Animal Housing in Hot Climates PDF

CIGR Workshop “Animal Housing in Hot Climate”, Cairo, Egypt, April 1-4, 2007
Workshop of CIGR Section II

April 1- 4, 2007 Cairo, Egypt
Animal Housing in Hot Climates
1
Organization Committee
Dr. Daniel Berckmans Belgium daniel.berckmans@biw.kuleuven.be
Dr. Mohamed Hatem Egypt hatem@cu.edu.eg
Dr. Panos Panagakis Greece ppap@aua.gr
Dr. Vasco Fitas da Cruz Portugal vfc@uevora.pt
Dr. Paolo Zappavigna Italy paolo.zappavigna@unibo.it
2
Table of Content
No. Title Page
1 Issues related to livestock housing under hot climatic conditions including the animals’ response to high 4
temperatures.
T. Banhazi, A. Aarnink, H. Thuy, S. Pedersen, J. Hartung, E. Maltz, H. Payne, B. Mullan and D.
Berckmans
2 Behavioural changes related to air temperature in sows kept in different housing conditions 25
M. Barbari and M. Bianchi
3 Use of different cooling systems by pregnant sows in experimental pen 30
M. Barbari, L. Conti and S. Simonini
4 Thermoregulatory responses to high ambient temperature in growing pigs: effects of temperature level 34
and breed.
D. Renaudeau, J.L., Gourdine, C. Anaïs.
5 Heat Stress in Dairy Cows: A Review of The Heat Load Effects on The Animal Response and Relief 39
Strategies.
P. Zappavigna, E. Maltz, S. D’Archivio
6 Method to evaluate and optimise climate control strategies in livestock buildings taking into account 55
dynamic behaviour of animal heat production
D. Berckmans, E. Vranken
7 Is it convenient to condition the resting area in dairy cows barn? 66
F. Calegari, P. D’Alessio and E. Frazzi
8 Influence of the thermal environment on the physiological responses of dairy goats breeding on deep 71
bedding systems.
Santos, C.R; Souza, C.F; Tinôco I. F.; Cruz, V. F.; Pereira, V.N; Acevedo, R.R.; Mendonça, H.V.
9 Dairy cows thermal comfort evaluation in hot climates using temperature humidity index 74
Maurício Perissinotto; Vasco Fitas da Cruz; Daniella Jorge de Moura
10 A Virtual Animal for Climate Control Design: Static and Dynamic Simulations of Heat Losses 77
J.-M. Aerts, D. Berckmans
11 An Approach to Evaluate the Suitability of an Evaporative Pad Cooling System in Animal Housing to 89
the Summer Climate in China
Baoming Li*, Chaoyuan Wang, Wei Cao and Zhengxiang Shi
12 Developing a set of strategies, in Portugal, to monitor and prevent damages in animal housing, due to hot 95
climate conditions
Vasco Fitas da Cruz, José Carlos Barbosa, João Santos e Silva
13 Modeling of Technology Implementation and Dairy Farm Foundation in Hot Climates 97
M. Samer, H. Grimm, P. Epinatjeff 1, M. Hatem, R. Doluschitz, and T. Jungbluth
14 Analysis of The Effects of The Roofing Design on Heat Stress in Dairy Cow Housing. 99
P. Zappavigna, P. Liberati
15 Technical Solutions for Reduction of Heat Stress in Livestock Buildings in Germany 108
H.-J. Mueller
16 Evaluation of fogging in a mechanically ventilated pig facility 110
Angelika Haeussermann, Eberhard Hartung, Erik Vranken, Jean-Marie Aerts, and Danie Berckmans
17 New Trends in Animal Housing in Greece: Green housing Type Livestock Buildings. 115
C. Nikita – Martzopoulou
18 Diagnosis of the Air Quality for Matrices Raised in Collective Stalls and Individual Cages 117
Campos, J.A., Tinoco, I.F.F., Silva,J.N., Baêta, F.C., Cruz, V.F., Mauri, A.L.
19 Diagnosis of the CO and CO2 concentration in the broiler chicken production in half-acclimatized 119
poultry facilities
Menegali, I.; Tinôco, I.F.F.; Baêta, F.C.; Guimarães, M.C.C.; Cordeiro, M.B.
3
Issues related to livestock housing under hot climatic conditions including

the animals’ response to high temperatures
T. Banhazi1, A. Aarnink2, H. Thuy3, S. Pedersen4, J. Hartung5, E. Maltz6, H. Payne7, B. Mullan7
and D. Berckmans8
1
Livestock Systems Alliance, South Australian Research and Development Institute, Roseworthy
Campus, Adelaide University, Roseworthy, SA 5371;
2
Animal Sciences Group van Wageningen UR, Divisie Veehouderij, Postbus 65, 8200 AB Lelystad;
3
HCMC Department of Animal Health, Ministry of Agricultural and Rural Development, 151 Ly
Thuong Kiet ward 7, distr. 11, HCMC, Vietnam,
4
Danish Institute of Agricultural Sciences, Research Centre Bygholm, DK-8700 Horsens, Denmark;
5
Institute of Animal Hygiene, Welfare and Behaviour, University of Veterinary Medicine Hanover,
Foundation Buenteweg 17 P 30559 Hannover;
6
Volcani Research Center P.O. Box 6, Bet Dagan, 50250, Israel;
7
Animal Research and Development, Department of Agriculture and Food, Locked Bag No 4, Bentley
Delivery Centre, Western Australia 6983,
8
M3-BIORES, Katholieke Universiteit Leuven, Kasteelpark Arenberg 30, Leuven - Belgium;
Key words: heat stress, critical temperatures, intensive livestock, agricultural buildings,
environmental modelling, heat production, cooling methods
Introduction
Temperature is one of the most important environmental variables that can affect the health,
welfare, the production efficiency of domesticated animals and thus the profitability of
livestock farming. It is therefore essential to understand the response of intensively housed
livestock species to the thermal environment to minimize the negative effects of sub-optimal
temperatures. Thus, the main aim of this article is to review issues related to heat stress in
animals kept in livestock buildings; with a focus on animals’ responses to hot environments.
This review will specifically focus on animals kept in livestock buildings, such as pigs, poultry
and dairy cattle, and will not deal with issues related to heat stress experienced by livestock
kept outdoors. This review is also intended to encourage research that will improve building
management and climate control practices and thus alleviate the negative effects of heat stress
on domesticated animals. Although we aimed to include a variety of livestock species in the
review, due to the expertise of the review team most emphasis was placed on issues related to
the thermal environment of domesticated pigs.
Heat and moisture production

Heat and moisture production of domestic animals has been examined over the last five
decades (Brown-Brandl et al., 2003; Heitman and Hughes, 1949; Holmes and Mount, 1967;
Monteith and Mount, 1974; Strøm, 1978; Xin et al., 2001a, b). One of the first studies
establishing the relationship between heat production, evaporative and non-evaporative heat
loss in homoeothermic animals was published in the early seventies (Monteith and Mount,
1973). Total heat production is independent of environmental temperature within a certain
range of environmental temperature. At low environmental temperatures, total heat production
has a negative relationship with ambient temperature (Monteith and Mount, 1974). However,
there are conflicting opinions in relation to heat production of animals at high environmental
temperatures.
Under typical production conditions in hot climate countries the outdoor temperature changes
slowly over days and thus the indoor temperature also changes gradually. Under such
4
circumstances, the animals have the opportunity to adapt to the higher temperatures by
decreasing their feed intake, resulting in lower total heat production and metabolic rate.
However, in situations where animals do not have the capacity to adapt to a rapidly changing
environmental temperature (for example, sudden increase in temperature caused by abrupt
failure of ventilation system) it is likely that total heat production in some cases will increase
due to increased animal activity and continuation of unabated feed consumption.
A general model for animal heat production was published by (CIGR, 1984), based mainly on
studies undertaken by (Strøm, 1978). The graph representing the total heat production of all
species and ages is shown in Figure 1. The heat production unit (HPU) corresponds to 1000 W
of total heat at 20°C.
1200
Animal heat production, W
T otal heat
1000
800 Latent
600
Sensible
400 heat
200
0
10 20 o
30 40
T emperature, C
Figure 1 Total heat production. General model (Strøm 1978, CIGR 1984)
The graphs for total, sensible and latent heat are continuously changing and thus there is no
clear indication of a thermo-neutral zone. However, at around 20 oC, (approx. 15 to 25 oC) the
total heat production is nearly constant (Brown-Brandl et al., 2004; Huynh et al., 2005b). The
literature over the last two decades indicates that confinement buildings under normal
production conditions do not show a typical thermo-neutral zone, even if it theoretically exists
(Quiniou et al., 2001; Wachenfelt et al., 2001). Total heat production at increasing
temperatures is much greater for large animals (i.e. cattle) than for smaller animals (i.e.
poultry). In addition, the decrease in total heat production at increasing ambient temperatures
depends on the type of feeding (restriction or ad-libitum) and housing of animals (group-
housed or individual-housed). That is why individual graphs need to be developed for different
species and sizes or use made of computer models to calculate for each particular example.
Another issue is the difference between animal heat production at house level as compared to
animal heat production at animal level. The latent heat is normally higher at house level than
on animal level due to evaporation from spilt water and manure. Figure 1 is to be considered as
heat production at animal level because it is based mainly on laboratory measurements. Over
the last decade many field tests have been carried out in commercial production units. In
Figure 2 the total heat production of pigs at house level shows (CIGR, 2002) that the total heat
production decreases by 1.2 % per oC as the indoor temperature increases. More figures are
available for cattle, pigs and poultry in a recent CIGR publication (CIGR, 2002).
5
1200
Animal heat production, W

