Ecological total merit index for an Austrian dual purpose cattle breed

Organic farming plays a considerable role in Austrian agriculture. Therefore it seems to be worthwhile to make suggestions about a more ecological breeding goal for cattle breeds. Because ofthe fact that Simmental is the most frequent dual purpose cattle breed in Austria, this breed is chosen to show the impact of selection under an ecological total merit index. With model calculations the efTect on selection response in milk production beef production and functional traits under selection with the current economic total merit index and more ecological index variants, e.g. with higher economic values for fitness and ftinctional traits, is shown. As a basis for future decisions the efficiency of more or less ecological indices is compared. The results indicate that an increase of the current economic values of functional traits of about 50 percent does not present a great risk, expected selection responses for milk production traits are still high.


•
Introduction In breeding objectives traits should be included aecording to their economic importance.MIESENBERGER (1997) estimated such economic weights for production traits and functional traits using a herd model for Austrian cattle breeds.The economic weights of several functional traits (e.g.longevity) turned out to be high (MIESENBERGER et al., 1998).An improvement in one ofthose traits increases the efficiency of an animal not by higher Output of products but mainly by reduced input of costs (GROEN et al, 1996;MIESENBERGER, 1997).Some authors (DEMPFLE 1992;GROEN et al.,1996) point out that in addition to economic reasons for including functional traits in the breeding programmes there are several other reasons, for example ethical reasons and consumer concern, which become more and more important.For this reason and because of the fact that organic farming plays a considerable role in Austrian agriculture, it seems worthwhile to make suggestions about a more ecological breeding goal for cattle breeds.In the following "ecology" is defined in a broad sense.The ecological system includes not only the manifold interactions between organisms and their environment but also physiological relationships within the organism itself (CAPRA, 1997).Therefore an ecological total merit index has to deal with characters which are indicators for the inner balance of an organism like health and fertility traits.This study shows different possibilities how to create a more ecological orientated total merit index for the dual purpose Simmental population, which is the most frequent cattle breed in Austria.Further, the effect on the expected selection response in milk production, beef production and functional traits under selection with the current economic total merit index and more ecological index variants, e.g. with higher economic weights ofthe functional traits, is shown.The efficiency of different indices is compared to get a basis for future decisions.

2.
Material and methods To estimate the economic weights and calculate a total merit index a herd model and methods described by MIESENBERGER et al. (1998) and MIESENBERGER (1997) were used.

2.1
Estimation of the economic weights A herd including milk production, bull fattening and heifer rearing was simulated in a steady State over an infinite planning term.A Computer program developed by AMER et al. (1994), which was adapted by MIESENBERGER et al. (1998) was used to compute the economic weights for the various traits by calculating herd profit per year before and after a genetic change.The resulting difference in herd profit was divided by the number of cows per herd.All costs were treated as variable.For a more detaiied description see MIESENBERGER et al. (1998).All economic weights presented in this study were calculated under the assumption of a quota on fat and protein yield.A reference Situation was defined with respect to the Simmental population in Austria.The average milk yield per cow place and year was 5321 kg milk, 223 kg fat and 180 kg protein.The age structure ofthe herd modelled approximated the Situation in the present Austrian Simmental population (see MIESENBERGER et al., 1998).Prices and costs in the basis Situation were average prices in Austria in 1996.Some ofthe prices considered are given in Table 1 (scenario I)-

2.2
Modelling of the breeding programme The program ZPLAN (KARRAS et al., 1994) was used to describe the Austrian Simmental breeding programme.For predicting the annual genetic gain a pure deter-ministic approach is employed.One round of selection is considered with its impact on a given time horizon with specific discount rates.All selection groups in the whole population are to be defined, each with a specific selection intensity and with particular sources of information (type and number of relatives) for index selection.The ZPLAN method applies the gene flow method described by MC CLINTOCK and CUNNINGHAM (1974), HILL (1974) and ELSEN and MOCQUOT (1974) to calculate a number of criteria such as annual monetary genetic gain for the aggregate genotype, annual genetic gain for single traits, discounted return and discounted profit over the given investment period.These criteria allow the evaluation ofthe genetic and economic efficiency of breeding programmes.Using this program different genetic improvement schemes have been evaluated by GRASER et al (1994) RIEDL (1996) MUELLER (1995), KOMINAKIS et al. (1997), WÜNSCH et 'al. (1999) and SÖLKNER et al. (2000).In the following some of the most important imput parameters for the Austrian Simmental population are given.For a more detaiied description see SÖLKNER et al. (2000).

