| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Sustainable Livestock Systems Group, Scottish Agricultural College, Bush Estate, Penicuik, Midlothian, EH26 0PH, United Kingdom
AbacusBio Ltd., PO Box 5585, Dunedin, New Zealand
1 Corresponding author: eileen.wall{at}sac.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
Key Words: body energy • economic value • genetic selection • body tissue mobilization
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
Feed intake is currently only accounted for in UK dairy cattle breeding indices by implicit reductions in relative economic values of fat and protein yield because of associated feed costs (Stott et al., 2005). Economic efficiencies or inefficiencies arising through body energy loss or gain are not considered. Cows that have greater milk production must eat more food to produce this milk (and therefore increase feed costs), utilize their own body energy, or both, potentially at the expense of some metabolic functions. This information may be harnessed for selection purposes in terms of developing a breeding goal trait with an economic value.
Wall et al. (2007) produced daughter energy balance (EB) genetic profiles across first lactation for sires with type-classified daughters using the effective energy system (Emmans, 1994). This system describes the amount of energy processed when lipid and protein weights change, and assumes that more energy is required to gain or replenish body lipid and protein than is released when they are mobilized. Therefore, if a cow loses body energy through losses of body lipid and protein, more feed energy is required to replenish that body energy than was provided when the body energy was lost. Depending on the feed costs at various stages of the lactation and dry period, genetic changes in the pattern of body energy mobilization and deposition are expected to have consequences for the aggregate feed costs for a cow across the calving cycle.
The aim of this study was to propose descriptors of first-lactation daughter body mobilization curves for sires. A method for the derivation of economic weights for these traits based on their effects on aggregate feed costs across the calving cycle was also developed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
. The cost to feed a cow (FC) on day t is modeled as a function of a vector g of genetic characteristics (e.g., genotype or breeding values) and, assuming that energy is the limiting feed nutrient, is as follows:
|
 |
where on each day t, and expressed as functions of genetic characteristics g, Eprod, Eret, and Emaint are cow ME requirements for milk production, for retention of body energy, and for maintenance, respectively; Emob is the metabolic energy released from tissue mobilization; and EC is the cost of feed on day t expressed per unit of ME provided to the cow.
Total feed costs between 2 successive calvings (CIFC) for cows having genetic characteristics, g, can be computed through integration of the above function as
 |
where t is the last day of the calving interval. The feed cost component of an economic value (EV) for a specific genetic characteristic, gi, effecting lactation EB through body tissue mobilization, retention, or both can be computed as the partial derivative of the above function with respect to a unit change in the genetic trait variable.
It is reasonable to assume that production and maintenance energy requirements are independent (at the metabolic level) of genetic traits affecting changes in body energy. Therefore, in theory, the economic value of any defined genetic trait (EV_gi) that measures body tissue mobilization and recovery can be computed as
 |
Although there is no reason to assume that the above formulation cannot be computed for complex functions of energy costs, energy mobilization, and retention profiles, this study evaluated economic values for 2 special cases of this general theoretical model. The resulting formulas for economic values are therefore more straightforward and more transparent.
Genetic Characterization of Body Tissue Mobilization
Figure 1a describes key reference points on a simplistic body energy content (BEC) profile though the calving interval (lactation and dry period) of a dairy cow. These include the base BEC (BEC_base) at the start of the lactation (which we take to be the point to which the cow needs to return by the next calving), the minimum level of BEC (BEC_min), which occurs at time LDAY_min, and the BEC at the end of a standard 305-d lactation (BEC_305). Body energy changes between these key reference points are assumed to be linear with respect to time. Thus, there are linear representations of the average rates of body energy mobilization (b_early) and body energy recovery (b_late) during the lactation. Although this representation is unrealistic, when computing economic values it is the modeling of the change in the profiles arising from a genetic change that is most relevant, and this is addressed below. The profile considered in Figure 1a characterizes a representative national average body tissue mobilization profile across first parity, and so although some cows may not enter energy deficit at all and others may reach positive body energy before the dry period, the national average profile is assumed to have the following properties: BEC_min < BEC_base; BEC_305 > BEC_min; and BEC_305 
BEC_base.
