calculated as the sum of daily DMI from day 1 to maximum DIM per cow per lactation.

Efficiency measurements

Estimates of efficiencies used in this study were described by Prendiville et al. (2009). Three measurements of efficiency were investigated: biological efficiency (BE), calculated as total lactation MS per 100 kg of LW; dry matter intake capacity (DMIC), calculated as total DMI per 100 kg of LW; and feed conversion efficiency (FCE), as total lactation MS per total DMI expressed as g of MS per kg DMI.

Statistical analysis

Estimates of regression coefficients of the third-order Legendre polynomial for each cow in each lactation, total lactation yields of milk, fat, protein and milk solids, lactation persistency, mean of SCS, BCS and LW, BE, FCE and DMIC were analysed using the MIXED procedure of SAS version 9.3 (SAS Institute Inc). The mixed linear model included fixed effects of season, breed group, the interaction between season and breed group, the linear and quadratic effect of lactation number, and the linear effect of days from median calving date, and the random effect of the cow–lactation combination.

Results

Table 1 presents descriptive statistics for each trait considered in this study. Across the breeds, the predictive ability of the lactation curve methodology was high for the milk traits (r ≥ 0.93), meanwhile for LW and BCS linear correlations between actual and predicted values ranged between 0.82 and 0.91. Those correlations are presented in Table 2.

Table 1. Mean, standard deviation (SD), minimum and maximum values for the total lactation record of milk, fat and protein yield, and average somatic cell score, live weight and body condition score of all dairy cows at Massey University dairy farm No. 1 during the dairy seasons 2013–2014 and 2014–2015.

Trait	Mean	SD	Minimum	Maximum
Total milk yield (kg)	3991	1117	684	7429
Total fat yield (kg)	204.7	48.3	34.4	324.0
Total protein yield (kg)	159.1	41.1	25.8	266.9
Average somatic cell score	6.32	1.44	1.58	13.58
Lactation length (days)	261	37	52	333
Average live weight (kg)	495	67	272	700
Average body condition score^a	4.67	0.58	2.50	7.00

^aBody condition score on a 1 to 10 scale.

Table 2. Estimated Pearson linear correlation and standard error for actual and predicted daily milk, fat and protein yield, somatic cell score, live weight and body condition score modelled with a third-order orthogonal polynomial fitted to Holstein-Friesian (F), Jersey (J) and F × J crossbred cows under once-a-day milking at Massey University dairy farm No. 1.

	Holstein-Friesian	Crossbred	Jersey
Milk yield (kg)	0.98 ± 0.04	0.98 ± 0.03	0.97 ± 0.04
Fat yield (kg)	0.96 ± 0.04	0.95 ± 0.03	0.95 ± 0.04
Protein yield (kg)	0.98 ± 0.04	0.98 ± 0.03	0.97 ± 0.04
Somatic cell score	0.93 ± 0.04	0.93 ± 0.03	0.95 ± 0.04
Live weight (kg)	0.91 ± 0.03	0.88 ± 0.02	0.72 ± 0.03
Body condition score	0.87 ± 0.04	0.89 ± 0.03	0.87 ± 0.05

Table 3 details the estimates of regression coefficients describing the lactation curves of milk yield traits, SCS, LW and BCS of the different breed groups milked OAD. At the beginning of the lactation (intercept), F cows had greater values for MY and PY than J and F × J cows. However, FY, the intercept was greater in the crossbred cows. For SCS, the intercepts did not differ among the breeds studied.

Table 3. Least squares means and standard errors of the estimates of regression coefficients of the lactation curves for milk, fat and protein yields, somatic cell score, live weight and body condition score modelled with a third-order orthogonal polynomial fitted to Holstein-Friesian (F), Jersey (J) and F × J crossbred cows under once-a-day milking at Massey University dairy farm No. 1.

