| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL PRODUCTION |
,2,3
* Department of Animal Health, Ministry of Agriculture and Rural Development, Vietnam;
and
Livestock Environment Group,
and
Animal Nutrition Group, and
and
Adaptation Physiology Group, Wageningen University and Research Center, The Netherlands; and
and
¶ Ho Chi Minh City Natural Science University, Vietnam
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Heat Stress • Physiological Response • Pig • Relative Humidity • Temperature
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). At higher ambient temperature (T; upper part of the aforementioned range), the pig may already show increased respiration rate (RR), rectal temperature (RT), and decreased voluntary feed intake (VFI; Nienaber and Hahn, 1996; Quiniou et al., 2000). In addition to experiencing increased heat loss, the pigs increase the exposure of their bodies to cool air or cool and wet floors (Aarnink et al., 1996, 2001). Furthermore, the effects of high T on pigs are expected to be more pronounced at high RH.
If the T exceeds the point above which the balance between heat production (HP) and heat loss can be maintained, evaporative heat loss is at its maximum and respiratory evaporation will be inadequate for sufficient heat loss to keep the RT constant. Under such conditions, the animals supposedly are above the thermoneutral zone. As a consequence, RT increase and then there is an adaptive depression of the HP (Quiniou et al., 2001). Thus, the decrease in the associated thermal effect of feeding is an efficient mechanism to decrease heat load (Verstegen et al., 1987).
Little information is available about the T above which group-housed pigs start to adapt their mechanisms for balancing heat loss (evaporative heat loss, behavioral, and physiological adaptation) and HP. The T above which the responses change (also called critical temperature or inflection point temperature [IPt]) may well differ, depending on the measurement studied. In addition, the effect of RH on these measurements is unknown. This study was designed to quantify the short-term responses of growing-finishing pigs to increased T at different RH.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Housing and Climatic Control
Respiration Room.
In this study, two identical 80-m3 climate respiration rooms were used (Verstegen, 1987). Each chamber had an inner room of 6 x 4 x 2.2 (length x width x height). Air was drawn from the chamber by means of a centrifugal fan. Air velocity at animal level was approximately 0.2 m/s. The volume of air exhausted from the chamber was replaced by the same amount of outside air. This was equivalent to 3% of chamber volume capacity per minute. The outside air was added to the air conditioning circuit through motorized valves to maintain a constant flow under pressure. The amount of recirculated air was 158 m3/min, which equaled an air exchange rate of twice per minute. Condensate in the heat exchanger was collected and measured.
Pen.
Groups of 10 gilts were assigned randomly to one of the two respiration rooms. Pen size inside the room was 2.50 x 4.50 m. Of the pen floor, 60% was solid (2.50 x 2.80 m) with 4% slope, and 40% was metal slatted floor (2.50 x 1.70 m). The thickness of the insulated concrete solid floor was 10 cm. Under the slatted floor was a slurry tank in which urine, feces, and spilt water collected. The slurry tank was emptied on d 13 of each trial. The slats had 15-mm-wide tribar metal bars with 5-mm intervening spaces. A dry feeder and a bowl drinker were installed to provide the pigs with ad libitum feed and water.
Climate Control.
The air temperature was kept constant within a day, but during a 9-d period, it was increased stepwise in the morning by 2°C/d from 16 to 32°C. Due to the data collection procedures carried out by the researchers, there was a 30-min time lag between Rooms 1 and 2 in temperature increase and the switching on or off of the lights. In Room 1, the temperature was increased daily at 0900. Daylight was provided from 0600 to 1800 using nine fluorescent tubes, which produced 400 to 450 lx at floor level. Two light bulbs of 25 W, which was 4 to 5 lx, were left on during the night (1800 to 0600). Throughout the 13 d, the RH was fixed at one of three levels (50, 65, or 80%), and it was kept constant by a humidifier (one per chamber). The circulating air was heated or cooled depending on the deviation from the set point temperatures. Temperature and RH were well controlled; their fluctuation was less than 0.5°C and 5%, respectively.
