Developing pasture grass system on seasonally waterlogged and dry soil and effect of cottonseed cake and Trichanthera gigantea on milk production in Khmer ethnic farms 

Nguyen thi Hong Nhan, Nguyen Van Hon, Nguyen Thiet and Somchai Chanpongsang*

Cantho University, Cantho, Vietnam
nthnhan@ctu.edu.vn
*Chulalongkorn University, Thailand

Abstract

Experiment 1 was set up in a field of 500 m² to measure the productivity and nutritional quality of Hymenachne acutigluna and Paspalum atratum grasses planted on waterlogged soils and to examine the productivity, quality and persistence. The two plants are similar in biomass and nutritive value.

Experiment 2 was conducted on a former rice paddy field in an area of 800 m². Paspalum grass was cultivated in wet land with approximate 20 cm in water depth. Four different plant spacing (20 x 50 cm; 30 x 50 cm; 40 x 50 cm and 50 x 50 cm) showed that plant spacing were not significantly different during four crops in fresh and dry yield and qualitative parameters of the grass

Eperiment 3 was carried out in dairy farms in Soctrang province, where the farmers have become familiar with the practice of raising dairy cattle. Fifteen F1 (Holstein x Sindhi) lactating cows were allocated in a randomized block design and the trial period was 90 days. Cows were allocated to treatments on the basis of milk yield, parity and days of lactation. There were three households involved in the study to provide 5 replicates per treatment. In each family, 5 cows are housed in individual stall and offered one of the three different treatments: Control (Paspalum grass ad lib and 0.4 concentrate/kg milk), PC (Paspalum grass ad lib 4 kg/day cottonseed cake) and PCT (Paspalum grass ad lib 1 kg/day cottonseed cake and 1 kg/day Trichantera in DM basis). Milk production, milk quality and feed conversion to milk were not affected by the different forms of supplementation.

Key words: Hymenachne acutigluna, lactating cows, Paspalum atratum, plant spacing, , 

Introduction

Many programs and policies have been made recently by local governments in the Mekong Delta to help farmers in alleviating their poverty, one of which is the practice of raising dairy cattle. In Soctrang province, for example, more than 250 Khmer farmers are members of a dairy cooperative known as EVERGROWTH. This model has been successful and the numbers of dairy cows appropriately reach 2,000 heads (personal communication). When the dairy herd expands, farmer is faced by inadequacy of feed resources on farm to meet the animals' requirement. Among the contributing factors is the land shortage for pasture production and seasonal variation in terms of quantity and quality of forages, with scarcity being intensified in the wet and dry season. These factors lead to introduction  of grass as a pre-requisite fodder to establish  zero grazing systems.

There has been no detailed pasture research carried out on seasonally waterlogged and dry soil where dairy farming is now expanding. Currently, only 2 grass species are suited to these soils. These are Hymenachne acutigluna and Paspalum atratum. They have been proven to be valuable forage grasses for low fertility and acid soils which become waterlogged during the wet season. They also produce good production in dry season on low lying sites if fertilized well.

However, sole feeding of grass requires protein supplementation for better performance of the animals. Protein supplementation through concentrate feeding in the smallholder dairying has been limited due to high costs of industrial by products and low prices of milk. Alternative recommendation has been integration of forage through intercropping.  The supplementation of rich-protein materials like cottonseed cake is therefore of importance to satisfy the nutrient requirements of the host animals. However, the use of cottonseed cake is also limited because (i) a high level of gossypol contained in cottonseed cake may negatively influence milk production (Rogers et al 1995) and (ii) in many cases the use of cottonseed cake does not ensure net profits to farmers due to its high cost. Hence, the search for alternative feeds for supplementation is still needed.

The multi-purpose tree Trichantera gigantea (Preston and Murgueitio 1994), introduced into Vietnam from Colombia in 1991, has adapted readily to a wide range of ecosystems throughout Vietnam (Ha and Phan 1995; Nhan et al 1996). The crude protein content of the foliage (leaves and the thin stems, which are also consumed by the animals) varies from 18 to 20 % in DM. This kind of tree has been used to partly replace the protein source for laying hens and ducks and positive results were obtained (Nhan et al 1997). In ruminants, the use of Trichantera was also reported in lactating goats (Duyen et al 1996), but so far data on the supplementation of Trichantera to dairy cattle diets has been lacking.

The objectives of the research:

(1) To examine the productivity, quality and persistence of Hymenachne acutigluna and Paspalum atraum on seasonally waterlogged and dry soil

(2) To determine an optimal plant spacing (plant density) on Paspalum atratum production in waterlogged condition.

(3)To determine the nutritional value of Paspalum atraum as a basal diet for dairy cattle when supplemented with foliage of Trichanthera or cotton seed cake.

Materials and methods

Trial 1: Evaluation of productivity, quality and persistence of Hymenachne acutigluna and Paspalum atraum on seasonally waterlogged and dry soil.