Trait	Breed	α0	α1	α2	α3
MY	F	15.31^a ± 0.201	−9.16^c ± 0.193	−0.96 ± 0.108	1.50^a ± 0.116
	J	11.61^c ± 0.207	−6.66^a ± 0.198	−0.87 ± 0.111	1.00^b ± 0.119
	F × J	14.11^b ± 0.138	−8.22^b ± 0.132	−1.18 ± 0.074	1.27^a,b ± 0.079
FY	F	0.70^b ± 0.010	−0.34^a,b ± 0.009	−0.04 ± 0.004	0.04 ± 0.006
	J	0.70^b ± 0.011	−0.32^a ± 0.009	−0.05 ± 0.004	0.05 ± 0.006
	F × J	0.74^a ± 0.007	−0.35^b ± 0.006	−0.05 ± 0.003	0.05 ± 0.004
PY	F	0.58^a ± 0.008	−0.29^b ± 0.008	−0.02 ± 0.005	0.03 ± 0.004
	J	0.51^b ± 0.008	−0.23^a ± 0.008	−0.03 ± 0.005	0.03 ± 0.004
	F × J	0.57^a ± 0.005	−0.28^b ± 0.006	−0.04 ± 0.004	0.02 ± 0.003
SCS	F	6.44 ± 0.091	1.11 ± 0.082	0.17^b ± 0.046	−0.05 ± 0.063
	J	6.44 ± 0.093	0.98 ± 0.085	0.28^a,b ± 0.047	−0.09 ± 0.064
	F × J	6.37 ± 0.062	0.98 ± 0.057	0.29^a ± 0.031	−0.09 ± 0.043
LW	F	524.16^a ± 3.259	37.40^a ± 1.436	−16.68^c ± 0.947	19.80^a ± 1.024
	J	438.46^c ± 3.473	17.36^c ± 1.530	−4.20^a ± 1.009	5.17^c ± 1.091
	F × J	497.59^b ± 2.254	29.76^b ± 0.993	−13.67^b ± 0.655	15.81^b ± 0.708
BCS	F	4.61^b ± 0.034	0.03^a ± 0.012	−0.25 ± 0.014	−0.41^a ± 0.006
	J	4.55^b ± 0.035	0.03^a ± 0.013	−0.23 ± 0.015	−0.41^a ± 0.006
	F × J	4.71^a ± 0.023	−0.02^b ± 0.008	−0.25 ± 0.010	−0.44^b ± 0.004

MY, milk yield; FY, fat yield; PY, protein yield; SCS, somatic cell score; LW, live weight; BCS, body condition score.

^a,b,cMeans with different superscripts in the same column are significantly different (P < 0.05).

Breed group had a significant effect on the traits studied except for SCS, peak day, and lactation persistency. Holstein-Friesian cows yielded 376 kg more milk than F × J cows and 1101 kg more milk than J cows per lactation. There was no significant difference in total lactation MS between F and F × J cows (Table 4).

Table 4. Predicted means and standard errors of lactation length, total lactation yields of milk, fat and protein, average somatic cell score, live weight and body condition score, peak and persistency of milk yield of Holstein-Friesian, Jersey and crossbred cows under once-a-day milking at Massey University dairy farm No. 1.

	Holstein-Friesian	Crossbred	Jersey	P value
Lactation length (days)	256 ± 4	253 ± 3	253 ± 4	0.734
Total milk yield (kg)	4437^a ± 76	4061^b ± 53	3336^c ± 79	<0.0001
Total fat yield (kg)	199.7^a,b ± 3.9	208.2^a ± 2.7	197.5^b ± 4.1	0.046
Total protein yield (kg)	165.8^a ± 3.1	160.3^a ± 2.1	142.4^b ± 3.2	<0.0001
Average somatic cell score	6.22 ± 0.10	6.15 ± 0.07	6.18 ± 0.11	0.834
Average live weight (kg)	518.2^a ± 3.2	492.5^b ± 2.2	435.0^c ± 3.4	<0.0001
Average body condition score^1	4.63^b ± 0.04	4.74^a ± 0.02	4.56^b ± 0.04	0.001
Peak DIM (day)	27 ± 3	30 ± 2	31 ± 3	0.524
Peak milk yield (kg)	22.9^a ± 0.3	20.9^b ± 0.2	17.2^c ± 0.3	<0.0001
Milk persistency^2 (%)	77.3 ± 0.7	78.9 ± 0.5	78.6 ± 0.8	0.186

DIM, days in milk.

¹Body condition score on a 1 to 10 scale.

²Milk persistency calculated as total milk yield divided by peak daily milk multiplied by lactation length, expressed as percentage.

^a,b,cMeans with different superscripts in the same row are significantly different (P < 0.05).

Lactation persistency (%) for MY was similar among the breed groups (P > 0.05) (Table 4). Peak daily MY occurred at the end of the first month after calving (Figure 1) and was greater in F, followed by F × J and J cows (P < 0.05). The typical milk lactation curve, with a rapid increase in MY up to about 30 days post-calving and then a gradual decline to the end of the lactation (Grossman et al. 1999) was found in all three breeds. A less well-defined peak, however, was also observed in lactation curves for MS of F and F × J cows (Figure 2).

Figure 1. Actual herd test record and predicted lactation curves of milk yield (kg/day) of A, Holstein-Friesian; B, F × J crossbred; and C, Jersey cows milked once daily at Massey University dairy farm No. 1.

