Effects of substitution of groundnut cake with poultry litter on productive and reproductive performance of crossbred dairy cows fed a basal diet of maize stover

Yoseph Mekasha and Mengistu Urge

School of Animal & Range Sciences, College of Agriculture and Environmental Sciences, Haramaya University,
P.O. Box 138, Dire Dawa, Ethiopia
ymekasha@gmail.com

Abstract

The effect of substitution of poultry litter (PL) by groundnut cake in a dairy ration was evaluated using 20 crossbred (Holstein Frisian x Zebu) dairy cows at Haramaya University dairy farm, Ethiopia. The mean (±SD) initial body weight of the animals was 437.2±54.6 kg. Experimental animals were randomly assigned to one of the four dietary treatments, which included supplementation of concentrate mix containing PL and groundnut cake (GNC) varying at different proportion. Animals were fed a basal diet of chopped maize stover ad libitum. Treatments were: PL45= 45% PL and 0% GNC; PL35=35% PL and 10% GNC; PL22= 22%PL and 17% GNC; PL0=0% PL and 29% GNC.

Dairy cows fed PL22 diet had the highest (P<0.05) basal and total Dry Matter Intake (DMI) followed by PL35, while cows in PL0 had the highest (P<0.05) supplement DMI followed by PL22. Inclusion of PL at 22% (PL22) and 35% (PL35) of the concentrate mix improved (P<0.05) basal DMI by 13% and 9%, respectively. However, supplement and total CPI was the highest (P<0.001) for cows in PL0 followed by PL22. Dairy cows fed PL0 diet produced the highest (P<0.05) daily and 305-days milk yield followed by PL22. Including PL at 35% (PL35) and 22% (PL22) of the concentrate mix significantly extended lactation length. Dairy cows in PL35 produced milk more efficiently (P<0.05) per unit DMI. But no difference (P>0.05) noted between cows in PL22 and PL0. Efficiency of protein utilization for milk production was also the highest for cows in PL35 followed by PL22, but least for cows in PL0. In general, inclusion of PL beyond 22% resulted in loss of milk yield, body weight change, and reproductive wastage. The economic efficiency of dairy cows for milk production was highest for PL45 followed by PL22 but lower for other treatments. It is, therefore, concluded that PL could substitute GNC up to 22% of the concentrate mix for dairy cows fed a basal diet of maize stover, and this could be considered as an alternative feeding strategy to improve the performance of high-grade dairy cows.

Key words: economic-efficiency, milk-production, tropics

Introduction

Milk production in tropics and sub-tropics is mostly based on native pasture and crop residues. However, these feed resources are deficient in most of the required nutrients. Thus, it is hardly possible for ruminants to meet their nutrient requirement from this diet. Feed resources that contain less than 7% CP do not support optimum rumen fermentation (Whitman 1980). The fact that crossbred and high-grade dairy genotypes require high nutrients worsens the situation. It has been indicated that high yielding dairy cows have a greater capacity for intake and use of absorbed nutrients and thus can only achieve their full productive potential through maximal nutrient intake (Reynolds et al 1997). 

In order to efficiently exploit the genetic potential of crossbred dairy cows under tropical environment, there is a need to look for biologically and economically sound supplementation. Supplementation of low quality feeds with concentrate, especially that composed of grains, may not be practical as the latter is the major staple diet for human beings in developing countries and expensive to afford for smallholders. Instead, agro-industrial by-products such as flour mill and oil extracting factories by-products have been widely used as energy and protein supplements. Nevertheless, the rising cost of these by-products, especially oil seed cakes such as groundnut and noug seed cake, increase in the number of farms keeping crossbred dairy cows, and competition with mono-gastric animals made it less reliable. It is, therefore, imperative to search for cheap, nutritious and easily available non-conventional feed resources that would complement deficient nutrients and compensate the nutritive value of low quality roughage. Poultry litter, to this effect, is one of the non-conventional feed resources available in urban and peri-urban areas where improved dairy genotypes dominate. It contains high crude protein, ranging from 15 to 35% (DM basis), and is an excellent source of minerals such as calcium, phosphorus, potassium, magnesium and sulfur (Yoseph 1999; Goetsch and Aiken 2000). Studies confirmed positive effect of poultry litter supplementation on performance of ruminants (Tesfaye and Beyene 1998). It has been observed that urban and peri-urban dairy farms in tropical countries such as Ethiopia include poultry litter in dairy ration (Yoseph 1999). However, there is no documented information on the biological and economic optima of poultry litter inclusion in dairy cows ration. Thus, we hypothesized that inclusion of poultry litter at a given level would replace groundnut cake in dairy ration and thereby reduce the cost of milk production. The objectives of this study, therefore, were to investigate the effects of including different levels of poultry litter to maize stover based diet on the productive and reproductive performance of dairy cows and to evaluate the biological and economic efficiency of poultry litter. 

