variation (c) and temperature-humidity index (d) from 2002 to 2015
| Livestock Research for Rural Development 31 (12) 2019 | LRRD Misssion | Guide for preparation of papers | LRRD Newsletter | Citation of this paper |
This research aimed to characterize changes in average temperature (AT), relative humidity (RH), diurnal temperature variation (DTV), and temperature-humidity index (THI), and to determine the effects of stress levels (SL) on milk yield (MY) in Thai multibreed dairy cattle. The AT and RH records (n = 85,824) measured by 17 meteorological stations across Thailand were used to calculate DTV and THI. Climate variables were analyzed using a linear model. The effect of SL on MY was estimated by using milk records (n = 46,442) from 5,080 first-lactation cows raised in 456 farms with a model that included calving herd-year-season, breed group (BG), days in milk, SL, calving age, and interaction between BG and SL as fixed effects, and residual as random effect. All fixed effects were important for all climate variables (p < 0.01). The highest percentages of THI were in the moderate stress level (67%) for all regions. The SL, BG and interaction between SL and BG were important factors for MY (p < 0.01). Cows in BG4, BG5, and BG6 tended to have higher MY than other breed groups including purebred H in the comfort zone and mild stress level, and also had higher MY than BG1, BG2, and BG3 under a moderate stress level. Cows from all breed groups had higher MY than under mild and moderate stress levels. Thus, to increase MY per cow under increasingly stressful climate conditions, farmers would need to enlarge their investment in active cooling solutions to reduce temperature and humidity
Keywords: climate change, temperature-humidity index, stress, dairy cattle, tropics
Global climate is changing, and it is manifesting itself through various kinds of environmental changes. Extreme droughts, high ambient temperatures, heat waves, floods, and freak storms occurred across the globe and affected agriculture and productivity of livestock systems (IPCC 2007), especially in tropical areas. The tropical land mass covers over 20% of the earth’s total land surface, and as many as 62 countries are in this region (Oliver and Hidore 1984). Most agriculture and livestock production are located in tropical regions characterized by hot and humid weather conditions (Zhao et al 2005; Key et al 2014). Thailand considered a tropical country with consistently high ambient temperatures and high relative humidity throughout the year (OEPP 2000), has two monsoons every year (TMD 2014): a northeast monsoon between October and February and a southwest monsoon between May and October. High temperature and humidity could create environmental stress and affect dairy cattle productivity, especially in open-house operations which are the predominant dairy cattle housing system in Thailand.
Commercial dairy cattle farming in Thailand began in 1957. To increase milk production, Holstein and other European dairy cattle breeds (Jersey, Brown Swiss, and Red Dane) have been used in crossbreeding and upgrading mating schemes. The current Thai dairy population is multibreed (Koonawootrittriron et al 2009) and most animals (91%) are crossbred and have a Holstein fraction higher than 87.5% (Ritsawai et al 2014). High ambient temperature and relative humidity create uncomfortable and stressful conditions that have a negative effect on feed intake, reproduction and production of dairy cattle, especially lactating cows (Bouraoui et al 2002; Kadzere et al 2002).
The Temperature-Humidity Index (THI) is an indicator to assess the level of stress caused by high ambient temperature and humidity, and THI values are examined to determine whether animals are in the comfort or stress zones (de la Casa and Ravelo 2003; Bohmanova et al 2007). Dairy cows are likely to begin experiencing heat stress when THI values exceed 72 units (Armstrong 1994). Considering proportion of THI by stress level of dairy cows per season would help increase dairy production efficiency and prevent loss of annual income in dairy operations (St-Pierre et al 2003). Thus, the objectives of this research were:
Climate variables consisted of 85,824 daily records of average temperature (AT; ºC), diurnal temperature variation (DTV; ºC), relative humidity (RH; %), and temperature-humidity index (THI) obtained from January 1, 2002 to December 31, 2015 at 17 meteorological stations of the Thai Meteorological Department (TMD). The DTV were calculated as differences between maximum and minimum daily temperatures (MNT; ºC). The THI was calculated for all pairs of AT and RH records using the expression (NOAA 1976): THI = (1.8 × AT + 32) – (0.55 – 0.0055 × RH) × (1.8 × AT – 26), where AT is the dry-bulb average temperature in ºC, and RH is the relative humidity from a hygrometer (%). The meteorological stations, located in five regions of Thailand (latitudes 5° 37’ N to 20° 27’ N and longitudes 97° 22’ E to 105° 37’ E), were chosen because they were in high-density dairy population areas. The characteristics of each region are as follows (OEPP 2000):
The CT, ET, NE, and NT are classified as tropical savanna (AW), whereas the ST is classified as tropical rainforest (AF) in the Köppen climate classification system. Thailand has three seasons: winter (November to February), summer (March to June), and rainy (July to October).