Animals.
One hundred twenty 90-d-old crossbred females were used for the trials. For each trial, 20 gilts were purchased at approximately 25 kg of BW and assigned to one of two groups based on BW. The composition of the groups was not changed. Before the experiment, the pigs had spent 40 d in growing pens that were similar to the pens used during the experimental period. At 130 d of age, the group of 10 pigs was moved into a chamber. At this time, the average initial BW was 61.7 kg (range 58.0 to 65.5 kg). Starting 14 d before the experimental period, the pigs were offered the experimental pelleted diet, containing 157 g/kg of CP and 16.13 MJ/kg of GE (on an as-fed basis). For data collection, 3 of the 10 pigs in each room were chosen randomly and each was assigned a unique identifying number, which was painted on their backs. Twice daily (at 0900 and 1400 and at 0930 and 1430 for Rooms 1 and 2, respectively), the physiological variables of these three pigs were measured during the 9-d trial, as described below. Finally, weights were taken in the morning of d 13.
Measurements.
The RH and temperature of the incoming and outgoing air were continuously recorded automatically. Feed intake was recorded twice daily by weighing the dry feeder. Water intake was measured by reading the water meter twice a day. All physiological measurements on the animals were done twice a day, for 30 min per chamber before feeding time from 0800 to 0900 in the morning and from 1400 to 1500 in the afternoon. During collection of these individual daily data, the pigs were not restrained. The RT of marked pigs was taken per rectum with a digital thermometer (Barchen YS-723, Taipei, Taiwan). The RR was observed by means of a stopwatch and by counting the flank movements of the marked pigs. Skin temperature (ST) was measured with a radiant thermometer (Chino IR-AH, Tokyo, Japan). The emission factor of the radiant thermometer was set as instructed from the manual. The radiant thermometer was calibrated regularly by means of a black hole apparatus. During calibration, the calibration room temperature was set at different levels within the range of the chamber temperatures. A linear regression model was applied to adjust the measured ST to the calibrated temperatures. Four fixed points on the skin surface were measured (on each of the three marked pigs): one point on the area behind the ear, three points on the lateral side of the pig.
Heat Production.
Throughout the 9-d experimental period, O2 consumption and CO2 and CH4 production were measured at 6-min intervals as described by Verstegen et al. (1987). From these data, HP was calculated as described by Brouwer (1965).
Evaporative Water and Evaporative Heat Loss.
Water evaporation mostly by panting is a vital route of heat loss in pigs at high T. To quantify the evaporative heat loss of the pigs during the experimental period, complete air-water balance measurements were performed, as described below. Calculations were done daily and for each replicate:
 | [1] |
In this equation, A is the volume of water in outgoing air, calculated daily after measuring the volume and humidity of the outgoing air. The RH and T of the outgoing air were measured every minute (HMT320, Vaisala Oyj, Helsinki, Finland), and the volume of air was measured continuously with dry gas meters (G65, Actaris-Schlumberger, Dordrecht, The Netherlands; see also Verstegen, 1987). Variable B is the amount of water that condensed on the heat exchangers. Water was collected in a tank outside the chamber, which was positioned on top of a weighing device and was recorded daily. To prevent evaporation from this tank, a layer of soy oil was poured onto the water. Variable C is the volume of water in incoming air calculated daily after measuring the volume and humidity of the incoming air. The RH and T of incoming air were measured every minute with the same devices as mentioned in A, and the volume of air leaving the chambers was set equal to the volume entering the chamber (Verstegen et al., 1987). Variable D is the volume of water evaporated from wet solid floor. The wet area on the solid floor was determined using video images taken at hourly intervals to estimate the proportion of the solid floor wetted with urine. The mean surface of solid wet floor then was calculated and multiplied by the evaporation per square meter as calculated in E (see below). Variable E is evaporation of water was measured daily as the weight loss of a tray with water with a known surface area (approximately 0.20 m2). As this contributes to the air water balance, it should be accounted for in Eq. [1]. The evaporated water was then divided by the area of the trays to calculate evaporation in kg·m–2 to use in the parameter D. Variable F is the volume of water to maintain a specific humidity in the climate respiration chambers additional water had to be added to the air, especially at the higher T and RH set points. The volume of water sprayed into the air by a humidifier was measured daily.