1000 Total heat
800
Latent
600
Sensible
400 heat
200
0
10 20 30 40
o
Temperature, C
Figure 2 Total heat production for pigs, CIGR 2002
Another important factor is the diurnal variation in animal heat production. Figure 3 shows a
typical diurnal variation in animal heat production in confinement buildings, where the animal
heat production is 20% higher during daytime than at night (Pedersen, 2005).
Activity
1.5
1.0
Activity factor
0.5
0.0
0 4 8 12 16 20 24
Time of the day
Figure 3 Standard correction of animal heat production due to diurnal variation (dromedary model).
The diurnal variation in animal heat production is related to diurnal changes in human activity,
such as feeding, cleaning and other work activities. For example it was demonstrated that it is
possible to reduce daytime peak in heat production in broilers, when animal activity was kept
constant over 24 hours by feeding the birds ad libitum and using constant artificial light
(Pedersen, 2005). This knowledge should be utilized more extensively in relation to reducing
heat stress under commercial conditions; e.g. by avoiding feeding the animals in the middle of
the day during summer.
6
In addition, US researchers claimed that ASABE standards currently used for calculating heat
and moisture production are probably outdated as they are based on information collected in
the late 60’s (Brown-Brandl et al., 2001). Feeding regimes, the environment maintained in
livestock buildings, management practices, growth rate, genotype of pig breeds used currently,
including expected body composition have all changed considerably in the last few decades.
These changes would have substantially altered expected heat and moisture production rates.
The authors also argue that due these changes it is likely that the newer genetic lines of pig
breeds are more susceptible to heat stress than their older counterparts (Brown-Brandl et al.,
2003). Thus effective reduction of heat stress, especially for modern pig breeds becomes a
very important management issue (Brown-Brandl et al., 2004).
Thermal comfort
Thermal Comfort Zone (TCZ) or Thermo Neutral Zone (TNZ) was defined as the range of
ambient temperature over which physiological functions of animals are maintained with
minimum energy utilization (Mount, 1968). This concept is not restricted to the domestic pig
but it is typical for all endotherm vertebrate species. The temperature bordering the TNZ on
the lower end is called the Lower Critical Temperature (LCT) and the upper limit of the TNZ is
called the Evaporative Critical Temperature (ECT) (Black et al., 2001; Black et al., 1993).
These temperatures vary according to the complex interactions between environmental
conditions and animal related factors. The thermal environment of intensively housed pigs is
markedly influenced by air temperature, humidity and airspeed (Black et al., 1999; Boon, 1978,
1982; Riskowski and Bundy, 1995). However, additional factors such as the age of animals,
floor type, stocking rate, extent of skin wetness and nutrition also have an important influence
on how individual animals are affected by the thermal conditions in the building (Botermans
and Andersson, 1995; Geers et al., 1989; Jones and Nicol, 1998). Gates et al (1991) referred to
this subjective thermal comfort as the ‘perceived’ thermal environment (Gates et al., 1991).
Therefore, accurately defining air temperature requirements of the different livestock species
kept under different housing conditions is very difficult. Although many publications
(Bockisch et al., 1999; Kruger et al., 1992; Le Dividich and Herpin, 1994; Seedorf et al., 1998)
provide tables of recommended temperatures for different livestock species in various life
stages, these recommendations should only be used as a guide. Due to the difficulties
associated with precisely determining the upper and lower temperature limits of the TNZ under
commercial conditions it is suggested that animals might be provided with housing where they
can freely choose an area in the pen within their TNZ. That can be achieved, for example, by
zone heating/cooling to alter ambient air temperature in a few positions within the livestock
building (Aarnink et al., 1996; Zhang et al., 2001).
Effects of high temperatures
Several definitions exist for upper critical temperature (UCT). It was suggested that the UCT
is the point above which the core temperature and frequency of respiration will rise (Heitman
and Hughes, 1949) or a temperature at which a pig with a dry skin can maintain its maximal
rate of heat loss (Holmes and Mount, 1967). Furthermore, the UCT was defined as a distinct
temperature above which there is a strong decrease in voluntary feed intake (Nienaber et al.,
1997). While UCT can be defined in a number of ways, most of the authors agree that at the
UCT different behavioural or physiological changes will occur facilitating the animals’ attempt
to maintain homeostasis.
A study on acute heat stress on 84 kg high lean barrows showed breaking temperatures of 180C
at which respiration rate increased, and at 280C a decrease in feed intake occurred (Brown-
Brandl et al., 2001). A broken line model to derive inflection point temperatures for lying and
excretion behaviour of pigs was constructed recently (Figure 4). The authors reported pigs
7
altered their lying and excreting behaviour when ambient temperature was above 250C at 25 kg
to 200C for a 100 kg animal (Aarnink et al., 2006).
18
2100
IPt VFI
-1
-1
Total heat production, MJ pig d
Voluntary feed intake, g pig d

-1
-1
16
IPt HP 1800
14
1500
12 1200
14 16 18 20 22 24 26 28 30 32 34
0
HP 50%RH HP 65%RH HP 80%RH Ambient temperature, C
VFI 50%RH VFI 65%RH VFI 80%RH
Figure 4. Broken line responses to temperatures of feed intake and total heat production. (IPt VFI,
inflection point temperature for voluntary feed intake at 65 % relative humidity; IPt HP; inflection
point temperature for total heat production at 65 % relative humidity)
Following the broken line model, inflection point temperatures for various animal responses
such as respiration rate, rectal temperature, heat production, feed intake, wallowing, and lying
behaviour were established (Figure 5) (Huynh et al., 2005a; Huynh et al., 2006). For pigs of
60 kg these authors found an increase in respiration rate in the ambient temperature range from
21.30C - 23.40C. The lower range corresponded with a high humidity of 80%, while the upper
range corresponded with a lower humidity of 50%. Likewise, voluntary feed intake decreased
between 25.40C – 25.60C and eventually rectal temperature increased in the ambient
temperature range from 24.60C – 27.10C.
Decrease Decrease
heat feed
production intake
0 0 0 0 0 0 0 0 0
16 C 18 C 19 C 20 C 22 C 23 C 24 C 25 C 26 C
Start to Increase Start to Increase Increase Increase

wallow, lying on defecate water respiration rectal
reduce slatted on solid intake rate temperature
huddling floor floor
Figure 5. Adaptation of pigs to increasing temperatures; (the temperature scale should be read from
left to right)
8
Animal response to heat stress

Behavioural responses
The first visible sign of how the pig reacts to an increasing ambient temperature is a change in
behaviour (Huynh et al., 2005a). Pigs facing environmental temperatures above the TNZ can
employ a number of behaviour strategies to maintain deep body temperature. A pig in a hot
environment can; (1) reduce activity, (2) spread out, (3) modify laying behaviour, (4) reduce
contacts with other pigs and (5) seek to wet their skin.
Heat-stressed pigs will generally spend more time lying down when resting in groups and they
will spread out at high temperatures (Blackshaw, 1994; Ekkel et al., 2003; Olsen et al., 2001).
If the floor temperature is cooler than air temperature, pigs will change their lying position to
ensure that heat loss via the floor is maximised. It was demonstrated that at ambient
temperatures of 16 to 190C pigs lie more on their lateral side and avoid contact with other pigs
(Huynh et al., 2005a). Recent studies demonstrated that the pigs altered their lying behaviour
from solid floor to slatted floor above an average temperature of 22.60C (Aarnink et al., 2006).
Pigs will also seek to rest on the coolest part of the pen such as slatted areas where air
movement is greatest. As a result, the lying areas within pens (solid concrete floor) tend to
become soiled at higher temperatures.
Pen dirtiness is an important factor influencing ammonia emission (Hacker et al., 1994). A
study on ammonia emission showed that the extent of urine contamination on the body of
animals and pen floors positively related with ammonia emission (Aarnink and Elzing, 1998;
Aarnink et al., 1997). The average ammonia emission from a partially slatted pen floor was
40% of the total emission. In another study, pen fouling was higher in summer than in winter
and pen fouling increased towards the end of growing period (Aarnink et al., 2000). Reduction
in the level of pen hygiene usually leads to increased concentrations of a number of airborne
pollutants (Banhazi et al., 2004b; Banhazi et al., 2005a, b), providing immunological challenge
to heat stressed animals (Banhazi and Cargill, 1998; Wathes et al., 2004).
The domestic pig is very lightly covered by hair that allows evaporation from the skin.
However, as pigs do not sweat when they are exposed to heat, body cooling is based on
wallowing or skin wetting. If adequate facilities are not provided to pigs to allow them to wet
their skin they will wallow in their own excrements in order to cool down at higher
temperature thus making pens unhygienic (Huynh et al., 2005a). They will also seek any other
means to wet their skin (i.e. splash water from drinkers over themselves) in order to maximise
evaporative heat loss from the body surface. In the wild, wallowing occurs at low temperatures
(Schein and Hafez, 1969) and in captive pigs wallowing also occurs at relatively low ambient
temperatures at around 16 to 170C (Huynh et al., 2005a).
Physiological responses
The first physiological indicator that pigs are reacting to high ambient temperatures is an
increase in respiration rate. This occurs at 22.40C for pigs of approx. 60 kg (Brown-Brandl et
al., 2004; Huynh et al., 2005a). As air temperature approaches the upper limit of the TCZ; pigs
will increase their respiration rate (panting) to maximise evaporative heat loss via the lungs. In
a field study under tropical conditions, it was found that respiration rate of the pigs in the
afternoon (average temperature 320C) was significantly higher than the respiration rate in the
morning (average temperature 260C): 64.8 breathes min-1 vs. 36.9 breathes min-1 (Huynh et al.,
2005a; Huynh et al., 2006).
9
Various authors have suggested that rectal temperature is an important indicator of heat-
stressed animals (Close, 1971; Holmes, 1973; Kadzere et al., 2002). Rectal temperature will
rise above ambient temperatures of 26.10C (Huynh et al., 2005a). A higher rectal temperature
is a sign the pig is experiencing thermal discomfort, indicating heat stress on organs and the
digestive tract.
Consequences of high temperatures