|
, 3 genetic traits were defined, which characterize the body tissue mobilization profile. These are 1) BEC_305 = BEC_min –BEC_base: the loss in body energy from calving to the nadir (MJ); 2) b_early: the average rate of loss of body energy from the start of lactation to the point of minimum body energy (MJ/d); and 3) b_late: the average rate of recovery of body energy from the point of minimum body energy until the start of the dry period (MJ/d).
|
graphically represent the definitions of these 3 body energy traits in such a way that they are mutually independent. It was assumed in the model that the difference between BEC at d 305 and BEC at the start of lactation must be recovered in the dry period. Cow BEC will inevitably be different from BEC_base at the beginning of the following lactation (Coffey et al., 2004). Both growth toward mature size and a general tendency for average BCS to decline through a cow’s productive life contribute to this difference. However, these differences are assumed to be independent of any factors associated with body energy mobilization in our economic model. The economic values calculated in the model framework are partial economic values and calculate the economic value per unit change in the trait of interest assuming that all other traits are held constant. This is consistent with the methodology applied to deriving economic weights for other traits in the UK. Average BEC at 305 d for the national herd can be simplistically defined as a linear projection from BEC predicted at 2 points at or after BEC_min. The point at DIM 200 was chosen because conformation data are sparse after that point in the lactation. That is,
 |
where LDAY_min X < 200 and where (BEC_200 –BEC_X)/(200 – X) is a prediction of the slope b_late as depicted in Figure 1, such that BEC_305 = BEC_min + (305 – LDAY_min) x b_late .
From Figure 1 and the energy cost pattern through the lactation, the calving interval can be divided into 3 segments that deal with body tissue mobilization. These segments are described below.
Segment 1 (early): the period when average BEC is decreasing until it reaches a minimum. During this period, feed costs are essentially saved because the energy obtained from mobilization of body energy would have been obtained from feed (assuming feed intake is not limiting).
Segment 2 (late): the period during which average BEC is increasing back toward BEC_ base until the end of the lactation. During this period, there are additional feed costs associated with the recovery of BEC.
Segment 3 (dry): the dry period when the remaining shortfall of BEC relative to BEC_base has to be recovered using feed supplied during the dry period. As with segment 2, there are additional feed costs associated with the recovery of BEC during this period.
Feed-Cost Models
Rather than mathematically define a continuous feed-cost function throughout the calving interval, we consider 2 more simple representations (models A and B) of the feed-cost function during which costs per unit of ME remain unchanged over long periods of the calving interval.
Model A. Step Changes Through the Lactation in Feed Cost per Unit of ME Supplied.
The cost of feed per energy unit supplied on any day t is assumed to take separate constant values for days within the 3 periods: d 0 to LDAY_min (EC_early), LDAY_min until the start of the dry period (EC_late), and the dry period (EC_dry).
The segment-specific and overall feed-cost consequences of BEC change can be computed using the following formulas:
Segment 1: BEC declining to a minimum BEC (BEC_min):
 |
where k is the efficiency of energy mobilization relative to energy retention and can be calculated as the energy released from lipid divided by the energy cost of replenishing that lipid: 39.6/56 = 0.707, under the assumption that all body tissue mobilized in the national herd is from lipid reserves (Emmans, 1994). This efficiency figure can be modified downward to reflect the much poorer efficiency of energy mobilization relative to energy retention when the mobilization and retention are from protein. Further, FC1 should be a negative value, representing temporary savings in feed requirements resulting from energy obtained for production from mobilization of BEC.
Segment 2: BEC increasing through the lactation:
 |
where FC2(A) should be a positive value reflecting the extra feed requirements associated with recovering BEC over and above normal requirements for maintenance and milk production and so on.
Segment 3: Recovering BEC to base levels in the dry period of the calving interval:
 |
where FC3(A) should be a positive value reflecting the extra feed requirements associated with recovering BEC during the dry period.
The overall feed-cost consequence (FCT) of BEC change is computed as:
 |
Model B. No Change Through the Calving Interval in Feed Cost per Unit of ME Supplied.
This model is identical to model A above with the exception that EC_early, EC_late, and EC_dry are identical.
Derivation of Economic Values
Trait 1. Maximum Absolute Change in BEC.
The economic value of BEC_drop (BEC_min – BEC_base) can be computed as:
 |
Taking necessary derivatives under feed cost model A gives:
 |
such that,
 |
Note that
 |
because the slope b_late is unchanged by a change in BEC_min (as shown in Figure 1b).