Evaporation heat was 2,260 kJ/kg at 373K = 100°C. In our case, the RT of finishing pig was, on average, 39°C. Specific heat of water was 4.18 kJ/(kg·K), between 293 and 373K = boiling point, or from 20 to 100°C. From 39°C RT to 100°C = (100 – 39) x 4.18 = 254.98 kJ/kg. Hence, energy of evaporative water from a pig, kJ/(pig·d) = (2,260 kJ/kg + 254.98 kJ/kg) x evaporative water, expressed in kJ/(pig·d).
Statistics
For determining the effect of temperature and RH on physiological data (RR, RT, and ST) the mean of individual data per pig per day (three pigs in each chamber) was used. Data analysis of VFI and ADG were based on means per group per day. Data were initially subjected to a broken-line model (Aarnink et al., 2001) to validate whether an IPt could be determined. If the model failed to converge, a linear regression model was used. This REML technique (Genstat 5, Release 6.1) was used to analyze effects of T, RH, and their interaction on ST, as well as to establish the relationship between RR and water intake and RR and EW. The 
2 test was used to test the significance of the model, and the Student’s t-test was used to test differences between treatments.
Broken-Line Model.
The broken-line analysis was performed to calculate the inflection point temperatures for different relative humidities above which response variables changed. The broken-line analysis proved to be a suitable model for RR, EW, RT, ratio of water to feed (rWF), feed intake, and HP. The model can be described as follows:
 |
where Y is the response variable (e.g., RR, EW, RT, rWF, feed intake, and HP); and C is a constant over a range of temperatures at each level of humidity (50, 65, and 80%); Z is a regression coefficient at each level of humidity (50, 65, and 80%); T is the chamber temperature in °C (16 to 32°C); and Ipt is the inflection point temperature in degrees Celsius at each level of humidity.
An F-test was used to test the significance of the broken-line model, and a Student’s t-test was used to test the differences within measurements.
Linear Regression Model.
The linear regression model was as follows:
 |
whereµ, 
, and ß are regression coefficients; Ti = room temperature, (16 to 32°C); RHj = the effect of RH (50, 65, and 80%); and (T x RH)ij = the interaction between room temperature and humidity. This component was excluded from the model when it was not significant; and 
ijk = residual error.
All analyses were performed with Genstat software (Genstat 5).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
shows that the broken-line model fits for RR, EW, RT, VFI, rW:F, and HP. Treatment effects on RR are presented in Figure 1. With increasing T, RR remained constant at on average 32 breaths/min until the inflection point (on average 22°C), after which it increased by on average 13 breaths·min–1·°C–1. Humidity affected the IPt (23.1, 22.6, 21.3°C for 50, 65, and 80% RH, respectively; P < 0.05) as well as the increase (regression coefficient) above IPt (15.1, 14.8, and 12.0 breaths/min, for 50, 65, and 80% RH, respectively; P < 0.05).
|
|
shows the effects of treatments on EW. Above an average of 20.4°C, the EW increased. Below that inflection point, there were differences in volume of water evaporated at different humidity (1.26, 1.12, 0.89 g·pig–1·min–1, respectively; P < 0.05). For each degree Celsius above the IPt, EW increased by approximately 0.08 g·pig–1·min–1, or approximately 115.2 g·pig–1·d–1. Humidity had no effect on IPt and the increase of EW after the IPt had been exceeded.