Animals utilized in this research were 5,080 first-lactation cows. Cows were from 456 dairy farms in Central (61 farms), Northeastern (226 farms), Northern (79 farms), and Southern Thailand (90 farms). Cows were daughters of 771 sires and 4,385 dams. The main breed represented in an individual cow was Holstein (H), and most cows (94%) were crossbred (75% H and above). The other breeds represented in an individual animal were various Bos indicus (Brahman, Sahiwal, Red Sindhi, Sahiwal, and Thai Native) and Bos taurus breeds (Jersey, Brown Swiss, and Red Dane).
Cows were housed in open barns with access to roughage, concentrate, and mineral supplement. Available fresh grasses were Guinea (Penicum maximum), Ruzi (Brachiaria ruziziensis), Napier (Pennisetum parpureum), and Para (Brachiaria mutica). Concentrate (14 to 22% of crude protein and 63 to 83% of nitrogen-free extract) was provided to cows twice a day during milking (4:30 a.m. to 7:00 a.m. in the morning, and 2:30 p.m. to 4:30 p.m. in the afternoon). Additionally, cows were given silage and agro-industrial by-products (rice straw, pineapple peel, and sweet corn husk) when fresh grass was insufficient in winter and summer. The production trait was test-day milk yield (MY) from first-lactation cows. The MY were collected monthly from 2002 to 2015 (n = 46,442). The mean for MY per cow was 13.9 kg/d (SD = 4.83 kg/d).
The statistical model used to analyze the four climate variables (AT, RH, DTV, and THI) included the main effects of region, year, season, and year-region, year-season, region-season, and region-year-season interactions as fixed effects, and residual as a random effect. Least squares means (LSM) of fixed effects were computed using the GLM procedure of SAS (SAS Inst. Inc., Cary, NC, USA).
To assess the effect of stress level (SL) on MY, SL were classified into three categories using THI values (Armstrong 1994):
Percentages of this SL were determined for each region, month, season, and year. Then, the THI values from the closest meteorological stations to each dairy farm (based on postal code) were used to calculate an average of daily THI three days before each MY. Lastly, these average THI values were used to classify the stress level of dairy cows for each MY at each farm.
The statistical model used to analyze MY included the fixed effects of contemporary group (herd-year-season), breed group (BG; BG1 = H < 0.50, BG2 = 0.50 ≤ H < 0.75, BG3 = 0.75 ≤ H < 0.875, BG4 = 0.875 ≤ H < 0.9375, BG5 = 0.9375 ≤ H < 0.9687, BG6 = 0.9687 ≤ H < 1.0, and BG7 = Holstein), days in milk, stress level (SL; comfort zone, mild stress, and moderate stress), calving age, and BG-SL interaction. The random effect was residual. LSM of fixed effects were computed using the GLM procedure of SAS (SAS Inst. Inc., Cary, NC, USA). Significant differences were considered at α = 0.05. LSM of fixed effects were compared using adjusted Bonferroni t-tests.
The distributions of AT, RH, DTV, and THI data during the entire period of the study (January 1, 2002, to December 31, 2015) are shown in Figure 1. Values ranged from 14.6 to 35.5 ºC for AT (Figure 1a), from 29 to 99 % for RH (Figure 1b), from 0.3 to 19 ºC for DTV (Figure 1c), and from 58.3 to 88.8 units for THI (Figure 1d). Means were 27.9 ºC (SD = 2.56 ºC) for AT, 73.6 % (SD = 9.65 %) for RH, 9.9 ºC (SD = 2.94 ºC) for DTV, and 78.7 units (SD = 4.03 units) for THI. The mean of LSM for THI was within the range of values previously estimated in Thailand (73 units to 80 units from January 1990 to December 2008; Boonkum et al 2011). However, all of them were higher than THI estimated under temperate climate conditions in Europe (40.8 units to 79.9 units in May 2010 to October 2012; Schüller et al 2014).