Production, welfare and health effects
When temperatures exceed ECT for extended periods, growing pigs reduce feed intake by 10%
to 20% with a corresponding reduction in growth rate. This reduction in feed intake can be
disadvantageous in a number of ways. Apart from the immediate reduction in growth; pigs
that had reduced feed intake often compensate by eating more later on and thus may be fatter at
slaughter (Trezona et al., 1999). Experiments conducted by Brown-Brandl et al. (1997)
confirmed that a significant decrease of feed intake occurs in barrows exposed to temperatures
of 28 0C and 32 0C for 22 h, when compared to the same length of temperature exposure at 18
0
C and 24 0C. Correspondingly, water intake also increased at 32 0C (Brown-Brandl et al.,
1997).
Figure 4 shows inflection point temperatures for total heat production changes and voluntary
feed intake (VFI) changes. For each degree Celsius ambient temperature increase above
inflection point temperature, VFI declined steadily by an average of 95.5 g (Huynh et al.,
2005a). Other researchers reported that for 45 and 85 kg pigs an increasing ambient
temperature from 20 to 300C reduced VFI from 65 to 74 g d-1 per degree Celsius (Nienaber et
al., 1997). For the same temperature range the decrease in a separate study (Huynh et al.,
2005a) was 43 g. d-1, which was lower than the reduction in VFI found by the previously
mentioned research team (Nienaber et al., 1997). An effective result of panting was that the
pigs could maintain a constant VFI until a few degrees above the IPt for respiration rate (+
2.00C at RH 50 %, + 3.00C at RH 65 % and + 4.30C at RH 80 %) (Huynh et al., 2005a). In
other words, the pigs were able to maintain their level of VFI by increasing RR and exploiting
respiratory evaporative heat loss. Figure 4 shows that total heat production remained constant
until the IPt occurred.
A popular definition of welfare is the individual animal’s state as it attempts to cope with its
environment (Barnett and Hemsworth, 1990; Barnett et al., 2001; Hemsworth et al., 1995).
Therefore, the welfare of animals can be regarded as ‘poor’ when they have difficulties coping
with their environment. For example, the postural behaviour of heat-stressed pigs can be used
to assess the thermal environment and the welfare status of pigs. When temperature rose, the
pigs display signs of discomfort becoming inactive, avoiding physical contact with other pigs
and trying to spread out when resting. Thus with increasing temperature, the welfare of the
animals decreases, especially if high stocking densities are maintained.
A number of authors argue that due to the confinement housing used in intensive animal
production, (such as pigs, poultry and to some extent dairy cattle production systems), the
quality of environment will have a dominant effect on animal welfare (Feder, 1999; Hicks et
al., 1998; Minton, 1994; Salak-Johnson and McGlone, 2006). Unsuitable environments may
induce stress, which in turn will affect the animals’ ability to efficiently digest feed, direct
available nutrients to production and withstand disease challenges (Minton, 1994). The
reduction in digestive efficiency and immunological vigour is the result of the complex
humeral and bio-chemical processes that are induced by any stress, including heat stress.
10
Stressors can be classified as any outside influence that provokes an adaptive response in the
animal, thus engaging their biological reserves during the adaptive process, making these
biological resources unavailable for efficient production and maintenance of health. For
example, a study evaluating the effects of heat and social stress on immune indices of pigs
demonstrated that heat-stress (33 oC) interacted with social stress. Socially intermediate pigs
had higher levels of lymphocyte proliferation and antibody production than socially dominant
or subordinate pigs (Morrow-Tesch et al., 1994). Therefore it can be concluded that there is an
immunological cost from heat and social stress for pigs, which can direct energy away from
muscle growth (Hicks et al., 1998; Salak-Johnson and McGlone, 2006).
High ambient temperatures are also associated with a reduction in carcass fatness, which is
thought to be as a direct result of the decrease in voluntary feed intake (Le Dividich and
Herpin, 1994; Safranski et al., 1997). Ambient temperature can also have an effect on fat
distribution in the growing pig. High ambient temperatures are associated with a decrease in
the proportion of internal fat (viscera and leaf fat) due to changes in the activity of lipogenic
enzymes. However, the precise mechanisms by which ambient temperature can alter body fat
distribution is unclear.
Reproductive effects
Dairy cattle
The decrease in fertility of postpartum dairy cows inseminated in the summer compared to
cows inseminated in winter is a well-known phenomenon. However, the precise mechanism of
this effect has not been conclusively identified. Different factors contribute to this situation;
the most important are a consequence of increased temperature and humidity that result in a
decreased expression of overt oestrus and a reduction in appetite and dry matter intake.
(De Rensis and Scaramuzzi, 2003) summarised in their review that “There appear to be two
distinct and largely independent pathways by which heat stress leads to infertility. The first is a
direct effect of hyperthermia on the reproductive axis. The second is an indirect effect related
to the effects of heat stress on appetite and dry matter intake, both of which are produced by
heat stress. The consequence is a worsening energy balance and since the postpartum dairy
cow tends to be in negative balance, the consequences of heat stress on fertility are more likely
to be severe.” Numerous studies (see Rensis and Scaramuzzi 2003) indicate that both
pathways take the blame for this physiological phenomenon that has an enormous economical
impact. The dramatic decrease in conception rate during the hot season can range between 20-
30% compared to 65-75% in winter (Cavestany et al., 1985; De Rensis and Scaramuzzi, 2003;
Rensis et al., 2002). Obviously, this can lead to fluctuating annual production (with summer
milk shortages) and economical weakness.
Heat stress reduces the degree of dominance of the selected follicle and this can be seen as
reduced steroidogenic capacity of its theca and granulosa cells and a fall in blood estradiol
concentrations. Plasma progesterone levels can be increased or decreased depending on
whether the heat stress is acute or chronic, and on the metabolic state of the animal. These
endocrine changes reduce follicular activity and alter the ovulatory mechanism, leading to a
decrease in oocyte and embryo quality (Wolfenson et al., 2000). The uterine environment is
also modified, reducing the likelihood of embryo implantation. Appetite and dry matter intake
are both reduced by heat stress thus prolonging the postpartum period of negative energy
balance and increasing the calving-conception interval, particularly in high producing dairy
cows. The utilization of cooling systems may have a beneficial effect on fertility (Kadzere et
al., 2002) but dairy cows cooled in this way are still unable to match the fertility achieved in
11
winter. Recent studies suggest that the use of gonadotropins to induce follicular development
and ovulation can decrease the severity of seasonal postpartum infertility in dairy cows.
Domesticated pigs
High temperatures can also have a major impact on the reproductive performance of
domesticated pigs. When temperatures remain high for long periods boar fertility may decline,
with a slight reduction in ejaculate volume, decreased sperm motility, reduced sperm numbers,
increased abnormal sperm and lower conception rates (Kunavongkrit et al., 2005;
Suriyasomboon et al., 2004). It may take six weeks for the effects of heat stress on boar semen
to become apparent and a further six weeks for semen quality to return to normal (Wetterman
and Bazer, 1985). The effects of short bursts of high temperature can be worse than a
sustained period of hot weather where there is an opportunity for the animal to adapt to the new
conditions. Fertility may be reduced by up to 30% when sows and gilts are mated in summer.
Embryonic deaths may occur early in gestation, resulting in poor returns to service or very
small litters. The effect of heat stress is one component of seasonal infertility which can have a
dramatic impact on overall pig supply (Suriyasomboon et al., 2006). One study suggested that
individual pigs susceptible to stress are more likely to suffer from seasonal infertility than pigs
with a placid nature (Wan et al., 1994).
The main effect of heat stress on lactating sows is to reduce feed intake, which reduces the
amount of energy for growth and production. A number of experiments have shown that feed
intake of sows might be reduced by over 40% when air temperature increases from 18°C to
30°C (Black et al., 1993; Quiniou et al., 2000). The major effect of reduced feed intake is loss
of body weight and condition, which may increase the weaning-to-mating interval and the
number of delayed returns to oestrus. Milk production may also be reduced by heat stress,
possibly because blood is diverted from the udder to the skin to assist in heat loss (Black et al.,
1993). More importantly, the negative effects of heat stress on lactating sows is greater than
the expected effects of simple feed intake reduction, as it was demonstrated that the reduction
in milk yield of sows exposed to high temperatures is usually greater than would be expected
from an equivalent decline in food intake for sows housed under thermoneutral conditions
(Black et al., 1993). Weaning weights of piglets are reduced when sows are exposed to
prolonged periods of temperatures above 25°C (Quiniou et al., 2000; Quiniou et al., 1999;
Quiniou et al., 2001).
Primiparous sows lactating during the summer have a longer interval between weaning and
mating than do those during winter (Clark et al., 1986; Cox et al., 1987). Season does not,
however, appear to have the same effect with multiparous sows (Clark et al., 1986) and it has
been suggested by (Fernandes et al., 1990) that this is because the younger animal mobilises a
greater proportion of its more limited body reserves if voluntary food intake is low during
lactation. High ambient temperatures during lactation cause a decrease in luteinising hormone
pulse frequency and this might be responsible for the delay in breeding after weaning (Barb et
al., 1991), but low nutrient intakes per se can also have a direct effect on the normal secretory
pattern of luteinising hormone (Mullan et al., 1989) and hence subsequent reproductive
performance.
Traditional ways of alleviating heat stress