Materials and Methods

Study site 

The study was conducted at Haramaya University, the then Alemaya University, dairy farm, Ethiopia. The site is located 515 km east of the capital, Addis Ababa, and situated at 1950 m above sea level. It has an average annual rainfall of 790 mm and temperature of 16.0 ºC (Mishra et al 2004). 

Animals and management 

Twenty high-grade dairy cows (Holstein Frisian X Zebu) were secured from the University dairy farm and joined the experiment two weeks after calving. The animals had initial average (±SD) body weight of 425.2±13.4 kg.  The animals were inspected for routine health and condition of parturition before joining the experiment. Then they were randomly assigned to one of the four dietary treatments. Since all experimental animals did not calve at the same time, randomization considered period of calving as a block where a group of four cows calved during the same period were randomly allocated to four treatments. Exotic gene level and parity were equally distributed across all the treatments.  

Experimental animals were housed individually in a loose type barn (1.5m x 3.0m) with concrete flour and tied with a neck chain. The basal diet and test feed were offered in a separate feeding trough located in-front of the animals. The animals were given two-week’s adaptation period to the experimental feeds. Clean tap water was provided with a bucket fixed next to the feeding trough, and the water was changed regularly. The barn was cleaned daily and bedding material was also refreshed every day.  

The basal diet consisted of maize stover (stem, leaves, sheath) harvested right after removal of the maize grain and stored in a shed. The feed was chopped manually into approximately 1-2 inch length, packed into sack and moistened with water and left overnight before offering to soften the stover and improve feed intake. Feed preparation was handled by the same individuals and feed material was sampled every week to determine dry-matter content of the feed for the entire experimental period. Poultry (layers) litter was purchased from the University poultry farm. After removal from layers house, it was packed into sacks and transported to the University dairy farm, where it was sun-bathed for two weeks followed by deep-stacking for three months before it was included in the ration. Similar methods have been found to be effective in minimizing risks associated with parasites in poultry litter (Yoseph et al 2004).  

Dietary treatments 

Dietary treatments were prepared from poultry litter and groundnut cake at different proportions, while wheat bran and wheat short were included to make the diet balanced (Table 1).  

The quantity of concentrate mix offered to each animal was based on the level of milk production (0.5 kg concentrate for 1 kg milk). Feed adjustment was made every week based on the previous week milk yield. The basal diet (maize stover) was offered for ad libitum consumption ensuring 20% refusal.

Measurements 

The daily feed offer was divided into two equal portions and given to the animals at 8:00 and 16:00 hours separately. Orts were collected every 24 hours. Both feed offer and orts were measured and recorded daily. Behavioral oestrus manifestations were monitored twice a day (9:00-11:00 and 16:00-18:00 hours) and all observed estrus manifestations were recorded. Cows were mated to bulls following observed behavioral oestrus. The body weight of the animals were measured every week using fixed weighing bridge while the body condition score was assessed subjectively using the scale 1-5 (1=emaciated and 5=Obese) by an experienced technician. The experimental animals were milked twice a day using bucket type milking machine assisted with hand milking for stripping. Both the AM and PM milk yield were measured immediately after milking and recorded for each animal. The economic analysis of supplementing different levels of poultry litter was done by employing partial budget analysis system.

Sampling, sample processing and chemical analysis 

Samples of feed offer and oarts were taken every week in a labeled individual bag according to animal identification. Except for maize stover, samples of other feed types were bulked for a month, sub sampled and ground to pass through 1mm sieve screen. The chemical analysis was made after pooling and sub-sampling the ground material by feed type and animal identification. Since maize stover was rinsed with water before offer, samples were dried in an oven at 60 ^oC to constant weight on weekly basis and then pooled for a month before sub-sampling. The moisture content of the feed types was measured instantly after sampling using a forced draft oven at 105 ^oC overnight. Feed samples were analyzed for Nitrogen (N) content using the Kjeldahl procedure. The crude protein (CP) content of the feed was determined as Nx6.25. The energy content of the feed materials was estimated from literature (Annindo et al 1994; Kearl 1982, Seyoum and Zinash 1989).  

