The study aimed to determine the effect of temperature and humidity (ITH) variations on cow dairy production in Las Tunas, Cuba. The work was carried out between January 2017 and December 2018. The statistical analysis was carried out, using the multivariate MLG procedure through a variance analysis, and a binary logistic regression analysis to estimate the risk factors affecting milk production, concluding that the problems caused by caloric stress for livestock are evident and accentuated by climate change, reflected in variations in temperature and humidity that negatively impact on dairy production.

Heat stress has been widely recognized as a factor affecting the productivity, reproductive efficiency and health of production animals (Allen et al., 2015; Boni and Cecchini, 2014; Dash et al., 2016; Ruiz-García et al., 2018). In cattle ranching, temperature and humidity indices (Allen et al., 2015) and relative humidity (HR) have been used as indicators of animal welfare; considering rates greater than 74 as stressful (Aguilar et al., 2014; Ruiz and Sandoval-Monzón, 2019).

A slight decrease in production can be observed in dairy cattle in short periods of heat stress, but the consequences can be serious in prolonged periods of heat stress (St-Pierrre and Schnitkey, 2003; Ruiz-García and Sandoval-Monzón, 2018). Cattle respond to this type of stress by reducing feed consumption and increasing water consumption, in loss of water by evaporation and with increased respiratory rate and body temperature (West et al., 2003; Wheelock et al., 2010; Allen et al., 2015).

The great specialization of animals in milk production, translated by their great efficiency in the use of ingested food, triggers a high production of metabolic heat, making them more sensitive and susceptible to heat stress (Cerqueira et al., 2016). Furthermore, as a result of its action on food intake, heat stress influences the metabolism of the mammary gland and the composition of milk (Cerqueira et al., 2016). The objective of the present work was to determine the affectation of milk production by the effect of variations in temperature and humidity.

This work was carried out between January 2017 and December 2018 in the Vaquería 17 of the Basic Unit of Agricultural Production (UBPC) Maniabo, located in the Jobabo municipality. Jobabo is one of the eight municipalities in the province of Las Tunas, it is located to the southeast of the province, bordering the Gulf of Guacanayabo to the south, the municipalities of Tunas and Río Cauto to the east, Colombia and Guáimaro to the west and the north with the political divisions of Guáimaro and the Tunas municipality; Las Tunas province, Cuba.

The area has a tropical climate, with relatively humid summers, with long dry periods prevailing. Relative humidity is 85%. The duration of the rainy period is just over 80 days, with June and May being the most intense months, with rainfall averages of 219 and 179 mm respectively, while the least rainy are December and January, with 15 and 20 mm, due to their order. The circulation of the winds predominates from north to south, although its monthly behavior is variable, presenting maximum values in the months of September and October, with registered average speeds of 9.9 km/ h, especially in this last month.

Trade winds predominate, counteracted by breezes and reinforced by storms. The temperature is warm, predominantly partly cloudy skies in the period from April to August and slightly lower between December and March. The annual average temperature is 33ºC. The climate behaves with two well defined periods: consistent summer or rainy that was divided from June 1 to October 31 and dry or winter from November 1 to May 31.

The dairy population during the study period is shown in Table 1. The dairy has an area of 131 ha, of which 31 are planted with Morerus alba mulberry, 20 ha with sugar cane Saccharum officinarum and 15 ha with titonia, and the rest are occupied by natural pastures. The paddocks are divided at a rate of 1 ha by electric fences. The plants with the highest protein content are dried and used mainly as a supplement at a rate of 2 kg / animal. The cane is ground and supplied together with the protein plants at a rate of 1 kg / animal. The supplement is provided during the afternoon-night when the animals are housed.

The exploited dairy genotype is Siboney de Cuba. Calves are sent to a rearing center and milking is carried out twice a day. The water supply is continuous. The animals start grazing after the first milking (04: 00-05: 00). The second milking is around 17: 00-18: 00.

