The effects of herbage allowance on pasture characteristics and milk production of dairy cows

ABSTRACT

This study aimed to evaluate the effect of herbage allowance on forage productivity, botanical composition and milk yield of Holstein cows. The treatments comprised three conditions of pasture allowance, high (38.4 kg DM cow⁻¹ day⁻¹), medium (30.3 kg DM cow⁻¹ day⁻¹) and low (26.8 kg DM cow⁻¹ day⁻¹) forage supply during autumn, and throughout the pasture allowance conditions of high (49.8 kg DM cow⁻¹ day⁻¹), medium (33.7 kg DM cow⁻¹ day⁻¹) and low (27.6 kg DM cow⁻¹ day⁻¹) herbage supply during spring. Significant differences (P < .05) were observed for forage accumulation, total biomass production and animal stocking rate according to the seasons. Herbage allowance affected (P < .05) the botanical composition and milk production, registering a decrease in Festuca in the summer, and a continuous reduction in legume proportion. Pasture management with a medium herbage allowance throughout the year provided a greater total annual milk production per hectare.

KEYWORDS:

• Botanical composition
• Festuca arundinacea
• growth rate
• pasture management
• stocking rate

Introduction

Pasture management is one of the tools used to control forage allowance and intake (Fonseca et al. 2013). In this way, the goals of pasture management, such as controlling the frequency and intensity of defoliation, result in potential tools to achieve a high level of milk production per cow and per hectare, impacting the global efficiency and profitability of livestock production.

Increased stocking density has been the most common path to boosting production in the dairy systems of Uruguay, with adjustments in pasture management made according to the intensity and frequency of defoliation (Mattiauda et al. 2003). Incremental increases in stocking rate, without acknowledging the ecophysiological limits of the forage plant, bring about a consequent decrease in pasture productivity (Zanine et al. 2016), which leads to a degradation process caused by a decrease in plant population or even an increase in the amount of uncovered soil (Volaire et al. 2014).

Farming production on dairy farms in Uruguay is based on annual crop rotation with perennial pasture. The most widely used rotation system combines a cycle of annual crops (winter + summer) with a pasture cycle of 3 years. In rotational schemes, the technical approach proposes that 25% of the area should be cultivated in summer and 50% in autumn. However, the commercial information registered systematically shows that from 50% to 70% of the area is cultivated in autumn and from 30% to 50% of the area is cultivated in spring (Chilibroste 2002).

The high proportion of the crop to pasture transitional area and vice versa in autumn determines the reduction of the ‘effective’ area for grazing, which leads to grazing pressure two to four times greater than that accepted as optimal for animal production in temperate climates, causing excessive exploration of forage resources (Chilibroste 2002). The intensification of forage systems generates some changes in the botanical composition and reduces species diversity (Schonbach et al. 2009; Dourado et al. 2015). These results, according to Bakker et al. (2002), negatively influence the productivity of pasture systems.

This background indicates the necessity of quantifying these phenomena, establishing response curves that integrate the effect of defoliation intensity on the productivity and durability of forage resources regarding animal performance. Technical-scientific information from the Uruguay community has, however, been unable to quantify the relationship between milk production and the frequency and intensity of defoliation. It is possible that the lack of information about the short- and long-term effects of different strategies of using forage resources and their effects on the production of forage and milk have been the result of faulty technical approaches.

Therefore, the objective of this study was to evaluate the effect of herbage allowance on forage production, botanical composition and milk production of lactating dairy cows throughout the year.

Materials and methods

Experimental site and pastures

The experiment was conducted at the Antonio Cassinoni Experimental Station (Estação Experimental Mario Antonio Cassinoni), Paysandú, Uruguay, which belongs to the Faculty of Agronomy at the Universidad de La República.

The soils of the experimental area are described in the CONEAT 11.3 group of soils (Duran 1987), classified according to American Soil Taxonomy (Soil Survey Staff 2003) as a typical Argiudoll associated with Argiaboll, and typical Argiudoll associated with typical Hapludoll.

