| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,1
* OntarBio, RR5 Guelph, Ontario, Canada, N1H 6J2
Agriculture and Agri-Food Canada, Dairy and Swine Research and Development Center, Sherbrooke, Quebec, Canada, J1M 1Z3
Canadian Dairy Network, Guelph, Ontario, Canada, N1G 4T2
CanWest Dairy Herd Improvement, Guelph, Ontario, Canada, N1K 1E5
1 Corresponding author: miglior{at}cdn.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: organic farming • survey • total merit index
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Whereas research in organic dairy production has mostly focused on health, alternative medicine, welfare, and nutrition, there has been little research in the area of genetic selection. On a national basis, conventional dairy breeding has followed a philosophy of "one size fits all," with the Holsteinization of dairy production worldwide and the pursuit of the same goal of maximum production per cow per year. More recently, however, there are signs of a change in emphasis with the inclusion of an increasing number of functional traits in the national breeding goals (Miglior et al., 2005). Also, international genetic evaluations have followed this trend and now include production, type, udder health, longevity, and calving performance traits (Mark, 2004; Interbull, 2006), and research is under way for other functional traits, especially female fertility traits (Jorjani, 2005).
Gamborg and Sandøe (2005) reviewed the difficulties of agreeing on a definition of sustainable animal breeding, in spite of the work carried out by the Sustainable European Farm Animal Breeding and Reproduction project. They concluded that a pragmatic definition should be used based on concerns (e.g., animal welfare) addressed by criteria or the direction of change (e.g., improve animal welfare) and measured by appropriate indicators. Boelling et al. (2003) discussed organic animal breeding and listed genetic diversity, genotype-by-environment interaction (G x E), and selection goals as the main areas of research.
Genotype-by-environment interaction between conventional and organic production should be estimated because, if significant, it would decrease the effectiveness of using breeding values estimated in conventional herds (Nauta, 2001; Pryce et al., 2001). Nauta et al. (2006b), from an analysis of 188 organic and 152 conventional Dutch dairy farms, found genetic correlations significantly lower than unity between organic and conventional farms: 0.80 for milk (P < 0.01) and 0.78 for protein yield (P < 0.05). However, the standard error of the estimates was still too large to conclude that a separate organic breeding program was necessary.
In general, estimating G x E between organic and conventional herds can be problematic because of the small size of many organic populations and the heterogeneity among organic farms within and across countries. Therefore, results from a given population or region cannot be generalized. For example, in Switzerland and Denmark, organic farms are very similar for herd size and feeding to conventional farms (B. Bapst, FiBL, Research Institute of Organic Agriculture, Frick, Switzerland; personal communication; Kristensen and Pedersen, 2001), whereas in the Netherlands and Great Britain there are large differences among organic dairy farms (Nauta, 2001; Pryce et al., 2001). A reliable estimate of G x E is presently unfeasible in Ontario because of the small organic population involved. Boettcher et al. (2003) studied the effects of different feeding systems on Canadian genetic evaluations and found no significant G x E between pasture-based and conventional herds in Ontario and Nova Scotia. If the main differences between conventional and organic herds were only due to pasture availability, then little or no G x E interaction would be expected. However, this may not be the case because organic requirements also affect other aspects of dairy production.
The choice of the breeds best suited for organic production is a subject of much discussion. Sustainable animal production should adjust to local conditions and it may require different types of animals to fit different production situations. Therefore, genetic diversity, choice of breed, and mating system are important for organic dairy breeding. Organic production tends to increase the amount of forage in the ration and to require feeding on pasture. Research in Ireland (Dillon et al., 2003a,b) and in New Zealand (Lopez-Villalobos et al., 2000; Harris and Kolver, 2001) has shown that under a grazing system, Holstein cows of high genetic potential might have more reproductive problems and lower lifetime performance than other breeds or crosses. Concerns about loss of biodiversity and of robustness due to the conventional selection programs have prompted parts of the organic sector to try to conserve genetic resources, such as local or traditional breeds. In Germany, at the organic research station in Frankenhausen, a herd of traditional Friesian cows is maintained, and organic farmers in the Netherlands use Blaarkop and conventional Friesian cows (Nauta, 2001).
Olesen et al. (2000) discussed the definition of animal breeding goals for sustainable systems and showed how animal breeding should contribute to optimize the whole production system. Therefore, different characteristics of agricultural production will affect selection goals. For example, when high energy and nutrition costs force production onto marginal land, then selection should aim to improve utilization of local feeds and to increase intake of roughage and adaptation to low-energy-input systems. When production must adapt to diverse local conditions, selection should increase livestock robustness and adaptation to different environments. When chemical medications are less available or too costly, selection should improve genetic resistance to disease and parasites. Therefore, the focus may shift on different groups of traits, depending on production characteristics and constraints. Then, these different conditions could apply to organic production and affect selection goals. Total merit indices specific for organic dairy farmers are currently available in a few countries: in Switzerland (Bapst, 2001) and, for dual-purpose populations, in the Bavarian region of Germany (Krogmeier, 2003) and in Austria (Baumung et al., 2001).