In the special case when EC_dry = EC_early = EC_late (i.e., feed-cost model B where constant feed price per unit of energy is assumed throughout the calving interval), then
 |
Thus, in the situation of a constant feed energy price throughout the lactation, it is economically disadvantageous to reduce the point of minimum BEC because of the inherent biological inefficiency associated with mobilization.
Trait 2. Rate of BEC Loss.
The economic value of b_early can be computed as:
 |
Taking necessary derivatives under feed cost model A gives the following;
 |
Note that
 |
because the equivalent energy and therefore feed costs is lost in the period to BEC_min but just over a shorter time period (as shown in Figure 1c).
Note also that
 |
so that
 |
Thus, slower loss of body energy (more positive b_early) is advantageous unless the cost of ME supplied from feed is greater in the dry period than during later lactation.
When EC_early = EC_late = EC_dry (i.e., feed cost model B), then EV_b_earlyB = 0.
Trait 3. Rate of BEC Recovery.
The economic value of b_late can be computed as:
 |
Taking necessary derivatives under feed cost model A and noting that 
BEC_305/b_late = 305 – LDAY_min, gives the following:
 |
such that
 |
Note that
 |
because feed costs in the first period (early) are not affected by a change in b_late (as shown in Figure 1d).
In the special case where EC_early = EC_late = EC_dry (Feed cost model B), EV_b_lateB = 0.
Data
Information on feed costs and practices in the UK were required to parameterize the feed-cost models described above. This information on the feeding regimen of first-lactation dairy cows was extracted from the dynamic programming model used to calculate economic weights for the national dairy index (£PLI) in the UK (Stott et al., 2005). Feed costs were calculated for 15 possible yield states (low to high yields) across 12 lactations (lactation 1 to 12), and the values for each yield state of lactation 1 were extracted. Feed costs were calculated based on least-cost ration formulation to meet daily energy requirements (Veerkamp et al., 1995). This least-cost ration was based on the type of feed available (grass, silage, concentrates) and the parameters presented in Table 2.
|
Other population parameters required for the model such as calving interval, dry period, and lactation length were derived using production and fertility data from nationally milk-recorded Holstein-Friesian cows from January 2003–December 2005. A Mathcad (Mathsoft, 1999) module of the economic model framework described was built for derivation of economic weights for body tissue mobilization during lactation. First, a generalized scenario was examined using the predicted feeding regimen (e.g., grazing pattern, energy costs) obtained from the model of Stott et al. (2005). However, the other scenarios were examined to investigate economic weights of the traits using different production system assumptions (e.g., feed energy costs throughout the year, grazing pattern, and calving pattern). This included examining the impact of calving pattern on the economics of body tissue mobilization.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
. The monetary cost of 1 MJ of ME was predicted for each day of lactation for the average of the 15 yield states (Figure 2).
|
coinciding with the period when cows were out at grass. The cost of 1 MJ of ME during the housed period was relatively constant, irrespective of the relative proportions of silage to concentrates in the diet. This is partly because of modest differences in the proportions of concentrate in the diet during the housed period and partly because of the costs per MJ of ME from 1 kg (DM) of silage and concentrate being similar (Table 2). Thus, the model was parameterized assuming 1 MJ of ME had an economic cost of 0.002 pence (p) when cows were at grass and 0.008 p at other times (Table 3).
|
) ranged from 5% occurring in December to 12.4% in May. This reflects a degree of seasonal calving in UK dairy herds, favoring spring calving when cows will be out at grass. However, it is clear that a large proportion of cows calving for the first time (2003–2005) are calving in all-year-round systems, a departure from the classic spring–autumn seasonal calving associated with temperate climates that can produce milk off grass for a large proportion of the year. The day that the daughters of bulls reached the lowest BEC varied with a mean of 57.9 d (SD = 26.16), which is close to, but just after, the peak of first lactation. Studies have shown that cows tend to be thinnest (as described by BCS) at this stage of lactation (Coffey et al., 2001; Berry et al., 2002). Approximately 5% of sires had their lowest BEC on the first day of the data range (DIM = 10) and were therefore predicted to have daughters that gained body energy through first lactation. For these sires it was not possible to calculate a rate of body energy loss; therefore, it was set to 0.