2.3
Development of an ecological total merit index There are several possibilities to alter an index in direction of a more ecological breeding goal.Different traits can be included in the total merit index and the economic weights of these traits can be changed.Both possibilities are discussed subsequently.

2.3.1
Choice of traits Which traits should be taken into account in an ecological total merit index has to be clarified when the breeding goal is defined.In case of a meaningful ecological breeding goal one supposition should be to include traits for health and fertility into the index.MIESENBERGER et al. (1998) showed that functional traits should be included in the aggregate genotype also from an economic point of view.Selection for such a total merit index will result in a higher economic efficiency than selection for production traits only (see also SÖLKNER et al., 2000).In the total merit index in Austria the following traits are incorporated: fat and protein yields, daily gain, dressing percentage, EUROP grading score, longevity, persistency, paternal and maternal fertility, paternal and maternal calving ease, paternal and maternal stillbirth and somatic cell count.Fat and protein yields are combined in the subindex "milk value", daily gain, carcass percentage and EUROP grading score give the "beef value" and the combination of all other traits, the functional traits, results in the "fitness value".Because of the consideration of the functional traits the current economic total merit index in Austria can be judged positively with regard to an ecological orientated choice of traits.The scenarios considered subsequently include only traits currently used in the total merit index in Austria.Feed intake capacity and efficiency of feed utilisation are not included in these indices although improvement of such traits may lead to a reduction of environmental costs , e.g.due to less nitrogen secretion per unit of product.GROEN et al. (1996) point out that feed intake is a very complex trait, which can not be treated separately but should always be considered in relation to milk production and body weight.The question whether a reduction or an increase in (residual) feed intake (capacity) should be considered, is not solved yet.

2.3.2
Relative economic weight of the traits The calculation of economic weights with the herd model requires assumptions with regard to the costs of input.Under ecological aspects costs for the environmental pollution and consumption of fossil energy have to be taken into account.It is assumed that the consideration of such external costs leads to higher prices for feedstuffs such as concentrate and com silage.From a social Standpoint the value of agricultural products (milk, beef) has to be higher.Therefore economic weights were calculated under different scenarios with regard to costs and returns (Table 1).The resulting weights were used to construet 4 different selection indices.The following scenarios were assumed: Scenario I: Costs and returns used by MIESENBERGER et al. (1998) were assumed.
The economic weights correspond to those assumed for the current total merit index in Austria.Scenario II: Costs of 1 kg of dry matter of concentrate (barley and soya) were 50 percent higher than under scenario I (see Table 1) Scenario III: The same assumptions as in scenario II, but the costs of com silage are increased additionally.One energy unit (MJ NEL) of com silage is as expensive as one energy unit of grass silage (Table 1) Scenario IV: The same costs as in scenario III were assumed, but no com silage was used in cattle feeding.
To take social aspects into consideration returns of agricultural products were increased in scenario III and IV (see Table 1).The milk price per kg was increased by 0.015 EUR, beef price per kg carcass weight by 0.360 EUR.These are typical differences between prices paid to organic farmers in comparison to conventional products.In deriving economic weights for functional traits, especially reproduetive and health traits related to animal welfare, it is important to consider public opinion and consumer attitude towards animal production (GROEN et al., 1996).To provide some indication of the effects of selection aecording to a total merit index where ethical aspects with regard to animal welfare are taken into consideration, two further index variants were investigated.The weights ofthe traits, which constitute the subindex "Fitness value" in the current Austrian total merit index were increased arbitrarily.Scenario V: Economic weights for all traits ofthe "Fitness value" 50 per cent higher than in scenario I Scenario VI: Economic weights for all traits ofthe "Fitness value" 100 per cent higher than in scenario I Table 2 gives an overview of the economic weights of the traits included in the 6 indices.3.
Results and discussion 3.1 Relative economic importance of subindices Table 3 shows that the consideration of extemal costs leads only to a marginal shift of the relative importance of the three subindices "Milk Value", "Beef Value" and "Fitness Value", favouring the "Fitness value" (scenario II to IV in comparison with scenario I).The reason for this shift was a higher economic weight ofthe functional trait persistency under scenario II and III.The weight of persistency increases due to higher concentrate costs.This confirms the results of MIESENBERGER (1997), who showed an increase of the economic weight of persistency under increasing costs for concentrate.Generally higher feeding costs lead to lower economic weights of functional traits, where an improvement is coupled with increased feedstuff consumption.
Table 3 Relative economic importance ofthe 3 subindices "Milk Value", "Beef Value" and "Fitness Value" under the 6 scenarios (Relative wirtschaftliche Bedeutung der 3 Subindizes "Milchwert", "Fleischwert" und "Fitnesswert" unter den 6 Szenarien) The arbitrary increase of the weights of the functional traits caused a stronger shift of the relative importance of the subindices towards the traits combined in the "Fitness value" (scenario V and VI compared to scenario I).