The experiment started at the beginning of January 2010, and ended in January 2011, totaling 365 days. In April 2009, an existing 18-ha Festuca arundinacea pasture was intercropped in with the legumes white clover (Trifolium repens) and bird's foot trefoil (Lotus corniculatus); the seeding rates of the grass and the two legumes were 15, 2 and 8 kg ha⁻¹, respectively. Fertiliser with 18 kg N and 46 kg P₂O₅ ha⁻¹ was applied at sowing; in the second year, 13 kg N and 32 kg P₂O₅ ha⁻¹ were applied in autumn and spring. Additional applications of 50 kg of N ha⁻¹ as ammonium sulphate were applied after each grazing period in autumn, winter and spring, totaling 200 kg of N ha⁻¹ over the year.

Grazing system and animals

The total area was divided into four blocks of 4.5 hectares, and each block was divided into three paddocks of 1.5 hectares. A total of 36 purebred Holstein cows (24 multiparous and 12 primiparous) of American origin were used. Experimental cows had previously grazed on paddocks with a similar structure to that of the experimental pastures, and were free of sanitary and reproductive problems, and had an average body weight (BW) of 540 ± 72 kg, milk production of 27.02 ± 0.38 and a body score (BS) of 2.84 ± 0.33 (according to the methodology of Ferguson et al. 1994) at 14 ± 10 days post-calving. Cows were blocked by lactation number, calving date and BS, and randomly assigned within block to one of three allowance treatments: high (HA), medium (MA) and low (LA) allowance. Allowance was determined above ground level and varied depending on the season: during the spring, the herbage allowances were: 49.8, 33.7 and 27.6 kg DM cow⁻¹ day⁻¹ for HA, MA and LA, respectively; during the rest of the year, allowances were 38.4, 30.3 and 26.8 kg DM cow⁻¹ day⁻¹ for HA, MA and LA, respectively. Experimental paddocks were rotationally grazed over 42-day cycle whereby cows spent approximately 14 days in each paddock, moving to the next paddock according to forage availability and stocking rate as described by Astigarraga et al. (2002). Different forage allowance conditions were achieved by adjusting the stocking rate. The final stocking rate was determined by calculating the number of animals grazing and the duration of occupation for each paddock throughout the year, for each pasture management treatment.

During spring, animals grazed from 8:00 to 14:30 h and from 17:30 to 4:30 h, totaling 17.5 h of daily access to pasture. Cows were milked twice a day (5:00 and 15:00 h). During milking, cows were individually supplemented with 2 kg DM cow⁻¹ day⁻¹ of concentrate (consisting mainly of grain corn, wheat, expeller soybean and sunflower meal).

During autumn and winter, cows grazed from 7:00 to 14:00 h, totaling 7.0 h of daily access to pasture. Cows were milked twice a day (7:30 and 14:30 h). After afternoon milking, cows were individually supplemented with 3.0 kg DM cow⁻¹ day⁻¹ of F. arundinacea and T. repens silage (16.1 ± 0.7% CP, 52.6 ± 2%NDF, 33.2 ± 1.6%ADF on DM basis) and 5.5 kg DM cow⁻¹ day⁻¹ of concentrate. To measure milk production, daily evaluations were undertaken individually in two milking sessions, using automated meters (Waikato Milking Systems^® LP, Hamilton, New Zealand).

The forage mass pre- and post-grazing was estimated using the double sampling protocol described by Haydock and Shaw (1975). The forage mass and pasture compressed height (Rising Plate Meter^®: Ashgrove Co., Palmerston North, New Zealand) in each paddock were calculated from 300 measurements per treatment in a zigzag pattern. Plate meter readings were converted to kg DM ha⁻¹ using a calibration equation (average compressed pasture height × 251 + 248) across all sites. The values obtained from the equation were calculated based on the forage harvests. Growth rate and forage accumulation were calculated from the differences between weekly forage mass measurements that were carried out in all paddocks.

The botanical composition was determined in autumn, winter, spring and summer by the BOTANAL method, at 40 points per paddock, using visual observations in a quadrant of 0.3 × 0.3 m (Tothill et al. 1994). At each of these observation points, the areas occupied by grass, legumes, invasive species and dead material, and the proportion of uncovered soil were measured.

