Previous Page Table of Contents Next Page


Feeding systems for milk production in the high potential areas of Kenya: on-farm trials

A. Abate1, A.N. Abate2 and S. Gacugia3

1 Department of Animal Production, University of Nairobi, Kenya
2 Kenya Agricultural Research Institute, Muguga, Kenya
3 Bayer (East Africa), Nairobi, Kenya


Abstract
Introduction
Materials and methods
Analyses
Results
Discussion
Conclusion
Acknowledgements
References

Abstract

Composition of various types of feeds offered to Friesians, Ayrshires or their crosses were studied in several farms practicing zero-grazing, semi-zero-grazing and grazing systems of milk production. The experiments were carried out in the high potential areas of Kenya during the wet season. There was uniformity in the chemical composition of concentrate fed within and across farms except for the level of phosphorus (P) which differed significantly (P<0.01) between farms. Composition of fodder varied significantly (P<0.01) between farms with regard to crude protein (CP), fibre (ADF), calcium (Ca) and P. Farms differed significantly (P<0.05) only in the level of P in pastures. Pasture or fodder contributed most to dry-matter (DM) intake averaging about 12.1 kg per animal each day under grazing and about 10.0 kg when animals were fed in confinement. A wide range of fodder was fed under semi-zero-grazing and intake of DM from such supplements averaged 1.9 kg per animal per day. Fodders fed under semi-zero-grazing were superior to that fed under zero-grazing in terms of levels of critical nutrients. Daily concentrate consumption differed significantly (P<0.01) between farms and was highest (4.3 kg DM) and least (0.44 kg DM) per animal under zero-grazing and grazing respectively. Average milk production per farm was 12.0, 11.6 and 10.2 kg for zero- grazing, semi-zero-grazing and grazing systems respectively. Given the level of production, all the feeding systems were deficient in at least one nutrient. It was concluded that for all systems, protein, Ca and P supplementation seem desirable particularly when milk production per animal is in excess of 10 kg per day.

Introduction

The high potential areas of Kenya are defined as those characterised by between 1200 and 2000 mm of rainfall per annum with dairying as a major farm activity carried out in small holdings. The historical development and ecological distribution of the systems of milk production in these areas have been reviewed (Cinema, 1984) and corresponding milk output including measures for increased yields discussed (Abate et al., 1987b). There is also information on the economics of the different production features of small holder dairying (Stotz, 1983). The type of feeding practiced in each system is generally known but data on the quality and quantity of material fed particularly fodder are lacking (Abate et al.,
1987a). This study was initiated to assess the wet season variation in nutrient supply to milking animals managed under different systems of production.

Materials and methods

The experiments were carried out in 5 farms representing 3 systems of production as summarised in Table 1. Friesians, Ayrshires or their crosses were selected on the basis of age, parity and production and used in the experiments. The zero-grazing trial was divided into two 28 day and two 18 day periods. All other experiments consisted of 3 periods each of 30 days.

Table 1: Features of production system and experimental conditions.

Farm

Production systems

n

Days on trial

labour

Water frequency per day

1

Zero-grazing

4

92

Hired

x 3

2

Semi-zero-grazing

5

90

Family

x 4

3

Semi-zero-grazing

3

90

Family

x 2

4

Semi-zero-grazing

3

90

Hired

x 2

5

Grazing

3

90

Hired

x 2

Types of feed offered in each system are given in Table 2. Under zero-grazing, each animal was fed individually from concrete troughs twice a day at 08.00 and 15.00 hours. Each time concentrate was first fed after which chopped Napier grass was fed to appetite; the amount offered was weighed using a spring balance of a 50 ± 0.5 kg capacity. Little concentrate was also fed at milking. Refusals were weighed every afternoon and morning and samples accumulated over a week or 5 days.

In the semi-zero-grazing and grazing treatments animals were individually fed from half-cut drums or basins. Feeding of fodder was once a day either in the morning or afternoon; the rest of the time, the animals grazed. All fodder was chopped into 3-8 cms lengths before feeding. A balance similar to the one described above, was used to weigh feeds offered and those rejected daily. Concentrates, were rationed and fed twice a day at milking. Intake from pasture was estimated based on the acreage available for grazing, grass cover, time spent on grazing and ranged from 2.0 to 3.0% of body weight. A mineral supplement Baymix maziwa, was mixed in the concentrate and provided in all trials to each animal. The frequency of watering from piped, bore hole and standing rainwater is as shown in Table 1.