The study was conducted in dairy farms in Soctrang province where the system was expanding on former rice paddy land. The soil was acid (pH 4 – 4.5) which was waterlogged for long periods during the wet season and then dry out in the dry season.  The land was first cleared from weeds, and then ploughed by tractor to a depth of 20 - 25 cm to loosen the soil. Weeds were also removed twice during the establishment period.

The experiment was designed in a completely randomized design consisting of two treatments (Hymenachne acutigluna and Paspalum atratum grass.) and five replications (equivalent to 10 plots).

The experiment was set up in a field of 700 m², of which 500 m² were used for planting and 200 m² as border areas. Hymenachne acutigluna stems and Paspalum atratum tillers were planted at spacing of 50x50cm and on the same day.

Each plot measured 5 x 10 m. The first harvest was made at 90 days after planting and 45 days for the next cutting. At each harvest, four 0.25 m² quadrates were cut 5 cm from ground level in each plot. The fresh samples were weighed and a 200 g sub sample from each plot dried at 70^o C for 48 hours. The dried samples were bulked across replicates and 3 samples per treatment were used for analysis.

After each sampling cut, the remaining forage in the plots was cut 5 cm above ground level before applying fertilizer.

All biomass from each plot was weighed to determine the fresh yield.  The fresh biomass was sampled and pooled from the 3 replicates (1.5 kg fresh weight each), and was placed in a porous paper bag for dry matter determination and chemical analyses.  A similar sample was collected to analyze on a DM basis.  The contents of crude protein (CP) in the samples was determined according to the procedure of AOAC (1990).

Trial 2: Effect of plant spacing on Paspalum atratum production in waterlogged condition

The experiment was similar in location and data collection of trial 1, but carried out in the wet land with approximate 20 cm in water depth. A randomized complete block design was used in this experiment. There were four treatments in each of 4 blocks which were randomly assigned. Therefore, a total of 16 plots was used in a size about 800 m². There were four treatments corresponding with four different plant spacing (20 x 50 cm; 30 x 50 cm; 40 x 50 cm; 50 x 50 cm).

Trial 3: Effect of cottonseed cake and Trichantera gigantea on dairy cattle performance in the Khmer ethnic farms.

The experiment was carried out in dairy farms in Soctrang province, where the Khmer farmers have become familiar with the practice of raising dairy cattle. Fifteen F1 (Holstein x Sindhi) lactating cows were allocated in a randomized complete block design. Cows were allocated to treatments on the basis of milk yield, parity and days of lactation.

There were three households involved in the study each to provide 5 replicates per treatment. In each family, five cows were housed in individual stalls and were offered one of three different treatments as described below. Collection parameters were nutrient intake, milk production, milk composition, live weight change. The study lasted 60 days. During the first 15 days all the cows were fed the control diet. The milk yield at this time was used as covariant in the statistical analysis. During the next 7 days the cows on treatments BCT and PCT were adapted to the new diet. Milk yield was measured on all diets for a further 6 weeks.

Treatments:

Control: Paspalum grass ab lib and 0.4 kg concentrate/kg milk.

PCT: Paspalum grass ab lib and 4 kg/day cottonseed cake.

PCT: Paspalum grass ab lib plus 1 kg/day cottonseed cake and 1 kg Trichantera in  DM basis.

Before the experiments started, cows were drenched against internal parasites. The animals were housed individually in a  shed separated from their calves and received free water and home-made mineral lick blocks at all time. They were milked twice daily at 07:00 and 15:30 by hand milking followed by suckling of  residual milk by the calves. All ingredients fed to cows were divided into three portions per day, grass and Trichantera were mixed together to prevent selection; molasses was mixed with urea and cottonseed cake was given as its normal form.

Feed intake of grass and Trichantera were estimated daily by the difference between DM of amounts offered and refused. The samples were dried at 65^oC and bulked together for the whole period. The determination of dry matter (DM) was by drying at 100^oC for 24 h.

Milk production was recorded daily as the sum of sucked and milked raw milk. Milk intake of the calves was determined by the weight-suckle-weight technique every week (Williams et al 1979). Milk samples were collected twice weekly at consecutive 0500 a.m. and 1500 p.m. milking and analyzed for total protein, butterfat  lactose and total solids by Milkotester (model-Master) at Evergrowth cooperative.

Economic analysis was made using partial budget analysis based on increased costs and increased returns of the treated animals.

Photo 2. Paspalum atraum grass on on former rice paddy land	Photo 3. Trichanthera gigantea attracted the attention of many farmers
Photo 4. The Khmer farmers have become familiar with the practise of raising dairy cattle

Statistical analysis:

The effects of treatments on milk yield and feed intake were subjected to ANOVA using the GLM procedure of Minitab 13.2. Covariance analysis was used with initial milk yield on standard diet as the covariant. When the F-test was significant, the Tukey test for paired comparisons was used to compare means. All the data were coded for subsequent statistical analysis using Microsoft Office Excel program. 

Results and discussion

Experiment 1: Evaluation of productivity, quality and persistence of Hymenachne acutigluna and Paspalum atraum on seasonally waterlogged and dry soil.

There were not significantly differences between treatments in both types of grass at different time of cutting (Table 1).

