Optimizing replacement of dairy cows: modeling the effects of diseases.

We modified an existing dairy management decision model by including economically important dairy cattle diseases, and illustrated how their inclusion changed culling recommendations. Nine common diseases having treatment and veterinary costs, and affecting milk yield, fertility and survival, were considered important in the culling decision process. A sequence of stages was established during which diseases were considered significant: mastitis and lameness, any time during lactation; dystocia, milk fever and retained placenta, 0-4 days of lactation; displaced abomasum, 5-30 days; ketosis and metritis, 5-60 days; and cystic ovaries, 61-120 days. Some diseases were risk factors for others. Baseline incidences and disease effects were obtained from the literature. The effects of various disease combinations on milk yield, fertility, survival and economics were estimated. Adding diseases into the model did not increase voluntary or total culling rate. However, diseased animals were recommended for culling much more than healthy cows, regardless of parity or production level. Cows in the highest production level were not recommended for culling even if they contracted a disease. The annuity per cow decreased and herdlife increased when diseases were in the model. Higher replacement cost also increased herdlife and decreased when diseases were in the model. Higher replacement cost also increased herdlife and decreased the annuity and voluntary culling rate.

Evaluation of timed insemination during summer heat stress in lactating dairy cattle.

We wished to compare the effect of summer heat stress on pregnancy rate in cows that were inseminated at a set interval associated with a synchronized ovulation vs those inseminated upon routine estrus detection. The study was carried out on a commercial dairy farm in Florida from May to September 1995. Lactating dairy cows were given PGF2 alpha (25 mg i.m.) at 30 + 3 d postpartum and randomly assigned to be inseminated at a set time (Timed group) or when estrus was detected (Control group). Cows in the Timed group were synchronized by sequential administration of Buserelin (8 micrograms i.m.) on Day 0 at 1600 h, PGF2 alpha (25 mg i.m.) on Day 7 at 1600 h and Buserelin (8 micrograms i.m.) on Day 9 at 1600 h. They were inseminated on Day 10 between 0800 and 0900 h (Day 9 + 16 h). Cows in the Control group were given PGF2 alpha at 57 + 3 d postpartum and inseminated when detected in estrus. Estrus detection or insemination rate for control insemination cows was 18.1 +/- 2.5% versus 100% for time inseminated cows (P < 0.01). Mean interval from PGF2 alpha to insemination was shorter for time inseminated cows (3 +/- 2.1 d < 35.5 +/- 1.9 d; P < 0.01). Pregnancy rate was greater for time inseminated cows (13.9 +/- 2.6 > 4.8 +/- 2.5%; P < 0.01) as was overall pregnancy rate by 120 d postpartum (27.0 +/- 3.6 > 16.5 +/- 3.5%; P < 0.05). Number of days open for cows conceiving by 120 d postpartum was less for time inseminated cows (77.6 +/- 3.8 < 90.0 +/- 4.2 d; P < 0.05), as was interval to first service (58.7 +/- 2.1 < 91.0 +/- 1.9 d; P < 0.01). Services per conception were greater for time inseminated cows (1.63 +/- 0.10 > 1.27 +/- 0.11; P < 0.05). The timed insemination program did improve group reproductive performance. However, the timed insemination program will not protect the embryo from temperature-induced embryonic mortality, but management limitations induced by heat stress on estrus detection are eliminated. An economical evaluation of the timed insemination program indicates an increase in net revenue per cow with implementation of timed insemination for first service during the summer months.

A network simulation model of large herringbone and parallel milking parlors.

A network simulation model of herringbone and parallel milking parlors was built. Parlor performance was predicted based on the number of cows milked per hour and the amount of milk harvested per shift. Modeled parlors operated using a task-oriented milking routine. The simulation also modeled milking personnel, milking system vacuum pressure and pulsation ratio, and milk yield per cow per milking. The probability distributions of the model components that were used for validation were fitted from data collected from four dairies with the following parlors: double-16 herringbone, double-20 herringbone, double-35 parallel, and double-40 parallel. All fitted distributions were continuous, non-negative, and skewed to the right. The mean number of cows milked per hour (simulated and observed, respectively) was 1) double-16 herringbone, 164.8 and 164.6;2) double-20 herringbone, 207.6 and 206.6; 3) double-35 parallel, 319.2 and 320.2; and 4) double-40 parallel, 361.6 and 362.5. Mean milk yields harvested per shift were 1) double-16 herringbone, 9020 and 9030 kg; 2) double-20 herringbone, 14,800 and 14,746 kg; 3) double-35 parallel, 20,900 and 20,945 kg; and 4) double-40 parallel, 27,100 and 26,974 kg, for simulated and observed yields, respectively. Differences between simulated and observed means for either variable were not significant.

