Copyright is owned by the Author of the thesis. Permission is given for
a copy to be downloaded by an individual for the purpose of research and
private study only. The thesis may not be reproduced elsewhere without
the permission of the Author.
irl. Da..i:ry Ca..ttle
A thes i s
presented in parti a l ful f i lment o f the requi rements for the degree of
Doctor of Phi losophy in Anima l Breeding at
Massey Univers ity
�ii9�
/I) (f .
ACKNOWLEDGEMENTS
I grateful ly acknowledge the guidance and support of my Chi e f Supervisor Professor Robert Anderson in my endeavours in mathematical and animal breeding statistics .
I would l ike to thank Professor Hugh B l a i r for the guidance of my work , h i s valuable comments on the manuscript and h i s friendly messages .
In particul ar , I would l ike to express my gratitude to Dr Brian Wickham of Livestock Improvement . His encouragement and support in many different ways made this study possibl e .
I would l ike to thank my supervisors Professor Al Rae and Dr Dori an Garrick for the ir help and advi se .
I am grateful to my former col leagues at Livestock Improvement , Dr Pat Shannon , Rob Jackson and Dr John Rende l for sharing thei r ideas and knowledge with me .
I would. l ike to thank Professor B i l l Hohenboken , Professor Leo Dernpfle and the late Dr Reuven Bar-Anan for the ir support and inspirations in matters animal breeding . I am indebted to Dr Bernhard Breier for h i s encouragement and practical advice and our stimulating academic discussions .
I am most grateful to Livestock Improvement to have provided me with this opportunity to further my education and contribute to dairy cattle breeding in New Zeal and .
I am indebted to my parents who encouraged me to search for answers and taught me not to give up .
Fina l l y , I would l ike to thank my husband Karl Matthys for h i s valuable comments on the manuscript of thi s thes i s anc h i s patience , encouragement and understanding during thE f inal phase of my studies .
i ABSTRACT
Economic merit of d a i ry cattle is an important consideration in d a i ry cattl e breeding . I n this thes i s , genetic and economic aspects were combined to quantitatively derive an aggregate genotype with the aim to improve the economic merit of New Zealand da i ry cattle .
Re l ative economic va lues for production traits , l i vewei ght and survival rate were estimated . A profit function was defined for a pasture based milk production system . The effects of changes in several genetic , biologica l and economic parameters on various components of farm profit were s imul ated with a bioeconomic computer model .
The economic value for protein yield was $ 4 . 6 4/kg and
$ 4 . 5 8/kg for Holstein-Fries i an and Jersey cows , fol lowed by
$ 1 . 8 4/kg and $ 1 . 7 5/kg , respectively , for mi lkfat yield . Livewei ght had a negative economic va lue of $ O . 4 9/kg and
$ O . 5 3/kg for Holstein-Friesian and Jersey cows . An increase in survival rate by 1 % had a pos itive economic va lue of
$ 9 . 2 5 and $ 9 . 2 9 for Holstein-Friesian and Jersey cows . A sens i tivity ana lysis showed that decl ining returns for milk reduced economic values for all milk components and s imultaneously increased economic values for l ivewe ight .
The effects of several traits on survival rate were quant i f i ed . An increase in surviva l rate increases a cow ' s prof i tabi l i ty as cows contribute to net farm income only during their l actation years and constitute a cost during the rearing period . After production traits , the f armer ' s overal l opinion of the cow and the traits udder support and udder overal l had the greatest impact on survival rate . These traits were used as selection criteria for survival rate in the s e l ection indexes .
i i variance and covariance components were estimated with
mul ti trai t REML procedures for a l l production , management and conformation traits used in the current selection index . Heri tabi l i ties were
0 . 2 5
for prote in yield and0 . 2 1
for mi lkfat yield for Holstein-Fries ians and 0 . 1 7 for both trai ts in Jerseys . Genera l l y , parameter estimates did not d i f f e r substantial ly from other popu l ations .An aggregate genotype as \'I1e l l as sel ection indexes for Hol stein-Fries ians and Jerseys were constructed and we ighting factors calcul ated based on the economic va lues and genetic parameters estimates .
The aggregate genotype included the traits protein yi e l d , milkfat yield , volume , l iveweight and survival rate .
The selection index for Hol stein-Friesians included thE trai ts prote in yield , m i l kfat yield , volume , I i veweight I
udder overal l , overa l l opinion , temperament , milking speec and udder support .