Table 2- Types of feed offered in each feeding system


Zero-grazing


Semi-zero-grazing


Grazing

Feed type







1a

2

3

4

5

Napier grassb

Cabbage

Cabbage

Napier grass + molasses



Maize fodder

Maize fodder

Napier grass + molasses



Weeds

Potatoes

Banana stems and leaves + molasses



Potatoes

Oat fodder

Banana stems and leaves + molasses



Kale


Banana stems and leaves + molasses





Green bananas


Pasture



Mixed species

Mixed species

Kikuyu grass

Kikuyu grass


Kikuyu grass


Mixed species

Chloris gayana

Cocentrate



Daily meal

Wheat bran

Daily meal + wheat bran

Crushed maize + fodder yeast

Crushed maize + daily meal


Dairy meal maize germ + wheat bran



Wheat bran + daily meal





Wheat bran + dairy meal + crushed maize

a Farm
b Usually maize fodder is also fed.

Samples of Napier grass offered and rejected in the zero-grazing trial were taken once a week and prepared for analysis. In the other treatments, samples of all feeds offered and rejected were collected once a month. Grazing was sampled by clipping grass from randomly selected areas once monthly. All concentrates were sampled periodically, accumulated over each period and sub-sampled for analysis. Milk produced was weighed and recorded daily for each animal. Heart girth measurements were taken once in each period to estimate animal weight changes.

Analyses

Samples were assayed for dry matter (DM), crude protein (CP), fibre (ADF) and lignin using approved methods (A.O.A.C., 1980; Van Soest, 1965). Calcium (Ca) and phosphorus (P) were determined after wet ashing using atomic absorption spectrophotometry and a Beckman Spectrophotometer respectively. Only data for 3 periods were subjected to least square analysis and separation of means using the mixed model package of Harvey (1987).

Results

Types of feed offered to animals in each feeding system are presented in Table 2. They included a variety of fodders, pasture and a number of concentrates. In some farms fodder was fed without additives, in others it was mixed with molasses. Concentrates consisted of cereal grains, commercially compounded feeds and by-products of industrial processing. Very often a number of concentrates were mixed together before feeding as a meal or in the crushed form as in the case of maize grains. Some form of rotational grazing was practiced during the day; at night animals were grazed nearer the homesteads.

The DM content of fodder increased with period in ail farms except where a fodder of particularly low DM was introduced into the feeding system (Tables 3 and 4).

Chemically, the nutrient content of fodder varied with time within a feeding system but the differences were not significant (P>0.05). Under zero-grazing (Table 3) where Napier grass was the only fodder fed this variation was in the direction of lower CP. There was no recognised pattern of change in nutrient levels under semi-zero-grazing; only in one farm was there a significant (P<0.05) difference in the levels of ADF with period (Table 4). The trend of higher DM content with time was consistent in pastures of all farms (Tables 4 and 5). Since only one observation was recorded per period no statistical test was possible to detect period differences with regard to nutrient content of pasture. There were, however, fluctuations of differing degrees in all nutrients. Concentrate DM content was about uniform from period to period (Tables 4 and 5). The CP content of the concentrates was also uniform throughout the experimental period; mineral levels, however, varied.

Table 3: Variation with period in the chemical composition of feeds offered under zero-grazing.

Feed type

Period

DM %

CP

ADF

Ca

P




% DM

Fodder



1

12.4

10.4

45.2

0.13

0.27

2

15.7

9.4

41.2

0.26

0.21

3

17.4

7.0

44.3

0.28

0.18

Concentrates



1

88.9

17.0

-

0.88

0.67

2

86.2

16.4

-

0.89

0.64

3

86.7

16.1

-

0.83

0.65

(-) denotes not determined

Table 4: Variation with period in the chemical composition of feeds offered under semi-zero-grazing.


