CLOACAL AND SURFACE TEMPERATURES OF BROILERS SUBJECT TO THERMAL STRESS

The cloacal and surface temperatures are important physiological parameters to quantify the thermal comfort of animals, its variations may occur as a function of thermal variables. In this context, the goal of the present study was to evaluate the cloacal (tclo) and surface (tsur) temperatures of Cobb 500 broilers, in the second week of life, subjected to different air dry-bulb temperature (24, 27, 30 and 33 °C) and stress durations (1, 2, 3 and 4 days). The experiment was carried out in four climate-controlled wind tunnels. Air relative humidity and air velocity were set at 60% and 0.2 m s, respectively. The tclo and tsur were measured daily through digital thermometer and thermographic camera, respectively. Analysis of variance and Scott Knott mean test at 5% probability were applied to the data. The results indicate that the stress by low air temperature (24 and 27 °C) resulted decreased of tclo and tsur (p<0.05, Scott-Knott test). When subjected to high air temperature stress (33 °C), there was no significant difference (p>0.05, Scott-Knott test) compared to the control treatment (30 °C). Acclimation of broilers to thermal stress occurred from the second day of stress.


INTRODUCTION
Brazil, a country with a large territorial extension presents great climatic diversity with temperature amplitudes during the year, in which promotes complications in the broilers breeding due to the variations that occur in the environment production (Nääs et al., 2010).
Thus, in the current scenario of Brazilian poultry development it is necessary that the poultry be raised in appropriate environments so they can achieve high performance maximizing their genetic potential.Under these conditions, characterized as thermal comfort, obtain low energy expenditure to control the homoeothermic thus increasing the performance (Nascimento et al., 2014).
The cloacal (tclo) and surface (tsur) temperature are among the main physiological indicators for thermal comfort on broilers, as they change with the variation of air dry bulb temperature (tdb), relative humidity (RH) and luminosity (Costa et al., 2012;Ferreira et al., 2012).When tsur and tclo variation occur, the broilers dissipate or retain heat, and part of the energy that should use to gain weight will be applied in the thermoregulatory process reducing productive responses.
Any variation in body temperature of broiler indicates that the performed changes on their surface were not enough to maintain the thermal equilibrium (Nascimento et al., 2013) indicating physiological changes.
The increase of tclo is proportional to the age (Marchini et al., 2007), and the stress caused by high tclo results in decrease in weight gain (Costa et al., 2012).According to Oliveira et al. (2006) the body core temperature for the second week of life varies between 41 to 42 °C consistent with the values found by Bueno et al. (2014).
According to Nääs et al. (2014), through skin analysis temperature it is possible to determine the physical healthy state of the animal.One way to evaluate the tsur on broilers is through infrared thermography which is a procedure that does not cause interference in the ecological niche of the evaluated animal (Nascimento et al., 2011a).
Although feathers obstruct the infrared emission of the skin, the thermography allows the thermal mapping of the body (Castilho et al., 2015;Souza Junior et al., 2013) and is used in several studies to obtain the thermal responses (Barnabé et al., 2015;Mayes et al., 2014;Roberto & Souza, 2014), since the variation of the tsur is instantaneously modified when thermal changes occur in the thermal environment of the breeding.
The aim of this study was to evaluate the average surface and cloacal temperature on broilers in the second week of life submitted to different intensities and duration of thermal stress, adjusting correlation equations between these variables.