The objectives of this study were to identify the priorities of selection for organic dairy production in Ontario and to formulate a total merit index for organic dairy farmers based on their subjective priorities. Given the limits imposed by the small size of the organic dairy population, the use of the index was not to formulate an alternative selection program, but to select among proven bulls those with the highest breeding values for the chosen traits.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
A short survey of 20 farms was first carried out to identify breeds and mating plans most commonly used in organic herds. A more extensive questionnaire was then developed to identify production characteristics and breeding strategies of organic dairy farms in Ontario. Because the majority of Ontario organic farms participate in milk recording (DHI), DHI records were also used for the study.
For this research a sample representative of organic dairy farms in Ontario was identified. The sample included 18 farms, or 40%, out of a total 46 organic Ontario farms. These farms were spread over the main agricultural regions of the province and had different breeds, breeding strategies, and different levels of milk production. All farms were certified organic between 1978 and 2002, except for 1 farm still in transition. Response rate was 100%. The same researcher visited all 18 farms and conducted individual interviews, and all producers except one agreed to release their DHI data for research purposes. Information on production, crops, feeding, culling, health problems, and breeding policies based on the last year were collected through the survey. Producers in the survey were asked to score different factors affecting their management. Scores ranged from 0 (not important) to 5 (most important) and answers were averaged across farmers to determine the relative weight for each trait. Differences in average scores were tested with Friedman’s multiple comparison test (Hollander and Wolfe, 1973). Six years of complete DHI data were included, from 1998 to 2003. All participating herds were enrolled in DHI services over this period, except for 1 that had 4 instead of 6 yr of data.
An alternative approach based on the declared preferences of organic farmers, rather than economic weights, was taken to build a total merit index for organic dairy producers in Ontario. Such an index was intended as a tool to select from available proven bulls that best met the perceived needs of organic producers. Farmers were asked to choose and score which traits to select. A total merit index was formulated as follows:
 |
where Hi = total merit index for ith bull, v1 = relative subjective weight for trait 1, and A1i = EBV of ith bull for trait 1.
Average EBV of bulls selected by the various indices and correlations between indices were used to compare the organic total merit index with the conventional index—the Lifetime Profit Index (LPI), which is the selection index for all dairy breeds in Canada (Canadian Dairy Network, 2005).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Among the factors limiting milk production, organic farmers in Ontario felt that the milk quota was by far the most important, with an average score of 4.2 (5 being the maximum score). Barn size, labor, and grain and forage production were all scored around 2.5 and were considered very limiting by few farmers (Table 1
). Differences between average scores were significant only between milk quota and all other factors (P < 0.05). Even though organic agriculture tends to be more labor intensive than conventional, only 2 producers saw labor as a serious constraint for milk production.
|
Crops and Feeding.
Tillable land was on average 101 ha per farm, ranging from 36 to 243 ha, with hay, haylage, and pasture on 68% of the land (Table 2). Small grains and forage were grown on average on 88% of the land and corn on 7% of the land. Only 28% of surveyed farms grew soybeans and 61% grew corn. Corn was less common in organic rotations because it needed more nutrients and was more prone to weed infestation. Also, due to the widespread use in Canada of genetically modified varieties, it was difficult for organic farms to find required genetically modified-free corn and soybean seeds on the market. Compared with a sample of 169 conventional Ontario farms, pasture, hay, and haylage were higher in organic: 68% instead of 55% of tillable land, whereas corn was lower: 7% instead of 14% (OMAFRA, 2003).
|
When reasons for culling were scored, the most important were fertility and mastitis, followed by feet and legs. Milk production and old age were secondary, whereas overall conformation, calving problems, injuries, or temperament received a very low score (Table 3). Average scores for fertility and mastitis were significantly higher than for all other reasons for culling (P < 0.05), except for feet and legs. Except for fertility, culling reasons seemed to differ between organic and conventional herds. In conventional Ontario herds, fertility was the main reason for culling, followed by low production, mastitis, sickness, udder breakdown, and feet problems (CanWest DHI, 2003).
|
). Mastitis and calving problems were the most widespread across farms, affecting 11 (60%) of surveyed farms. Milk fever and ketosis did not occur in half of the farms and together affected 2% of all cows. All interviewed farmers recognized that after becoming organic, the incidence of health problems, particularly regarding metabolic disorders, had definitely decreased and that cases of acute mastitis had almost disappeared.
|
Organic Dairy Breeding in Ontario
Breeds and AI Usage.