Climate measures

Climatic variables such as maximum, minimum, and mean temperature, relative air humidity and precipitation were recorded daily by the automatic weather station of the experimental station (Figure 1).

Figure 1. Weather conditions during the experiment (January 2010–December 2010).

Statistical analysis

Milk production, stocking rate, forage growth rate, forage accumulation in the pasture and botanical composition data were analysed using the paddock as a sampling unit. All statistical procedures were performed using the MIXED procedure of SAS version 9.2 (2010). When significant differences were found (P < .05), means were compared using Tukey's test.

All data were analysed with a GLM considering the following statistical model:$Y_{ij}=\mu +{\alpha }_i+{\pi }_j+{\epsilon }_{ij},$ μ is the overall mean; α_i is the effect of treatment i; π_j is the effect of animal block j; ε_ij is the random error associated with each observation.

Results

Pasture characteristics

There was a significant effect of herbage allowance (P < .05) on the growth rate in the autumn and spring: a greater autumn growth rate was observed for HA compared with LA (Table 1). Conversely, a greater growth rate was observed for LA than for HA during the spring. Herbage allowance did not affect (P > .05) the annual mean growth rates (Table 1).

Table 1. Growth rate by season (kg DM ha⁻¹day⁻¹) and annual means of high (HA), medium (MA) and low (LA) herbage allowance.

Season	HA	MA	LA
Autumn	60 ± 5.2 a	50 ± 5.2 ab	46 ± 5.2 b
Winter	43 ± 5.2	55 ± 5.2	42 ± 5.2
Spring	38 ± 5.2 b	57 ± 5.2 ab	56 ± 5.2 a
Summer	18 ± 5.2	19 ± 5.2	24 ± 5.2
Annual means	40.0 ± 4.4	45.4 ± 4.4	42.2 ± 4.4

Notes: Means followed by the same letter in rows do not differ by the Tukey test (P < .05). Values are expressed as mean ± SD.

In the autumn, there was greater (P < .05) forage accumulation in HA compared with LA (Table 2). The opposite behaviour was observed in the spring: LA showed greater forage accumulation than HA. Herbage allowance did not affect (P > .05) annual forage accumulation (Table 2).

Table 2. Forage accumulation (kg DM ha⁻¹) by season and annual means for high (HA), medium (MA) and low (LA) herbage allowance.

Season	HA	MA	LA
Autumn	5491 ± 475 a	4522 ± 475 ab	4211 ± 475 b
Winter	4022 ± 475	5117 ± 475	3898 ± 475
Spring	3462 ± 475 b	5168 ± 475 ab	5051 ± 475 a
Summer	1645 ± 475	1770 ± 475	2214 ± 475
Annual accumulation	14,620 ± 571	16,576 ± 571	15,373 ± 571

Notes: Means followed by the same letter in rows do not differ by the Tukey test (P < .05). Values are expressed as mean ± SD.

No significant differences (P > .05) in the botanical composition were observed between the pasture management regimes during the seasons (Figure 2).

Figure 2. Botanical composition of high (HA), medium (MA) and low (LA) herbage allowance during seasons.

The percentage of uncovered soil per season and total annual means were not significantly different (P < .05) between treatments (Table 3).

Table 3. Percentage of uncovered soil (%) and annual means for high (HA), medium (MA) and low (LA) herbage allowance.

Season	HA	MA	LA
Autumn	17.9 ± 8.0	19.8 ± 8.0	24.2 ± 8.0
Winter	17.6 ± 8.0	19.8 ± 8.0	13.1 ± 8.0
Spring	16.2 ± 8.0	12.7 ± 8.0	7.7 ± 8.0
Summer	32.3 ± 8.0	45.9 ± 8.0	55.1 ± 8.0
Means	21.0 ± 4.0	24.2 ± 4.0	25.6 ± 4.0

Notes: Means followed by the same letter in rows do not differ by the Tukey test (P < .05). Values are expressed as mean ± SD.