|
| Figure 1. Distribution of average temperature (a), relative humidity (b), diurnal temperature variation (c) and temperature-humidity index (d) from 2002 to 2015 |
Variation in all climate variables (AT, RH, DTV, and THI) in Thailand was affected by differences among regions, years, seasons, and year-region, year-season, region-season, and region-year-season interactions (p < 0.01). Significant interactions among main effects (region, year, and season) indicated that each of them influenced the value of the other two in this population, and none of them could be estimated unbiasedly. The LSM estimates of climate variables for within region-year-season subclasses and their standard errors ranged from 23.4 ± 0.07 ºC (Northeastern-2008-winter) to 31.2 ± 0.08 ºC (Central-2010-summer) for AT, from 58.2 ± 0.41 % (Northern-2010-summer) to 86.6 ± 0.71 % (Eastern-2013-rainy) for RH, from 7.1 ± 0.16 ºC (Southern-2007-rainy) to 15.5 ± 0.13 ºC (Northern-2004-winter) for DTV, and from 71.2 ± 0.11 units (Northeastern-2008-winter) to 83.4 ± 0.24 units (Eastern-2013-summer) for THI.
Trends for year-region LSM for AT, RH, DTV, and THI are shown in Figure 2 for the summer season, Figure 3 for the rainy season, and Figure 4 for the winter season. In the summer and winter seasons, all climate variables (AT; Figures. 2a and 4a, RH; Figures. 2b and 4b, DTV; Figure 2c, and THI; Figures. 2d and 4d) tended to have wider fluctuations in the last seven years (2009 to 2015) than during the first seven years of the study (2002 to 2008), except for DTV in the winter season that had wider fluctuations during the first seven years (2002 to 2008; Figure 4c). Conversely, in the rainy season, all climate variables tended to have small fluctuations during the entire period of the study. The expansion of variability in climate variables was widespread between 1981 and 2010, and it was more evident in Asia probably due to the unusually strong El Niño Southern Oscillation (ENSO) in 1997 to 1998 (Hansen et al 2012). Although increasing trends in THI did not occur during the years of the study, variability in THI increased since 2009. This variability was likely associated with ENSO, which was a primary factor for temperature fluctuations in large areas of Southeast Asia. Moreover, ENSO affected many countries in temperate and tropical Asia causing frequent severe droughts and floods in the 20th century. El Niño Southern Oscillation is associated with high temperatures and droughts whereas another Southern Oscillation, called La Niña, is associated with heavy precipitation and flooding. The combined ENSO phenomena of El Niño and La Niña are the main factors leading to climate changes generating uncertain temperature and rainfall patterns (Díaz et al 2002). Further, Segnalini et al (2011) indicated that THI changes in the Mediterranean basin caused extreme climate events between 1998 and 2007. Dairy industry losses in the United States due to climate change amounted to more than $1 billion (Collier and Zimbelman 2007). To help decrease the negative effects of El Niño and La Niña Southern Oscillations, farmers should implement selection and management strategies suitable for unpredictable climate changes.
 |
| Figure 2. Least squares means trends per regions; Central region (CT), Eastern region (ET), Northern region (NT), Northeastern region (NE) and Southern region (ST) during the year 2002 to 2015 for average temperature (a), relative humidity (b), diurnal temperature variation (c), and temperature-humidity index (d) that occurred in the summer season |
 |
| Figure 3. Least squares means trends per regions; Central region (CT), Eastern region (ET), Northern region (NT), Northeastern region (NE) and Southern region (ST) during the year 2002 to 2015 for average temperature (a), relative humidity (b), diurnal temperature variation (c), and temperature-humidity index (d) that occurred in the rainy season |
 |
| Figure 4. Least squares means trends per regions; Central region (CT), Eastern region (ET), Northern region (NT), Northeastern region (NE) and Southern region (ST) during the year 2002 to 2015 for average temperature (a), relative humidity (b), diurnal temperature variation (c), and temperature-humidity index (d) that occurred in the winter season |
Trends for year-season LSM for AT, RH, DTV, and THI in each region (Figures 2 to 4) indicated that the highest AT (Figure 2a) and THI (Figure 2d) occurred in the summer season, the highest RH (Figure 3b) in rainy season, and the highest DTV (Figure 4c) in the winter season. Further, most of the highest LSM for AT, RH, and THI estimated within seasons were from the Eastern region, except for the winter season in the Southern region where LSM for RH was higher than in other Thai regions. However, the highest LSM for DTV was from the Northern region in all three seasons. Interestingly, all regions in Thailand had THI LSM values higher than 72 units indicating that the weather in all regions of the country created stressful conditions to dairy cows. Ravagnolo et al (2000) found that MY decreased by 0.2 kg per unit of increase in THI when THI exceeded 72 units. Moreover, the negative impact of increases in temperature not only includes milk production, but also the availability of food, water, and the spread of infectious diseases. Nearly all dairy farms in Thailand kept lactating cows in open barns; thus, both ambient temperature and relative humidity directly affect the stress level of dairy cows. Utilization of THI information should indicate farmers when to provide cows with appropriate cooling solutions to mitigate climate stress.
