CamDairy ration formulation and analysis model.

Abstract

CAMDAIRY RATION FORMULATION * AND ANALYSIS * MODEL R.C. KELLAWAY and D.J. HULME SUMMARY CAMDAIRY is a package of computer programs which is designed to help advisers, farmers, students and research workers who are involved in the feeding management of dairy It can be used to predict performance and to explore the profitability of alternative management decisions relating to The core program is a the nutrition of dairy cows. mathematical model of a lactating cow, which incorporates functions to predict nutrient requirements, feed intake, substitution effects when feeding concentrates, tissue mobilisation and partition of nutrients between milk production and growth. cows l (Key Words: Computer model, dairy / cow nutrition) INTRODUCTION There has always been a large gap in application between the information available on the nutrition of dairy cows and the practice of feeding management. Milk production is determined by a complex interaction of feed and animal factors. Whilst standard texts on feeding dairy cows (MAFF 1975tNRC 1978;ARC 1980) provide information on requirements for individual nutrients, they do not provide the means of integrating feed and animal factors to predict milk production. Similarly, >these texts cannot be used to calculate rations They assume that the relationship which maximise profit. between energy intake and milk production is linear, whereas it When farmers is in fact a negative exponential relationship. wish to maximise profit, the optimum feeding level has to be determined by the shape ,of the response curve, feed costs and The computations involved are only returns from milk sales. feasible with a computer and this has severely restricted the application of current knowledge in the feeding management of dairy cows. The recent availability of powerful and inexpensive microcomputers makes it possible to access and apply available The complex mathematics of information on dairy cow nutrition. ration formulation and analysis are no longer a problem. Micro-computerti can do these calculations in a few seconds. The starting point for CAMDAIRY was the California model (Dean et al, 1972) from which we used data and file handling routines and screen layouts.* This saved about one man year of development time. Since then, we have spent in excess of three man years to develop the current model. In CAMDAIRY we have -------------------------------------------------------------e- M.C.Franklin Laboratory, Department of University of Sydney, Camden, NSW 2570 152 * Animal Husbandry, used what we believe to be the most relevant; i'nformation available relating to nutrient intake and utilisation by lactating cows. The model is carefully designed to .be 'user friendly' so that it can be rapidly understood and applied by anyone with a basic knowledge of dairy cow nutrition. CAMDAIRY STRUCTURE The model is comprised of three modules: 1. Profit maximising 2 . Performance prediction and ration analysis 3. Feed library Profit Maximisin_q -w--v --w The object of this program is to determine the allocation of available feed resources which maximises profit from cows of specified potential and body condition. Feed &sources are specified in terms of the types of pasture, hay, silage and concentrates available, constraints on their availability and their costs. Nutrient contents and edibilities of a wide range of feeds are given in the feed library, which can be edited or expanded,, Cows are considered in two herds (usually early, and mid to late lactation) and are specified in terms of breed, age, liveweight, condition score, potential peak milk yield, milk fat level, stage of lactation and pregnancy. Milk quotas for the farm are specified in terms of litres Per day, cents per litre and production required above the quotas. Feed intake is predicted from liveweight, potential milk production, stage of lactation, edibility of roughages and substitution effects of concentrates. Partition of feed nutrients is predicted from stage of lactation, condition score and breed. Energy requirements for maintenance are calculated according to Corbett (1987). Energy requirements for milk production are based on milk response curves derived from the data of Jensen et al. (1942). Energy contents of liveweight change are calculated from data on the chemical composition of cows with a wide range of body condition. Energy requirements for pregnancy are based on ARC (1980). Protein requirements are based on those of ARC (1980), with the addition of a requirement for metabolic faecal nitrogen and a reduction in efficiency of utilisation of absorbed amino acids. The net effect of these changes is that total protein requirements are more in accord with lactation trials and with NRC (1978). Mineral requirements -'are calculated according to NRC (1978). Linear programming techniques are used to calculate the ration which maximises profit, whilst satisfying nutrient requirements, quota (or production) requirements, and constraints on feed and animal resources. 