Factors affecting the performance of simulated large herringbone and parallel milking parlors.

A simulation model of double-16 and double-20 herringbone and parallel milking parlors and double-32 and double-40 parallel milking parlors was used to examine the effects of size, design, operating characteristics of the milking system, management strategies, and milk yield on parlor performance. Analysis of factorial experiments indicated that smaller parlors were more efficient. Turns per hour and milk per stall per hour for double-16, -20, -32, and -40 parallel parlors were 5.87, 5.91, 5.21, and 5.00 turns/h and 56.19, 56.46, 49.66, and 47.94 kg/h, respectively. A wider pulsation ratio (60:40 to 70:30) increased performance measures about 4%, and increased vacuum pressure (46.6 to 50.8 kPa) increased performance measures > 6%. Parallel parlors outperformed herring-bones by nearly 8%. Abbreviated milking procedures resulted in a > 6% increase in performance measures over standard milking procedures. Performance response was significantly diminished when the amount of milking labor exceeded deficit amounts (20 to 32 units per milker) for abbreviated milking procedures or standard amounts (13.3 to 16 units per milker) for standard milking procedures. When milk yield increased, turns per hour decreased, but milk per stall per hour increased.

Evaluation of equations based on animal factors to predict intake of lactating Holstein cows.

The accuracy of seven DMI prediction equations based only on animal factors was evaluated with 11 independent data files. Mean square prediction error was used to compare equation accuracy, which was considered to be unsatisfactory when the square root of the mean square prediction error was greater than +/-20% of the observed mean DMI. Robust intake equations that have a tolerable level of prediction errors for most data files would be less risky for practical use than models that are highly accurate for some data files but highly inaccurate for others. The number of independent data files for which equation accuracy was unsatisfactory was used to measure lack of robustness. No equation evaluated was able to predict individual cow DMI with a prediction error that was consistently lower than +/-20% of the observed mean intake. The most robust equation in this study predicted intake unsatisfactorily for 3 of the 11 evaluation data files. Unsatisfactory accuracy for this equation was mainly due to mean bias.

Evaluation of a stochastic dynamic replacement and insemination model for dairy cattle.

An approach to evaluating results from a stochastic, dynamic model for insemination and replacement of dairy cattle was devised using sensitivity and behavioral analyses. Sensitivity analysis was defined as the quantification of the various outputs resulting from uncertain price and production inputs. Behavioral analysis determined how outputs changed when model specifications were changed. The variation in outputs that were identified by sensitivity analysis was used as an objective measure to assess variation caused by behavioral analysis scenarios. A contemporary model was modified, and a base run was defined using Florida values for input variables. The model specifications that were varied were decision horizon, number of milk production levels, and number of days open classes. Effects of including seasonality of milk production, milk price, and conception rate were examined. The necessary decision horizon was shorter than other works would suggest. Optimal policies were influenced greatly by the number of days open classes, but not by the number of milk production levels. Removal of seasonality of milk production and conception rate resulted in meaningful changes in seasonal patterns of all outputs measured. Our results suggest that model specifications could affect results and should be evaluated objectively when models are being developed.

Macromineral element interrelationships and lactational performance: empirical models from a large data set.