Rectal temperature was affected by increasing temperature (Table 1; Figure 2). Until the T reached 26.1°C, the RT of pigs was constant at an average of 39.3°C. Above that IPt, it increased 0.13°C/°C. Relative humidity affected the IPt (26.6, 27.1, 24.6°C, respectively; P < 0.05).
|
and Figure 3. As the T increased, the VFI remained constant until an average of 25.5°C. After that IPt, per degree Celsius there was a decrease of an average of 95.5 g·pig–1·d–1. Relative humidity had no statistically significant effect on either the IPt or the decrease.
|
; Figure 4). Relative humidity had no effect on the rW:F; however, the increase was affected by RH (0.22, 0.37, and 0.58 at 50, 65, and 80% RH, respectively; P < 0.05).
|
). Above this IPt, each degree Celsius increase brought about a decrease in HP of 8.40 kJ·BW–0.75·d–1. No significant effect of RH on HP was found.
Table 2 shows that ST clearly increased with T (P < 0.001; Figure 5). On average, ST increased by 0.25°C for every degree Celsius increase in T. Average ST was lower in the 80 RH group than in the 50 and 65% RH groups (P < 0.05; Table 2). There was no interaction effect on ST.
|
|
. With each extra respiration stroke per minute the water intake increased by 12.7 g·pig–1·d–1 (P < 0.001). Water intake was lower at 80% RH than at 50 and 65% RH (P < 0.05). A similar relationship was found between EW and RR (Table 2): EW increased by 10 mg·pig–1·min–1 per extra respiration stroke per minute (or 14.4 g·pig–1·d–1; P < 0.001). In this study, at 32°C and at 180 breaths/min of RR, the EW reached a maximum value of 2.1 g·pig–1·min–1. There were no effects of interaction on these relationships. We found that pigs kept at 50% RH grew faster than pigs at 65 and 80% RH (721 vs. 644 and 630 g·pig–1·d–1, respectively; P < 0.05).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
According to Mount (1979), the thermal neutral zone of pigs can be defined as the range of environmental temperature within which metabolic rate is minimum, constant, and independent of temperature. The thermoneutral zone can be divided into two distinct parts, based on the evaporative activity of the pig. One part is the comfort zone, which covers the range from the lower border of the thermoneutral zone to the point where pigs activate their evaporative function. The second is the zone from the upper border of the comfort zone to the upper border of the thermoneutral zone. Within this zone, evaporative heat loss increases considerably. With regard to the thermoneutral zone of finishing pigs, Verstegen and Henken (1987) stated that the lower border of the thermoneutral zone for 40-kg pigs housed in groups was in the range of 19 to 20°C. Little information is available on the upper border of the thermoneutral zone of pigs, especially for finishing pigs. Quiniou et al. (2000) found that pigs of 30 to 90 kg had decreased VFI in the temperature range from 23 to 25°C. In this study, there were no differences between humidities in IPt for VFI. Therefore, the upper border of thermoneutral zone as measured by VFI did not depend greatly on RH. For pig production, and thus for economic reasons, it is important to assess heat stress on the basis of depression of VFI and HP; however, for reasons of animal welfare, increasing RR (panting) followed by increasing RT may be more important. When T increases, sensible heat loss (radiation, convection, conduction) decreases rapidly, and evaporative heat loss becomes a vital route for pigs to eliminate heat load. As they do not sweat, pigs must pant. Above an IPt of approximately 22.4°C, the pigs had increased RR with increasing T. As expected, the IPt was somewhat lower at higher RH. Christon (1988) mentioned an RR in tropical finishing pigs of 120 breaths/min at an average T of 29°C in a RH range of 69 to 91%. Brown-Brandl (2000) found the RR of pigs exposed for 22 h to temperatures of 18 and 32°C were 56.7 and 100.7 breaths/min, respectively. At high temperature, the RR we measured was higher than the results of Christon (1988) and Brown-Brandl (2000). It should be noted that our pigs had greater feed intake below IPt, which caused more heat to be produced. Another reason for these different results could be unreported differences in radiant temperature between the studies. In our study, the temperature of floor surfaces, walls, and ceilings was similar to the T; thus, T was the same as air temperature. There are two other possible reasons for the higher RR values in our pigs. First, our pigs were housed in groups, and therefore they may have been more active. Furthermore, group-housed pigs are less able to lose heat by radiation because they are surrounded by the warm bodies of other pigs. Moreover, our pigs experienced constant heat stress throughout the whole day, whereas in the other studies mentioned, the pigs were exposed to cyclic temperatures changes. Because pigs can influence HP within a day by the feed intake pattern, they can dissipate more heat during cooler periods.