Cooling methods
It is possible to reduce the effects of high temperatures using different cooling methods (Dong
et al., 2001; Huynh et al., 2006). Investment in simple cooling systems is usually cost-
effective with a short payback period and helps to dramatically alleviate heat related problems
12
such as reduced feed intake, weight gain and breeding performance. For example, a study
assessing the production benefits of spray cooling of grower pigs demonstrated that a 5.8%
increase in daily feed intake, a 4.5% improvement in feed conversion and an 11.2% increase
(p<0.05) in average daily gain could be achieved on Australian pig farms (Banhazi et al.,
2001a).
Cooling a solid floor during hot periods in the summer improves the feed intake and daily gain
(Bull et al., 1997; Cui et al., 2002; Eigenberg et al., 2002; Huynh et al., 2006; Lucas et al.,
2000). The cool floor also improves the lying behaviour: more pigs choose to lie on the cool
solid floor instead of the slatted floor (Huynh et al., 2005a). The temperature of the cooled
solid floor should be the lower than that of the slatted floor to ensure that pigs are attracted to
the rest area (Shi et al., 2006).
Pigs kept in a hot humid climate used the water bath and sprinklers intensively (Huynh et al.,
2005a). The same authors reported that on average, each pig used sprinklers 4.7 times of the 12
sprinkling periods daily between 10.00 and 16.00. The pigs used the water bath on average 7.4
times per day. The intensive bathing time was between 1400 and 1700. In addition, the pigs
made use of the wet floor under the sprinklers to wet themselves when the sprinklers were not
operational.
It is essential (and economically justifiable) to install well-controlled sprinkling systems in

piggery buildings (Lucas et al., 2000). The amount of water required is about 330 ml per pig
per hour for spray cooling dry sows, boars, growers and finishers, and drip cooling of sows.
For weaners, 65 ml per hour is recommended. A typical spray cycle for growers and finisher
pigs would be five minutes spraying followed by a 45 minute delay. Adequate ventilation is
essential for drip and spray cooling to be effective. A minimum air speed of 0.2 metres per
second at pig level is essential, but there is no advantage in exceeding 1.0 metre per second.
Too much air movement can chill even fully grown pigs (Hahn et al., 1987; Riskowski and
Bundy, 1995; Riskowski et al., 1990). Therefore it is important that only part of the pen area is
under sprays, so pigs can choose to stay dry, as the animal itself is the best sensor of heat
stress. Spray cooling is also viable cooling method for poultry (Tao and Xin, 2003) and has
been shown to positively influence egg production (Ikeguchi and Xin, 2001; Xin and Puma,
2001).
Nutritional/dietary solutions
Increasing the nutrient density of pig diets by decreasing the fiber and adding fat to compensate
for lower feed intake and to reduce digestive heat production might be an option at higher
temperatures (Black et al., 1993). This will improve growth rate and feed conversion but may
increase carcass fatness depending on genotype. A study by (Spencer et al., 2005) showed that
in a hot environment, decreased crude protein (CP) content improved finishing pig average
daily gain (ADG) when dietary fat supplementation was low. High dietary fat inclusion during
heat stress improved ADG, especially when CP level was elevated. High-fat diets fed in a hot
environment increased pork color intensity by decreasing the glycolytic potential at slaughter
and elevating muscle pH.
The adverse effects of heat stress on poultry production could also be eliminated by a variety of
means, including nutritional manipulation of the diet. For example, the use of Virginiamycin
has been promoted in some countries to provide an effective aid for reducing mortality rate
under heat stress conditions (Teeter and Belay, 1996). Cooling birds by providing cool
drinking water may also be beneficial to welfare and production efficiency in hot weather
(Barnett and Newman, 1997) and might be beneficial for improving growth rate in pigs
(Banhazi et al., 2001c).
13
Analysing the detrimental effect of heat stress on dairy cows; (West, 2003) pointed out that
increasing nutrient density and maintaining normal rumen function might be beneficial during
episodes of high temperatures.
Improved housing design and management practices

Well designed housing, including appropriate orientation, insulation and shading can have a
major impact of the resulting thermal environment. Identifying the statistically important
aspects of housing design is important, as it can help the practical implementation of best
practice management and building practices into housing design.
The objective of a large project recently undertaken in Australia was to determine the effects of
housing and management factors on the resulting thermal environment in piggery buildings.
Statistical models developed identified key housing and management factors associated with
sub-optimal thermal conditions. The results highlighted the need for innovative warm climate
building designs, which should (according to the results of this study) incorporate automated
ridge vents, good quality roof insulation, steeper roof pitch and smaller compartments. The
existence of either heating or cooling equipment in the sheds significantly improved the control
capacity of buildings. It was also shown that the thermal control capacity of study buildings
declined with the age (Banhazi et al., 2004a). Evidence collected on commercial farms also
indicated that overcrowding could result in reduced production efficiency in grower finisher
pigs, especially under heat stress conditions (Banhazi et al., 2001b).
A similar statistical modelling study was undertaken recently in relation to cattle housing
(Shoshani et al., 2006) and the results indicated that improved lay-out of barn could minimize
heat stress in dairy cattle. Based on the study outcomes, it was concluded that an optimal barn
design for high milking cows should be a loose house type barn, with orientation perpendicular
to the dominant wind, with opened roof, opened ridge, with roof margins above 5 m, roof slope
at least 19% and width between 30 to 51 m (Shoshani et al., 2006).
A strong association was found between weight gain and housing orientation in a Brazilian
study (Moura and Naas, 2001). The study developed a mathematical model that associated the
use of shade, forced ventilation and different solar orientation with broiler weight gain, thus
confirming the important affect of these building parameters on production efficiency.
These studies could help improving building construction practices, optimising building design
and management.
Innovative technologies to alleviate heat stress