The distribution of absolute lowest BEC in cows (Figure 4) was normal with an average of 4,751 MJ (SD =171.17). The distribution for the BEC_drop (drop in BEC from calving until BEC_min) was skewed with a mean of 330 MJ (SD = 229.78). The average value of BEC at DIM 10 and DIM 200 was 5,081 and 5,190 MJ, respectively (SD = 293.44 and 162.67, respectively). The average rate of loss, until the minimum was reached (b_early), was also negative (–3.1 MJ/d) and the rate of recovery (b_late) was 3.09 MJ/d (SD = 1.32).
|
). The economic value for BEC_drop for the majority of the calving year was negative with the lowest value (–0.39 p/MJ) occurring for cows calving in early September. The economic value for BEC_drop was only positive for cows calving from mid March to mid April reaching a maximum of 0.09 p/MJ. The economic value for b_early also varied conditionally on date of first calving, ranging from –15.13 p/ MJ per d for cows calving during April through June to 10.92 p/MJ per d for all other cows. The largest economic value was for b_late and ranged from –66.34 p/ MJ per d for calvings from September to the end of January to 96.08 p/MJ per d for calvings from April through June.
|
. The model output, using these assumed parameters produced economic values for the 3 traits that define body tissue mobilization in Figure 1. The output from the model in Table 3 indicated that the economic value for the maximum drop in BEC (BEC_drop) was small and negative at –0.14 p. The economic value for the rate of loss of body energy in early lactation (b_early) was also negative (–3.1 p) and for the rate of body energy gain in later lactation (b_late) was positive (19.7 p). The overall range in the index of daughter body tissue mobilization was small (– £ 1.94 to £2.17 for the bulls analyzed in this study) with a mean of 0.4 (SD = 0.54). However, there was variation in the index and therefore, the index could be altered via genetic improvement. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
, the economic weight per phenotypic SD of BEC_drop, b_early, and b_late would be – £ 0.32/MJ, – £ 0.09/MJ per d, and £0.26/MJ per d, respectively. The weights for traits in the current UK index are of larger magnitude. For example, Stott et al. (2005) showed that the economic value per phenotypic SD improvement in kilograms of fat and protein were £36.05 and £79.38, respectively. The weight of lifespan was £86.70 per phenotypic SD improvement in functional lifespan. Although there is an economic aspect to the utilization of body tissue reserves in terms of feed energy costs, the relative impact of this in terms of overall importance on the economic performance of the system is small. Therefore, the relative weight of these traits in a selection index for a broad breeding goal (production, longevity, health, fertility, and body tissue mobilization) is small. The national index for the UK, £PLI, was updated in August 2007 (www.mdc.org.uk) with a relative weight of production:functional traits of approximately 50:50. The relative weight of body tissue mobilization in this index, if added, would be 1%. However, if the economic values for a spring-calving system were used, the relative weight would increase to 2%. The overall effect on reranking of sires was negligible. The economic weights account for the economic costs of body tissue mobilization above and beyond its impact on functional traits, which is accounted for in the economic weight for functional traits directly. The magnitude of the economic values calculated is similar to that of the economic values calculated for BCS by Pryce et al. (2006).
The results show that the economic cost, and therefore economic value, for body tissue mobilization is very dependent on the calving season system used (Figure 5). Generally, when the economic value for b_early is positive, the economic value for b_late is negative. The calving pattern by month in the UK has flattened out over time as calving intervals increase and farmers move away from strict seasonal-calving systems. Therefore, the average economic value for body tissue mobilization traits based on the flatter national calving pattern is appropriate. However, some farmers may still operate predominantly spring or autumn seasonal-calving systems. Table 4 shows that the economic values for the traits vary dramatically between the 2 systems. The economic value, for example, for b_early is positive in an autumn-calving system and negative in a spring-calving system. It would not be cost effective for a cow to lose body energy in a spring-calving system when the cost of energy from feed is low, because any loss in early lactation will need to be recovered toward the end of lactation when feeding takes place indoors and energy is more expensive. The opposite is true for autumn calving when it is economically sensible for a cow to lose body energy when feed is expensive and regain it when turned out to grass when feed costs are low. Table 4 also shows that the economic values for the b_early and b_late have reduced over recent years. This is due to the change in calving pattern in that period. In 1985 the calving pattern in the UK had a clear peak around autumn calving. However, in 2005 (Figure 3) the calving pattern was flatter with calvings occurring year-round with only a slight peak in the autumn months.