The biomass averages of Paspalum atratum and Hymenachne acutigluna were 23.8 and 24.6 tonnes/ha/cutting in fresh matter, 5.25 and 4.86 tonnes/ha/cutting in DM. Both of grasses were well adapted with waterlogged soils and stayed constant in production with time. In the previous studies we reported that the production of Paspalum atratum and Hymenachne acutigluna was 16.6 tonnes/ha/cutting in fresh biomass which was lower compared to the current research on (Nhan et al 2009). This is due to the grass being influenced by flood for longer period and the first harvest was shorter  (60 days compared to 90 days). Conversely, this study was similar to the results of Hare et al. (1999). They reported that Passpalum atratum grew well on wet and waterlogged acid soils in Thailand and produced about 20 tonnes DM/ha during 6 wet months.

In this experiment we found that Hymenachne acutigluna was a native grass that has been used by farmers in fattening cattle (Nhan et al. 2005) and planted for long time in Mekong delta provinces. But Paspalum atratum has just been used in a few years and it appears to be well suited for smallholder dairy farmers who have many choices in their feeding systems. There is also much potential for development in waterlogged soils and this needs  more research. From this point, we carried out the second experiment for evaluation of plant spacing on Paspalum atratum production in waterlogged condition.

 Experiment 2: Effect of plant spacing on Paspalum atratum production in waterlogged condition

There were influences of spacing on the first and second harvest but not in third and fourth cutting  (Table 2).

In the first two harvests, treatment 20 x 50 had higher yield compared to others, particularly in treatment 50 x 50. This reason came from the difference in density. Narrower spacing had more plants at the same plot in comparison with wider spacing. But this tendency changed in the third and fourth harvest. High density of grass requires more fertilization and may be easily N deficient.  This phenomenon has affected Paspalum production in the following year and made it to maintain the same biomass among treatments (Phaikaew et al. 2001). Hare et al. (1999) showed that Jarra digit (Digitaria milanjiana cv.) swards planted in narrow rows produced more dry matter (DM), were twice as dense and had fewer weeds than swards planted in wide rows. At the second cut 6 months after planting, row spacing had no influence on DM yield of Jarra digit.

Experiment 3: Effect of cottonseed cake and Trichantera gigantea on dairy cattle performance in the Khmer ethnic.

Dry matter intake was the same level on all diets (Table 3). Similarly, there were no influences on milk yield, milk composition and FCR among treatments.

Treatment PCT had higher profit compared to the Control and did not differ with PC. This difference was partly a reflection of replacing the protein sources, especially with Trichanthera.  They did not pay money or only a small amount for  the transport. The price of protein from foliages such as Trichanthera was cheaper than from concentrates. Thus, there was lower investment per kg milk and more benefit from selling their milk. uggested that the most effective way to enhance energy intake and performance of animals fed on crop residues was to provide them with good quality forages, including forage legumes (Topps 1997).     

Conclusions

• Paspalum atratum and Hymenachne acutigluna were well-adapted in waterlogged condition and gave the same production.

• The plant spacing had no influence on Paspalum atratum production

• Supplementation with cottonseed cake solely or in combination with Trichanthera as a replacement of concentrate in dairy diet gave the same milk yield..

Acknowledgments

The study  was carried out at the farm of dairy cooperative farmer known as EVERGROWTH, Vietnam and was supported by the MEKARN project, financed by sidaSAREC for which the authors are greatly indebted.

References

Hare M D, Booncharern P, Tatsapong P, Wongpichet K, Kaewkunya C and Thummasaeng K 1999  Performance of para grass (Brachiaria mutica) and Ubon paspalum (Paspalum atratum) on seasonally wet soils in Thailand. Tropical Grassland. 33: 75–81.

Hare M D, Tatsapong P, Lunpha  A and Wongpichet K 2004 Effect of plant spacing, cutting and nitrogen on establishment and production of Digitaria milanjiana cv. Jarra in north-east Thailand. Tropical Grasslands. 38: 217 –226.

Nhan N T H, Man N V and Preston T R 2009: Biomass yield of Hymenachne acutigluna and Paspalum atratum in association with Sesbania sesban on seasonally waterlogged soils and their use as feeds for cattle in the Mekong delta, Vietnam. Livestock Research for Rural Development. Volume 21, Article #121. http://www.lrrd.org/lrrd21/8/nhan21121.htm

Nguyen Thi Hong Nhan, Nguyen Trong Ngu, Vo Van Son, Preston T R and Leng R A 2007: Effects of oil drench on growth rate of cattle fattened on grass, supplemented with molasses, rice bran or rice straw. Livestock Research for Rural Development. Volume 19, Article #133. http://www.lrrd.org/lrrd19/9/nhan19133.htm

Phaikaew C, Khemsawat C,  Tudsri S,  Ishii Y,  Numaguchi H and Tsuzuki E 2001 Effects of plant spacing and sowing time on seed yield and seed quality of Paspalum atratum in Thailand. Tropical Grasslands 35: 129–138.

Go to top