Modelling animal responses to thermal environment
While we understand many of the effects that high temperatures have on the physiology of
farm animals it is far more difficult to calculate how the various factors interact with each
other, and even more difficult to then accurately predict what might happen under particular
circumstances. Computer models have been developed by several groups to help us
understand the impact of high temperatures, and other factors, on animal performance. One of
the most sophisticated models is the AUSPIG model developed by (Black et al., 1986) for the
pig, which takes into account such factors as ambient temperature, wind speed, nutrient intake,
stocking density, genotype and live weight. As an example, AUSPIG was then used by Mullan
(Mullan et al., 1992) to calculate the lower critical temperature of the sow during gestation,
taking into account feed intake, wind speed, group versus single housing, bedding and floor
14
type. The combined effect that these factors have on the LCT is depicted in Figure 6 and the
same calculation can be done to look at the effect of high temperatures. Such models can then
be used to predict the benefits of, for example, installing cooling equipment in an animal
facility and to predict the likely improvement in performance and hence economic benefit.
Ultimately computer models could be linked to environmental monitoring equipment and
changes made automatically to diet and feeding level so as to minimise the impact of high
temperatures on animal performance and hence profitability.
Feeding 2 maintenance 1.5 maintenance
Flooring Straw Concrate
Air speed Still Draught Still Draught
Stocking Group Single Group Single
15 16 17 18 19 20 21 22 23 24 25
o
Estimated LCT ( C)
Figure 6. Effects of environment and management on LCT of pregnant sows (Mullan, 1991)
A Belgian research group used a different, so called dynamic data-based modeling approach to
control the metabolic response of broiler chickens to the building environment. The group
demonstrated that the time-variant response of animal growth to food supply could be
predicted on-line with a maximum prediction error of 5% and approximately 3-7 days ahead
depending on the type of feeding schedule. The study also highlighted the potential conflicts
between environmental, financial and biological pressures on sustainable poultry production
that might be effectively resolved via the development of a futuristic management systems
incorporating modern process control techniques (Aerts et al., 2003).
As there could be some difference in the outcomes based on the modelling approach, a study in
the US was undertaken to evaluate different modeling techniques (such as statistical, fuzzy
inference and neural network models) aimed at describing the physiological responses of
feedlot cattle to environmental conditions. Results of this specific study showed that the neural
network modeling approach described the most variation in the test data (68%) (Brown-Brandl
et al., 2005).
Mathematical models are also frequently used to predict animal response. A model, based on
heat transfer principles, was developed to predict the thermal status of pigs and broiler
chickens in livestock buildings under different environmental conditions. The model was tested
by comparing predicted values generated by the model with actual measured values of heat
loss, skin and body temperature recorded by previous studies. For a single newborn pig, the
model underestimated (overall error of -9%) heat loss over the range of wind speeds likely to
be experienced indoors. On the other hand, the model over-predicted heat loss by an average of
15
20%, probably due to the absence of huddling response incorporated in the model (Turnpenny
et al., 2000).
These previous publications demonstrated that a variety of modeling approaches can be used to
model the animals’ response to environmental conditions. The accuracy of the outputs from
these models depends on the quality of input data and the precision of algorithms developed by
the various research teams.
Modeling buildings and climate control systems

A number of models describing the behaviour of climate control systems and micro-
meteorological processes within various animals buildings were also developed due to the fact
that model based climate control is increasingly favoured as compared to simple control
systems based on maximum and minimum limits (Soldatos et al., 2005; van 't Klooster et al.,
1995; Van Wagenberg et al., 2005).
For example, in order to improve the effectiveness of control techniques; an adaptive systems
for simultaneous temperature/humidity control of livestock buildings and a control algorithm
based building temperature, the velocity and the direction of wind were developed recently
(Daskalov et al., 2005; Daskalov et al., 2006). In addition, the dynamic auto-regressive
moving average models developed by the same team (Daskalov, 1997) are promoted as
potential tools to calculate the required supplementary heating and cooling capacity of piggery
buildings.
Results of studies undertaken by North American research teams indicated that existing staged
ventilation control systems could be significantly improved by applying a simple modification
to allow the incorporation of fuzzy logic based controller method (Gates et al., 2001; Gates et
al., 1997). Furthermore, it was also demonstrated that the incorporation of time integrated
control algorithms in ventilation system (that takes the thermal history of both the air space and
the animal into consideration) offers additional benefits (Timmons et al., 1995).
Combined models
A radically new climate control method is also proposed by Belgian workers. They argue that
although a range of new technologies has become available for ventilation, heating and cooling
of livestock buildings, limited progress (such as the ones described before) has been made in
the development of adequate control algorithms. To obtain more benefit from the new
technologies developed, more knowledge about the interaction between the animal responses
(e.g. heat production) and the control actions has to be integrated into the applied control
algorithms. Although several authors tried to give some general guidelines about the use of
ventilation control equipment for livestock buildings (Andonov et al., 2003; Ouwerkerk and
Pedersen, 1994; van 't Klooster et al., 1995; Van't Klooster and Heitlager, 1994; Zhang and
Barber, 1995), there is no internationally accepted procedure and there is no agreement
between the guidelines for tuning ventilation controllers in livestock buildings. Consequently,
in many livestock buildings the indoor climate control is sub-optimal.
An appropriate technique to develop and optimise control algorithms is the use of simulation
models. In the last decades, different authors have described mathematical models to simulate
the indoor climate in livestock buildings (Mitchell, 1993; Ouwerkerk and Pedersen, 1994;
Overhults and Gates, 1997; Timmons et al., 1995; Zhang and Barber, 1995) but most of them
assume constant steady state heat, moisture and gas balances applied to a perfectly mixed
ventilated space. By combining steady state equations with the equations that describe the
controller actions, the resulting indoor climatic conditions at different outdoor temperatures can
16
be calculated, but they do not take into account the dynamic responses of animal heat
production. Such an approach is only acceptable for evaluating the indoor temperature if the
temperature difference between the indoor and the outdoor temperature is high and the daily
fluctuations of the outdoor temperature is relatively low (Berckmans, 1986).
In the evaluation of the efficiency of ventilation and cooling strategies, a more realistic
dynamic modelling approach is suggested by the Belgian team. This means that, due to the
dynamic changes of inputs (animal heat production, ventilation rate, cooling rate, etc.) and
disturbing factors (outdoor climate); the calculated indoor conditions rapidly change over time.
In order to describe the dynamics of such a complex process like the temperature and humidity
in a livestock building, a set of differential equations can be written. However, previously
published research papers revealed that other authors (van't Klooster et al., 1995; Zhang et al.,
1992) did not incorporate the dynamic responses of animal heat and moisture production into
their model analysis and did not perform validation work under real field conditions.
Consequently the reliability of such simulation models remains questionable. The section
below will briefly describe the global simulation model proposed by the Belgian research team
to control the rapidly changing of indoor environment of livestock buildings.
The global climate simulation model, adapted from (Berckmans et al., 1992), consists of
different sub models, as presented in figure 7.
Outdoor climatic data

Control settings
Xo To
2
Temperature Qs
Heating system Xi
controller
Temperature
Ti
set point
Φa Φa
Fan
1 3
Process
4
5
Temperature sensor
Figure 7: The different sub models in the climate dynamic simulation model’s, heat supply of
heating element; Ti , indoor temperature; To ,external temperature; Xi, indoor absolute humidity; Xo,
outdoor absolute humidity; Φa, air flow
These components describe (by a set of mathematical equations) the dynamic behaviour of
different sub-systems of a mechanically ventilated pig house, such as the (1) controller, (2) the
heating system, (3) the fan, (4) the process of heat and mass exchange within the ventilated
structure and (5) the temperature sensor. The inputs of the model are the control settings as a
function of time. Disturbance variables are the outdoor climatic data of temperature and
humidity, originating from a reference year.
17
The mechanistic model simulates the dynamic behaviour of outside temperature and humidity,
the sensible and latent heat production of the animals, and hence calculates the dynamic
behaviour of the resulting inside temperature and humidity and the corresponding energy use
for heating and ventilation.
Input variables in the different sub-models are outside weather conditions, number and weight
of animals, data on their feed intake in relation to maintenance requirements, building thermal
characteristics, as well as control settings for ventilation and cooling. Building characteristics
include wall and floor heat transfer coefficients and their thermal capacities, as well as the
dimensions of the compartments.
The model described allows users to conduct dynamic analyses to evaluate the effect of control
systems on the dynamic behaviour of indoor climate. Moreover it provides a vehicle to account
for the dynamic behaviour of animal heat and moisture production during modelling. This type
of analysis might be used for design of ventilation control equipment, such as fans, heating and
cooling equipment.
Summary and conclusions