The objectives were to model empirical interrelationships among dietary macromineral elements and cation-anion difference influencing DMI, milk yield, 4% FCM yield, and milk composition of dairy cows. The database consisted of 1022 cow-period means collected from 326 midlactation cows in experiments with incomplete block designs. Dietary concentrations of elements ranged from below to above current recommendations, and cation-anion difference (Na + K-Cl) was from +5.8 to +61.2 meq/100 g of dietary DM. Interrelationships were detected for dietary concentrations of Na by K for DMI, 4% FCM yield, and milk fat percentage; Na by Ca for DMI, milk yield, and 4% FCM yield; K by Cl for DMI, 4% FCM yield, and milk protein percentage; K by Ca for 4% FCM yield, milk protein, and fat percentages; Na and P for DMI and milk protein percentage; and K by P for 4% FCM yield and milk fat percentage. Maximal DMI and milk yield were observed with .58% Na, .40% Mg, and +38 meq/100 g of dietary DM. Agreement was close for DMI and milk yield responses to dietary cation-anion difference from our database and independent data sets. The models developed are empirical, do not necessarily establish cause and effect, and should be evaluated prudently; however, they provide useful information to design future research.

Simulating individual cow milk yield for milking parlor simulation models.

A method of simulating individual cow milk yield per milking as a function of herd milk yield and month was formulated for milking parlor simulation models. Milk yield per milking was modeled for each month in three herd milk yield categories: 8165, 8845, and 9525 kg/yr of milk per cow. Actual individual cow DHIA test day milk weight data for three Florida dairy herds in each herd milk yield category and month were adjusted to the mean of their respective actual milk shipped per cow on test day then pooled and converted to a basis of three times per day milk yield per milking. After minor truncation, Weibull probability distributions fitted to these data sets adequately modeled milk yield per milking per cow. Analysis of simulation results for milk yield per milking per cow indicated no significant differences between actual and simulated means for any herd milk yield category or month. Simulations of monthly and yearly total herd milk yield for each herd indicated that fitted Weibull distributions also adequately modeled monthly and yearly herd milk yield characteristics and reflected seasonal herd milk yield patterns typical of Florida.

Optimizing model: insemination, replacement, seasonal production, and cash flow.

Dynamic programming to solve the Markov decision process problem of optimal insemination and replacement decisions was adapted to address large dairy herd management decision problems in the US. Expected net present values of cow states (151,200) were used to determine the optimal policy. States were specified by class of parity (n = 12), production level (n = 15), month of calving (n = 12), month of lactation (n = 16), and days open (n = 7). Methodology optimized decisions based on net present value of an individual cow and all replacements over a 20-yr decision horizon. Length of decision horizon was chosen to ensure that optimal policies were determined for an infinite planning horizon. Optimization took 286 s of central processing unit time. The final probability transition matrix was determined, in part, by the optimal policy. It was estimated iteratively to determine post-optimization steady state herd structure, milk production, replacement, feed inputs and costs, and resulting cash flow on a calendar month and annual basis if optimal policies were implemented. Implementation of the model included seasonal effects on lactation curve shapes, estrus detection rates, pregnancy rates, milk prices, replacement costs, cull prices, and genetic progress. Other inputs included calf values, values of dietary TDN and CP per kilogram, and discount rate. Stochastic elements included conception (and, thus, subsequent freshening), cow milk production level within herd, and survival. Validation of optimized solutions was by separate simulation model, which implemented policies on a simulated herd and also described herd dynamics during transition to optimized structure.

Effects of acute thermal stress and amount of feed intake on concentrations of somatotropin, insulin-like growth factor (IGF)-I and IGF-II, and thyroid hormones in plasma of lactating Holstein cows.

Our objective was to evaluate effects of acute thermal stress, independent of reduced feed intake caused by elevated temperatures, and of reduced feed intake in thermal comfort on plasma concentrations of somatotropin, insulin-like growth factors I and II, thyroxine, and triiodothyronine. Six Holstein cows (averaging 475 +/- 18 kg BW, 2.3 +/- .3 parities, and 96 +/- 12 d in lactation) surgically fitted with catheters in the hepatic portal vein, mesenteric vein, and intercostalis posterior artery were exposed to treatments of thermal comfort environments with ad libitum or restricted (75% of ad libitum) DM intake and a thermal stress environment with ad libitum intake in two balanced 3 x 3 Latin squares. Thermal stress increased rectal temperatures and respiration rates. Dry matter intake of the thermal-stressed cows offered feed ad libitum (11.1 +/- .7 kg/d) was similar to the experimentally imposed reduction in DM intake of the thermal comfort restricted group (11.5 +/- .7 kg/d). Dry matter intake of cows in thermal comfort was 15.1 +/- .7 kg/d. Plasma somatotropin concentrations tended (P less than .08) to decrease during thermal stress but were unchanged by amount of feed intake in thermal comfort environments. Concentrations of IGF-I were not affected by treatments. Concentrations of IGF-II tended (P less than .14) to increase with thermal stress compared with thermal comfort treatments. Thyroxine concentrations tended (P less than .15) to increase in the thermal stress treatment compared with the thermal comfort restricted intake treatment. Triiodothyronine tended (P less than .11) to decrease with restriction in feed intake in the thermal comfort environment.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of thermal stress and level of feed intake on portal plasma flow and net fluxes of metabolites in lactating Holstein cows.