Herbage allowance significantly affected (P < .05) the botanical composition during the seasons, with a decrease in F. arundinacea in the summer, and a continuous reduction of legume proportion throughout the year (P < .05). In the summer, greater proportions of invasive species, dead material, and uncovered soil were recorded (Table 4).

Table 4. Botanical composition (%) and percentage of coverage of uncovered soil (%) according to high (HA), medium (MA) and low (LA) herbage allowance.

	Autumn	Winter	Spring	Summer
Fescue	58.0 ± 2.4	54.5 ± 2.4 a	56.0 ± 2.4 a	40.7 ± 2.4 b
Legume	19.8 ± 1.3 c	26.4 ± 1.3 b	36.4 ± 1.3 a	5.4 ± 1.3 d
Invasive species	8.9 ± 3.0 b	5.8 ± 3.0 b	2.1 ± 3.0 b	35.6 ± 3.0 a
Dead material	12.7 ± 1.8 b	12.0 ± 1.8 b	10.3 ± 1.8 b	31.7 ± 1.8 a
Uncovered soil	20.6 ± 2.9 b	16.5 ± 2.9 b	12.8 ± 2.9 b	44.5 ± 2.9 a

Notes: Means followed by the same letter in rows do not differ by the Tukey test (P < .05). Values are expressed as mean ± SD.

There was a significant effect of herbage allowance (P < .05) on stocking rate in the autumn and winter: a lower stocking rate was observed in the autumn for HA than for LA (Table 5). On the other hand, a greater stocking rate was obtained for MA and HA compared with LA in winter. Herbage allowance did not affect (P > .05) the annual mean stocking rates. Due to the severe dry period during the summer, it was not possible to start grazing and the stocking rate was zero.

Table 5. Stocking rate (cows ha⁻¹) per season and for whole year of high (HA), medium (MA) and low (LA) herbage allowance.

Season	HA	MA	LA
Autumn	1.8 ± 0.2 b	2.3 ± 0.2 ab	2.8 ± 0.2 a
Winter	1.7 ± 0.2 a	1.5 ± 0.2 a	0.7 ± 0.2 b
Spring	3.3 ± 0.2	3.7 ± 0.2	3.5 ± 0.2
Summer	0 ± 0.2	0 ± 0.2	0 ± 0.2
Means	1.7 ± 0.1	1.9 ± 0.1	1.7 ± 0.1

Notes: Means followed by the same letter in rows do not differ by the Tukey test (P < .05). Values are expressed as mean ± SD.

Animal performance

Herbage allowance affected (P < .05) milk production in autumn, winter and spring among the herbage allowance management regimes (Table 6). In the autumn and spring, there was lower milk production under the HA regime; during winter, LA presented lower milk production (P < .05) than MA and HA. Total milk production was significantly greater (P < .05) for MA than for the other treatments.

Table 6. Milk production (L ha⁻¹) by season, and total for the year of high (HA), medium (MA) and low (LA) herbage allowance.

Season	HA	MA	LA
Autumn	2872 ± 147 b	3730 ± 147 a	3998 ± 147 a
Winter	3541 ± 147 a	3541 ± 147 a	1180 ± 147 b
Spring	5129 ± 147 b	5985 ± 147 a	6210 ± 147 a
Summer	0 ± 147	0 ± 147	0 ± 147
Total	11,542 ± 290 B	13,256 ± 290 A	11,392 ± 290 B

Notes: Means followed by the same letter in rows do not differ by the Tukey test (P < .05). Values are expressed as mean ± SD.

Discussion

The greater growth rates and forage accumulation for the HA and MA management regimes compared with LA in autumn can be attributed to the greater photosynthetically active area remaining in HA and MA, and to the very low remaining leaf area and late regrowth in LA (Lemaire and Agnusdei 2000).

In the spring, growth rate and forage accumulation were lower for the HS management regime. This can cause a rapid loss of pasture structure with a high proportion of inflorescences, associated with a decrease in the proportion of green leaves and basal and vegetative tillers (Zanine et al. 2013; Mezzalira et al. 2014). Holmes et al. (2002) observed that a high forage growth rate, as found in this study for the HA management regime in spring, increases the accumulation of dead material. This causes greater shading and leaf death at the plant base, thus decreasing tillering and stimulating flowering during the spring.