• The knowledge acquired about heat and moisture production of domesticated animals over
a number of decades confirmed that although Thermal Comfort Zone (TCZ) do exist in
theory, under practical conditions it is extremely difficult to determine for different
domesticated species living under a variety of housing conditions.
• However, if intensively hosed animals are forced to live outside of their TCZ; their
behaviour and physiology will be negatively affected. Hence, the behaviour of the animals
(if objectively measured) might be used to indicate if the animals within their TCZ.
Overall, it is likely that production efficiency, welfare, health, value of the carcass and
reproductive capacity of animals will be reduced. The space requirement of various
livestock species will increase and the hygiene level of pig growing facilities will likely to
decrease, creating to additional health challenges and negative environmental
consequences.
• Traditional technologies, such as the use of different cooling system, implementation of
improved nutrition regimes, genetic selection and better building design can be used to
alleviate the negative effects of heat stress on animals.
• However, there are real opportunities to further develop and improve advanced (PLF
related) climate control technologies and create intelligent environmental control systems,
which will be able to predict and therefore control both the responses of animals and the
buildings in relation to selected control interventions.
• There might be an exciting possibility of combining the best available animal and buildings
models into a “universal” system, which can be then used to reduce the negative impact of
heat stress on domesticated animals, improve animal welfare and environmental outcomes.
References
Aarnink, A. J. A., and A. Elzing. 1998. Dynamic model for ammonia volatilization in housing with
partially slatted floors for fattening pigs. Livestock Production Science 53: 153-169.
Aarnink, A. J. A., J. W. Schrama, M. J. W. Heetkamp, J. Stefanowska, and T. T. T. Huynh. 2006.
Temperature and body weight affect fouling of pig pens. Journal of Animal Science 84.
Aarnink, A. J. A., J. W. Schrama, R. J. E. Verheijen, and J. Stefanowska. 2000. Effect of ambient
temperature on pen fouling by pigs. In: 10th International Congress on Animal Hygiene,
Maastricht, The Netherlands. p 933-937.
18
Aarnink, A. J. A., D. Swierstra, A. J. van den Berg, and L. Speelman. 1997. Effect of type of slatted
floor and degree of fouling of solid floor on ammonia emission rates from fattening piggeries.
Journal of Agricultural Engineering Research 66: 93-102.
Aarnink, A. J. A., A. J. van den Berg, A. Keen, P. Hoeksma, and M. W. A. Verstegen. 1996. Effect of
slatted floor area on ammonia emission and on the excretory and lying behaviour of growing
pigs. Journal of Agricultural Engineering Research 64: 299-310.
Aerts, J.-M., C. M. Wathes, and D. Berckmans. 2003. Dynamic data-based modelling of heat
production and growth of broiler chickens: Development of an integrated management system.
Biosystems Engineering 84: 257-266.
Andonov, K., P. Daskalov, and K. Martev. 2003. A new approach to controlled natural ventilation of
livestock buildings. Biosystems Engineering 84: 91-100.
Banhazi, T., and C. Cargill. 1998. An assessment of the benefits of age-segregated rearing and all-in/all-
out management in respiratory disease free herds. In: Proceedings of the 15th IPVS Congress,
Birmingham, England. p 387.
Banhazi, T., C. Cargill, Z. Harper, and P. Glatz. 2001a. Assessing the benefits of spray cooling of pigs.
In: AgriBuilding 2001, Campinas, Brazil. p 152-156.
Banhazi, T., C. Cargill, Z. Harper, and P. Glatz. 2001b. The effect of stocking rate on pig production.
In: AgriBuilding 2001, Campinas, SP, Brazil. p 163-166.
Banhazi, T., C. Cargill, Z. Harper, J. Wegiel, and P. Glatz. 2001c. The effects of drinking water
temperatures on pig production. In: AgriBuilding 2001, Campinas, Brazil. p 157-162.
Banhazi, T., D. Rutley, and P. Glatz. 2004a. Factors effecting thermal conditions in piggery buildings in
south Australia. In: New trends in farm buildings, Evora, Portugal. p CD publication.
Banhazi, T., J. Seedorf, D. L. Rutley, and W. S. Pitchford. 2004b. Factors affecting the concentrations
of airborne particles in Australian piggery buildings. In: Proceedings of ISAH Conference, St
Malo, France. p 193-194.
Banhazi, T. M., J. Seedorf, D. L. Rutley, and W. S. Pitchford. 2005a. Statistical modelling of airborne
bacteria and endotoxins concentrations in Australian piggery buildings. In: ILES, Bejing,
China. p 72-78.
Banhazi, T. M., J. Seedorf, D. L. Rutley, and W. S. Pitchford. 2005b. Statistical modelling of gas
concentrations in Australian piggery buildings. In: ILES, Bejing, China. p 64-71.
Barb, C. R. et al. 1991. Endocrine changes in sows exposed to elevated ambient temperature during
lactation. Domestic Animal Endocrinology 8: 117-127.
Barnett, J. L., and P. H. Hemsworth. 1990. The validity of physiological and behavioural measures of
animal welfare. Applied Animal Behaviour Science 25: 177-187.
Barnett, J. L., P. H. Hemsworth, G. M. Cronin, E. C. Jongman, and G. D. Hutson. 2001. A review of the
welfare issues for sows and piglets in relation to housing. Australian Journal of Agricultural
Research 52: 1-28.
Barnett, J. L., and E. A. Newman. 1997. Review of welfare research in the laying hen and the research
and management implications for the Australian egg industry. Australian Journal of
Agricultural Research 48: 385-402.
Berckmans, D. 1986. Analyse van de klimaatbeheersing in dierlijke produktie-eenheden ter
optimalisering van de regeling - analysis of climate control in animal houses for optimal control
purposes. Ph.D, K.U of Leuven, Leuven.
Berckmans, D., M. Van Pee, and V. Goedseels. 1992. Evaluation of livestock environment by
simulation technique. In: International Summer meeting ASAE, Charlotte, North Carolina,
USA. p Paper no. 92-4055.
Black, J. L., H. J. Bray, and L. R. Giles. 1999. The thermal and infectious environment. In: Kyriazakis
(ed.) A quantitative biology of the pig. p 71-97. CABI Publishing.
Black, J. L., R. G. Campbell, I. H. Williams, K. J. James, and G. T. Davies. 1986. Simulation of energy
and amino acid utilisation in the pig. Research and Development in Agriculture 3: 121-145.
Black, J. L. et al. 2001. A review - factors limiting the performance of growing pigs in commercial
environments. In: Manipulating Pig Production VIII, Adelaide, Australia. p 9-36.
Black, J. L., B. P. Mullan, M. L. Lorschy, and L. R. Giles. 1993. Lactation in the sow during heat stress.
Livestock Production Science 35: 153-170.
19
Blackshaw, J. K. 1994. Behavioural and production studies of space utilisation by grouped dry sows,
farrowing sows and weaner pigs. Final report, University of Queensland.
Bockisch, F.-J., T. Jungbluth, and A. Rudovsky. 1999. Technical indicators for evaluation if housing
systems for cattle, pigs and laying hens relating to animal welfare. Zughtungskunde 71: 38-63.
Boon, C. R. 1978. Airflow patterns and temperature distribution in an experimental piggery. Journal of
Agricultural Engineering Research 23: 129-139.
Boon, C. R. 1982. The effect of air speed changes on the group postural behaviour of pigs. Journal of
Agricultural Engineering Research 27: 71-79.
Botermans, J., and M. Andersson. 1995. Growing-finishing pigs in an uninsulated house. 2. Pen
function and thermal comfort. Swedish Journal of Agricultural Research 25: 73-82.
Brown-Brandl, T. M., R. A. Eigenberg, J. A. Nienaber, and S. D. Kachman. 2001. Thermoregulatory
profile of a newer genetic line of pigs. Livestock Production Science 71: 253-260.
Brown-Brandl, T. M., D. D. Jones, and W. E. Woldt. 2005. Evaluating modelling techniques for cattle
heat stress prediction. Biosystems Engineering 91: 513-524.
Brown-Brandl, T. M., J. A. Nienaber, and L. W. Turner. 1997. Acute heat stress effects on heat
production and respiration rate in swine. ASAE Annual International Meeting, Minneapolis,
Minnesota, USA, 10 11 August, 1997 No. 974009: 11.
Brown-Brandl, T. M., J. A. Nienaber, H. Xin, and R. S. Gates. 2003. A literature review of swine heat
and moisture production. In: Swine Housings II, North Carolina USA. p 31-40.
Brown-Brandl, T. M., J. A. Nienaber, H. Xin, and R. S. Gates. 2004. A literature review of swine heat
production. Transactions of the ASAE 47: 259-270.
Bull, R. P., P. C. Harrison, G. L. Riskowski, and H. W. Gonyou. 1997. Preference among cooling
systems by gilts under heat stress. Journal of Animal Science 75: 2078-2083.
Cavestany, D., A. B. El-Whishy, and R. H. Foot. 1985. Effect of season and high environmental
temperature on fertility of Holstein cattle. Journal of Dairy Science 68: 1471-1478.
CIGR. 1984. Climatization of animal houses, Scottish Farm Building Investigation Unit, Craibstone,
Aberdeen, Scotland.
CIGR. 2002. Climatization of animal houses: Heat and moisture production at animal and house. DIAS,
Bygholm, Denmark.
Clark, J. R., A. Komkov, and L. F. Tribble. 1986. Effects of parity, season, gonadotropin releasing
hormone and altered suckling intensity on the interval to rebreeding in sows. Theriogenology
26: 299-308.
Close, W. H. 1971. The influence of environmental temperature and plane of nutrition on heat losses
from individual pigs. Animal Production 13: 295 - 302.
Cox, N. M., J. L. Ramirez, I. A. Matamoros, W. A. Bennett, and J. H. Britt. 1987. Influence of season
on oestrous and luteinizing hormone responses to estradiol benzoate in ovariectomized sows,.
Theriogenology 27: 395-405.
Cui, S. R., K. Y. Wang, X. W. Miao, R. Geers, and V. Goedseels. 2002. Experimental study on cooling
effect of growing pig housing in southern china. In: ASAE Annual International Meeting,
Chicago, USA. p Paper No 024178.
Daskalov, P., K. Arvanitis, N. Sigrimis, and J. Pitsilis. 2005. Development of an advanced microclimate
controller for naturally ventilated pig building. Computers and Electronics in Agriculture 49:
377-391.
Daskalov, P. I. 1997. Prediction of temperature and humidity in a naturally ventilated pig building.
Daskalov, P. I., K. G. Arvanitis, G. D. Pasgianos, and N. A. Sigrimis. 2006. Non-linear adaptive
temperature and humidity control in animal buildings. Biosystems Engineering 93: 1-24.
De Rensis, F., and R. J. Scaramuzzi. 2003. Heat stress and seasonal effects on reproduction in the dairy
cow - a review. Theriogenology 60: 1139-1151.
Dong, H., X. Tao, J. Lin, Y. Li, and H. Xin. 2001. Comparative evaluation of cooling systems for
farrowing sows. Applied Engineering in Agriculture 17: 91-96.
Eigenberg, R. A., J. A. Nienaber, G. L. Hahn, and S. D. Kachman. 2002. Swine response to misting
synchronized with meal events. Applied Engineering in Agriculture 18: 347-350.
Ekkel, E. D., H. A. M. Spoolder, I. Hulsegge, and H. Hopster. 2003. Lying characteristics as
determinants for space requirements in pigs. Applied Animal Behaviour Science 80: 19-30.
20
Feder, M. E. 1999. Organismal, ecological and evolutionary aspects of heat-shock proteins and the
stress response: Established conclusions and unresolved issues. American Zoologist 39: 857-
864.
Fernandes, L. C. O., A. C. Altafini, H. C. Kunert, and F. Carnino. 1990. Summer feed intake and weight
loss of primiparous and pluriparous sows associated with weaning to estrus interval. In:
International Pig Vet Society Meeting. p 412.
Gates, R. S., K. Chao, and N. Sigrimis. 2001. Identifying design parameters for fuzzy control of staged
ventilation control systems. Computers and Electronics in Agriculture 31: 61-74.
Gates, R. S., M. B. Timmons, D. G. Overhults, and R. W. Bottcher. 1997. Fuzzy reasoning for
environment control during periods of heat stress. In: Livestock Environment V - 5th
International Symposium, Bloomington, Minnesota. p 553-562.
Gates, R. S., J. L. Usry, J. A. Nienaber, L. W. Turner, and T. C. Bridges. 1991. An optimal misting
method for cooling livestock housing. Transactions of the ASAE 34: 2199-2206.
Geers, R. et al. 1989. An assessment of optimal air temperatures in pig houses by the quantification of
behavioral and health-related problems. Animal Production 48: 571-578.
Hacker, R. R., J. R. Ogilvie, W. D. Morrison, and F. Kains. 1994. Factors affecting excretory behavior
of pigs. Journal Of Animal Science 72: 1455-1460.
Hahn, G. L., J. A. Nianaber, and J. A. DeShazer. 1987. Air temperature influences on swine
performance and behaviour. Applied Engineering in Agriculture 3: 295-302.
Heitman, H. J., and E. H. Hughes. 1949. The effects of air temperature and relative humidity on the