The s e l ection index for Jerseys included the traits proteir y i e l d , mi lkfat yield , volume , I i vewe ight , udder overa l l overal l opinion , temperament , mi lking speed and fore udder,
These s e l ection indexes improve the economical ly importan' trai t survival rate included in the aggregate genotyp' through indirect selection us ing tra i ts which can be measure, in the first l actation . This a l l ows the current generatio interval to be mainta ined when selection for survival rate i incorporated in the aggregate genotype .
Abstract
Table of Contents
L i st of Figures
L i st of Tables
1 . INTRODUCTION
TABLE OF CONTENTS
2 . SELECTION FOR MULTIPLE TRAITS - A REVIEW
2 . 1 . Methods o f selection for multiple traits 2 . 1 . 1 . Tandem sel ection
2 . 1 . 2 . Independent cul l ing leve l s 2 . 1 . 3 . Sel ection o f extremes
2 . 1 . 4 . Sel ection index methods 2 . 1 . 4 . 1 .
2 . 1 . 4 . 2 . 2 . 1 . 4 . 3 . 2 . 1 . 4 . 4 . 2 . 1 . 4 . 5 . 2 . 1 . 4 . 6 . 2 . 1 . 4 . 7 . 2 . 1 . 4 . 8 .
Optimum selecti on index
Best Linear Unbi ased Prediction Def inition o f aggregate genotype Multiple-stage selection
E f f iciency o f s e l ection index Estimated selection index Substitution i ndex
Indices based on phenotypes
i i i
Page i
i i i
vi i i
x
1 4
5 5 6 8 8 1 1 1 4 1 6 1 7 1 9 2 0 2 1 22 2 . 1 . 4 . 9 . Selection indices with constraints 24 2 . 1 . 4 . 1 0 . Alternatives to index selection 2E
2 . 2 . Estimation of relative economic values 2,
2 . 2 . 1 . Discount rates 2�
2 . 2 . 2 . Profit function - obj ective functi on 3:
2 . 2 . 3 . Methods o f estimating reI . economic values 3' 2 . 2 . 4 . Use of economic indices 4:
2 . 2 . 5 . Non- l inear economic values 4:
2 . 3 .
Effects of errors in parameters on expected genetic gain2 . 3 . 1 .
Errors in genetic parameters2 . 3 . 2 .
Errors in economic values2 . 4 .
The current breeding obj ective for New Zealandiv
4 4 4 4 4 8
dairy cattl e
4 8
2 . 4 . 1 .
Selection obj ective4 8
2 . 4 . 2 .
Current selection criteria 513 . A BIOECONOMIC MODEL
3 . 1 .
I ntroduction6 1
3 . 2 .
A bioeconomic mode l for a pasture based mi lkproducti on system
6 2
3 . 3 . Base l ine parameters and variables 6 3
3 . 3 . 1 .
Dairy farm production system parameters6 4 3 . 3 . 2 .
Metabo l i sable energy avai l abl e6 5 3 . 3 . 3 .
Uti l i zation/requirements o f metabo l i sableenergy
65
a ) Ma intenance
6 7
b ) Lactation
6 9
c ) Prenatal growth
7 4
d ) Postnatal growth and weight gain
75 3 . 3 . 4 .
Biological animal variablesa ) Mi l k production b ) Liveweight
7 6 7 6 7 7
c ) Replacement rate
7 8
d ) Age distripution of herd
7 8
da ) Estimation o f age distribution7 9
db ) Estimation of age-production factor8 2
dc ) Effect on energy requirement for mi l k8 3
dd ) Effect o n revenue from cul l cows8 4
e ) Surv ival rate o f cows8 5
f ) Survival rate of calves ( 0- 4
days )8 6
g ) Survival rate of repl aceme�ts( 4
days- 2
years )8 6
h ) Dress ing percentage
8 6
i ) Ratio of cal f weight at birth to cow
l ivewe ight 8 8
v
3 . 3 . 5 .
Economic variables8 9
a ) Price for m i l k fat , protein and volume
8 9
b ) Price for carcass of mature cow90
c ) Price for carcass of c a l f91
d ) Price of cal f sold for rearing92
e ) Var iabl e costsf ) Fixed costs g ) Discount rate 3 . 3 . 6 . Obj ective variabl es
3 . 4. Va l idation of bioeconomic model
3 . 4. 1 . Va l idation of conceptua l model 3 . 4 . 2 . Ver i f ication of computeri zed mode l 3 . 4. 3 . Operationa l va l i di ty
4 . RELATIVE ECONO MIC VALUES ESTI MATED WITH THE BIOECONO MIC MODEL
4. 1 .
Introduction4 . 2 .
Material and methods4. 3.