153 Detailed technical information, including data equations used, is given by Hulme et al. (1986). 2. sources and Performance Prediction and - Ration - Analysis This program determines the milk production which is Feed possible from defined resources of feeds and cows. Nutritional intakes have to be known or estimated. requirements, substitution effects and nutrient partitioning are calculated in the same way as in the profit maximising This program is useful when first assessing feeding program. management practice on a dairy farm. 3. -m Feed Library Because there has been so little application of information on feeding dairy cows, there has been no pressure to determine the nutrient content of pastures, conserved forages and concentrates. There is very little information on the nutrient content of pastures and conserved forages grown in Australia and information on concentrates is based largely on overseas data. The feed library contains the nutrient analysis of a large range of feeds, which are classified as concentrates or roughages. In many cases interpolation of data were necessary The program allows you to fill gaps in information available. to change the data for a feed in the library, to replace a feed, to delete a feed or to add new feeds. CAMDAIRY The following CAMDAIRY. 1. Central w--Coast APPLICATION applications of examples illustrate in Summer Dairy Assumptions on the main resources and returns are as follows: cows: - - 100 Friesians with year-round calving Feed supply: kikuyu grass ($17/t as fed) and a commercial concexEte ($150/t as fed) Quota: 1000 litres/day @ 36 cents/l; surplus milk @ 14 -cents/l &estion: In order to maximise profit, --should be fed and to which cows? how much concentrate Information on the cows, feeds and quota are entered into the computer, and when edited to the user's satisfaction, appear as shown in Tables 1 and 2. 154 Screen display of herd Table 1 characteristics and quota details Screen CENTRAL THE COAST display DAIRY FEEDS of Table 2 nutrient content February, of feeds available 1987 RATION: FOLLOWING WERE AVAILABLE FOR THIS It was assumed that 30 of the 100 cows (identified as herd 1) would be in early lactation (average week of lactation 6), and the other 70 (identified as herd 2) would be mid-late lactation (average week of lactation 26). Cows were assumed to be in calf by an average of 12 weeks after calving. When the maximum-profit ration is formulated (this took 8 seconds on an Ariel AT computer), the following information is displayed: ration 155 Screen display of Table 3 predicted milk production, income and feed 156 Screen display of Table 5 mineral content MINERAL of ration components ESTIMATED ANALYSIS - DM BASIS Answer: --me- Results in Table 3 show that, with the resources available, profit is maximised when concentrates are fed at rates of 6.7 and 5.0 kg/day to cows in herds 1 and 2 respectively. It is profitable to produce 967 litres/day in excess of the quota of 1000 litres/day. The nutrient analysis and constraints in Table 4 show that crude protein was the limiting nutrient. It may be more profitable to supply protein from a protein concentrate than from the commercial concentrate; this possibility could be examined. As well as crude protein (cp), the model has the facility to run with rumen-degradable protein (RDP) and undeqraded dietary protein (UDP) or with CP and UDP. The mineral analyses in Table 5 indicate that magnesium, cobalt and zinc concentrations in the ration are lower than recommended. The accuracy of the ingredient analyses could be checked and mineral supplements made available if necessary. The effect of potential peak milk yield on both requirements for concentrates and gross profit is illustrated in Table 6. Table 6 157 2 - . Victorian - Dairy Farm -- Without Quota Assumptions on the main resources and returns are as follows: cows: --v 100 Friesian x Jersey; 500 kg liveweight Pastures: w--- Perennial ryegrass, late vegetative, costing $30/t DM White clover costing $30, $36 or $42/t DM; assumed edibility 26% higher than that of ryegrass Milk - - return: 14 cents/l &estion: What proportion of the pasture should be white clover inxrder to maximise profit? At the present time there is a lot of interest in the role of white clover in pastures for dairy cows in Victoria. A unique characteristic of clover is its high relative edibility. This feed property is taken into account when predicting intake in CAMDAIRY. Effects on gross profit of increasing the white clover component of the pasture are determined by the relative cost of producing ryegrass and white clover, as illustrated in Table 7. Table 7 Answer: Output from the model in Table 7 indicates that, on the basis of the assumptions made, gross profit decreases with increases with the amount of white clover in the pasture, when clover costs are 60% or more than ryegrass costs and cows are To the feed costs must be 20 weeks or later in lactation. added animal health costs associated with the use of clover, such as drugs used in bloat prevention and losses associated Without considering these with cows succumbing to bloat. costs, it appears that it would be useful to increase the percentage of white clover in the pasture, provided that costs of production are no more than 20-403 higher than those of producing perennial ryegrass. 