When evaporation is decreased by high humidity (80 compared with 65 and 50%), the RR and RT increase at lower temperatures (Table 1). Figure 6 demonstrates that the increase in RR occurred at a lower temperature than the increase in RT. Panting allows pigs to maintain a constant RT for an increase in temperature of approximately 3 to 4°C after RR increases. Above IPt, RT increases linearly with room temperature. Some authors have suggested that RT is an important indicator of heat-stressed animals (Close, 1971; Holmes, 1973; Kadzere et al., 2002; Ricalde and Lean, 2002; Silanikove, 2000). These authors reported that cattle and sheep showed increased RT above T of 24 to 26°C. In the Mount (1979) diagram, the thermoneutral zone is defined as the temperature zone with minimal HP at constant RT. Above this zone, core temperature increases and pigs become very heat stressed. This has a direct effect on pig productivity (McDowell et al., 1976). Our results show that at 80% RH, the IPt for RT and for RR were approximately 2.0°C lower than at 50% RH (Table 1). Earlier, it was suggested by Curtis (1983) that at 30°C, an increase of 18% in RH is equivalent to an increase in air temperature of 1°C, which is supported by our results.
|
reported that in an 85-kg pig kept by itself in a respiration room, HP decreased above 28°C, but then increased above 32°C. However, Close (1971), Holmes (1973), Nienaber et al. (1987), and Ricalde et al. (2002) reported that at high temperatures, a decrease in HP was found in relation to increasing rectal temperature. The differences between the different studies can be explained by many factors. First, one should distinguish between studies with restricted and ad libitum intake. Our animals were fed ad libitum, so when T rose above the thermoneutral zone, causing increased RT, VFI was largely decreased, with the consequence that HP decreased as well. Secondly, a distinction must be made between studies with group-housed animals and those with individually housed animals. Thirdly, because total HP is connected to muscle activity, at high temperature, HP changes markedly, especially for heavy pigs (van Milgen et al., 1998). Heavy pigs will lie more and decrease activity. A final factor confounding the comparison is that the experimental protocols differed between the studies. In the other studies, animals often had been subjected to constant thermal conditions for longer than in our study. In this study, for each degree Celsius increase in T above IPt, VFI decreased steadily by an average of 95.5 g. Nienaber et al. (1983) reported that for 45- and 85-kg pigs, an increase in T from 20 to 30°C decreased VFI from 65 to 74 g·d–1·°C–1. For the same temperature range, the average decrease in our study was 43 g/d, which was lower than the decrease in VFI found by Nienaber et al. (1983).
An effective result of panting was that the pigs could maintain at constant VFI until a few degrees above the IPt for RR (+2.0°C at RH = 50%, +3.0°C at RH = 65%, and +4.3°C at RH = 80%). In other words, the pigs were able to maintain their level of VFI by increasing RR and exploiting evaporative heat loss. The first physiological adaptive reaction to heat stress in our study was RR. Panting pigs increased their total oxygen consumption at maintenance or constant feed intake and, therefore, an increase in physical activity was accompanied by an increase in total HP (Ingram, 1965). In our study, VFI fell so much that, although panting activity increased, total HP decreased.
