Heat Stress and Milk Production in the South Carolina Coastal Plains1 D. E. LINVILL2 and F. E. PARDUE3 Clemson University Clemson, SC 29634 ABSTRACT
A model developed for the South Carolina coastal plains relates hours with temperature-humidity index values above 74 and 80 to summer season daily milk production. When tested on an independent production data set for 1985, the root mean square model error was less than 1.3 kgld per cow. The model can be used to develop expected summer season dairy production climatologies. Realtime milk production forecasts obtained using daily predicted maximum and minimum temperatures can be used in herd management to reduce effects of heat stress on productivity. (Key words: heat stress, milk production, computer models) Abbreviation key: humidity index.
THI
=
temperature-
INTRODUCTION
Heat stress is defined by Buffington et a1. (2) as any combination of environmental parameters producing conditions that are higher than the temperature range of the animal's thermal neutral zone. The survival and performance of an animal during heat stress periods depend on several weather factors, especially temperature and humidity. Thompson (15) reported that extended exp~ sure to excess heat reduced feed consumption, milk production, and breeding efficiency. Milk production and nonfat milk solids were signifi-
Received July 29, 1991. Accepted April 13, 1992. ITechnicai Contribution Number 3184 of the South Carolina Agricultural Experiment Station, Qemson University, Clemson, SC. 2Departrnent of Agricultura1 and Biological Engineering. 3Departrnent of Animal, Dairy, and Veterinary Science. 1992 J Dairy Sci 75:2598-2604
cantly decreased by thermal stress in a study by Roussel et al. (11). The National Research Council (7) documented dramatic decreases in milk production as temperatures rose above 30·C. The vulnerability of dairy cows to heat stress caused by hot, humid weather is well established. The need to establish dairy herds in warm climates necessitates decisions on methods for heat stress relief. Location of the farm, shade, and supplemental cooling system designs are considerations when herds are located in warm climates. In the experience of South Carolina milk producers, hot, humid weather is more detrimental to production in early summer, when rapid changes in weather conditions are likely to occur, than in mid or late summer. This type of weather is also more detrimental during any season following an extended cool period. Therefore, effective summer season management of dairy herds requires advance warning of impending heat stress periods. Guidance for these warnings must be based on well-accepted heat stress parameters, and warnings must be triggered by operationally useful algorithms. Cattle have the ability to acclimate to changes in the environment. The acclimatization process, however, requires time (5). A cow's response to weather is a complex reaction affected by breed, animal size, level of production, stage of lactation, present and recently past weather, and other factors, such as nutrition and caloric content of feed. Florida researchers (13) studied milk production and breeding efficiency under climatically controlled conditions. Cows in airconditioned facilities produced almost 10% more 4% FCM than cows in facilities that were not air-conditioned. Cows in cooled facilities also displayed increased fertility. In a study conducted by Roman-Ponce et a1. (9), cows placed under shade to remove solar radiational heating produced more milk and had higher conception rates than unshaded cows.
2598
2599
HEAT STRESS EFFECT ON MILK PRODUcrlON
Other researchers related production decline of lactating Holstein dairy cows to the temperature-humidity index (THI) (14) and normal production levels (1). Hahn (4), in turn, used those results to produce maps of expected production from June through September in the United States. Pardue and Linvill (8) reported adverse relationships between high temperature and humidity stress values and milk production in a coastal South Carolina study.
depends on maximum daily temperature (Tmax) occurring 2 h after solar noon and the shape of the daily heating curve responding to the daily solar cycle. This allows the temperature wave from sunrise to sunset to be described by Equation [2]:
MATERIALS AND METHODS
where T(t) is temperature at time t hours after sunrise, Tmin is the morning minimum temperature, and DL is day length in hours. Outgoing radiation from the earth is not balanced by incoming radiation until approximately .5 h after astronomical sunrise (10). Thus, minimum daily temperatures are recorded near sunrise. Thermograph records for Clemson, SC were used to develop a logarithmic expression for the nighttime cooling curve from sunset to sunrise using the assumption that minimum temperature occurs at sunrise:
Production Dat!l Sources
Daily milk production and herd size data were obtained from a large South Carolina coastal plains herd for 1979 through 1985. Total per cow production from June through September ranged from a low of 2216 kg in 1984 to 2487 kg in 1982. These data were converted to average production per cow for each day, and data for 1979 through 1984 were used in model development, and data for 1985 were used for model testing. No attempt was made to account for production changes that were due to freshening cycles or age of the herd or for year to year fluctuations in normal production levels. Temperature records from a nearby National Weather Service Cooperative Station and hourly data from the National Weather Service Office in Charleston, SC were used to characterize weather on the farm during these years. Calculation of the Environmental Variables
The THI equation exists in three forms. One form expresses the relationship using air temperature and relative humidity. A second form uses air temperature and wet bulb temperature. The relationship used to calculate THI in the present study is THI
= Ta +
.36 x Tdp + 41.2
[1]
where T a is air temperature and Tdp is the dew point temperature, both in degrees Celsius. The methods for determining daytime and nighttime hourly temperatures were published in an earlier paper (6). The calculation scheme
T(t)
= (Tmax
- Tmin) x sin[(1t x t)/(DL + 4)] + Tmin
[2]
T(t)
= Ts -
[(Ts - Tmin)nn(24 - DL)] [3] x In(t)
where Ts is the sunset temperature obtained from Equation [2] with t set to DL, and T(t) is temperature through the night when t = 1 at sunset. Other terms in this equation are as defined for Equation [2]. Stuff and Dale (12) published an algorithm to calculate the climatological day number (CD). Climatological days are numbered from March I rather than January 1 to avoid the February 29th day numbering problem in leap years. Months are numbered, starting with March as 3. This algorithm is as follows: If month
A model developed for the South Carolina coastal plains relates hours with temperature-humidity index values above 74 and 80 to summer season daily milk production. When tested on an independent production data set for 1985, the root mean square model error was less than 1.3 kgld per cow. The model can be used to develop expected summer season dairy production climatologies. Realtime milk production forecasts obtained using daily predicted maximum and minimum temperatures can be used in herd management to reduce effects of heat stress on productivity. (Key words: heat stress, milk production, computer models) Abbreviation key: humidity index.