3.2
Expected selection response under selection with different total merit indices.The results given in this chapter are intended to provide some indication of the effects of selection aecording to the 6 scenarios presented in this paper.When interpreting the expected selection responses one should be aware ofthe assumptions.Table 4 contains the selection response per year in genetic Standard deviations for each trait under the 6 different scenarios.Selection with the current total selection index in Austria (scenario I) lead to an improvement of all considered traits except four functional traits.Small negative responses were observed for patemal and maternal fertility, patemal calving ease and patemal stillbirth.Selection with indices where extemal costs were considered (scenario II, III and IV) yielded similar results though there is less deterioration in the functional traits mentioned above.The picture changed under scenario V and VI where a higher economic weight is put on the functional traits, but even under scenario VI a small negative response of patemal calving ease and patemal stillbirth was observed.The application of the indices under scenario V and VI allowed no improvement of the trait dressing percentage.Table 4 also illustrates that with higher weights for the functional traits the selection response for traits combined in the "milk value" decreased by 19 and 38 percent, respectively.

3.3
Economic efficiency ofthe different total merit indices The economic efficiency expresses the robustness and efficiency of a certain set of economic weights used in a total merit index with regard to the monetary total selection response (GIBSON, 1995).The economic efficiency is calculated by multiplying the natural selection response ofthe Single traits under selection with the "used" total merit index with the corresponding economic weights of the "true" total merit index.These products are summed up and divided by the expected monetary total selection response with the "true" total merit index.Table 5 shows the proportion of the monetary total selection response, which can be achieved by use of an index with the economic weights ofthe index for the scenario in the respective line under the assumption that the weights ofthe index in the respective column are "true".
Table 5 Economic efficiency with regard to the monetary total selection response per year of the used index scenario relative to the "true" index scenario (bold) in the respective line (ökonomische Effizienz bezogen auf den jährlichen monetären Gesamtzuchtfortschritt des verwendeten Index relativ zum "wahren" Index (fett)  In the first line of Table 5 the economic weights used in the current Austrian total selection index are assumed to be "true".From an economic point of view the indices under scenario II to IV were of the same value.A selection index with 50 % higher economic weights (scenario V) lead to marginally reduced economic efficiency (3% less efficiency).Doubling the economic weights of functional traits in the selection index under scenario I reduced the efficiency by 11 %.Irrespective of the assumed "true" scenario the economic efficiency ofthe index under scenario V was about 97 to 98 %.

Conclusions
The derivation of economic traits with a herd model with and without consideration of extemal costs leads to almost identical economic weights for the single traits.Selection responses under selection with indices based on such economic weights did not differ very much.Recording of extemal costs and a subsequent calculation of economic weights is not the most effective way to construet a more ecological total merit index, because the economic weights tumed out to be relative insensitive to the assumed extemal costs.Interpreting the calculated expected selection responses for the single traits one should be aware of the fact that the assumed correlations between the traits play a considerable role.Under the assumed genetic and phenotypic correlations it could be shown that increasing the economic weights of functional traits directly is a reasonable possibility to take some ecological values into account.This way to alter economic weights pays regard to extemal costs indirectly.SUNDRUM (1997) showed that increasing the average age of cows from 5.1 to 5.6 years was more effective with regard to mediane emission than increasing the mean milk Performance by 1.000 kg per year.From an economic point of view there is no great risk in increasing the economic weights of functional traits by 50 %.The economic efficiency of such a selection is reduced marginally compared to a selection using an index with currently used economic weights of the single traits.The lower selection response of milk traits is nearly compensated by the higher response of functional traits.Nevertheless, due to the high subjective value of the milk traits in practice such an ecological total merit index may be at a competetive disadvantage beyond economic reasons.

Table 1
Some returns and costs in the 6 different index scenarios in Euro (Erlöse und Kosten unter 6 unterschiedlichen Indexszenarien in Euro)

Table 4
Natural selection response per year in genetic Standard deviations under the 6 scenarios (Naturaler Zuchtfortschritt pro Jahr in genetischen Standardabweichungen unter den 6 Szenarien)