Farm 1

Farm 2

Farm 3

Feed type

11

2

3

1

2

3

1

2

3

Fodder

Dry matter, %

14.3

17.3

17.7

16.3

16.6

21.2

31.1

14.6

10.9

2CP, %

11.3

12.9

13.2

10.6

10.6

9.6

7.3

6.3

3.1

ADF, %

27.8

31.8

27.8

28.3

27.3

30.0

44.6a

27.2b

17.1bc

Ca, %

0.56

0.48

0.55

0.34

0.35

0.88

1.08

1.40

1.34

P. %

0.26

0.22

0.21

0.31

0.33

0.32

0.25

0.10

0.08

Pasture

Dry matter, %

21.2

26.0

36.3

26.5

31.8

37.1

16.3

24.5

25.6

2CP, %

6.9

8.6

7.2

7.8

8.5

7.1

8.0

6.3

7.2

ADF, %

30.9

35.2

39.7

32.6

37.5

36.3

32.4

43.4

36.4

Ca, %

0.46

0.46

0.45

0.45

0.41

0.32

0.43

0.42

0.48

P, %

0.10

0.10

0.09

0.24

0.28

0.18

0.29

0.18

0.21

Concentrate

Dry matter, %

89.0

90.8

88.6

89.6

90.2

89.0

89.1

89.6

90.5

2CP, %

16.1

16.0

16.0

16.8

17.0

16.5

16.3

16.2

16.4

Ca, %

0.16

0.47

0.71

0.58

0.61

0.54

0.78

1.45

0.44

P, %

1.61

0.94

1.12

1.04

1.03

1.06

0.50

0.50

0.39

1Period
2% DM
Means with different superscripts along the same row are significantly different (P<0.05).

Table 5: Variation with period in the chemical composition of feeds offered under grazing.

Feed type

Period

DM %

CP

ADF

Ca

P


% DM

Pasture

1

17.0

13.7

33.9

0.35

0.67

2

26.1

7.8

43.5

0.20

0.64

3

31.8

7.2

45.7

0.62

0.65

Concentrate

1

69.1

17.0

-

0.58

0.82

2

89.6

17.3

-

0.31

1.09

3

91.7

15.0

-

0.62

0.24

(-) denotes not determined.

Differences between farms in the levels of nutrients in feeds offered are shown in Table 6. Crude protein, ADF, Ca and P of fodder varied significantly (P<0.01) between farms. Crude protein was significantly higher (P<0.01) under semi-zero-grazing than in zero-grazing and grazing systems. Analysis of variance showed that, overall, nutrient concentrations in pastures of all farms was fairly similar (Table 6). Linear contrasts of the farms showed them, however, to differ significantly (P<0.05) in the content of P. The CP and Ca levels in the concentrate were not significantly affected (P>0.05) by farm but P concentrations were (P<0.05).

In Table 7 are shown animal weights, DM intake and milk production per animal in each farm. In all systems, fodder or pasture contributed most of the DM ingested. Intake of DM from concentrate differed significantly (P<0.01) being highest under zero-grazing and least under grazing. System of feeding also significantly affected (P<0.01) the quantities of fodder consumed per animal daily. Dry-matter intake from pasture was similar in the semi-zero-grazing farms and significantly lower (P<0.05) than under grazing. Animals in farm 4 were the lightest and produced the least milk.

Animal requirements were met for protein and P but not Ca in the zero-grazing system. Under semi-zero-grazing feeding, animals were deficient in all nutrients when they produced at least about 11 kg of milk, their requirements were, however covered with production of about 6 kg of milk. Under grazing the animals experienced protein and Ca deficits but had sufficient intake of P.

Table 6: Variation with farm in the chemical composition of feeds offered.


Farm






11

2

3

4

5

Fodder

Dry matter, %

15.2

16.5

18.0

18.9

-

2CP, %

9.0ac

12.5b

10.2ab

5.6c

-

ADF, %

43.6a

29.1b

28.6b

29.6b

-

Ca, %

0.22a

0.53b

0.52a

1.3b

-

P, %

0.22a

0.23a

0.32b

0.14a

-

Pasture

Dry matter, %

-

27.8

31.8

22.1

25.0

2CP, %

-

7.6a

7.8a

7.2a

9.6a

ADF, %

-

35.3a

35.5a

37.4a

41.0a

Ca, %

-

0.46a

0.41a

0.44a

0.39a

P, %

-

0.10a

0.23ab

0.23ab

0.33b

Concentrate

Dry matter, %

87.3

89.5

89.6

89.7

83.5

2CP, %

16.5a

16.0a

16.8a

16.3a

16.4a

Ca, %

0.84a

0.45a

0.58a

0.89a

0.50a

P, %

0.65a

1.07b

0.04b

0.46ac

0.72ab

1 period
2% DM
Means with different superscripts along the same row are significantly different (P<0.01 and P>0.05).