physiological well being of swine. Journal of Animal Science 8: 171 - 181.
Hemsworth, P. H., J. L. Barnett, L. Beveridge, and L. R. Matthews. 1995. The welfare of extensively
managed dairy cattle: A review. Applied Animal Behaviour Science 42: 161-182.
Hicks, T. A., J. J. McGlone, C. S. Whisnant, H. G. Kattesh, and R. L. Norman. 1998. Behavioral,
endocrine, immune and performance measures for pigs exposed to acute stress. Journal of
Animal Science 76: 474–483.
Holmes, C. W. 1973. The energy and metabolism of pigs growing at a high temperature. Animal
Production 16: 117 - 133.
Holmes, C. W., and L. E. Mount. 1967. Heat loss from groups of growing pigs under various conditions
of environmental temperature and air movement. Animal Production 9: 435 - 452.
Huynh, T. T. T. et al. 2005a. Thermal behaviour of growing pigs in response to high temperature and
humidity. Applied Animal Behaviour Science 91: 1-16.
Huynh, T. T. T., A. J. A. Aarnink, C. T. Truong, B. Kemp, and M. W. A. Verstegen. 2006. Effects of
tropical climate and water cooling methods on growing pigs' responses. Livestock Science 104:
278-291.
Huynh, T. T. T. et al. 2005b. Effects of increasing temperatures on physiological changes in pigs at
different humidities. Journal of Animal Science 83: 1385-1396.
Ikeguchi, A., and H. Xin. 2001. Field evaluation of a sprinkling system for cooling commercial laying
hens in Iowa. Applied Engineering in Agriculture 17: 217-221.
Jones, R., and C. J. Nicol. 1998. A note on the effect of control of the thermal environment on the well-
being of growing pigs. Applied Animal Behaviour Science 60: 1-9.
Kadzere, C. T., M. R. Murphy, N. Silanikove, and E. Maltz. 2002. Heat stress in lactating dairy cows: A
review. Livestock Production Science 77: 59-91.
Kruger, I., G. Taylor, and F. Crosling. 1992. Summer cooling. NSW Agriculture, Tamworth, NSW,
Australia.
Kunavongkrit, A., A. Suriyasomboon, N. Lundeheim, T. W. Heard, and S. Einarsson. 2005.
Management and sperm production of boars under differing environmental conditions.
Theriogenology
Proceedings of the V International Conference on Boar Semen Preservation 63: 657-667.
Le Dividich, J., and P. Herpin. 1994. Effects of climatic conditions on the performance, metabolism and
health status of weaned piglets: A review. Livestock Production Science 38: 79-90.
Lucas, E. M., J. M. Randall, and J. F. Meneses. 2000. Potential for evaporative cooing during heat
stress periods in pig production in Portugal (Alentejo). Journal of Agricultural Engineering
Research 76: 363-371.
21
Minton, J. E. 1994. Function of the hypothalamic-pituitary-adrenal axis and the sympathetic nervous
system in models of acute stress in domestic farm animals. Journal of Animal Science 72:
1891-1898.
Mitchell, B. W. 1993. Process control system for poultry house environment. Transactions of the ASAE
36: 1881-1886.
Monteith, J. L., and L. E. Mount. 1973. Heat loss from animals and man. University of Nottingham
Twentieth Easter School in Agricultural Science, London, UK.
Monteith, J. L., and L. E. Mount. 1974. Heat loss from animals and man. University of Nottingham
Twentieth Easter School in Agricultural Science, London, UK.
Morrow-Tesch, J. L., J. J. McGlone, and J. L. Salak-Johnson. 1994. Heat and social stress effects on pig
immune measures. Journal of Animal Science 72: 2599-2609.
Mount, L. E. 1968. The climatic physiology of the pig. Edward Arnold, London.
Moura, D. J., and I. A. Naas. 2001. Development of a mathematical model to determine broiler weight
gain as a function of poultry housing orientation, shading and forced ventilation. In: Livestock
Environment VI. Proceedings of the Sixth International Symposium, Louisville, Kentucky. p
468-472.
Mullan, B. P. 1991. The response of the breeding sow to the climatic environment. In: Manipulating Pig
Production III, Albury, NSW, Australia. p 167-177.
Mullan, B. P., W. Brown, and M. Kerr. 1992. The response of the lactating sow to ambient temperature.
In: Proceedings from the Nutrition Society of Australia. p 215.
Mullan, B. P., W. H. Close, C. E. Sharpe, and A. M. Galer. 1989. The influence of feeding level and
litter size on the plasma concentration of lh during lactation and following weaning in first-
litter sows. Journal of Reproduction and Fertility Control of Pig Reproduction III: 101.
Nienaber, J. A. et al. 1997. Genetic and heat stress interaction effects on finishing swine. In: 5th
International Symposium - Livestock Environment V, Bloomington, Minnesota. p 1017-1023.
Olsen, A. W., L. Dybkjaer, and H. B. Simonsen. 2001. Behaviour of growing pigs kept in pens with
outdoor runs: Ii. Temperature regulatory behaviour, comfort behaviour and dunging
preferences. Livestock Production Science 69: 265-278.
Ouwerkerk, E. N. J., and S. Pedersen. 1994. Application of the carbon dioxide mass balance method to
evaluate ventilation rates in livestock buildings. In: XII World Congress on Agricultural
Engineering, Milano, Italy. p 516-529.
Overhults, D. G., and R. S. Gates. 1997. Energy requirements for operating tunnel ventilated broiler
houses. In: 5th International Symposium - Livestock Environment V, Bloomington, Minnesota.
p 859-865.
Pedersen, S. 2005. Climatization of animal houses - a biographical review of three decades of research.
66, DIAS, Bygholm, Denmark.
Quiniou, N., S. Dubois, and J. Noblet. 2000. Voluntary feed intake and feeding behaviour of group-
housed growing pigs are affected by ambient temperature and body weight. Livestock
Production Science 63: 245-253.
Quiniou, N., J. Noblet, J.-Y. Dourmad, and J. van Milgen. 1999. Influence of energy supply on growth
characteristics in pigs and consequences for growth modelling. Livestock Production Science
60: 317-328.
Quiniou, N., J. Noblet, J. van Milgen, and S. Dubois. 2001. Modelling heat and energy balance in
group-housed growing pigs exposed to low or high ambient temperatures. British Journal of
Nutrition 85: 97-106.
Rensis, F. D. et al. 2002. Fertility in postpartum dairy cows in winter or summer following estrus
synchronization and fixed time AI after the induction of an lh surge with GNRH or HCG.
Theriogenology 58: 1675-1687.
Riskowski, G. L., and D. S. Bundy. 1995. Effect of air velocity and temperature on growth performance
of weanling pigs. Transactions of the ASAE 33: 1669-1675.
Riskowski, G. L., D. S. Bundy, and J. A. Matthews. 1990. Huddling behavior and hematology of
weanling pigs as affected by air velocity and temperature. Transactions of the American
Society of Agricultural Engineers 33: 1677-1685.
22
Safranski, T. J., D. L. Harris, J. A. Nienaber, G. L. Hahn, and R. L. Korthals. 1997. Heat stress and limit
feeding pigs: Effects on growth and composition. In: 5th International Symposium - Livestock
Environment V, Bloomington, Minnesota. p 772-780.
Salak-Johnson, J. L., and J. J. McGlone. 2006. Making sense of apparently conflicting data: Stress and
immunity in swine and cattle. Journal of Animal Science 85: E81-88E.
Schein, M. W., and E. S. E. Hafez. 1969. The physical environment and behaviour. In: E. S. E. Hafez
(ed.) The behaviour of domestic animals. p 65 - 84. Bailliere, Tindall & Cassell, London.
Seedorf, J. et al. 1998. Temperature and moisture conditions in livestock buildings in northern Europe.
Shi, Z. et al. 2006. Using floor cooling as an approach to improve the thermal environment in the
sleeping area in an open pig house. Biosystems Engineering 93: 359-364.
Shoshani, E., A. Hetzroni, A. Levi, and R. Brikman. 2006. Optimal planning of barns for high yielding
cows under arid conditions. In: ISAE Annual Conference - Advances in agricultural
technologies, Tel Aviv, Israel. p 45.
Soldatos, A. G., K. G. Arvanitis, P. I. Daskalov, G. D. Pasgianos, and N. A. Sigrimis. 2005. Nonlinear
robust temperature-humidity control in livestock buildings. Computers and Electronics in
Agriculture 49: 357-376.
Spencer, J. D., A. M. Gaines, E. P. Berg, and G. L. Allee. 2005. Diet modifications to improve finishing
pig growth performance and pork quality attributes during periods of heat stress. Journal of
Animal Science 83: 243-254.
Strøm, J. S. 1978. Heat loss from cattle, swine and poultry as basis for design of environmental control
systems in livestock buildings (in Danish), Danish Building Research Institute, SBI-
Landbrugsbyggeri, Denmark.
Suriyasomboon, A., N. Lundeheim, A. Kunavongkrit, and S. Einarsson. 2004. Effect of temperature and
humidity on sperm production in Duroc boars under different housing systems in Thailand.
Livestock Production Science 89: 19-31.
Suriyasomboon, A., N. Lundeheim, A. Kunavongkrit, and S. Einarsson. 2006. Effect of temperature and
humidity on reproductive performance of crossbred sows in Thailand. Theriogenology 65: 606-
628.
Tao, X., and H. Xin. 2003. Surface wetting and its optimization to cool broiler chickens. Transactions
of the ASAE 46: 483–490.
Teeter, R. G., and T. Belay. 1996. Broiler management during acute heat stress. Animal Feed Science
and Technology
N. American Nutrition Conferences 58: 127-142.
Timmons, M. B. et al. 1995. Simulation analysis of a new temperature control method for poultry
housing. Journal of Agricultural Engineering Research 62: 237-245.
Trezona, M., B. P. Mullan, R. H. Wilson, and I. H. Williams. 1999. Pattern of food intake during the
growing phase determines depth of backfat in pigs at slaughter. In: Manipulating Pig
Production VII, Adelaide, South Australia. p 119.
Turnpenny, J. R., C. M. Wathes, J. A. Clark, and A. J. McArthur. 2000. Thermal balance of livestock:
2. Applications of a parsimonious model. Agricultural and Forest Meteorology 101: 29-52.
van 't Klooster, C. E., J. Bontsema, and L. Salomons. 1995. Dynamic model to tune a climate control
algorithm in pig houses with natural ventilation. Transactions of the ASAE 38: 911-918.
Van Wagenberg, A. V. et al. 2005. Climate control based on temperature measurement in the animal-
occupied zone of a pig room with ground channel ventilation. Transactions of the ASAE 48:
355-365.
van't Klooster, C. E., J. Bontsema, and L. Salomons. 1995. Dynamic model to tune a climate control
algorithm in pig houses with natural ventilation. Transactions of the ASAE 38: 911-918.
Van't Klooster, C. E., and B. P. Heitlager. 1994. Determination of minimum ventilation rate in pig
houses with natural ventilation based on carbon dioxide balance. Journal of Agricultural
Engineering Research 57: 279-287.
Wachenfelt, E. V., S. Pedersen, and G. Gustafsson. 2001. Release of heat, moisture and carbon dioxide
in an aviary system for laying hens. British Poultry Science 42: 171-179.
Wan, S. S., D. P. Hennessy, and P. D. Cranwell. 1994. Seasonal infertility, stress and adrenocortical
responsiveness in pigs. Animal Reproduction Science 34: 265-279.
23
Wathes, C. M. et al. 2004. Production responses of weaned pigs after chronic exposure to airborne dust
and ammonia. Animal Science 78: 87-97.
West, J. W. 2003. Effects of heat-stress on production in dairy cattle. Journal of Dairy Science 86:
2131-2144.
Wetterman, R. P., and F. W. Bazer. 1985. Influence of environmental temperature on prolificacy of
pigs. Journal of Reproductive Fertility Supplement 33: 199-208.
Wolfenson, D., Z. Roth, and R. Meidan. 2000. Impaired reproduction in heat-stressed cattle: Basic and
applied aspects. Animal Reproduction Science 60-61: 535-547.
Xin, H., I. L. Berry, G. T. Tabler, and T. A. Costello. 2001a. Heat and moisture production of broiler
chickens in commercial housing. In: Livestock Environment VI, Louisville, Kentucky. p 309-
318.
Xin, H., I. L. Berry, G. T. Tabler, and T. A. Costello. 2001b. Heat and moisture production of poultry
and their housing systems: Broilers. Transactions of the ASAE 44: 1851-1857.
Xin, H., and M. C. Puma. 2001. Cooling caged laying hens in high-rise house by fogging inlet air. In:
Livestock Environment VI. Proceedings of the Sixth International Symposium, Louisville,
Kentucky. p 244-249.
Zhang, G., S. Morsing, J. S. Strom, and P. Ravn. 2001. Air motion and temperature distribution within
covered pig creep zones in rooms. In: Livestock Environment VI, Louisville, Kentucky. p 262-
269.
Zhang, Y., and E. M. Barber. 1995. An evaluation of heating and ventilation control strategies for
livestock buildings. Journal of Agricultural Engineering Research 60: 217-225.
Zhang, Y., E. M. Barber, and S. Sokhansanj. 1992. A model of the dynamic thermal environment in
livestock buildings. Journal of Agricultural Engineering Research 53: 103-122.
24
Behavioural changes related to air temperature in sows kept in different