THI
=
temperature-
INTRODUCTION
Heat stress is defined by Buffington et a1. (2) as any combination of environmental parameters producing conditions that are higher than the temperature range of the animal's thermal neutral zone. The survival and performance of an animal during heat stress periods depend on several weather factors, especially temperature and humidity. Thompson (15) reported that extended exp~ sure to excess heat reduced feed consumption, milk production, and breeding efficiency. Milk production and nonfat milk solids were signifi-
Received July 29, 1991. Accepted April 13, 1992. ITechnicai Contribution Number 3184 of the South Carolina Agricultural Experiment Station, Qemson University, Clemson, SC. 2Departrnent of Agricultura1 and Biological Engineering. 3Departrnent of Animal, Dairy, and Veterinary Science. 1992 J Dairy Sci 75:2598-2604
cantly decreased by thermal stress in a study by Roussel et al. (11). The National Research Council (7) documented dramatic decreases in milk production as temperatures rose above 30·C. The vulnerability of dairy cows to heat stress caused by hot, humid weather is well established. The need to establish dairy herds in warm climates necessitates decisions on methods for heat stress relief. Location of the farm, shade, and supplemental cooling system designs are considerations when herds are located in warm climates. In the experience of South Carolina milk producers, hot, humid weather is more detrimental to production in early summer, when rapid changes in weather conditions are likely to occur, than in mid or late summer. This type of weather is also more detrimental during any season following an extended cool period. Therefore, effective summer season management of dairy herds requires advance warning of impending heat stress periods. Guidance for these warnings must be based on well-accepted heat stress parameters, and warnings must be triggered by operationally useful algorithms. Cattle have the ability to acclimate to changes in the environment. The acclimatization process, however, requires time (5). A cow's response to weather is a complex reaction affected by breed, animal size, level of production, stage of lactation, present and recently past weather, and other factors, such as nutrition and caloric content of feed. Florida researchers (13) studied milk production and breeding efficiency under climatically controlled conditions. Cows in airconditioned facilities produced almost 10% more 4% FCM than cows in facilities that were not air-conditioned. Cows in cooled facilities also displayed increased fertility. In a study conducted by Roman-Ponce et a1. (9), cows placed under shade to remove solar radiational heating produced more milk and had higher conception rates than unshaded cows.
2598
2599
HEAT STRESS EFFECT ON MILK PRODUcrlON
Other researchers related production decline of lactating Holstein dairy cows to the temperature-humidity index (THI) (14) and normal production levels (1). Hahn (4), in turn, used those results to produce maps of expected production from June through September in the United States. Pardue and Linvill (8) reported adverse relationships between high temperature and humidity stress values and milk production in a coastal South Carolina study.
depends on maximum daily temperature (Tmax) occurring 2 h after solar noon and the shape of the daily heating curve responding to the daily solar cycle. This allows the temperature wave from sunrise to sunset to be described by Equation [2]:
MATERIALS AND METHODS
where T(t) is temperature at time t hours after sunrise, Tmin is the morning minimum temperature, and DL is day length in hours. Outgoing radiation from the earth is not balanced by incoming radiation until approximately .5 h after astronomical sunrise (10). Thus, minimum daily temperatures are recorded near sunrise. Thermograph records for Clemson, SC were used to develop a logarithmic expression for the nighttime cooling curve from sunset to sunrise using the assumption that minimum temperature occurs at sunrise:
Production Dat!l Sources
Daily milk production and herd size data were obtained from a large South Carolina coastal plains herd for 1979 through 1985. Total per cow production from June through September ranged from a low of 2216 kg in 1984 to 2487 kg in 1982. These data were converted to average production per cow for each day, and data for 1979 through 1984 were used in model development, and data for 1985 were used for model testing. No attempt was made to account for production changes that were due to freshening cycles or age of the herd or for year to year fluctuations in normal production levels. Temperature records from a nearby National Weather Service Cooperative Station and hourly data from the National Weather Service Office in Charleston, SC were used to characterize weather on the farm during these years. Calculation of the Environmental Variables
The THI equation exists in three forms. One form expresses the relationship using air temperature and relative humidity. A second form uses air temperature and wet bulb temperature. The relationship used to calculate THI in the present study is THI
= Ta +
.36 x Tdp + 41.2
[1]
where T a is air temperature and Tdp is the dew point temperature, both in degrees Celsius. The methods for determining daytime and nighttime hourly temperatures were published in an earlier paper (6). The calculation scheme
T(t)
= (Tmax
- Tmin) x sin[(1t x t)/(DL + 4)] + Tmin
[2]
T(t)
= Ts -
[(Ts - Tmin)nn(24 - DL)] [3] x In(t)
where Ts is the sunset temperature obtained from Equation [2] with t set to DL, and T(t) is temperature through the night when t = 1 at sunset. Other terms in this equation are as defined for Equation [2]. Stuff and Dale (12) published an algorithm to calculate the climatological day number (CD). Climatological days are numbered from March I rather than January 1 to avoid the February 29th day numbering problem in leap years. Months are numbered, starting with March as 3. This algorithm is as follows: If month