MATERIAL AND METHODS
This research was carried out in an animal environment laboratory equipped with two air conditioning systems 18,000 BTUs each used to control and maintain room air temperature.It also contains four climate-controlled wind tunnels.Inside each wind tunnels there was a cage with dimensions of 0.40 x 0.60 m divided into three sub-distributions equal to 0.08 m² each, equipped with feeders and water fountains.
Two hundred and forty males and females broilers Cobb 500 ® lineage were used from same commercial hatchery, of one day of life.For each stage were used sixty broilers with distribution of five broilers per division during the first seven days of life, that is, fifteen broilers per tunnel.Subsequently, in the second week of life, only four chicks were kept in each replicate maintaining the breeding density recommended by the manual (Cobb, 2012).
During the experimental period the broilers were fed with a balanced diet based on corn and soybean, following the nutritional recommendations by Rostagno et al. (2011).Feed and water were available ad libitum.The hygienization of the wind tunnels and cages was carried out every day to avoid formation of gases, providing an adequate environment for broilers performance.The tclo and tsur of the poultries were also daily evaluated.
The experiment was carried out in different stages varying the intensity of tdb applied on broilers on second week of life.In the first seven days, the birds were kept in thermal comfort at 33 °C (Cony & Zocche, 2004).From the eighth day, the tdb levels at each stage were 24, 27 and 33 °C, in order to provide thermal stress at both low and high air temperatures (Cony & Zocche, 2004;Curtis, 1983).
The thermal stress was applied in four levels of duration and performed on days 8, 9, 10 and 11 of the experimental period of each stage.During the second week, the air temperatures returned to thermo neutrality zone.Table 1 shows how the treatments were divided.
For the experiment, the air relative humidity was maintained at 60 ± 1% and air velocity at 0.2 ± 0.1 m s -1 , characterized as comfort for broilers (Medeiros et al., 2005).The illuminance was adjusted by means of a luxmeter (LDR-380, accuracy of ± 3%), with intensities of 25 and 10 lux for the first and second week of life, respectively (Cobb, 2012).TABLE 1.The air dry-bulb temperature and days of maintenance of heat stress condition (in bold), applied to broilers from 1 to 21 days of age kept in climate-controlled wind tunnels.
Thermographic images were used to obtain surface temperatures.The images were acquired by means of a Fluke thermographic camera (model TI55FT20/54/7.5), positioned 1 meter high from the cage, in order to allow a wide view and comprehensiveness of all the animals.
The thermographic evaluation was elaborated using the Fluke Smart View 3.0.126.0 software, in which the adopted emissivity was 0.95 (Ferreira et al., 2011;Nääs et al., 2014).For each replicate of the cage a bird was randomly selected (Figure 1) and the average tsur was obtained as a function of the surface areas of three broilers (Case et al., 2012;Nascimento et al., 2011b).With respect to the tclo a broiler from each part of the cage was captured randomly to have its cycle measured, that is, twelve broilers per day totalizing forty-eight verifications per step.The measurement of tclo was performed by digital thermometer (INSTRUTHERM ® São Paulo, Brazil, Model THR-080, accuracy ± 0.01 °C), and before each use the sensor was sterilized and introduced For the statistical analysis of the variables was used completely randomized design (C.R.D.) with three replications, assembled according to the 4x4 factorial arrangement, that is, four tdb in the second week of broilers life (33, 30, 27 and 24 ˚C) and four stress durations (1, 2, 3 and 4 days), according to equation 1.The variables were treated by the statistical program SISVAR 5.3 (Ferreira, 2010) submitted to variance analysis and comparison means using the Scott Knott test at the 5% probability level.The data were submitted to clustering analysis (R Development Core Team, 2012), for the construction of the dendogram of broilers tclo using Ward's method which calculates the smallest sum of squares within the groups (Biagiotti et al., 2015).
In which, ijk Y is the observation of the repetition k of the i-th temperature in the j-th duration with i, j = 1, 2, 3, 4 and k = 1,2,3;  is the constant associated with each observation;