Percentages of THI by stress level and region during the entire period of the study are shown in Figure 5. The overall percentage of moderate stress level (67.4 %) was higher than the one for mild stress level (24.3 %) and comfort zone (8.3 %). Further, the percentage of THI in moderate stress level was higher than for other levels in all regions. Percentages of THI in the moderate stress level by region ranged from 52.2 % in the Northeastern region to 81.2 % in the Eastern region. The percentages of THI in the mild stress level in the Northeastern region (34 %) was higher than in the Northern (29.2 %), Southern (17.9 %), Central (17.7 %), and Eastern (16 %) regions. The percentage of THI in the comfort zone ranged from 2.1 % in the Southern region to 13.8 % in the Northeastern region. The lower percentages of THI for the moderate stress level in the Northern and Northeastern regions indicated that these regions were more favorable to dairy farming than the Central, Eastern and Southern regions. The Northern and Northeastern regions are higher than other regions and are more influenced by cold weather from the northeastern monsoon than the other regions in Thailand. These results pointed out that the need for more comprehensive strategies to reduce heat stress of dairy cattle in the Central, Eastern, and Southern regions is more significant than in the Northern and Northeastern regions.
 |
| Figure 5. Percentages of daily temperature-humidity indexes by heat stress level and region (Central region (CT), Eastern region (ET), Northern region (NT), Northeastern region (NE) and Southern region (ST)) of Thailand |
 |
| Figure 6. Percentages of daily temperature-humidity indexes by heat stress level and month and season of Thailand |
Figure 6 shows the percentages of THI by stress level, month, and season in Thailand. The percentages of THI in the stress levels by month ranged from 12.7 % (January) to 95.2 % (June) for moderate stress level, from 4.8 % (June) to 48.8 % (January) for mild stress level, and from 0 % (June, July, and August) to 38.5 % (January) for comfort zone. Trends for the percentage of THI by stress level and month (Figure 6) indicated that the frequency of THI in the moderate stress level increased after January until the June and then decreased after June until the December, which was contrary to the frequency of THI in the mild stress level and comfort zone. In addition, most of the months with high percentages of THI in the comfort zone and mild stress level were associated with the winter season. Thus, the winter season in Thailand has the highest frequency of THI in the comfort zone (23.7 %) and mild stress level (47.5 %) and the lowest frequency of THI in the moderate stress level (28.7 %; Figure 6). In the summer season, the percentage of THI in the comfort zone (1.2 %) and moderate stress level (87.1 %) are higher than in the rainy season. Conversely, the percentage of THI in the mild stress level (11.7 %) is lower than in the rainy season (0.2 % for comfort zone, 14.1 % for mild stress level and 85.7 % for moderate level). Variation of monthly percentages of THI in the three stress levels may have been associated with differences in MY per month of dairy cows in Thailand reported by Boonkum et al (2011) who found that more dairy cows with high MY occurred in the winter than in the summer season. However, trends for yearly percentages of THI by stress level within regions (Figure 7) indicated that all regions in Thailand had a higher frequency of THI values in the moderate stress level than in other stress levels. This result supported that the weather in all regions of the country created stressful conditions for dairy cows. Consequently, dairy farmers in all regions need to implement management strategies to mitigate the impact of high THI, such as the use of fans to reduce relative humidity and shades to reduce ambient temperatures.
