Final Report Summary - GREENHOUSEMILK (Developing Genetic Tools to Mitigate the Environmental Impact of Dairy Systems)
The EU25 produce approximately 132 million tonnes of milk from 24.3 million cows on 1.76 million farm holdings. Ruminant animals account for up to 20% of the world methane (CH4) production with the EU25 dairy population producing approximately 3.2 million tonnes of CH4 per year. Many EU countries have specific and binding commitments relating to the reduction of greenhouse gas (GHG) emissions, and all sectors of the economy are coming under increasing scrutiny in relation to their share in the overall emissions target. Little work has been done on the role of dairy cow genetics in dairy system emissions, particularly considering the role of genetics in the whole farming system, including feeding strategy and management policies (e.g. energy balance, housing periods, fertilisation and manure management). GREENHOUSEMILK helped us understand the role of energy efficiency and partitioning in the overall GHG output of dairy systems. Innovative tools were developed in GREENHOUSEMILK to help farmers’ select “environmentally friendly” bulls to suit their system and how to manage those bulls’ daughters in an appropriate manner. It harnessed statistical and genetic tools to elucidate the genetics of emissions in dairy cattle, and developed innovative and integrative tools that address the environmental impact of dairy farming, thus underpinning a high priority policy area.
Six ESRs were recruited in GREENHOUSEMILK to examine causes of variation in GHG emissions in dairy cows, genomic tools to help select for reduced GHG emissions, and the integration of animal GHG emissions into farm systems models.
ESR1 first built a database of published experiments that allowed calculation of energy and nitrogen efficiency according to differences in feed types and feeding regimes. The information in this database provided a good basic understanding of the components of efficiency, and was also used to carry out a meta-analysis of feed factors affecting energy and nitrogen efficiency. The next step was to develop a nutrient partition model that is capable of predicting efficiency across a whole lifetime. This work has built on an existing model, extending it to deal with the key factors affecting efficiency. ESR1 provided a tool to allow us to better match the selection goals for dairy livestock to a diverse range of environments that they may be confronted with, and through this achieve a better optimisation between production and GHG emissions. ESR2 compiled performance data from herds with CH4 measures, calibrated a prediction model and studied the phenotypic and genetic variability of CH4 indicators predicted by Mid-infrared spectroscopy (MIR) on Belgian and Irish data. ESR2 contributed to the development of individual CH4 prediction equation for dairy cows from milk MIR spectra. The biggest advantage of these prediction equations is that it is a trait that can relatively easy be made available for a national population. Currently, genetic selection for mitigated GHG emissions in dairy system is not a common practice, due to lack of data. The work of ESR2 helped addressing genetic selection as a mitigation tool. ESR3 mathematically defined feed efficiency based on body reserves usage and adaptation. Feed efficiency in dairy cows is extremely complex and the chosen approach has never been undertaken before. It applied a mixed model framework to estimate animal specific energy costs of different energy sinks. Two main outcomes from ESR3 will contribute to a global initiative to improve feed efficiency; (1) a novel and biologically more relevant method to model feed efficiency in growing animals, and (2) that heritable differences among animals exist in regression coefficients on the energy sinks which can be exploited to improve feed efficiency in a population through genetic selection. ESR4 focussed on optimal ways to use small sized reference populations, for difficult to measure traits such as CH4 emission, feed efficiency or energy partitioning. Reference populations are needed to allow estimating breeding values for important traits, thereby enabling improvement of those traits through selection. The main conclusion is that for traits such as CH4 emission, the optimal selection strategy is genomic selection based on reference populations consisting of cows on which CH4 emission is measured. The knowledge generated by ESR4 helped enabling implementation of genomic selection for CH4 emission and similar traits, while selection for such traits was impossible before. Uptake of this knowledge is already taking place, and recently many national and international research initiatives are investigating the possibilities to implement genomic selection for novel traits such as CH4 emission or feed efficiency. ESR5 developed a cattle model based on: (1) age, breed, body weight, genetics of the animal for the calf, heifer and cow; (2) the day in gestation and the number of inseminations (since the last calving for the cows) for the heifer and cow; and (3) the day in lactation and the body condition score (BCS) and BCS at calving for the cow. The model is capable of simulating the effects of different managements across farms. The model has been tested against real data, and in all situations the root mean square error was lower than 15% which suggests that the model is robust across feeding system and animal. The model allowed the development of the optimised mitigation strategies for grass based systems of milk production. This will help Irish (and European) agriculture to mitigate GHG emissions and will thus help to achieve the 50% increase in milk output and thus will result in increased employment and revenues in the Irish economy. ESR6 developed Marginal Abatement Cost Curves (MACCs) specific for the EU-15 dairy sector to present the cost-effectiveness of different management practices in abating GHG emissions. These MACCs can be used as a decision tool for policy makers and farmers to identify the suite of most cost effective mitigation measures at farm level or for the whole sector. ESR6 identified that GHG emission from the dairy sector can be reduced by 12% in 2020 (~15 Mt CO2e) with five win-win measures “Selection for higher yield”, “Feeding probiotics”, “Balanced fertilisation”, “Reduced fertilisation during the winter period” and “Reduced tillage”.