Statistical analysis 

The data on feed intake, body weight change, body condition score, milk yield, and reproduction performance were analyzed in a randomized complete block design using the General Linear Model procedure of SAS (SAS 1999). Means were separated using Least Significant Difference and declared significant at P<0.05. Quadratic equation was fitted to further assess the relationship between the different levels of poultry litter inclusion and response variables. 

Results and Discussion

Chemical composition of feeds 

Groundnut cake had the highest crude protein (CP) while maize stover had the least (Table 2). Poultry liter also had high CP but its energy content was medium. Wheat short had the highest energy content followed by groundnut cake and wheat bran. The CP content of groundnut cake, wheat bran and wheat short used in the present experiment was similar to the values reported earlier (Getnet 1997; Yoseph 1999). The CP content of poultry litter (layers litter) was similar to the value reported by Asrat (2003) and Tesfaye and Beyene (1998) for layers litter. However, it was lower than the value reported by Yoseph et al (2002) for broiler litter. The discrepancy might be due to differences in poultry production system and feed types offered to the birds. Besides, the litter removed from layers house at Haramaya University poultry farm contained more feed leftover which might have influenced its CP content.  

Feed intake 

Supplement, basal and total dry matter intakes (DMI) were differed (P<0.001) among treatments (Table 3). Dairy cows in PL22 had the highest DMI from the basal, while cows in PL0 had the highest DMI from the supplement. Total DMI, however, was the highest for cows in PL22. The finding is in agreement with Asrat (2003) who obtained high basal DMI at 28% poultry litter inclusion and attributed it to the development of optimum rumen ammonia N that promoted improved fiber fermentation. The combined effect of poultry litter and groundnut cake might have also contributed to such improvement. The highest supplement DMI for PL0 might be due to substitution effect as groundnut cake is more palatable than poultry litter since the latter has a bitter test due to burette content (Yoseph et al 2004).  Supplement and total CP intakes were the highest (p<0.05) for cows in PL0 followed by PL22. The mean CP intake from the supplement decreased as the level of poultry litter inclusion increased. These could be attributed to decreased efficiency of nitrogen usage at increase in level of broiler litter (Goetsch and Aiken 2000). Dairy cows in PL45 had the lowest supplement, basal and total DM and CP intakes compared to cows in other treatments. The lower acceptability of poultry litter, highest degradability in the rumen and high ash concentration could be the major contributing factors. Similarly, Nadeem et al (1993) reported depressed total DMI at 30% broiler litter inclusion in Barbari goats. Asrat (2003) also obtained low total DMI at 45% poultry litter inclusion in Hararghe highland goats. Cows in PL22 had the highest (p<0.05) DMI when expressed as percentage of body weight followed by cows in PL35 and PL0 while cows in PL45 had the lowest.  

Lactation performance 

The average daily milk yield, lactation yield, 305 days yield and lactation length of crossbred dairy cows differed (P<0.001) among treatments (Table 4). Dairy cows in PL0 produced the highest (P<0.05) average daily and 305-days milk yield followed by PL22 and PL35 while cows in PL45 had the least. This could be attributed to the high CP intake in the former treatment compared to the latter one. It could also be related to differences in rumen un-degradable protein composition since groundnut cake has 25% rumen un-degraded protein (RUP; NRC, 1988) compared to poultry litter which is completely rumen degradable material. It has been indicated that supplementing high producing dairy cows with high amount of RUP increased milk yield (NRC 1985). In spite of this, there was no difference (P>0.05) in feed conversion efficiency between cows in PL22 and PL0. The efficiency of CP utilization for milk yield was the lowest for cows in PL0 while the efficiency of feed utilization for milk production was the lowest for cows in PL45. This implies that supplementing dairy cows with poultry litter at 22% (PL22) to 35% (PL35) of the concentrate mix yielded optimum milk production with efficient utilization of feed DM and CP compared to the rest dietary treatments. Besides, 22 to 35% poultry litter inclusion as a replacement for groundnut cake in dairy cattle ration supplied the required amount of rumen degradable protein for optimum functioning of the rumen eco-system and fiber fermentation. The fact that poultry litter is rich in total ash, limited in its energy concentration and completely degraded in the rumen might not allow high level of inclusion in dairy cattle ration. Since dairy cows, particularly high producing genotypes, demand optimum energy and require rumen bypass protein for milk production, inclusion of poultry litter beyond 22% might not be biologically feasible as higher levels of inclusion in the present experiment showed depression in feed intake as well as milk production parameters. 