Figure 2. Actual herd test record and predicted lactation curves of milk solids yield (kg/day) of A, Holstein-Friesian; B, F × J crossbred; and C, Jersey cows milked once daily at Massey University dairy farm No. 1.

The three breed groups studied had similar average SCS. Holstein-Friesian cows were 25.7 kg and 83.2 kg heavier than F × J and J cows, respectively. Crossbred cows had a slightly greater BCS than F and J breeds (P < 0.05), with no difference between the pure breeds (P > 0.05).

On average, predicted total lactation DMI of F cows were 18% and 8.7% greater than J and F × J cows, respectively (Table 5). Jersey cows had greater FCE (g MS/kg DMI) and BE (kg/100 kg LW) than the other breed groups (10% and 12% greater for FCE and BE, respectively, compared with F and 5% for both FCE and BE compared with F × J). In contrast, DMIC (kg/100 kg LW) was similar among the breeds studied (Table 5).

Table 5. Least squares means and standard errors of predicted total dry matter intake, feed conversion efficiency, biological efficiency, dry matter intake capacity of Holstein-Friesian, Jersey and crossbred cows under once a day milking at Massey University dairy farm No. 1.

	Holstein-Friesian	Crossbred	Jersey	P value
DMI (kg/cow)	4599^a ± 68.0	4229^b ± 47.1	3894^c ± 72.5	<0.0001
FCE (g MS/kg DMI)	78.2^c ± 0.56	82.3^b ± 0.39	86.3^a ± 0.60	<0.0001
BE (kg MS/100 kg LW)	69.7^c ± 1.38	74.1^b ± 0.95	77.8^a ± 1.47	<0.0001
DMIC (kg DMI/100 kg LW)	887.9 ± 13.9	896.8 ± 9.6	897.7 ± 14.8	0.850

DMI, total dry matter intake; FCE, feed conversion efficiency; BE, biological efficiency; DMIC, dry matter intake capacity; MS, milk solids (fat + protein).

^a,b,cMeans with different superscripts in the same row are significantly different (P < 0.05).

Discussion

Total lactation MY per cow estimated in the current study was greater than those reported under research conditions by Clark et al. (2006) with F cows (2914 kg) and J cows (2211 kg) and more recently in commercial herds milked OAD in New Zealand by Lembeye et al. (2015) in F (3198 kg), J (2637 kg) and in F × J cows (3014 kg).

Total lactation MS per cow were also greater than those reported for OAD cows in previous studies (Clark et al. 2006; Lembeye et al. 2015). Clark et al. (2006) and Lembeye et al. (2015) reported values ranging between 222–278 kg/cow, compared with the range of 340–369 kg/cow found in the current study. Production performances found in the current study are similar to the TAD production of cows reported by Clark et al. (2006) and Lembeye et al. (2015).

The three breed groups had similar average SCS, which is in agreement with the study of Prendiville et al. (2010) in Ireland. In New Zealand, Berry et al. (2007) reported a greater SCS for J cows milked TAD, compared with the SCS found for J cows milked OAD in this study (Table 3). The SCS results of this study were similar to previously reported values for cows milked TAD (Clark et al. 2006; Lembeye et al. 2015) and lower than those observed in previous OAD studies (Clark et al. 2006; Lembeye et al. 2015). During the transition from TAD to OAD, an increase in SCS is expected (Lacy-Hulbert et al. 2005; Clark et al. 2006). This study indicates that switching from TAD to OAD systems does not necessarily imply an increase in SCS. The latter is attributed to better management practices and possibly to lower exposure to pathogens during the milking process as cows are milked just OAD (Dalley et al. 2007).

Farmers milking their herds OAD generally increase their SR by an average of 17% (Cooper & Clark 2001); however, the 2.2 SR at dairy farm No. 1 of Massey University is lower than the New Zealand average (2.86 cows/ha; LIC & Dairy NZ 2015). A reduced SR generally results in an increase in total MS per cow, but a decrease in total lactation MS per ha (Penno 1999; Macdonald et al. 2001, 2008, 2011). In the current study, the reduced SR may have led to higher availability of dry matter per cow, and more time available for grazing plus supplementary feeding may have allowed the cows to express more of their genetic potential. This may partly explain the higher total lactation milk yield traits compared with those from typical OAD systems in New Zealand. For instance, the average BW and PW of the three breed groups were in the upper quartile of herds in New Zealand (124 and 142, for BW and PW, respectively; LIC & Dairy NZ 2015). Total lactation MS of the herd (kg/cow), on average over the two seasons was slightly lower than the average yield in New Zealand under TAD systems for the same period (371–377 kg; LIC & Dairy NZ 2015). The results of the current study, however, do not necessarily imply that the optimal economic margin in herds milked OAD can be achieved with the SR used in this study. There is a quadratic relationship between SR and net income, being optimal at 3.0–3.5 SR in a typical TAD system in New Zealand (Macdonald et al. 2001). To date, no study has been conducted to estimate the optimum SR in OAD systems. Although the common practice is to increase SR, there is evidence that a 17% higher SR in OAD does not fully compensate for the decrease in milk production in the herds milked OAD compared with herds milked TAD (Clark et al. 2006).