Data from meteorological records were collected through the national agrometeorological bulletins of the Center for Agricultural Meteorology (Ministry of Science, Technology and Environment) (CITMA, 2017), and caloric stress was determined through the temperature and humidity index ( ITH) for months according to the modification proposed by Valtorta (1996), where ITH = (1.8 Ta + 32) - (0.55 - 0.55 HR / 100) (1.8 Ta - 26), where Ta = average air temperature (ºC) and RH = relative humidity (%).

The ITH scale defines critical points or levels of severity of caloric stress (Zimbelman and Collier, 2011; López et al., 2016):
- ITH - ITH = 72-79: Alert (moderate caloric stress)
- ITH = 80-89: Danger (moderate to severe heat stress)
- ITH> 90-98: Emergency (severe caloric stress)

The statistical analysis was carried out by means of the multivariate MLG procedure through an analysis of variance, posing as dependent variables the monthly milk production and daily average milk per cow; as a covariate, milking cows were used as independent variables, season and the temperature and humidity index (ITH). For statistical analyzes, the SPSS v. 23 for Windows, with a significance level of 5%.

The average milk production during the period 2017-2018 was 8962.1 ± 423.5 l / cow milking. Total milk production in 2017 was 81 213 L and in 2018 123 878 L, appreciating a difference of 32 665 L. It is interesting to note that the highest productive yield was obtained in the hottest months, a period that coincides with the higher yield of pastures and more abundance of water, as it is the spring season in Cuba.

The ITH was always greater than 72 in all the months (Table 2), especially in the summer months. The months with the highest ITH were June, July, August and September. On the other hand, caloric stress not only occurs in the summer months, but is a problem present most of the year. Normally, in August the average temperature increases with respect to June and July, and it is frequently a very hot month, the hottest of the year.

Table 3 shows the presence of significant differences in the two dependent variables considered in the model (monthly milk production and daily average milk per cow). The covariate milking cows showed significant differences when the monthly milk production was evaluated. The variables of factor, season and temperature and humidity index showed significant differences compared to the two dependent variables. Monthly milk production reached an adjusted R2 = 0,921 and the daily average milk per cow an adjusted R² squared = 0,775.

Caloric stress has been widely recognized as one of the factors that affect the productivity and reproductive efficiency of production animals in different parts of the world (Bernabucci et al., 2014; Molina, 2017; García and Monzón, 2018). Caloric stress has a significant impact on all livestock species causing economic losses (Olivera et al., 2015; Sandoval and Carcelén, 2017). St-Pierre et al., (2003) and García and Monzón (2018) propose annual economic losses produced by heat stress of the order of US $ 897-1500 million for the dairy industry in the United States and of $ 369 million for the meat production in Peru, respectively.

Bos indicus is known to be more thermotolerant when subjected to heat shock (Bó and Martínez, 2003; Espinoza et al., 2011). Perhaps, this aspect helps to explain the peak of estrous presentation of the Holstein x Cebu genotypes reported by Bertot et al. (2009) in the hottest months. Likewise, Peña et al. (2012) indicate that the highest productive yield is obtained with mixed breeds and not with pure breeds in the Cuban province of Camagüey.

The ideal climatic conditions for milk production occur at room temperature between 5° and 25°C, this interval being considered the thermal comfort zone; cows tolerance to temperatures below 5 ° C depends on age and level of milk production (Cerqueira et al., 2016).

Heat stress can reduce the conception rate to levels of 10% and milk production can decrease between 10 and 30%, in part due to a decrease in food intake, although the main reason is due to a direct effect of caloric stress (Flamenbaum, 2013). One of the alternatives to reduce the effects of high summer temperatures on animals of temperate breeds is the use of shades and cooling systems.

Heat stress in high-production dairy cattle results in less lactose synthesis and thus a reduction in milk production. Likewise, it affects the estrous cycle by altering the follicular selection process and increases the duration of the follicular waves, which results in a lower quality of the oocytes, as well as a reduction in the duration and intensity of the estrus (Pedersen, 2014).

Cerqueira et al. (2016) states that a significant effect of the ITH was registered in milk production per day, when the ITH value was above 78, revealing a lower production of the order of 1.8 kg / cow / day; results that coincide with the present study.

The problems caused by heat stress for livestock are evident and accentuated by climate change, reflected in variations in temperature and humidity that negatively impact milk production.