Body weight change  

Final body weight was the lowest (P<0.05) for dairy cows in PL45 compared to the rest dietary treatments although body weight loss was not observed in this study (Table 5). Inclusion of poultry litter in the concentrate mix either at 22% or 35% as a replacement for groundnut cake did not affect (P<0.05) final body weight as compared to the group in PL0. Mean body weight of dairy cows was also similar for cows supplemented with poultry litter either at 0% or 22%. The finding was in agreement with Asrat (2003) who reported depressed final body weight in goats supplemented with higher level of poultry litter (45%). Body weight change of dairy cows was the highest (P<0.05) for cows in PL0 followed by cows in PL22 while cows in PL45 had the least. However, the relationship between increasing level of PL and body weight change was negative and quadratic (R²=0.94). This could be due to supply of low level of rumen un-degradable protein in diets containing high proportion of poultry litter, which is highly rumen degraded material. In connection to this, McDonald et al (2002) revealed that rumen un-degraded protein was more useful in promoting daily live weight gain. This shows that inclusion of PL beyond 22% could be counter productive due to body weight loss. 

Reproductive performance 

There was no difference (P>0.05) in calving to first observed estrus (d), calving to fertile mating and the number of services per conception among the dietary treatments (Table 6). Nevertheless, the relationship between increase in level of PL in the concentrate mix and calving to observed oestrus and to fertile mating was quadratic (R2=0.97) although PL0 and PL22 were consistent. This implies that substituting groundnut cake with PL beyond 22% in the concentrate mix of dairy cattle might result in reproductive wastage, which is moderate for PL35 but large for PL45.  

Economic analysis 

Added return per day (BIRR) due to milk yield was the highest (P<0.05) for cows in PL0 and PL22 and the least for cows in PL35 (Table 7). The highest milk yield and total DMI in PL0 and PL22 relative to PL45 and PL35 are the major reasons for the difference. Added costs mainly due to feed expense was the highest (P<0.05) for cows in PL0 but the least for cows in PL45. Though feed costs due to the basal feed intake was the same (P>0.05) for cows in PL45 and PL0, feed costs incurred to supplement was the highest for cows in PL0. Thus, the highest added cost for cows in PL0 in the present experiment was due to the highest supplement cost. The net benefit was the highest for PL45 followed by PL22 and PL0. This is attributed to the lower feed cost for PL45 and moderate feed cost in the rest dietary treatments. Although economic evaluation suggests that economic efficiency was higher for PL45 followed by PL22 than for other dietary treatments, the recommendation should take into account biological optima to make sound conclusion.  

Conclusion

• Supply of adequate nutrients is crucial for optimum milk production without jeopardizing the reproduction capacity of dairy animals. However, tropical feed resources are of low quality and demand appropriate supplementation to boost milk production from improved dairy genotypes. This study indicated that supplementing dairy cows with a concentrate mix at 22% poultry litter (PL) as a replacement for groundnut cake increased basal and total dry matter intake compared to other treatments. Feed DM and CP conversion efficiency was the highest for cows fed PL at 35% followed by PL at 22%. Supplementing dairy cows with PL at 22% did not depress the body weight, body condition score and reproduction performance compared to cows fed supplement ration that contained only groundnut cake (PL0). Although economic efficiency of supplementing PL was higher for PL45, it negatively affected productive and reproductive parameters. However, inclusion of PL at 22% did not impair performance, and yielded higher net benefit compared to dairy cows supplemented with concentrate without PL. Thus, in terms of both biological and economic optima, this study elucidated that PL could substitute groundnut cake up to 22% of the concentrate mix for dairy cows fed a basal diet of maize stover. The study suggests that this could be considered as an alternative feeding strategy to improve the performance of high-grade dairy cows.

Acknowledgements

The authors would like to acknowledge the Federal Government of Ethiopia (through Ethiopian Institute of Agricultural Research) for financing this study and Haramaya University for providing experimental animals, facilities and administrative supports. Mullugeta Dessalegne, Marta Fikru, Desse Kebede and Merga Bayisa are highly appreciated for assisting in data collection. Our thanks are due to Mullugeta Bussa for availing maize stover for the experimental animals. Last but not least, our gratitude goes to Mihret Sissay for entering data. 