On this farm, total MS per cow from spring calving cows in the last season milked TAD were 427, 343 and 378 kg for F, J and F × J cows, respectively. Those values are 17%, 1% and 2.5% greater than the values presented in Table 3 for the same breed groups milked OAD. This suggests that the reduced SR applied when the farm switched to OAD has been effective in compensating for the reduction in milk yields by providing more energy per cow.

The low SR mentioned previously might also explain the greater average LW in the current study compared with the national average in New Zealand (458, 430 and 376 kg for F, F × J and J cows, respectively; LIC & Dairy NZ 2015), These values are also greater than those reported by Prendiville et al. (2011b) for F × J and J breed groups in Ireland.

Prendiville et al. (2009) reported 15% and 4.3% greater DMI in F compared with J and F × J cows, respectively, which is in agreement with the differences presented in Table 4 between F and J, although in that study, DMI was calculated as daily DMI, instead of total lactation DMI as in the current study.

Feed conversion efficiency (g MS/total kg DMI) was consistently higher in J compared with F and F × J. The values presented in Table 4 are similar to the findings by Prendiville et al. (2009) (79–88 g MS/kg DMI), but lower than those of Mackle et al. (1996) (114–132 g MS/kg DMI). Holstein-Friesian and F × J cows had greater milk trait production and predicted total lactation DMI than J cows. However, the utilisation of ME and total lactation DMI per unit of LW was higher in J cows. For this reason, cows with higher total lactation DMI (such as F cows) tend to be less suitable for grazing pastures (Veerkamp et al. 1995). Several experiments have demonstrated that J cows are more efficient at converting dry matter into yields of fat and protein (Grainger & Goddard 2004).

Biological efficiency (kg MS/100 kg LW) was greater for J compared with F and F × J cows (Table 4), reflecting greater utilisation of the ME and lower heat increment from J cows (L’Huillier et al. 1988). More efficient cows generally have lower BCS (Prendiville et al. 2011a), which is consistent with a reduced BCS found in the J cows compared with F, but BCS was greater in F × J cows compared with the pure breeds, suggesting that some degree of heterosis is possible (Prendiville et al. 2011a, 2011b; Table 3).

Dry matter intake capacity (kg DMI/100 kg LW) was similar among the three breeds groups (Table 4). This observation was unexpected given that J cows have a greater ability to consume more feed per unit of LW than F cows (2.4%–10.2% more) (Mackle et al. 1996; Prendiville et al. 2009, 2011a), due to a larger reticulo-rumen volume (Smith & Baldwin 1974) and greater digestive efficiency (Ferris et al. 2014).

In general, J cows were more efficient than F and F × J cows due to similar predicted DMIC and lower LW, while only producing slightly less MS per cow compared with the other groups. This implies that J cows required less energy than F cows to produce 1 kg of MS, which is consistent with previous studies (Prendiville et al. 2009, 2011a). However, differences in BE and FCE between J and F × J cows were only 5%, indicating that F × J cows can be effectively suited to OAD systems.

The results of this study indicate that J cows perform best under OAD systems, they have lower LW and DMI, greater FCE and BE, and slightly lower total lactation MS.

Conclusion

The results of this study indicate that there were differences in total MY traits, LW, BCS, total DMI, BE and FCE among F, J and F × J cows milked OAD. In general, J cows were more efficient per kg of LW than the other breed groups, which agrees with the higher ranking of this breed for OAD systems. However, F × J cows can be suitable for OAD systems, with greater MS and an intermediate FCE and BE. Comparing breed profit and efficiency per ha would provide additional data that would indicate the best breed for pasture-based OAD systems.

Acknowledgements

The principal author acknowledges information provided by Livestock Improvement Corporation (Hamilton, New Zealand).

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCiD

N López-Villalobos http://orcid.org/0000-0001-6611-907X

Additional information

Funding

The principal author acknowledges support from Programa Formación de Capital Humano Avanzado, Becas Chile, CONICYT doctoral scholarship (72120059).