158 3. B-- - Larqe Metropolitan s Dairy - w--- With - Quota ---and returns are as follows: Assumptions on resources cows: - - 1000 Friesian with year-round calving Feed su,: Limited grazing available - 10 ha/day of annual ryegrass, from which cows graze 2 tonnes dry matter/day; a contract supply of 6 tonnes/day brewers grain; a wide range of by-products and concentrates, with facilities for feed mixing. puota: 18,000 litres/day @ 36 cents/l; surplus milk @ 14 cents/l-The nutrient composition of the feeds available assumed to be as shown in Table 8. Table 8 Screen display of nutrient content of feeds available Question: In order to maximise profit, shzbe fed and how much of each? As in would be in lactation. an average which concentrates the first example, it was assumed that 30% of cows early lactation and 70% would be in mid-late Also, it was assumed that cows would be in calf by of 12 weeks after calving. When the maximum-profit ration is formulated (this ration took 32 seconds on an Ariel AT computer), the following information is displayed: 159 Screen display Table 9 of predicted milk production, allowances income and feed Screen display of Table 10 ration composition and opportunity prices 160 Answer: - - - Output from the model in Tables 9 and 10 gives the ration which maximises profit, together with opportunity costs. When the cost of feeds in the ration move outside the opportunity range, the ration should be re-formulated. When the cost of feeds not in the ration fall to or below the opportunity price, they would be included in the ration when re-formulated. In practice, the opportunity prices are a very effective bargaining tool in negotiations with feed suppliers. The model used cottonseed hulls as the cheapest source of fibre to meet the constraint of 170 g crude fibre/kg which is usually used (Table 4) to reduce the chance of low fat concentrations in milk. ACKNOWLEDGEMENTS Mr P.J. Booth contributed programming skills during early Dr L.R. Batterham gave valuable development of CAMDAIRY. guidance in linear programming techniques. M S A. Hodge, Dr C, Grainger and Dr A. McGowen of the Ellinbank Research Institute kindly supplied unpublished data on the chemical composition of cows # which is used in predicting partitioning of nutrients. Dr J. Moran of the, Kyabram Animal and Irrigation Research Institute kindly provided unpublished data on feed intake, which is used in predicting substitution effects. Dr H.J. Geddes drew our attention to the data of Jensen et al.(1942) which forms the basis of the milk response curves. Dr I. Lean provided valuable feedback from his application of the model in veterinary practice in the Hunter Valley. We are grateful to the Australian Dairy Research Committee, Veanavite Pty Ltd and the University of Sydney for financial assistance during the development of the CAMDAIRY model. REFERENCES Agricultural requirements Agricultural Agricultural Research Council (ARC)(1980). Nutrient of ruminant livestock. Technical review by an Research Council working party. Commonwealth Bureaux, Farnham Royal, Slough, 351 pp. Corbett,J.L.(1987). Energy requirements of the animal. In: Feeding standards for Australian livestock: Ruminants. 161 (Robards,G.E. & Radcliffe&IX. Eds),CSIRO,Melbourne/Standing Committee on Agriculture, Canberra, Dean,G.W, Carter,H.O,, Wagstaff,H.R., Olayide,S.O., Ronning,M. and Bath,D.L.(1972). Production functions and linear programming models for dairy cattle feeding, Giannini Foundation Monograph, 31. Hulme,D.J., Kellaway,R.C., Booth,P.J. and Bennett,L.(1986). The CAMDAIRY model for formulating and analysing dairy cow rations. Agricultural Systems 22, 81-108. Jensen,E., Klein,J.W., Rauchenstein,E., Woodward, T.E. Smith,R.H.(1942). Input-output relationships in milk production. USDA Technical Bulletin, 815. and MAFF(1975). Energy allowances and feeding systems for ruminants. Technical Bulletin 33. Her Majesty's Stationery Office, London. National Research Council (NRC)(1978). Nutrient requirements of domestic animals, No.3. Nutrient requirements of dairy cattle (4th edn), National Academy of Sciences, Washington, DC. 16 :2