 |
| Figure 7. The trend for percentages of daily temperature-humidity indexes by heat stress level occurred in each region of Thailand (2002 to 2015) |
The correlation between MY and THI was close to zero (r = 0.008) indicating that there was no correspondence between MY of animals and their stress to a unit increase in THI. However, the stress level of dairy cows based on THI values influenced MY in the Thai multibreed dairy population (p < 0.01). Cows in the comfort zone (14.2 ± 0.08 kg per day) and mild stress level (14.1 ± 0.05 kg per day) had higher LSM for MY than cows in the moderate stress level (13.9 ± 0.05 kg per day; p < 0.05). The lower MY of dairy cows in the are stress level (78 to 88 units for THI) may be due to heat stress which directly reduces dry matter intake, creating a negative energy balance that reduces milk synthesis (West 2003; Rhoads et al 2009; Wheelock et al 2010). In addition, heat stress can affect the rumen function and udder health of dairy cows resulting in lower milk production (Pragna et al 2016). Even though MY of Thai dairy cows trended to increase with decreased levels of stress, it was a little bit of amount of increased MY from moderate stress level to comfort zone (0.3 kg per day). The small difference in MY between cows from moderate stress level and comfort zone was consistent with previous findings in this population (Boonkum et al 2010) Holstein-Frisian in Serbia (Könyves et al 2017) and Holstein in Turkey (Duru 2018), who reported that MY declined 0.002 - 0.07 kg per unit increase in THI. Further, Boonkum et al (2011) indicated that first-lactation Thai multibreed dairy cows with H ≥ 93.6% had higher reductions in MY due to heat stress than cows with H percentages lower than 93.6%. Nearly all dairy cattle in Thailand have high Holstein percentages (Koonawootrittriron et al 2009), making them susceptible to high THI conditions. Thus, Thai dairy farmers need to utilize appropriate management practices to reduce dairy cow stress during high THI periods. In addition, Thai dairy farmers need to select dairy animals for both adaptability and productivity under high THI conditions.
Stress level defined in terms of THI values had nonsignificant effects on MY from dairy cows in BG1, BG2, BG3, and BG6 and purebred H, but it affected MY from cows in BG4 and BG5 (p < 0.05). The highest MY for cows in BG4 (14.6 ± 0.08 kg/day/cow) and BG5 (14.6 ± 0.10 kg/day/cow) occurred in the comfort zone. However, the MY LSM for cows in BG5 under mild stress was not significantly different from the MY LSM of cows in other stress levels. The lack of a significant effect of stress level on MY in crossbred cows with H percentages lower than 87.5% was in agreement with a previous study in the Thai multibreed dairy cattle population, where Holstein crossbred cows (H < 87%) were able to maintain milk production under high heat stress (Boonkum et al 2011).
Table 1 shows the LSM for MY per breed group and stress level of dairy cows. The MY LSM ranged from 13.3 ± 0.26 kg/day/cow (BG7; purebred H) to 14.6 ± 0.10 kg/day/cow (BG5) for comfort zone, from 13.7 ± 0.14 kg/day/cow (BG7; purebred H) to 14.3 ± 0.23 kg/day/cow (BG1) for mild stress level, and from 13.4 ± 0.18 kg/day/cow (BG1) to 14.2 ± 0.04 kg/day/cow (BG4) for moderate stress level. Breed group differences were significant for MY at each stress level (p < 0.01). Cows in BG4, BG5, and BG6 on the comfort zone and mild stress level had higher MY LSM than purebred H (p < 0.05). However, under moderate stress level, MY LSM for purebred H cows was not significantly different from MY LSM for crossbred cows (BG1 to BG6), but cows in BG4 and BG5 had higher MY LSM than cows in other breed groups including purebred H (p < 0.05). As previously reported in this population, crossbred cows in BG4, BG5, and BG6 (0.875 ≤ H< 1.0) tended to have higher MY than cows in other breed groups (Koonawootrittriron et al 2009; Boonkum et al 2011). Although crossbred cows with larger H fractions tended to have higher MY than cows with lower H fractions, they also tended to have lower MY as the stress level increased. Current cooling strategies would need to be improved if crossbred cows under mild and moderate stress are to achieve similar MY to cows in the comfort zone (Table 1). However, Koonawootrittriron et al 2009 and Seangjun et al 2009 indicated that not only farmers in Thailand have to manage for reducing stress due to hot and humid of dairy cows but also they have to provide appropriate nutrition and health care so that their cows can produce higher MY. Thus, to increase the amount of milk produced per cow under increasingly stressful climate conditions in Thailand, farmers would need to increase their investment in active cooling solutions to reduce ambient temperature and relative humidity as well as in nutrition and health care.