Table 7- Variation with farm in DM intake, animal weights and milk production


 

Farm

11

2

3

4

5

Animal weight, kg

438.9 ± 56.7

392.4 ± 40.4

385.3 ± 45.8

369.3 ± 25.6

401.8 ± 63.2

Dry matter intake, kg/d

Concentrate

4.3a

0.27bc

0.67b

1.80c

0.44d

Fodder

10.0a

1.2b

0.8bc

3.7d

-

Pasture

-

9.8a

9.6a

7.4a

14.1b

Total

14.3

11.7

11.1

12.9

14.5

Milk production, kg/d

12.0 ± 1.2

12.1 ± 1.4

11.2 ± 3.1

5.7 ± 1.5

10.2 ± 3.3

Means with different superscripts along the same row are significantly different (P0.05).
Mean ± SD

Discussion

The results of the present study show that in the high potential areas of Kenya, there are differences in the range of feeds offered to animals at farm level during the wet season. These differences are often environmentally determined so that farmers do not necessarily feed sufficient amounts of the types of feeds that would produce maximum performance in their animals in accordance with feeding standards. As a fodder, Napier grass has been popularized by the extension service (Stotz, 1983; Wouters, 1986; Abate et al., 1987a) and this explains its use in zero -grazing and semi-zero-grazing systems. The feeding of other fodders has developed over time through farmers own observations and exchange of ideas with each other. Cabbage was, for example, fed because it grew well in the area, was cherished by cows, was fed by other farmers in the neighbourhood and was believed to increase milk yield. Our observations, showed that many small holders, fed cabbage at least during milking. The farmers are, however, possibly unaware of the goitergenic effects of feeding large quantities of brassicas. At maturity, cabbage DM was still below 10% and this would limit the intake of DM from this fodder. Weeds were fed because it was noticed that during grazing cows concentrated around weed-dominated areas. On analysis the weeds were found to contain on average about 16% CP which is useful in the maintenance of the N economy of animals largely dependent on pasture. The farmers also fed potatoes and molasses as sources of readily available carbohydrates in addition to the wide range of concentrates offered. Mixing of concentrates was a common feature of the farms practicing semi-zero-grazing in this study. For some farmers the practice developed in an attempt to cut down costs by combining cheaper concentrates with expensive ones. For others it was simply a result of resource availability.

Dry matter of fodder offered to animals can be influenced by feeding at early stages of growth. Thus the increase in DM levels of fodders with time is an indication that in these experiments, farmers knowingly avoided the feeding of high moisture young fodders. This was achieved in one instance through a time-staggered planting regime so that not all fodders reached maturity at the same time. The near uniformity in the chemical content of fodders fed with time (Table 4) has practical implications. It shows that nutritionally, it is advisable to feed a combination of fodders inorder to maintain a high concentration of critical nutrients throughout the wet season. To prevent the alteration of such a favourable nutrient balance, it is desirable for farmers to plant and feed sufficient quantities of each fodder.

The increases in DM content of the pastures in these experiments were a reflection of maturity and consistent with results reported elsewhere in Kenya (Said, 1971; Karue, 1974; Abate, 1978). Fluctuations in chemical composition were a result of new regrowths induced by precipitation within a given period. Abate (1978) has shown similar changes in the protein and fibre components of a predominantly Chloris gayana pasture. By practicing some form of rotation the farmers allowed young vegetative growth to mature and hence showed that they had consideration for nutritive quality.

As purchased feeds, the DM content of concentrates can only be influenced et the point of formulation. A farmer can, however, affect the DM content by the method of feeding. Addition of water or molasses to concentrates to reduce dustiness etc., will reduce feed DM content (Table 5). Changing of the type of concentrate with period was responsible for the variation in the mineral levels with time.

In terms of nutrient concentration fodders fed under semi-zero-grazing were superior to that offered in zero-grazing. However, semi-zero-grazing resulted in intake of low DM from the fodders mainly because the content of DM of the fodders was often very low. Cabbage and banana stems were examples of such material. Intake under this feeding system can be improved by planting more fodders like maize and oat which are capable of accumulating DM with time.

Generally, all the pastures reported here were of sufficient nutrient level to promote satisfactory milk production if adequately consumed. Only in one farm was the level of P near the deficiency margin (Table 6); all other farms contained concentrations that are similar to those reported by Howard et al. (1962). Calcium levels were satisfactory and above the minimum below which deficiency is likely to occur.

The differences in P levels in concentrates with farm were because type of concentrate fed differed. There are also several possible reasons to explain the variation with farm in the intake of concentrate DM. These include farmers ignorance as to the importance of concentrate as a production ration, cost and availability, distance from source and the purchasing power of individual farms. Stotz (1979) noted an inverse relationship between the feeding of concentrate to cows and the distance from the supply centre. Efforts should, therefore, be made to produce concentrates nearer to the farmers and to encourage them to feed higher amounts. The animals used in these studies have higher genetic potential than was shown by their milking performance and would, therefore, respond to improved nutrition.