The Holstein breed was prevalent among the 18 surveyed farms, except for 2 farms: 1 with Jersey and 1 with Brown Swiss cows. The general feeling among the organic farmers was that Holstein cows have been selected for a different system of production and had problems adapting to a forage-based diet. There were also concerns about their health, fertility, longevity, grazing ability, loss of body condition, general fitness, and inbreeding. Research has shown that some of these concerns may be justified. Studies in New Zealand (Harris and Kolver, 2001) and in Ireland (Dillon et al., 2003a,b), where dairy production is based on pasture, have indicated that the most profitable cows for these environments were different from those selected under a high-concentrate regimen. Other researchers (Weigel et al., 2001; Kearney et al., 2004) have indicated a possible G x E interaction for milk production under intensive and extensive production systems. Kearney et al. (2004) have suggested that genetic correlations could be affected by a scaling effect between environments, whereas within environment, significant differences existed for milk and protein genetic correlations between the upper and lower grazing quartiles.
Crossbreeding can be very effective in eliminating inbreeding and improving fitness traits through hybrid vigor and decreased homozygosity. However, the choice of breed or breeds is critical, and there is little information available for dairy crossbreeding with respect to specific and general combining ability. Furthermore, the choice of bulls within breed is as important as the choice of the breeds, and it requires a good knowledge of the actual breeding values for each breed. In the organic sector, there is often the idea that going back to the "good, old breeds" is the approach to take, but some of these minor breeds may not have a developed selection program. Thus, the farmer is the one doing the progeny testing on his own herd, with all the risks involved.
Crossbreeding was more frequent in organic than in conventional herds. In fact, about 7, or 40%, of the 18 farmers in this research had crossbred some or all of their cows, whereas 9 had pure Holstein, 1 had pure Jersey, and 1 had pure Brown Swiss cows. Only 2 herds opted for crossbreeding all their cows, whereas in the other 5 herds, 17% of the cows were crossbred. The most frequent crosses were 3-way rotational crosses with Holstein, Brown Swiss, and Jersey (all in 1 herd), followed by crosses between Holstein and Dutch Belted, Milking Shorthorn, Simmental, Brown Swiss, Ayrshire, or Jersey (Table 5).
|
). Using natural service sires was also necessary when crossing with minor breeds because semen from these breeds was rarely available from AI centers, and this could have a negative impact on the herds’ genetic progress. There was also an association between percentage of Holstein in the herd and milk production level. The higher producing herds were pure Holstein, with average milk production of 9,492 kg, whereas crossbred herds had lower average production 8,094 kg for herds with 9% of cross-bred cows and 6,928 kg for herds of 43% of crossbred cows. However, fat and protein content were not affected by crossbreeding (Table 6).
|
Organic farmers were asked to score which traits, among those with a genetic evaluation in 2003–2004, were the most important for selection on their farm. Based on their average score, functional traits came first, with feet and legs and overall udder, followed by fat yield, body capacity, protein yield, and SCS (Table 7). The average scores for feet and legs and overall udder were significantly higher than those of lactation persistency, calving ease, and milk. Body capacity scores were high because more capacity was associated with higher forage intake, and similar to those of fat and protein yield. Somatic cell score, as an indicator of udder health and mastitis resistance and longevity, had similar scores. Milk production was the least important trait, with only 2 farmers scoring it as the most important, and 12 ignoring it. Only 2 farmers used the LPI extensively, and its average score was only 2.28. Dairy cattle selection is quite homogeneous in conventional herds in Canada, as dairy farmers and AI organizations in Canada use the LPI index extensively for their selection decisions. Chesnais and Van Doormaal (2006) have compared genetic level and progress of Holstein, Ayrshire, and Jersey cows for a series of traits including LPI from 1983 to 2003. They concluded that LPI had the highest genetic trend (2.89 genetic SD unit in Holstein) among all traits considered. In a separate analysis by Canadian Dairy Network (2003), Canadian provinces were compared by the average LPI of the cows, and Ontario was found to have the highest average LPI among all provinces.
|
Organic Total Merit Index.
The organic total merit index was built based on the traits chosen by organic farmers and their relative subjective scores (Table 7
). At the time of the survey, calving ease was chosen as a fertility trait because genetic evaluations for fertility were not available. However, when the organic selection index was put together, bull proofs for daughters’ fertility had become available in Canada. Calving ease was replaced by fertility because this was actually the trait that farmers wanted to improve if available, and because calving ease should be used to avoid problems rather than being selected for directly.