| Table 1. Least squares means and SE per heat stress level and breed groups (BG) for milk yield per day of the Thai multibreed dairy cattle population | |||||
| BG | Holstein Fraction | N | Milk yield (kg/d) | ||
| Stress level | |||||
|
Comfort zone (72 < THI) |
Mild stress (72 ≤ THI < 77) |
Moderate stress (77 ≤ THI < 88) |
|||
| 1 | H < 0.50 | 91 | 13.7 ± 0.37ab | 14.3 ± 0.23ab | 13.4 ± 0.18b |
| 2 | 0.50 ≤ H < 0.75 | 189 | 14.5 ± 0.29ab | 13.8 ± 0.14ab | 13.6 ± 0.13b |
| 3 | 0.75 ≤ H < 0.875 | 960 | 14.1 ± 0.12ab | 14.0 ± 0.06ab | 13.8 ± 0.06b |
| 4 | 0.875 ≤ H < 0.9375 | 1,752 | 14.6 ± 0.08a | 14.3 ± 0.05a | 14.2 ± 0.04a |
| 5 | 0.0.9375 ≤ H < 0.9687 | 1,202 | 14.6 ± 0.10a | 14.3 ± 0.05a | 14.2 ± 0.06a |
| 6 | 0.9687 ≤ H < 1.0 | 742 | 14.4 ± 0.12a | 14.2 ± 0.07a | 14.0 ± 0.08ab |
| 7 | Holstein (H) | 144 | 13.3 ± 0.26b | 13.7 ± 0.14b | 14.1 ± 0.18ab |
| a,b Least squares means within a column with different superscript letters differ (p<0.05) | |||||
The authors would like to thank the Graduate School of Kasetsart University and Kasetsart University Research and Development Institute (KURDI; S-K 21.56) for supporting the research scholarship, and Thai Meteorological Department for providing climate records in this research.
Armstrong D V 1994 Heat stress interaction with shade and cooling. Journal of Dairy Science 77(7): 2044-2050.
Bohmanova J, Misztal I and Cole J B 2007 Temperature-humidity indices as indicators of milk production losses due to heat stress. Journal of Dairy Science 90(4): 1947-1956.
Boonkum W, Duangjinda M, Pattarajinda V, Tumwasorn S, Sanpote J and Buaban S 2010 The critical point of temperature humidity index and Holstein fraction level affected on milk production and genetic parameters during heat stress condition. Khon Kaen Agriculture Journal 38: 275-284.
Boonkum W, Misztal I, Duangjinda M, Pattarajinda V, Tumwasorn S and Sanpote J 2011 Genetic effects of heat stress on milk yield of Thai Holstein crossbreds. Journal of Dairy Science 94(1): 487-492.
Bouraoui R, Lahnar M, Majdouc A, Djemali M N and Belyea R 2002 The relationship of the temperature-humidity index with milk production of dairy cows in a Mediterranean climate. Animal Research 51: 479-491.
Collier R J and Zimbelman R B 2007 Heat stress effects on cattle: what we know and what we don’t know. In: 22nd Southwest Nutrition & Management Conference, Arizona, USA.
de la Casa A C and Ravelo A C 2003 Assessing temperature and humidity conditions for dairy cattle in Córdoba, Argentina. International Journal of Biometeorology 48(1): 6-9.
Díaz J, García R, Velázquez de Castro F, Hernández E, López C and Otero A 2002 Effect of extremely hot days on people older than 65 years in Seville (Spain) from 1986 to 1997. International Journal of Biometeorology 46(3)-145-149.
Duru S 2018 Determination of starting level of heat stress on dairy milk yield in Holstein cows in Bursa city of Turkey. Ankara Üniversitesi Veteriner Fakültesi Dergisi 65: 193-198.
Hansen J, Sato M and Ruedy R 2012 Perception of climate change. Proceeding of the National Academy of Sciences of the United States of America 109(37): E2415- E2423.