While it is true that feeding standards have no relevance under certain feeding conditions, the results of this study show at least one nutrient to be deficient in any one feeding system. The effects of low dietary CP are manifested in decreased yields which is undesirable since it affects farmers' income. Supplemental protein is, therefore, recommended in the semi-zero -grazing and grazing systems except where the animals are poor milkers. Protein concentrates are expensive and this may militate against their use as supplements. Establishment of shrubs like Leucaena leucocephala on the farm is cheaper and can supply some of the protein required by the animals. Leucaena has been known to promote feed intake, utilisation and production in ruminant animals. If minerals were not provided in these experiments, the feeding systems were such that supplementation would be necessary for Ca and also P particularly when animals produced at least 10 kg of milk. This is important so that mineral deposition may be favoured during the wet season for use in periods of stress. If farmers are educated on the merits of supplementation, they are likely to adopt the practice. In these experiments, farmers who previously fed fodder in the long form agreed to continue chopping it because they realized it increased intake and reduced wastage. Kategile (1986) has noted that farmers acknowledge the benefits of improved feeding.

Conclusion

The following conclusion can be made from the results of the present study:

a. A variety of fodders, some of them with as low as 5% DM are fed during the wet season in feeding systems practiced in the high potential areas of Kenya.

b. Feeding a number of fodders together is superior to feeding only one type in terms of concentration of critical nutrients.

c. Concentrate fed under semi-zero-grazing is little but the amount could be improved by producing compound feeds nearer to the farmers and making the farmers aware of the benefits of concentrate supplementation.

d. All the feeding systems are deficient in at least one nutrient particularly when animals are good milkers.

e. For all systems protein and Ca supplementation seem desirable particularly at high levels of milk production.

Acknowledgements

The authors wish to thank Dr. R. Baptist of the Department of Animal Production, University of Nairobi for useful discussions on statistical matters.

References

Abate, A. 1978. Dry matter and nutrient intake by grazing dairy heifers from a predominantly Chloris gayana pasture. M. Sc. thesis, University of Nairobi, Kenya.

Abate, A., Kayongo-Male, H. and Wanyoike, M. 1987a. Fodder for high potential areas in Kenya. In: J.A. Kategile, A.N. Said and B.H. Dzowela (eds), Animal feed resources for small-scale livestock producers. Proceedings of the second PANESA Workshop held in Nairobi, Kenya, 11-15 November 1985. IDRC manuscript report. IDRC (International Development Research Center), Nairobi. pp 116-124.

Abate, A., Wanyoike, M. and Said, A.N. 1987b. Milk production under integrated farming systems in Kenya. Bull Int. Dairy Fed., 221:5-9.

A.O.A.C. 1980. Official methods of analysis. 13th Ed. Washington, D.C.

Chema, S. 1984. Milk production in Kenya. Paper presented at a Workshop on the Potential for Small-scale Milk Production in Eastern and Southern Africa held in Nairobi, 19-21 September, 1983. pp 34-47.

Harvey, W.R. 1987. User's guide for LSMLMW PC-1 version. Columbus, Ohio, USA.

Howard, D.A. Burdin, M.L. and Lampkin, G.H. 1962. Variation in mineral and crude protein content of pastures at Muguga in the Kenya Highlands. J. Agric. Sci. (Cambridge) 59:251-256.

Karue, C.N. 1974. The nutritive value of herbage in semi-arid lands of East Africa. I. Chemical composition. East Afr. Agric. For. J. 40:89-95.

Kategile, J.A. 1986. Need for more milk production in SADCC countries: a challenge to animal scientists. Paper presented at the SADCC-SACCAR Workshop on Livestock at Maseru, Lesotho, 24-26 November 1986.

Said, A. 1971. In-vivo digestibility and nutritive value of Kikuyu grass (Pennisetum clandestinum) with a tentative assessment of its yield of nutrients. East Afr. Agric. For. J. 37:15-21.

Stotz, D. 1979. Smallholder dairy development in past, present and future in Kenya. Ph.D. thesis, University of Hohenheim, Fed. Rep. of Germany.

Stotz, D. 1983. Production techniques and economics of smallholder livestock production systems in Kenya. Ministry of Livestock Development, Animal Production Division, Nairobi, Kenya.

Van Soest, P.J. 1965. Non-nutritive residues: A system of analysis for the replacement of crude fibre. J. Assoc. Off. Agric. Chem. 50:50-55.

Wouters, A.P. 1986. Dry matter yield and quality of Napier grass on farm level. Dairy Development Project, Ministry of Agriculture and Livestock Development, Nairobi.


Previous Page Top of Page Next Page