Overall, the outcome of GREENHOUSEMILK shows that it is possible to introduce new mitigation strategies, based on a better understanding of energy efficiency and partitioning, genetic selection and herd dynamic models that are either implemented at low costs or win-win mitigation measures, where the latter one increases the farmers’ income. All in all, this promises great perspectives for the livestock sector to feed the growing world population within the carrying capacity of planet earth.
A description of the project can be found at www.sac.ac.uk/greenhousemilk
Six ESRs were recruited in GREENHOUSEMILK to examine causes of variation in GHG emissions in dairy cows, genomic tools to help select for reduced GHG emissions, and the integration of animal GHG emissions into farm systems models.
ESR1 first built a database of published experiments that allowed calculation of energy and nitrogen efficiency according to differences in feed types and feeding regimes. The information in this database provided a good basic understanding of the components of efficiency, and was also used to carry out a meta-analysis of feed factors affecting energy and nitrogen efficiency. The next step was to develop a nutrient partition model that is capable of predicting efficiency across a whole lifetime. This work has built on an existing model, extending it to deal with the key factors affecting efficiency. ESR1 provided a tool to allow us to better match the selection goals for dairy livestock to a diverse range of environments that they may be confronted with, and through this achieve a better optimisation between production and GHG emissions. ESR2 compiled performance data from herds with CH4 measures, calibrated a prediction model and studied the phenotypic and genetic variability of CH4 indicators predicted by Mid-infrared spectroscopy (MIR) on Belgian and Irish data. ESR2 contributed to the development of individual CH4 prediction equation for dairy cows from milk MIR spectra. The biggest advantage of these prediction equations is that it is a trait that can relatively easy be made available for a national population. Currently, genetic selection for mitigated GHG emissions in dairy system is not a common practice, due to lack of data. The work of ESR2 helped addressing genetic selection as a mitigation tool. ESR3 mathematically defined feed efficiency based on body reserves usage and adaptation. Feed efficiency in dairy cows is extremely complex and the chosen approach has never been undertaken before. It applied a mixed model framework to estimate animal specific energy costs of different energy sinks. Two main outcomes from ESR3 will contribute to a global initiative to improve feed efficiency; (1) a novel and biologically more relevant method to model feed efficiency in growing animals, and (2) that heritable differences among animals exist in regression coefficients on the energy sinks which can be exploited to improve feed efficiency in a population through genetic selection. ESR4 focussed on optimal ways to use small sized reference populations, for difficult to measure traits such as CH4 emission, feed efficiency or energy partitioning. Reference populations are needed to allow estimating breeding values for important traits, thereby enabling improvement of those traits through selection. The main conclusion is that for traits such as CH4 emission, the optimal selection strategy is genomic selection based on reference populations consisting of cows on which CH4 emission is measured. The knowledge generated by ESR4 helped enabling implementation of genomic selection for CH4 emission and similar traits, while selection for such traits was impossible before. Uptake of this knowledge is already taking place, and recently many national and international research initiatives are investigating the possibilities to implement genomic selection for novel traits such as CH4 emission or feed efficiency. ESR5 developed a cattle model based on: (1) age, breed, body weight, genetics of the animal for the calf, heifer and cow; (2) the day in gestation and the number of inseminations (since the last calving for the cows) for the heifer and cow; and (3) the day in lactation and the body condition score (BCS) and BCS at calving for the cow. The model is capable of simulating the effects of different managements across farms. The model has been tested against real data, and in all situations the root mean square error was lower than 15% which suggests that the model is robust across feeding system and animal. The model allowed the development of the optimised mitigation strategies for grass based systems of milk production. This will help Irish (and European) agriculture to mitigate GHG emissions and will thus help to achieve the 50% increase in milk output and thus will result in increased employment and revenues in the Irish economy. ESR6 developed Marginal Abatement Cost Curves (MACCs) specific for the EU-15 dairy sector to present the cost-effectiveness of different management practices in abating GHG emissions. These MACCs can be used as a decision tool for policy makers and farmers to identify the suite of most cost effective mitigation measures at farm level or for the whole sector. ESR6 identified that GHG emission from the dairy sector can be reduced by 12% in 2020 (~15 Mt CO2e) with five win-win measures “Selection for higher yield”, “Feeding probiotics”, “Balanced fertilisation”, “Reduced fertilisation during the winter period” and “Reduced tillage”.
Overall, the outcome of GREENHOUSEMILK shows that it is possible to introduce new mitigation strategies, based on a better understanding of energy efficiency and partitioning, genetic selection and herd dynamic models that are either implemented at low costs or win-win mitigation measures, where the latter one increases the farmers’ income. All in all, this promises great perspectives for the livestock sector to feed the growing world population within the carrying capacity of planet earth.
A description of the project can be found at www.sac.ac.uk/greenhousemilk