IPPC (Intergovernmental Panel on Climate Change) 2007 Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt K B, Tignor M and Miller H L (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
Kadzere C T, Murphy M R, Silanikove N and Maltz E 2002 Heat stress in lactating dairy cows: a review. Livestock Production Science 77(1): 59-91.
Key N, Sneeringer S and Marquardt D 2014 Climate change, heat stress and U.S. dairy production. United States Department of Agriculture, USA.
Könyves T, Zlatković N, Memiši N, Lukač D, Puvača N, Stojšin M, Halász A and Miščević B 2017 Relationship of the temperature-humidity index with milk production and feed intake of Holstein-frisian cows in different year seasons. The Thai Journal of Veterinary Medicine 47(1): 15-23.
Koonawootrittriron S, Elzo M A and Thongprapi T 2009 Genetic trend in a Holstein × other breeds multibreed dairy population in Central Thailand. Livestock Science 122 (2-3): 186-192.
NOAA (National Oceanic and Atmospheric Administration) 1976 Livestock hot weather stress. Department of Commerce, NOAA, National Weather Service Central Region, Kansas City.
OEPP (Office of Environmental Policy and Planning) 2000 Thailand’s initial national communication under the united framework convention on climate change. Office of Environmental Policy and Planning, Thailand.
Oliver J E and Hidore J J 1984 Climatology: An introduction. Charles E. Merrill Publishing Company and Bell & Howell Company. Columbus, Ohio. p. 381.
Pragna P, Archana P R, Aleena J, Sejian V, Krishnan G, Bagath M, Manimaran A, Beena V, Kurien E K, Varma G and Bhatta R 2016 Heat stress and dairy cow: Impact on both milk yield and composition. International Journal of Dairy Science 12(1): 1-11.
Ravagnolo O, Misztal I and Hoogenboom G 2000 Genetic component of heat stress in dairy cattle, development of heat index function. Journal of Dairy Science 83(9): 2120-2125
Rhoads M L, Rhoads R P, VanBaale M J, Collier R J, Sanders S R, Weber B A, Crooker B A and Baumgard L H 2009 Effects of heat stress and the plane of nutrition on lactating Holstein cows: I. production, metabolism, and aspects of circulating somatotropin. Journal of Dairy Science 92(5): 1986-1997.
Ritsawai P, Koonawootrittriron S, Jattawa D, Suwanasopee T and Elzo M A 2014 Fraction of cattle breeds and their influence on milk production of Thai dairy cattle. In: Proceedings of 52nd Kasetsart University Annual Conference (Animal and Veterinary Medicine), Bangkok, Thailand.
Seangjun A, Koonawootrittriron S and Elzo M A 2009 Characterization of lactation patterns and milk yield in a multibreed dairy cattle population in Central Thailand. Kasetsart Journal (Natural Science) 43: 74-82.
Schüller L K, Burfeind O and Heuwieser W 2014 Impact of heat stress on conception rate of dairy cows in the moderate climate considering different temperature-humidity index thresholds, periods relative to breeding and heat load indices. Theriogenology 81(8): 1050-1057.
Segnalini M, Nardone A, Bernabucci U, Vitali A, Ronchi B and Lacetera N 2011 Dynamics of the temperature-humidity index in the Mediterranean basin. International Journal of Biometeorology 55(2): 253-263.
St-Pierre N R, Cobanov B and Schnitkey G 2003 Economic losses from heat stress by US livestock Industries. Journal of Dairy Science 86(Suppl.): E52-E77.
TMD (Thai Meteorological Department) 2014 Thailand weather. Thai Meteorological Department. Ministry of Information and Communication Technology, Thailand.
West J W 2003 Effects of heat-stress on production in dairy cattle. Journal of Dairy Science 86(6): 2131-2144.
Wheelock J B, Rhoads R P, Vanbaale M J, Sanders S R and Baumqard L H 2010 Effects of heat stress on energetic metabolism in lactating Holstein cows. Journal of Dairy Science 93(2): 644-655.
Zhao Y, Wang C, Wang S and Tibig LV 2005 Impact of present and future climate variability on agriculture and forestry in humid and sub-humid tropics. Climatic change 70(1-2): 73-116.
Received 9 August 2019; Accepted 11 November 2019; Published 2 December 2019