Six Holstein cows (averaging 475 kg body weight, 2.3 parities and 96 d in lactation) fitted with catheters in the hepatic portal vein, mesenteric vein and intercostalis posterior artery were exposed to treatments of thermal comfort environments with libitum or restricted (75% of ad libitum) DM intake and a thermal stress environment with ad libitum intake in two balanced 3 x 3 latin squares to evaluate effects of thermal stress on portal plasma flow and net fluxes of metabolites. Portal plasma flow was measured by administering a primed, continuous infusion of para-aminohippurate into a mesenteric vein and determining its concentration and dilution in portal vein plasma. Thermal stress treatment increased rectal temperatures and respiration rates. Dry matter intake decreased from thermal comfort ad libitum level (15.1 kg/d) to lower levels in thermal comfort restricted (11.5 kg/d) and thermal stress (11.1 kg/d) treatments. Portal plasma flow was related directly to level of DM intake, declining about 14% with thermal comfort restricted intake and thermal stress environment treatments compared with the thermal comfort restricted intake and thermal stress environment treatments compared with the thermal comfort ad libitum intake treatment. Net flux of alpha-amino N was reduced 20 and 35% by thermal comfort restricted intake and thermal stress treatments compared with the thermal comfort ad libitum intake treatment. Net fluxes of urea N, ammonia N and glucose were not affected by experimental treatments. A portion of the negative effects of thermal stress on milk production can be explained by decreased nutrient intake and decreased nutrient uptake by the portal-drained viscera of the cow.

Characterization of clinical mastitis records from one herd in a subtropical environment.

Patterns of occurrence and reoccurrence of clinical mastitis, duration of episodes, and number of quarters treated in a large north Florida dairy over 6.5 yr are described. Monthly incidence of milk disposal due to clinical mastitis varied within and between years and ranged from 1.3 to 4.7% of cow-days in milk. In first parity animals, 37.6% of lactations were treated for clinical mastitis, averaging 1.4 episodes per treated lactation. Almost half of these episodes occurred by d 35 of lactation. For second, third, and fourth or greater parities, 35.5, 48.1, and 56.6% of lactations were treated. These lactations averaged 2.1, 2.3, and 2.8 episodes. In parities 2, 3, and greater than or equal to 4, episodes initiated by d 35 of lactation were 28.2, 23.9, and 20.6% of parity episodes. Seventy-two percent of episodes were treated for less than or equal to 7 d. Sixty-seven percent of episodes were treated in one quarter and treatment in two, three, or four quarters accounted for 21.6, 5.5, and 5.9% of all episodes. When two quarters were treated simultaneously, udder halves (front, rear, right, left) were treated more frequently than were diagonals.

Climatic effects on occurrence of clinical mastitis.

Clinical mastitis records for 6.5 yr from a large north Florida dairy and corresponding daily weather data were analyzed. Monthly incidence of clinical mastitis was expressed as percent of cow-days in milk and graphed against monthly average daily maximum temperature humidity index values and monthly total rainfall. No trends were evident with rainfall. In 3 of 6 yr, monthly incidence of clinical mastitis increased more than 50% above annual incidence, and this followed high monthly temperature-humidity values. Least squares was used to estimate regression coefficients of temperature-humidity index categories based on 999,969 Holstein records. A temperature humidity index category represented the number of days used to calculate average daily maximum temperature-humidity index value. Sources of variation in observed occurrence of clinical mastitis were cow, parity, month, year, interaction of parity by month, and continuous effects of temperature-humidity index categories 2, 6, 15, 30, 60, each to third order and 60 by parity interaction. In all temperature-humidity index categories as the temperature-humidity index value increased, occurrence of clinical mastitis increased. When values rose from 55 to 80, twice as many for 2 d cows showed signs of clinical mastitis.

