Effects of feeding level and genetic merit on the efficiency of pasture-based dairy systems : field and modelling studies : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Animal Science at Massey University, Palmerston North, New Zealand
Loading...
Date
2011
DOI
Open Access Location
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Massey University
Rights
The Author
Abstract
The objective of this thesis was to develop and validate a dynamic and stochastic
whole-farm model that can predict physical and economic performance of pasture-based
dairy systems; explore interactions between cow genetic merit, feeding level
(supplementation and stocking rate) and market prices; and be applied to both ryegrassbased
and lucerne-based pasture dairy systems.
The effects of, and interactions between, stocking rate (SR), supplementation and
genetic merit of cows on grazing dairy systems were reviewed, approaching both the
physical and economic impact of these factors. The performance of strains of Holstein-
Friesian cows in experiments from Ireland, Australia and New Zealand was
summarised, and a meta-analysis that explores the relationship between herbage
allowance and herbage intake was included.
The development of the whole-farm model was completed in three steps. Firstly, a
model was developed and validated that predicts herbage intake at grazing for dairy
cows with or without supplementary feeding and that combines physical, metabolic and
ingestive constraints. Secondly, an animal model that predicts energy intake, milk yield
and live weight change of a single cow at grazing (e-Cow model) was developed and
validated. This model also integrates the above intake model, a mammary gland model
and a body lipid change model. The e-Cow model, which is available as a web-based
version, combines nutritional and genetic drives to control energy partitioning within
the cow. It also accounts for genetic differences between cows and is sensitive to
genotype by environment interactions. The third and final step was the development and
validation of a stochastic and dynamic whole-farm model that predicts physical and
economic performance of grazing dairy systems (e-Dairy model). Both the e-Cow and
the e-Dairy models simulate the performance of individual cows on a daily basis and
were developed using Visual Basic programming language.
The validation of the whole-farm model (e-Dairy) was conducted for ryegrassbased
dairy systems using existing data from a 3-year farmlet experiment comparing
five levels of SR (2.2 to 4.3 cows/ha) conducted in New Zealand, with cows offered
0.15 t dry matter (DM) supplement/cow/year. The validation of the model for lucernebased
dairy systems was performed with data from a 2-year farmlet experiment
designed and completed as part of the current research. This experiment compared three levels of SR (1.6 to 2.6 cows/ha) for cows offered 1.8 t DM supplements/cow/year in
Argentina. An indigestible intake marker developed from a purified enriched lignin
(LIPE®) was used to estimate individual herbage intake for a short-period within the
farmlet experiment.
Stochastic simulations (n=200) using the whole-farm model (e-Dairy) suggest that
for ryegrass-based New Zealand dairy systems (ratio $/kg milk to $/kg supplement of
1.1 ± 0.31), the increase in SR from 2.8 to 3.5 cows/ha together with an increase in
imported supplements from 0.15 to 1.45 t DM/cow/year can be profitable only when
milk price is higher than $NZ5.5/ kg MS ($US4.1). Simulations for lucerne-based dairy
systems in Argentina (ratio $/kg milk to $/kg supplement of 1.8 ± 0.55), suggest that the
increase in SR from 1.6 to 2.6 cows/ha, with a fixed amount of imported supplements
per cow at 1.8 t DM/cow/year would increase operating profit across the range of milk
prices tested ($US 3.3 ± 0.84/kg MS).
The e-Dairy model can be used to explore the effects and interactions of feeding
level and genetic merit of cows for grazing dairy systems with differing calving patterns
as well as evaluating the trade-offs between profit and the associated risk. It could also
potentially be used, after further development, to simulate the genetic evaluation of
cows and bulls under different selection objectives and selection schemes such as
progeny tests for bulls.
Description
Keywords
Dairy cattle, Feeding and feeds, Pasture, Computer simulation, Computer models, New Zealand