RESULTS AND DISCUSSION
To verify the tdb and RH which the broilers were submitted, the thermal environment inside the wind tunnels was monitored during the experimental period.The mean values and standard deviations observed during the second week of life are listed in Table 2.The observed air dry-bulb temperatures obtained low standard deviations, being close to the desired temperatures (Table 2).The highest standard deviations were 0.3 ° C for tdb and 1.7% for RH.These variations indicate the efficiency of climate-controlled wind tunnels in maintaining the desired thermal environment, corroborating with Schiassi et al. (2014).
The interaction between tdb and duration of thermal stress for the averages tclo was not significant (P <0.05, test F) however, the different tdb and thermal stress durations presented significant difference (P <0.01, Test F), and the means test is listed in Table 3.The lowest tclo (40.5 °C) occurred when the broilers were submitted to the lowest thermal stress of the tdb (24 °C) with the tclo observed being below the lower limit for the range considered as comfort which varies from 41 to 42 °C (Oliveira et al., 2006).Thus, it is observed that the tdb of 24°C provided a significant decrease in tclo of the broilers (P <0.05, Scott-Knott test), thus characterizing cold stress condition (Castilho et al., 2015).
For tdb of 27°C it is observed that the tclo is within the comfort range however, significant difference (P <0.05, Scott-Knott test) occurs in relation to the other temperatures of thermal stress.In the control treatment (tdb = 30 °C) that characterizes thermal comfort condition (Schiassi et al., 2015, Cony & Zocche, 2004), the value of the tclo is within the comfort range.However, when broilers were submitted to tdb of 33 °C it was found no significant difference (P <0.05, Scott-Knott test) in relation to the control treatment.This is due to the fact that the broilers have been able to dissipate the produced heat by the metabolism to the environment (Abreu & Abreu 2011), because this value of tdb is still not enough to provide significant increase of the tclo.Regarding the thermal stress duration there was a significant difference (P <0.05, Scott-Knott test) between the first day in relation to the others (2 nd , 3 rd and 4 th day), which are statistically the same (P >0.05, Scott-Knott test).In this way, the broilers' ability to acclimatize confirm as reported in several studies (Alves et al., 2010;Baêta & Souza, 2010;Schiassi et al., 2015;Silva & Vieira, 2010).
When applying the cluster analysis methodology on the tdb as a function of the tclo it is observed that the classifications of tdb associated with the distance scale in the dendogram (Figure 2) presented similar results to those obtained by the variance analysis and the mean test by Scott-Knott at the 5% probability level.By means of the thermographic images it was verified that the broilers average tsur reduced and increased when submitted to low and high stress temperatures respectively, corroborating with the results obtained by several authors (Welker et al., 2008;Yahav et al., 2005).
The interaction between tdb and the thermal stress duration for the means of the tsur was significant (P <0.05, F test) (Table 4).For one day of stress, it was observed significant difference (P <0.05, Scott-Knott test) for all used temperatures with the average variation of the tsur being 5.0 °C.For the durations of thermal stress of two, three and four days there was no significant difference (P> 0.05, Scott-Knott test) of the tsur between tdb at 24 and 27 °C (stress due to low temperatures) and also between tdb at 30 °C (comfort) and 33°C (stress with high temperature).However, there was a statistical difference (P <0.05, Scott-Knott test) from tsur between the two groups of tdb (24 °C / 27 °C and 30 °C / 33 °C).
Based on the results it can be verified that when the broilers were exposed to thermal stress at low air temperature (27 °C), in relation to comfort tdb (30 °C) significant difference was found (P <0.05, Scott-Knott test) for the tsur in four thermal stress duration.According to Cassuce et al. (2013) the air temperature of 27 °C is comfortable in the second week of life for broilers, but the recommended comfort value in the literature for the second week is 30 °C (Schiassi et al., 2015;Cony & Zocche, 2004), and can be confirmed through the tclo and tsur observed in this study.However, it should be emphasized that this analysis represents only the variation of two physiological parameters without considering the zootechnical responses.
Equations for estimation of the tclo and tsur were adjusted as a function of tdb (equations 2 and 3, respectively).The adjusted equations were significant (P <0.05, F test) and showed determination coefficients (R 2 ) of 0.75 and 0.76, respectively.The adjusted coefficients for eqs (2) (3) were significant (P <0.01, t test), and the standard errors of b0 constant were 0.39 and 0.87 and for b1 of 0.13 and 0,03, respectively. (2) (3) According to eqs (2) and (3) for each degree of variation in the tdb occurs the modification of 0.13 °C and 0.36 °C for the tclo and tsur, respectively.According to Carvalho et al. (2004), the surface temperature of broilers is subject to faster alteration, since the dissipation of the blood flow occurs through the convection.This behavior can be verified through eqs ( 2) and (3) in which for each degree of variation, tsur varies 2.77 times more than tclo.
Thus, when subjected to stress at low intensities (24°C) there is reduction of 0.78 and 2.16 °C for tclo and tsur respectively, and when stress occurs with high intensity (33 °C) the increase is 0.39 and 1.08 °C for tclo and tsur, respectively, when compared to comfort temperatures (30 °C).
Considering the importance of statistical analysis for correlation equations formulation between tclo as a function of tsur (Nascimento, 2010), [eq.( 4)] was adjusted in which R 2 was 0.75. (4) Through the [eq.( 4)] it is possible to determine the tclo in a non-invasive way, through the use of infrared sensors for the measurement of the tsur.Thus, it is possible to estimate tclo with minimum stress for broilers, and in the usual methodology for the determination of tclo it is necessary the direct contact of the thermometer with the cloaca of the broilers occurring in this way a stress factor.Equation 4 can be used in an algorithm to be shipped in microcontrollers to help the decision making regarding the activation of ventilation and evaporative cooling systems.

CONCLUSIONS
Based on the values of surface temperature (tsur) and cloacal temperature (tclo) it was found that the thermal comfort temperature for broilers in the second week of life is 30 °C.The acclimatization of broilers to thermal stress occurred from the second day of stress.The highest variation of the tclo occurred when the broilers were submitted to low stress temperatures, and the tsur varied according to the applied stress intensities.The correlation equations developed can be used to support the decision making.

FIGURE 1 .
FIGURE 1. Thermographic image of broilers subjected to determined thermal conditions inside the climate-controlled wind tunnels.

i
is the effect of the i-th temperature; j  the effect of the j-th duration; ij  interaction between dry air bulb temperature and duration, and ijk e error associated with each observation, being .

FIGURE 2 .
FIGURE 2. Analysis in dendrogram of air dry-bulb temperature (tdb) according to the cloacal temperature (tclo) of broilers subjected to different intensities of stress in the second week of life.

TABLE 3 .
Mean and standard deviations (in parentheses) of the cloacal temperature (tclo) of broilers submitted to different intensity and duration of thermal stress during the second week of life.
Means followed by different letters differ from one another by the Scott Knott test (P <0.05).

TABLE 4 .
Mean and standard deviations of surface temperatures (tsur) of broilers submitted to different durations of thermal stress during the second week of life.Averages with different letters (uppercase in the row and lowercase in the column) indicate statistical differences (P <0.05) by the Scott Knott test.