Work package 1

Work package 1. Optimum feeding strategies for dairy cows incorporating grazed grass with AM for various production systems in Europe


  • Establish the optimum proportion of grazed grass in the cows’ diet, for different production systems involving AM systems
  • Develop ‘best practice’ feeding strategies to meet variations in grass supply and quality, especially during periods of grass inadequacy
  • Define impact of nutritional inputs, milking frequency, number of cows per unit and stage of lactation on milk output from AM systems
  • Evaluate the influence of cow breed/ type on cow performance of AM systems in different production systems in EU

WP Leader: SLU, Dr. Eva Spörndly

Description of work

This Work Package (WP) focuses on challenges in feeding strategies when combining cow grazing with AM technology: seasonal variation in pasture quality and quantity, varying need of supplementary feeding and seasonal or staggered calving patterns. Controlled trials on research farms will form the base activities in this WP. Data from ‘Monitor Farms’ (described in WP 3) in Denmark will also be used.

The research performed in this current WP will target a wide range of production conditions representative of different consortium countries, e.g. from extensive systems using high proportions of pasture in the diet to intensive systems using relatively low proportions of pasture and high levels of supplements. Data on different levels of buffer feeding/supplementation on the ‘Monitor Farms’ will also be collated and analysed. The specific effect of and interaction of certain important parameters, such as level of production, pasture allowance, level of buffer feeding/supplementation and cow breed will be in focus.


Task 1.1 Maximizing milk output from integrated grazing and AM production systems where grazed grass forms different proportions of cow diet (M2-M28)

Leader: SLU, Eva Spörndly

Involved partners: TEAGASC, SLU

Two milk production systems, both incorporating AM and defined by ‘high’ and ‘low’ pasture proportions in the cow diet will be set up and compared on research farms in two countries, Ireland and Sweden. The experiments will be performed in research herds of a standard AM system size, 55-60 cow DeLaval VMS AM system in Sweden and a 70-80 cow Fullwood Merlin 225 AM system in Ireland.

The ‘high’ and ‘low’ diet in each country will be influenced by country specific practical norms, thus ranging from pasture/supplement ratios of 90/10 (high) to 75/25 (low) in Ireland and 25/75 (high) to 10/90 (low) in Sweden. Feeding in Sweden will be based on pasture and grass silage in different proportions with concentrates for high yielders; in Ireland it will be based on pasture and concentrates.

The experimental treatments will be implemented during the ‘standard’ pasture season in each country, i.e. 8 and 3 months in Ireland and Sweden, respectively. The following factors will be registered and analyzed in relation to the ‘high’ and ‘low’ pasture AM systems: milk production, milk composition, milk quality, feed intake, milking frequency, number of cows per unit and frequency and timing of passages between barn and pasture. The pasture proportion most suited with AM under the specific production conditions in two countries will be identified. Furthermore, discussion and analysis of the results in relation to the pasture and AM systems and prevailing conditions for pasture production in other partner countries will be performed. The data obtained in the experiments will be used in the sustainability and economic assessments in WPs 3 and 4, respectively.

Task 1.2 Develop ‘best practice’ feeding strategies to meet variations in grass supply and quality, especially during periods of grass inadequacy (M2-M28)

Leader: IDELE, Valerie Brocard


Periods of grass inadequacy regularly occur in regions of EU due to seasonal and climatic variation, e.g. spring, autumn or summer drought periods. Variations in seasonal grass production can also result in variation in grass quality. These variations represent a serious problem that can jeopardize milk production and is often mentioned by farmers as one of the major obstacles in incorporating grazing with AM systems. Various supplementation strategies will be tested and compared at research centres in France, Ireland and Belgium. Specifically, low and high levels of supplementation with maize silage will be compared during transition periods (between pasture and indoor feeding) in France (Derval Experimental farm with fixed AM robot, 75 Holstein cows and Trevarez Experimental farm with mobile AM robot, 45 cows and organic milk) and Belgium (mobile AM robot at Liège University). Different feeding strategies will be evaluated which will include level of grass and supplementation allocated, time of feed allocation, type of supplementation (forage or concentrates) and sequence of feeding strategy, e.g. robot-cubicles-water-buffer feed-pasture. The influence of level of supplementary feeding on cow-traffic will be evaluated in Ireland and in Belgium. Furthermore, supplementation strategies practiced on ‘Monitor Farms’ and on the SME end-user Farm (SME FARM DK) will also be monitored and analysed in Denmark. Results will be evaluated in terms of milk output and cow well-being.

Task 1.3 Optimum cow breed/ type for an integrated grazing and AM milk production system

Leader: TEAGASC, Bernadette O’Brien

Involved partners: TEAGASC, SLU

Cow breed and type could have a large influence on the performance of integrated grazing and AM production systems. Cow breeds and types differ in terms of milk production potential, udder conformation and suitability to grass-based systems. Specific cow breeds may interact differently under management that combines AM with grazing or may be more suited to either high or low feed supplementation systems and therefore, indirectly affect milk output. Comparisons between Holstein Friesian and Swedish Red (SLU) will be carried out in Sweden; while in Ireland (TEAGASC) the suitability of Holstein Friesian, Norwegian Red and Jersey x Friesian cow breeds will be compared on research farms. Superimposed on each genotype in each country will be ‘high ‘and ‘low’ input systems as described in Task 1.1. Under management with different proportions of pasture in the diet, this task will analyse factors such as milking frequency, distribution of milking and grazing behaviour, number of visits to pasture area and time at pasture in relation to milk yield and feed intake. Furthermore, within the project the performance and behaviour of Holstein Friesian cows in Ireland will be compared to that of Holstein Friesian cows in Sweden.