Little information is available on the changes of the ST of finishing pigs at high T. According to the thermal concept, at high T, animals may take up heat from the environment if the T is above the surface ST (Robertshaw, 1985). It is interesting that in our study, the pigs’ ST changed at moderate temperatures. Black et al. (1993) reported that ST rose promptly to 36.8°C in sows exposed to 28°C air temperature. Geers et al. (1987; cited from Fanger, 1972) reported that the comfort ST of homeothermic animals ranges from 32 to 35°C. This is confirmed by our results, with ST ranging from 33 to 35°C at T ranging from 16 to 22°C (Figure 2).
The rW:F in our study increased differently with higher T at the three RH. For each degree Celsius increase, pigs at 80% RH had the largest increase in ratio; the smallest increase was at 50% RH. The ratio increase at 80% RH was more than double that at 50%RH. This means that at high RH and high temperature, the pigs decreased feed intake and increased water intake more than at low humidity. For these two reasons, it can be expected that at high RH and high temperature, water consumption will be more independent from VFI than at low temperature and humidity. It should be noted that drinking water temperature was lower than room temperature. The water tank was located outside the respiration chamber, at a temperature of approximately 16°C. Thus, some direct cooling was also obtained by drinking the relatively cool water.
A general picture of the total HP, RR, and heat loss can be seen in Figure 7. This figure illustrates the theory of Mount (1979), but it is based on our measured data. Taking into account the HP, RR, and evaporation response of the pigs, ranges of thermal zone could be illustrated. The increased RR frequency could be interpreted as a sign of discomfort. The range between RR increase and HP decrease was narrow, only 1.2°C at 80% RH. It also can be concluded that EW increased with enhanced RR; however, a poorer relationship between evaporation and respiration can be seen at 80% RH (Table 2), presumably because at this high RH, less respiratory evaporative water could be added to the humid air by the pig. At high RH, pigs seem to find other means to get rid of the heat. One possibility is by means of behavioral changes (e.g. wallowing, defined as the behavior of rolling from side to side in urine and feces). Wallowing already occurred at low temperatures and increased with increasing T.
|
). Mount (1979) mentioned a decrease of 30 g/d in pigs’ ADG for each 1°C above the optimal temperature. Serres (1992) found that in the temperature range of 21 to 32°C, the growth rate of 70-kg pigs was decreased by 46%, which is consistent with our results. The decrease in ADG of the pigs seemed to be caused mainly by the drop in VFI during heat stress. From the results of this study, it can be concluded that in modern Western pigs, physiological signs of heat stress already occur at temperatures above 22°C. The modern Western pig has high metabolic activity and, in turn, high HP. Pig farmers should take this into account when designing and controlling indoor climate.
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
2 Current address: Livestock Environment, Bornsesteeg 59, P.O. Box 17, 6708 PD Wageningen, The Netherlands.
3 Correspondence: Bornsesteeg 59 P.O. Box 17, 6708 PD Wageningen (phone: + 31 317 47 65 54; fax: + 31 317 47 53 47; e-mail: andre.aarnink{at}wur.nl).
Received for publication August 10, 2004. Accepted for publication February 24, 2005.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Zumbach, I. Misztal, S. Tsuruta, J. P. Sanchez, M. Azain, W. Herring, J. Holl, T. Long, and M. Culbertson Genetic components of heat stress in finishing pigs: Development of a heat load function J Anim Sci, September 1, 2008; 86(9): 2082 - 2088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Zumbach, I. Misztal, S. Tsuruta, J. P. Sanchez, M. Azain, W. Herring, J. Holl, T. Long, and M. Culbertson Genetic components of heat stress in finishing pigs: Parameter estimation J Anim Sci, September 1, 2008; 86(9): 2076 - 2081. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Renaudeau, E. Huc, and J. Noblet Acclimation to high ambient temperature in Large White and Caribbean Creole growing pigs J Anim Sci, March 1, 2007; 85(3): 779 - 790. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, 
, and 
are means of all measured data of three marked pigs in each relative humidity (RH) treatment.