Influence of environmental temperature on reproductive performance of bovine embryo donors and recipients in the southwest region of the United States.

Superovulation and embryo transfer records on performance of embryo donor (n = 3,908; beef 99%, dairy 1%) and recipient (n = 19,936; beef 92%, dairy 8%) cattle from a commercial transfer unit were analyzed for environmental effects. Embryos (n = 42,428) were recovered on Days 5 to 8 postestrus from superovulated donors. Numbers of ova, fertilized ova and embryos of transferable quality were recorded. Transferable embryos were classified according to stage of development and morphological quality. Embryos (n = 19,936) were transferred nonsurgically. Responses were modeled with maximum-likelihood procedures using log-linear functions of independent variables and ANOVA. Fluctuations in daily maximum temperature (1 to 43 degrees C), for Days 0 to 7 of embryo development, had no effect on distribution of embryos classified as good (48%), fair (40%) and poor (12%). Temperature did not affect the percentage of donors flushed with recoverable ova (89%), mean number of ova (12.2 +/- 0.3), fertilization rate (76%) or percentage of transferable embryos (57%). Recipient pregnancy rate (56%) was not affected by mean maximum temperature for Days 0 to 10 posttransfer. Interactions between temperature and breed type (dairy vs beef), parity (cow vs heifer), or lactational status (lactating vs dry) on pregnancy rate were recorded. Elevated environmental temperature does not appear to adversely affect reproductive responses of donor and recipient cattle intensely managed in a commercial transfer unit.

Factors affecting days of discarded milk due to clinical mastitis and subsequent cost of discarded milk.

Records of clinical mastitis for 6.5 yr from one Florida dairy with 1050 to 1350 cows milking per month were used to predict duration of occurrence of clinical mastitis and to estimate dollar value of discarded milk per lactation with occurrences of clinical mastitis. Only 6.1% of lactations accrued more than 28 d of discarded milk and were responsible for 52.7% of days of discarded milk. Accurate maintenance of herd mastitis records enables calculation of actual days of milk available for sale and provides a useful tool for culling and treating decisions. Duration of each occurrence of clinical mastitis was modeled by fixed effects of breed, season of occurrence, current lactation history of mastitis, number of quarters with clinical signs, and average maximum temperature-humidity index for the 2 d prior to the occurrence (to second order). These effects explained 1.8% of observed variation in duration of occurrence. Random effects of cow explained 5 to 6% of variation. Prediction of duration of occurrence had a mean value of 6.6 d. When economic status of individual animals is evaluated, comparison of days of milk production available for sale is preferable to total days of milk production. Estimated average cost of discarded milk for parities 1, 2, 3, 4, and 5 ranged from $29.73 to 223.98 per lactation.

Occurrence and reoccurrence of clinical mastitis.

Clinical mastitis records for 6.5 yr (July 1977 through November 1983) from a large north Florida dairy were analyzed. Observed frequencies of clinical mastitis were calculated in 7240 Holstein and Jersey lactations. Least squares analyses used Holstein and Jersey lactations greater than 200 d and limited maximum parity to 5 (6732 lactations, 5738 episodes). Solutions for number of occurrences of clinical mastitis adjusted for month of parturition and breed effects were .47, .59, .94, 1.27, and 1.50 for parities 1 through 5. Solutions for number of occurrences across lactation in Holsteins was greater than for Jerseys (1.28 vs. .63). Solutions for occurrence of first episode in Holsteins increased from .35 in parity 1 to .71 in parity 5. Occurrence of episodes two through five increased as parity increased. Similar trends were found in Jerseys with the exception of parity 2. Correlation of residuals between specific episodes increased as episode increased. From episode 2 through 5 probability of another occurrence of clinical mastitis was greater than .75. Solutions for proportion of occurrences of clinical mastitis occurring during the first 35 d postcalving were .60, .36, .31, .30, and .28 for parities 1 through 5. Clinical mastitis in first parity cows occurred primarily during the first 35 d postcalving.