Tropical Grasslands (2001) Volume 35, 246–253

246

Effect of pasture production systems on milk production in the central plains of Thailand

S. TUDSRI1, S. PRASANPANICH2, S. SAWADIPANICH1, P. JARIPAKORN4 AND S. ISWILANONS 3 1 Department of Agronomy, and 2 Department of Animal Science, and 3 Department of Agricultural Economics, Kasetsart University, Bangkok, Thailand 4 The Dairy Farming Promotion Organisation of Thailand, Muaklek, Saraburi Abstract Three groups of dairy cattle were compared under different pasture production systems viz, stripgrazed on pure ruzi grass ( Brachiaria ruziziensis), strip-grazed on a ruzi-leucaena ( Leucaena leucocephala ) pasture or strip-grazed on lablab (Lablab purpureus) in the morning and on pure ruzi in the afternoon. In 1995, all cows received concentrate (14% CP; 70% TDN) at 1 kg per 3 kg of milk per day. Cows receiving leucaena or lablab in combination with grass produced higher daily milk yields (13.6 and 14.4 kg/cow, respectively) and fat percentage (4.2 and 4.5%) than those on pure grass alone (11.9 kg/cow at 4.0% fat). However, when all cows received concentrate at 4 kg/cow in the 1996 trial, pasture production systems had no effect on milk production. Milk composition (fat, solids-not-fat, total solids, protein and lactose) was unaffected by pasture system in 1996. Growing tree legumes with grass or sowing of pure herbaceous legume swards adjacent to grass swards will support satisfactory milk yields in dairy cattle in Thailand. The economics of the various systems are discussed. Introduction In the central plains of Thailand, small-scale dairy production plays a significant role in both Correspondence: S. Tudsri, Department of Agronomy, Kasetsart University, Bangkok, Thailand. E-mail: agrsat@ nontri.ku.ac.th

food production and income generation. The high genetic potential of European crossbred cows is, however, rarely achieved, mainly because of inadequate nutrition. Pasture production in Thailand comes primarily from 3 systems. The primary pasture production system is the growing of pure stands of cultivated grasses, primarily ruzi ( Brachiaria ruziziensis ), para (Brachiaria mutica ) and guinea ( Panicum maximum) grasses (Tudsri et al. 1991). These pasture species (except para grass) are usually broadcast on poorly prepared seedbeds at seeding rates of 6–18 kg/ha. They are cut or grazed 31–60 days after sowing and subsequently at 45 to 60-day intervals. Most dairy farmers apply little if any fertiliser (Tudsri et al. 1991). The quality of these pastures is usually low and yields are frequently less than 6250 kg DM/ha/yr with poor persistence (1-3 years). To maintain adequate milk yields, significant amounts of concentrates must be fed. Grass-legume pastures are the second most important source of fodder, the common mixtures being ruzi, para or guinea grass grown with Verano stylo (Stylosathes hamata ) and/or centro (Centrosema pubescens ) (Tudsri et al. 1999). However, these are not used widely as it is difficult to maintain pasture legumes in mixtures with grasses, particularly in more fertile areas. Even when sowing alternate rows of grass and legume, in an endeavour to reduce competition, the legume content in the mixtures still declines rapidly as reported after the second cycle approximately 3 months from sowing in the experiment of Wongsuwan and Watkin (1990). However, the introduction of the shrub legume leucaena ( Leucaena leucocephala ) has altered this situation somewhat and leucaena is becoming popular with farmers. As Tudsri et al. (1999) reported, it shows great potential for growth of high quality forage and is more persistent in mixtures with grasses. Hence, it has the potential to replace herbaceous legumes in our pasture production systems and produces higher quality mixed pasture.

Pastures and milk production in Thailand

The third system, suggested by Wongsuwan and Watkin (1990), is to sow the legume as a pure stand in a separate area and feed it daily, by cutting or grazing, along with the grass, thereby providing an important and valuable source of high protein feed to the animals. This system also allows farmers to achieve higher grass yields, by cutting at longer intervals and gaining greater benefit from added fertiliser nitrogen on the pure grass sward, without the problem of interspecific competition which plagues mixed grass-legume pastures. Although not a “perennial” source of fodder, annual forage crops such as hybrid sorghum and corn are an important source of fodder for livestock. However, in this study, the forages and forage systems being compared were restricted to the cheapest source of animal forage-perennial pasture species. This paper reports the results of a study to investigate milk production and milk quality from a representative pasture from each of the 3 systems, namely: ruzi grass alone; ruzi grassleucaena; and ruzi grass supplemented with lablab (Lablab purpureus). The experiment was conducted during the growing season (June– September) when grasses were growing rapidly and hence pasture quality could also drop rapidly. Materials and methods The experiment was conducted at the Dairy Farming Promotion Organisation of Thailand, located at Muaklek, Saraburi, 150 km north-east of Bangkok (14° 50′N, 101° 10′E; altitude 220 m). The soil is a clay loam of moderate fertility with pH 6.5. The experiment was carried out for 14 weeks in 1995 (June 7–September 14) and 15 weeks in 1996 (June 15–September 27). A preexperimental grazing period of 7-14 days was allowed for the animals to adjust on both occasions. Climatic conditions at the experimental site are monsoonal with the rainy season extending from May–October with peak precipitation in September. Mean annual rainfall is 1196 mm (10-year average 1987–1996). Rainfall during the studies was average in 1995 and above average in 1996 (Table 1). In both years, the rainfall was evenly distributed in the wet season except for a dry period in July 1995 when rainfall was less than half the medium-term mean. Thus, irrigation was applied to ensure good regrowth. The average monthly maximum and minimum temperatures for the 2 years were 28.9°C and 24.2°C, respectively, over the months June–September.