Subjective scores were averaged across respondents, and the relative weights transformed to a percentage scale (Table 8). Feet and legs, overall udder, body capacity, and SCS were the most important functional traits for Ontario organic farmers and together they had almost the same weight as all the other traits, production included. One of the major features of this index was the low weight of the production traits relative to functional: 28 to 72%. Even though there is a definite tendency toward increasing the relative importance of functional traits, worldwide emphasis on production traits in the Holstein breed ranges from 29 to 80%, with most countries placing at least 50% emphasis on production. Only those used in Scandinavian countries have a relative weight on production around 30%, very close to that of the organic index. Within production traits, protein is by far the most important trait, with a 3 to 1 ratio relative to fat, and milk is ignored in almost half of the countries (VanRaden, 2004; Miglior et al., 2005). This was in contrast with the weights in the organic index, where fat had a slightly higher emphasis than protein yield. Worldwide, the most important functional traits included in Holstein selection indices were, in decreasing order, longevity, SCS, overall udder, feet and legs, fertility, overall conformation, calving ease, growth, and milking temperament (Miglior et al., 2005). The group of traits considered in the different selection indices varied noticeably between countries. Only longevity was included in all indices, followed by udder traits, SCS, and feet and legs.
|
Ontario Organic Total Merit Index vs. Canadian LPI.
National genetic evaluations from May 2005 provided by Canadian Dairy Network were used to calculate the organic index using the weights from Table 8. Compared with the Canadian official selection index, LPI, the major difference was in the relative emphasis on production: 54% in the LPI compared with 28% in the organic index. On the other hand, when functional traits were grouped as durability and health related traits, health traits were 2.5 times more important in the organic index, whereas emphasis on durability traits was more similar. Correlations were estimated for all bulls officially proven in May 2005, and overall correlation between the 2 indices was 0.88. However, it decreased to 0.70 for the top 1,000 bulls for the organic index and to 0.65 for the top 100 bulls (Table 9).
|
), bulls selected for the organic index were about 0.5 SD lower for all yield traits, higher for body capacity, and lower for SCS. Those bulls were also about 0.25 SD better for herd life, feet and legs, mammary system, and lactation persistency, and about 0.15 SD lower for protein percentage and calving ease. Thus, by selecting proven bulls with the best organic index, organic producers could improve health traits and durability at the expense of production. However, even though this index was based on producers’ preferences, its use would depend on it being available to organic producers on a regular basis.
|
Organic selection indices have also been developed in the Bavaria region of Germany for the Simmental, Brown Swiss, and Gelbvieh breeds (Krogmeier, 2003) and in Austria for the Simmental (Baumung et al., 2001). All of these breeds have a beef component in their selection indices, either conventional or organic, with a weight of 10 to 15% in Bavaria and 13 to 19% in Austria. In Bavaria, among the functional traits, the most important were fertility (25%), longevity (15%), and lactation persistency (10%), whereas SCS accounted for only 5%. Such indices have been used as models for the Swiss organic indices (Bapst, 2001). Baumung et al. (2001) used a herd simulation approach and showed that changes in returns and costs slightly affected the weights in the organic selection index. Therefore, the organic indices were quite robust to different market scenarios. Only when the value of functional traits was arbitrarily increased by 50 and 100% did the weights change markedly. This could also suggest that an organic selection index could fit a wide range of farming systems.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Given the limited size of the organic dairy population in Ontario, an alternative selection program would not be viable. However, sires of organic cows could be ranked on a total merit index based on the preferences expressed by organic producers. Such an index was intended as a first step in addressing the needs of dairy organic producers; its limits are due to the fact that sires would be chosen from an existing population selected on a different index (LPI). The use of such index would have a very small effect on the overall genetic progress because it would affect the sire of cow path for 1% of the cow population.
In Canada the major difference between the Ontario organic index and the LPI index was the relative emphasis between functional and production traits. Among functional traits, body capacity, SCS, and lactation persistency had more emphasis in the organic index. Correlations between these 2 indices were 0.88 for all proven bulls and decreased to 0.65 for the top 100, indicating that a different group of bulls would rank at the top. However, given the current trend of conventional selection indices toward increasing the emphasis on functional and health traits, differences between conventional and organic indices could be reduced in the future.
Organic farmers surveyed in Ontario and in Switzerland expressed their need for genetic improvement of grazing traits and further research is needed to identify the traits that could improve production and longevity on a high forage diet. Such research may also be a priority for intensive graziers and possibly for conventional producers if a long-term increase of energy costs will force a change toward more extensive milk production systems.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication July 28, 2006. Accepted for publication October 16, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Togashi and C. Y. Lin Economic weights for genetic improvement of lactation persistency and milk yield J Dairy Sci, June 1, 2009; 92(6): 2915 - 2921. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Ruegg Management of mastitis on organic and conventional dairy farms J Anim Sci, April 1, 2009; 87(13_suppl): 43 - 55. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