Annual forage accumulation did not differ between treatments, as the decrease in forage production in autumn for LA was balanced by the lower production observed for the HA management regime in the spring. However, the MA management regime showed a high growth rate throughout the year. The greater pasture production during the winter observed for all treatments is a consequence of the high productive precocity of fescue (Festuca sp.) at the end of winter (Agnusdei et al. 2007). It is noteworthy that in the north of Uruguay, mean temperatures start to increase rapidly at the end of winter, accelerating floral induction at the end of August.

In summer, due to the severe dry period, both mean growth rate and forage accumulation were compromised. This impeded the entry of animals during this season, and thus caused a lack of milk production for all pasture management regimes.

Herbage allowance management did not affect the botanical composition or the proportion of uncovered soil in any season. These results suggest that the HA regime did not present undergrazing conditions, while the LA regime did not promote overgrazing that could modify the botanical composition and the proportion of uncovered soil. These data indicate the occurrence of degradation processes in the pasture (Schonbach et al. 2009; Zanine et al. 2016). However, some changes in the botanical composition and in the proportion of uncovered soil were observed between seasons. There was an increase in the proportion of legumes from autumn to spring, due to their proliferation in uncovered soil, which could be inferred by the decrease in the proportion of uncovered soil in the same period.

The reductions in the proportions of fescue and legumes associated with increases in the amount of dead material, invasive plants and uncovered soil from spring to summer are indicators that the striking climate conditions that occur at the end of spring and during the summer lead to pasture degradation (Schonbach et al. 2009), and not the herbage allowance managements that were adopted during the year.

It is noteworthy that the greater total milk production under MA pasture management was related to a greater stocking rate, and that the individual milk production of animals in this management strategy was significantly lower compared to that for HA pasture management; this is in agreement with the results of Delagarde and O’Donovan (2005).

The decrease in growth rate for LA in the autumn, and the decreased height post-grazing for the same pasture management regime, caused a longer waiting time for entry into the same paddock in winter (Brougham 1970). This resulted in reduced grazing in the winter, reflected in the lower stocking rate for LA compared with MA and HA, and decreased milk production in winter with LA pasture management. In spring, stocking rates did not differ, because it was possible to perform two and three more grazings for the HA and MA management regimes, respectively, after the spring experimental period. However, only one grazing was carried out after the spring experimental period in LA paddocks. This was a direct consequence of the reduced waiting time between grazings in MA paddocks, as they presented a greater post-grazing height than LA paddocks with the same growth rate.

The greater milk production from MA compared with HA and LA paddocks, even at the same stocking rate, indicates that the MA strategy may have resulted in a greater leaf/stem ratio, a greater proportion of young leaves, and better nutritional value in the sward. This is likely to have improved grazing efficiency and resulted in a higher dry matter intake and, consequently, greater milk production (Parga et al. 2002). This greater milk production from the MA management regime is in accordance with the results of another study employing similar management strategies (Chilibroste et al. 2012).

Conclusions

The results obtained in this study indicate that herbage allowances affect pasture characteristics and milk production. Pasture management with a medium herbage allowance throughout the year provided greater total annual milk production per hectare. However, the changes in forage accumulation and milk production indicate that high and medium forage allowance can be adopted during winter and autumn, respectively. In addition, during spring farmers should increase the stocking rate in order to achieve greater milk production and to optimise grazing systems in the mixed pasture.

Acknowledgements

The authors wish to thank the Coordination for the Improvement of Higher Education Personnel (CAPES) and the Experimental Station Mario Antonio Cassinoni (EEMAC) in Paysandú-Uruguay. The authors also wish to thank the FAPEMA (Maranhão State Research Foundation) for its financial support.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Anderson de Moura Zanine http://orcid.org/0000-0003-0100-3652

Ricardo Martins Araujo Pinho http://orcid.org/0000-0002-2816-1610

Additional information

Funding

This work was supported by Coordination for the Improvement of Higher Education Personnel (CAPES) [grant number 001].