Reference

Annindo D O, Said A N and Lahlou-Kassi A 1994. Chemical composition and nutritive value of feed stuffs for ruminant livestock in Sub-Saharan Africa. International Livestock Centre for Africa (ILCA), Addis Ababa, Ethiopia.p537. 

Asrat Gebretsadik 2003. Efficiency of utilization of poultry litter in the ration of goats. MSc. Thesis Submitted to the School of Graduate Studies, Alemaya University of Agriculture, Ethiopia. p72. 

Getnet Berhanu 1997. Performance of Somali goats supplemented with different proportions of ground nut cake and wheat bran. MSc. Thesis Submitted to the School of Graduate Studies, Alemaya University of Agriculture.  

Goetsch A L and Aiken G E 2000. Broiler litter in ruminant diets-Implications for use as a low-cost by-product feedstuff for goats. In: Proceedings of the Conference on the Opportunities and Challenges of Enhancing Goat Production in East Africa held at Debub University, Awassa, Ethiopia, from Nov. 10-12, 2000. Pp58-69 

Kearl L C 1982. Nutrient requirement of ruminants in developing countries. International feed stuffs institute, Utah Agricultural Experiment Station, Utah State University, Logan, Utah 84322, USA.  

McDonald P, Edward R A, Greenhalgh J F D and Morgan C A 2002. Anima Nutrition. 4^th Edition, Long Man Scientific and Technical. p693. 

Mishra B B, Heluf G/Kidan, Kibebew Kibret, Mohammed Assen and Bruk Eshetu 2004. Soil and land resource inventory at the Alemaya University research farm with reference to land evaluation for sustainable agricultural management and production. In: Heluf G/Kidan, and Mishra B B (Eds). Specific features and management options of soil and land resources of eastern Ethiopian highlands for sustainable agricultural production. Synthesis of working papers. Dire Dawa, Ethiopia: Alemaya University: Soil Science Bulletin No. 1. 

NRC (National Research Council) 1988. Nutrient Requirements of Dairy Cattle. 6^th Revised Edition. National Academy Press, Washington, D.C. p157. 

NRC (Nation Research Council) 1985. Ruminant Nitrogen Usage. National Academy Press, Washington, DC. Pp7-12. 

Nadeem M A, Ali A, Azim A and Khan A G 1993. Effect of feeding broiler litter on growth and nutrient utilization by Barbari goat. Australian Journal of Agricultural Science. 6(1):73-77.

Reynolds C K, Sotton J D and Beever  D E 1997. Effect of feeding starch to dairy cattle on nutrient availability and production. In: Garnsworthy, P.C. and Wiseman, J (Ed.). Recent Advances in Animal Nutrition. Pp:107-134. Nothingham University press. 

SAS 1999. Statistical Analysis System, Release 8.02, SAS Institute, Inc., Cary, NC, USA.  

Seyoum Bediye and  Zinash Sileshi 1989. The composition of Ethiopian feeds. Research report No. 6. Institute of Agricultural research, Ethiopia 

Tesfaye A and Beyene C 1998. Performance of native sheep fed on rations supplemented with poultry litter and/or monensin. In: IAR Proceedings. Second National Livestock Improvement Conference, Addis Ababa, Ethiopia, February 24-26, 1998. p281. 

Whitman P C 1980. Tropical Pasture Science. Oxford University Press, Oxford, p392. 

Yoseph Mekasha, Merkel R C, Goetsch A L, Sahlu T and Tesfai K 2004. Effects of method of offering broiler litter and level of prairie hay intake on growth of Boer x Spanish wethers. Small Ruminant Research. 55:123-133. 

Yoseph Mekasha, Azage Tegegne, Alemu Yami and Umunna N N 2002. Evaluation of non-conventional agro-industrial by-products as supplementary feeds for ruminants: in vitro and metabolism study with sheep. Small Ruminant Research. 44:25-35. 

Yoseph Mekasha 1999. Impact of feed resources on productive and reproductive performance of dairy cows in the urban and peri-urban dairy production system in the Addis Ababa milk shed and evaluation of non-conventional feed resources using sheep. MSc. Thesis.  Alemaya University of Agriculture, Ethiopia. p197.

Received 13 February 2013; Accepted 21 May 2013; Published 2 June 2013

Go to top