Milk production4. 3 . 1 .
Protein4. 3 . 2 .
Mi lkfat4. 3 . 3 .
Lactose4. 3 . 4.
Volume4 . 4. Liveweight
4. 4 . 1 .
Effects of l iveweight changes4. 4. 2 .
Economic value of l iveweight4. 5 .
Replacement rate4. 6 .
Compari son of economic va lues5 . SENSITIVITY ANALYSIS 5 . 1 .
Economic variables5 . 1 . 1 .
Returns for milk components92 96 99 1 0 1
1 0 1 1 0 2 1 0 2 1 0 3
1 0 7 1 0 7 1 0 8
III III
1 1 8 1 2 6 1 3 2 1 3 3 1 3 3 1 3 5 1 41 1 45
1 4S 1 5 C 1 5 C
a ) Protein b ) Mi lkfat c ) Lactose d ) Volume
5 . 1 . 2 .
Returns for carcass of mature cow5 . 1 . 3 .
Returns for carcass of ca l f5 . 1 . 4 .
Costs o f inputs 5 . 2 . Biologica l vari ables5 . 2 . 1 .
Metabol isable energy ava i l able5 . 2 . 2 .
Biological animal parameters5 . 3 .
Compari son of variables for sensitivity impact6 . EFFECTS OF TRAITS OTHER THAN PRODUCTION ON S URVIVAL RATES
6 . 1 .
I ntroduction6 . 2 . Material and methods
6 . 3 .
Results6 . 4 .
Conclus ion7 . AN AGGREGATE GENOTYPE AND SELECTION INDE XES FOR NEW ZEALAND DAIRY CATTLE·
EST I MATION OF GENETIC PARAMETERS FOR THEIR IMPLEMENTATION
vi
1 5 3 1 5 8
1 6 2
1 6 4 1 6 6 1 6 8 1 7 0 1 7 2 1 7 31 7 5
1 8 01 8 2 1 8 2
1 8 2 1 8 4 1 94
1 95
7 . 1 .
An aggregate genotype for New Zealand dairy cattl e1 95 7 . 2 .
Selection indexes for New Zeal and dairy cattle 1977 . 2 . 1 .
Hol stein-Fries ians1 97
7 . 2 . 2 .
Jerseys1 9E
7 . 3 .
Genetic parameters for selecti on obj ectives and selection criteria7 . 3 . 1 .
Introduction7 . 3 . 2 .
Material and methods7 . 3 . 3 .
Results and di scuss ion1 9�
1 9�
2 0 : 2 0 (
a ) Heritabi l ities
b) Phenotypic corre lations c ) Genetic corre lations 7 . 4 . Conclus ion
8.
SELECTION INDE XES 8 . 1 .
Introduction8 . 2 . Selection index equations
8 . 3 . Results and di scuss ion 8 . 4 . Conc lusion
9. CONCLUSION
1 0 . APPENDI X
1 0 . 1 .
APL-code of bioeconomic model1 0 . 2 . APL-code for calculating s e l ection index
v i i
2 0 8 2 1 2 2 1 4 2 1 9
2 2 0 2 2 0 2 2 0
2 2 8 2 2 9
2 3 0
2 3 4 2 3 4
coefficients 2 4 0
1 0 . 3 . Tables o f genetic parameters
for Hol stein-Fr i e s i ans , Jerseys and Ayrshires
2 4 2
10 . 4 . Publ ications 2 5 0
11 . PUBLICATIONS ARISING FROM THESI S 2 8 9
1 2 . BIBLIOGRAPHY 2 91
vi i i LIST OF F IG URES
Figure
Page
3 . 1 . Flowchart of bioecono mic model .
6 3 a4. 1 . Revenue per ha f ro m mi lk , sum of mi lk and bobby ca lves ,
sum of mi lk , bobby calves and cul l cows and pro f i t per ha def ined as sum of revenue from mi lk , bobby ca lves and cul l cows minus variabl e costs at dif ferent leve l s of protein yield per cow for Holstei n -Friesians and
Jerseys . 1 1 6
4. 2 . Revenue per ha from mi lk , sum o f mi lk and bobby calves , sum of mi lk , bobby calves and cul l cows and profit per ha defined as sum of revenue from mi lk , bobby ca lves and cul l cows minus variabl e costs at di f ferent leve l s o f mi lkfat yi eld per cow for Holstein-Friesians and
Jerseys .