247

Table 1. Rainfall at Muaklek, Saraburi during the study and the medium-term mean. Average1

Month

1995

1996

(mm) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 1 10-year

9 13 62 61 140 159 107 201 236 178 26 4 1196

2 0 20 85 186 149 38 275 347 113 2 0 1217

1 12 33 116 182 147 139 130 437 106 77 0 1380

average, 1987–1996.

The experimental area of 5.28 ha was subdivided into 6 paddocks each of approximately 0.88 ha, with 2 paddocks for each treatment. The pasture treatments consisted of pure ruzi grass (1.76 ha), ruzi-leucaena mixed pasture (1.76 ha) and ruzi plus lablab (each 0.88 ha). All paddocks were ploughed and cultivated to produce a firm fine seedbed before sowing on July 10, 1994. A basal fertiliser (15N:15P:15K) was applied before sowing at the rate of 200 kg/ha. For the pure ruzi grass pastures, seed was drilled in rows 50 cm apart at a seeding rate of 24 kg/ha. For the mixed pasture (ruzi-leucaena), leucaena was planted in rows 100 cm apart (50 cm apart within rows) on July 15, 1994 giving approximately 20 000 plants per hectare and ruzi grass was grown between the rows of leucaena by drilling the seed at the rate of 24 kg/ha. In the treatment incorporating the pure legume stand, lablab was sown on 0.88 ha on March 15, 1995 in rows (50 × 25 cm) at approximately 80 000 plants/ha. All paddocks, except the lablab, were cut to 15 cm for ruzi grass and 25 cm for leucaena on April 28, 1995 and fertiliser (15N:15P:15K) was applied at 200 kg/ha. 1995 trial In the 1995 trial, 24 Holstein–Friesian crossbred cows (62.5–75% Holstein–Friesian) with average weight 356 kg in their first to third lactation and their first and second months of lactation were selected and divided into 3 treatment groups of 8 cows per treatment. They were balanced for these factors and also previous milk yield across the following 3 treatments:

248

S. Tudsri, S. Prasanpanich, S. Sawadipanich, P. Jaripakorn and S. Iswilanons

• Strip-grazed on pure ruzi grass alone; • Strip-grazed on ruzi-leucaena mixed pasture, as for Group 1; and • Strip-grazed on ruzi grass from 1300 h–0500 h and on lablab from 0500 h–1300 h. Rotational strip grazing (2 days grazing) was adopted, with an average inter-grazing interval of 25–30 days giving 3 grazing cycles in the 98 days of trial. During the first cycle of grazing in 1995, all pasture treatments had 35 to 50-day regrowth periods but in the second and third cycles, the intergrazing interval was reduced to 25–30 days. Ruzi and lablab were grazed down to 20 and 25 cm above ground, respectively. All cows were also fed concentrate according to their individual milk production daily, at the rate of 1 kg concentrate per 3 kg of milk per day. The concentrate was fed at milking time (twice a day). The concentrate used was comprised of: kapok seed meal (24%), corn meal (18%), rice bran (17%), mungbean meal (15%), para rubber meal (15%) plus smaller proportions of soybean meal, cassava meal and minerals, giving a crude protein concentration of 14.0% and an estimated TDN of 70%. Pasture production and chemical composition were measured before and after grazing by cutting nine 1 m 2 quadrats in each treatment. The material from each harvest was dried at 60°C for 3 days and dry matter yield estimated. The dried samples were ground in preparation for nitrogen analysis using an autoanalyser. Grass and lablab were cut at 15 and 20 cm above ground level, respectively. Leucaena, mostly edible, was cut at 25 cm above ground level. Pasture intake was calculated as the difference between pasture dry matter on offer at the beginning and end of each grazing period. Milk production was recorded daily and milk fat was determined once a week.

No analyses of other milk constituents were made due to lack of facilities. 1996 trial In the 1996 trial, lablab was re-established on March 5, by drilling the seed (50 × 25 cm) at a rate similar to that used in the 1995 trial. In the ruzi-leucaena and pure ruzi treatments, ruzi was cut to 15 cm above ground and the leucaena to 25 cm on May 20, 1996. All pasture treatments were fertilised with 250 kg/ha of compound fertiliser (15N:15P:15K). Eighteen European cows (average weight 372 kg) were selected and the experiment was carried out as for the 1995 trial but all cows were fed concentrate (14% CP; 70% TDN) at the rate of 4 kg/cow/d throughout the experimental period. The average intergrazing interval was 30–35 days, with 3 grazing cycles in the 105-day study. Pasture production and quality were determined as were milk production and composition. Cows were milked twice daily and milk production was recorded at each milking. Subsamples of milk were collected once a week for analysis for milk protein, lactose, solids-notfat and milk fat by using the Milko scan tester. For statistical analysis, data for weekly milk yield and milk composition were analysed as for a completely random design with animals as the experimental unit and differences tested for significance at P0.05). 1996 trial. In the 1996 trial, however, when concentrate was fed at a constant level, and not according to milk production, average milk yields on the various treatments were not significantly different (Table 2). However, there was an indication that the normal decline in milk yield occurred more slowly in the legume-added treatments than in the pure ruzi treatment but differences between the slopes of the lines were not significant (P > 0.05). There were no significant differences between treatments in percentages of milk fat, protein, lactose, solids-not-fat and total solids (Table 3). Liveweight changes showed only minor differences between treatments (Table 2). Pasture production and quality 1995 trial. The amount of pasture on offer was greater in the first cycle but crude protein concentration was lower than in the later grazing cycles. The leaf percentage in the pre-grazed pas-

ture was lower in the first cycle in all treatments (41–50%) than in the later grazing cycles (53– 72%) (Table 4), and was consistently higher than in post-grazed pasture. Ruzi grass alone had more dry matter on offer than the grass-legume treatments (P