|
The national stylized curve of body tissue mobilization (not shown) differs from the theoretical curves described in Figure 1 in that cows are predicted to end first lactation with a greater BEC than at the start. Previous work using Langhill research herd data (Coffey et al., 2003) and UK industry data (Wall et al., 2007) has shown similar body tissue mobilization curves with a net increase in BEC by the end of first lactation. We can assume that this increase in BEC is due to a proportion of body energy being accumulated as normal growth during first lactation. Proportions of energy contained in the total weight of protein and lipid are calculated from both predicted liveweight and BCS on a given day; therefore, relative change in BCS is not taken into account (i.e., animals are putting on weight in terms of both body protein and lipid actually growing, but getting thinner because body protein is increasing relative to body lipid). However, the assumption that the average cow in the UK ends first lactation in poorer condition than she started has been shown not to be true. Coffey et al. (2003) and Wall et al. (2005) showed that the fixed population curve for BCS was greater at the end of first lactation than at the beginning. Although it would be feasible to model a system in which factors associated with growth toward mature size are neutralized, such an adjustment would just shift the average curve downwards because the values subtracted would be constant for all bulls. For selection it is how the bulls deviate from the national curve that will produce differences between sires in terms of index values for body tissue mobilization.
Our definitions for the 3 body tissue mobilization traits are simplistic, but when used together, they should be sufficient to describe the majority of genetic variation that exists within current populations of UK dairy cows. It is conceivable that some cows might lose BEC in such a way that b_early and BEC_drop change simultaneously without any change in LDAY_min. From Figure 1, it is clear that we cannot represent such a genotype with any partial change in an individual trait, and so our economic valuation of such a genotype will be approximated under feed-cost model A because we will not be accounting for the fact that our economic value EV_b_earlyA depends on the herd levels taken by itself and other traits (i.e., b_late, BEC_base, BEC_min). Inaccuracies due to this approximation are likely to be trivial in the context of this study, in which economic values are modest, and large changes in body tissue mobilization traits as a consequence of our derived economic values are not expected. However, the general model framework we propose could be formulated differently to more accurately account for interactions and nonlinearities, provided that the traits defined to have partial economic values could also be evaluated with reasonable accuracy for selection candidates.
The model assumes that all mobilization of body energy is a result of mobilization of lipid reserves (k value described earlier). However, an animal with very low fat reserves will start to utilize the energy stored in muscle for maintenance and maintain milk yield, and therefore this assumption may not hold. Estimating the proportion of animals in the UK that are mobilizing energy from their body protein is difficult. Daily daughter protein weight was calculated for the selected sires (Banos et al., 2006) and there was very little change in protein weight across lactation with the average across lactation fluctuating between 86 and 87.7 kg compared with fluctuations between 69.1 and 83.4 kg for lipid. Figure 6 shows that the proportion of bulls predicted to have daughters losing protein weight increases over the lactation. Up to 200 DIM, approximately 8% of bulls are predicted to have daughters that have had a cumulative loss of body protein. After 200 DIM, the cumulative loss of body protein begins to increase to 37% at the end of lactation; however, this may be an artifact of the technique of random regression modeling when the data are scarce in this part of the trajectory (Pool et al., 2000; Wall et al., 2005) and hence DIM 200 is chosen as a cut-off point, as previously discussed.
|
The steps in Figure 3 relate to the periods when cows are turned out to grass and return to housing. Feed costs jump when cows are indoors and the source of forage is silage. Concentrates are added to the diet to meet energy requirements based on the state of lactation yield. These grazing profiles throughout the year are based on the average UK practice and taken from Stott et al. (2005). However, there will be large variation in the practices of farmers with some not utilizing grass outdoors, but rather choosing to ensile it and feed it with concentrates indoors with forage quality varying throughout the year. Also, farmers may choose to feed a TMR based on homegrown or imported feedstuffs rather than a concentrate. The different feeding systems potentially have differing costs of 1 MJ of ME at different times of the year. However, the framework can accommodate the costs of differing energy sources throughout the year for different systems. Developing a model framework that allows for variations in milk yield, feed energy costs (and energy supplied by the feed), and calving pattern will allow farm-system–specific economic values for body tissue mobilization. It will also be easily adapted to account for any fluctuations in the feed prices.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
Received for publication June 6, 2007. Accepted for publication September 19, 2007.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