1 2 44 . 3 . Revenue per ha from mi lk , sum o f mi lk and bobby calves , sum of mi lk , bobby ca lves and cul l cows and pro f i t per ha def ined as sum of revenue from mi lk , bobby calves and cul l cows minus variable costs at di fferent leve l s of lactose yield per cow for Holstein-Fries ians and
Jerseys . 1 3 1
4. 4. Revenue per ha from mi lk , sum o f mi lk and bobby calves ,
sum of mi lk , bobby calves and cul l cows and pro f i t per
ha defined as the sum of revenue fro m mi lk , bobby ca lves
and cul l cows minus variable costs at different level s
of l iveweight f o r Hol ste in -Fries ians and Jerseys . 1 40
4.5 . Revenue per ha from mi l k , sum of mi lk and bobby calves ,
sum of mi lk , bobby calves and cul l cows and pro f i t per
ha de fined as the sum of revenue fro m mi lk , bobby calves
and cul l cows minus variabl e costs at di f ferent levels
of repl ace ment rate def ined as the reciprocal of
herdl ife for Hol stein-Friesians and Jerseys . 1 44
ix
5 . 1 . Real pri ces for mi lkfat and meat fro m cul l cows and bobby calves during the l ast 2 0 years . 1 5 1 5 . 2 . Changes in econo mic val ues for protein , mi lkfat , lactose and l iveweight for Hol s tein-Friesians and Jerseys with an increasing price for protein . 1 5 5 5 . 3 . Changes i n economic values for protein , mi lkfat , lactose and l iveweight for Ho lstein-Friesians and Jerseys with an increasing price for mi lkfat . 1 6 1 5 . 4. Changes in economic va lues for protein , mi lkfat , lactose and l iveweight for Holstein-Friesians and Jerseys due to
changes in volume penal ty . 1 6 5
x
LIST OF TABLESTable Page
Chapter 3
3 . 1 . Dairy farm production syste m para meters e mployed in the
bioeconomic model . 6 4
3 . 2 . Esti mates for ME re quirement for mi lk co mponents . 7 4 3 . 2 . Milk production statistics for 1 9 9 0 / 9 1 season for
Holstein-Fries ians and Jerseys .
7 73 . 3 . Statistical parameters for production per cow by age for
Hol stein-Fries ians and Jerseys . 8 0
3 . 4. Proportion of cows i n each age group for Holstein-
Fries ian and Jersey. 8 1
3 . 5 . Age-production factors for Holstein-Fries ian and Jersey
. � . ..,
.for d i fferent replace ment rates .
8 4
3 . 6 . Expected dressing percentages for Holstein-Friesian and
Jersey cows . 8 8
3 . 7 . Variable costs per cow.
9 :::3 . 8 . Total mi l ki ng t i me esti mates for Holstein-Fries ian anc Jersey cows .
3 . 9 . Co mpari son of production para meters derived fro m a l experi ment and the bioeconomic model f or Hol stein·
Friesians and Jerseys . 1 0
Chapter 4
4. 1 . Phenotypi c and geneti c para meters employed for Holste i n
Friesian s and Jerseys . 1 1
xi 4 . 2 . Changes in energy requirements , revenue and cost per cow and per ha for Hol stein-Fries i ans and Jers e ys for
,different levels of protein yi eld . 1 1 3 4 . 3 . Production para meters , profit per ha and econo mi c values at di f ferent levels of protein yield per cow for Holstein-Friesian and Jersey cows . 1 1 7 4 . 4 . Changes in energy re quirements , revenue and costs per cow and per ha for Hol stein-Friesians and Jers e ys with dif ferent l evels of mi lkfat yield . 1 2 0 4 . 5 . Production para meters , profit per ha and econo mic values at d i fferent leve l s of mi lkfat yi eld per cow for Holstein-Friesians and Jerseys .
1 2 54 . 6 . Changes in energy requirements , revenue and
cow and per ha for Hol stein-Friesian with l evels of l actose yield .
costs per d i fferent 1 2 8 4 . 7 . Production parameters , profit per ha and economic values at di f ferent level s of lactose yield per cow for
Holstein-Friesians and Jerseys .
1 3 24 . 8 . Changes i n energy re quirements , revenue and costs per cow and per ha for Hol stein-Fries ians and Jerse ys for d i fferent l evels of l iveweight . 1 3 7 4 . 9 . Production parameters , profit per ha and econo mic values at d i fferent l evels of l iveweight for Hol stein-Friesians
4 . 1 0 .
4 . 1 1 .
4 . 1 2 .
and Jerseys . 1 4 1
Production parameters , profit per h a and economi c values at different leve l s of replace ment rate E for Holstein-Friesians and Jerse ys .
Comparison o f decrease i n stocking rates pe l phenotypic standard deviation i ncrease in variou�
traits for Hol stein-Friesians and Jers e ys . 1 4 1 Economic values per cow for Holstein-Friesians an
Jerseys . 1 4
xi i
Chapter 5
5 . 1 . Prices for mi lkfat , cul l cow carcass and bobby calves . 1 5 2 5 . 2. Changes in pro fit per ha and per cow with increas ing protein prices for Hol stein -Friesians and Jerse ys . 1 5 7 5 . 3 . Changes in prof i t per ha and per cow and econo mic values for mi lk prices predicted for 2 0 0 1 and 2 0 1 1 . 1 5 8 5 . 4. Changes i n prof i t per ha and per cow resulting from increas ing mi l kfat prices for Hol stein-Fries ians and
Jerse ys .
1 6 25 . 5 . Changes in prof i t per ha and per cow due to an increased value for lactose for Holstein-Friesian and Jersey. 1 6 4 5 . 6 . Changes i n prof i t per ha and per cow with increased
costs for vol u me for Hol stein-Fries ian and Jersey. 1 6 6 5 . 7 . Changes i n prof i t per ha and per cow resulting from changes in returns for cow meat for Holstein -Fries ians
and Jerseys . 1 6 8
5 . 8 . Changes i n prof i t per ha and per cow resulting from changes in returns for ca l f meat for Holstein-Friesians
and Jerseys .
1 6 95 . 9 . Changes i n prof it per cow and per ha with a 1 0 % increase in variabl e costs , a 1 0 % increase in stock prices and a 2 % i ncrease i n interest rate for Holstein-Fries ians and
5 . 1 0 .
5 . 1 1 .
Jers e ys .
1 71Changes i n econo mic values per ha due to changes in
OMgrown , uti l i sation and energy content o f
OM . 1 7 4
Changes in stocking rate , profit per ha and economic val ues per ha resulting f ro m a
1 0%i ncrease in biological
variablesfor Holste i n -
Fr ies ians and Jers e ys . 1 7 8
x i i i Chapter 6
6 . 1 .
Number of cows with production , management and survival information at d i f ferent lactations . 1 8 36 . 2 .
Means and standard deviations o f production and nonproduction tra its of primiparous cows imcluded in the
analys i s .
1 8 5
6 . 3 .
Means and standard devi ations for survival rates betweensubsequent l actations . 1 8 7
6 . 4.
Part i a l regress ion coefficients for traits affecting survival rate f rom1 nt
to2 �
lactation .1 8 8 6 . 5 .
Standardi sed partial regression coefficients for traits affecting survival rate from1 rst
to2 nd
lactation .1 9 0 6 . 6 .
Part i a l regression coe f f icients for tra its affecting survival rate fromrst
to3 rd
lactation .1 9 1 6 . 7 .
Part i a l regress i on coefficients for traits affectin<?survival rate f rom
1 ut
to4�
l actation . 1 9 3Chapter 7
7 . 1 .
Number of records for estimating vari ance and covarianc(components for producti on and non-producti on traits . 2 0:
7 . 2 .
Heritabil ities of se lection obj ectives in the aggregat genotype and selection criteria in the selectioindexes . 2 0
7 . 3 . Phenotypic correlations between selection criteri
included i n the selection indexes . 2 1
7 . 4.
Genetic corre lations between traits in aggregatgenotype and criteria in selection index . 2 :
xiv Chapter 8
8 . 1 . Phenotypic var i ances and covari ances used for
determining we ighting factors for selection indexes . 2 2 4
8 . 2 .
Genetic vari ances and covariances used i n determiningweighting factors for se lection indexes .
2 2 6 8 . 3 .
Weighting factors ( $/cow/annum )i ncluded i n the selection Friesians ) and 1J ( Jerseys ) .
for selection criteria indexes 1M ( Holste in-
2 2 8
Chapter 1 0 - Appendix
1 0 . 1 .
1 0 .2 .
1 0 . 3 .
Adj usted means , phenotypic and addi ti ve genetic vari ances and heri tabi l i ty estimates with thei r standard deviations and standard errors for producti on and non-production traits for primiparous Holstein-Fries i an , Jersey and Ayrshire
cows .
2 42
Heritab i l i ty and phenotypic and genetic correlation estimates for primiparous Holste in
Fries i an , Jersey and Ayrsh i re cows .
2 45
standard errors for heritabi l i ty and phenotypic and genetic correlation estimates for primiparous Holstein-Fri e s i an , Jersey and Ayrshire cows .2 48