housing conditions
M. Barbari and M. Bianchi

University of Firenze, Department of Agricultural and Forest Engineering, Via San
Bonaventura 13, 50145 Firenze, Italy
Keywords: cooling systems, pregnant sows, heat stress, behaviour.
Introduction
The behaviour of pigs greatly changes in hot conditions. The pig takes the position of lateral
decubitus, giving the maximum body surface to the air and to the floor contact. It tries to find
isolated positions, far from other animals in the pen, and possibly in areas with air streams.
Furthermore, the pig reduces the feed ingestion but drinks high amounts of water. When
possible it uses the drinker as a “shower” and lies on the floor wetted with water or dung.
Particularly the latter behaviours are of great importance in reducing the heat stress in adult
pigs, if housed in pens with solid floors. On the other hand they involve a considerable
worsening of hygienic conditions of the pen, an increase of ammonia emissions and a rise of
air humidity.
The study aims to contribute to definition of temperature values, which can cause changes in
the behavioural patterns of the sows. The individuation of thermal levels connected to
behavioural changes can be useful to set the working parameters of cooling systems in pig
houses.
Materials and Methods

Three different experimental trials were carried out in a pig farm located in Po Valley (Italy),
during summers 2004 and 2005.
A) In a first survey a static group of 4 pregnant sows was arranged; 6 cycles were repeated
during summer. In the pen 4 areas were realized, treated with different cooling systems (A: not
cooled; B: stream of air; C: stream of air and water on the floor; D: water on the floor). The use
of the different areas by the sows was checked in relation to air temperature and THI. This first
survey is topic of a specific paper of the Workshop, titled “Use of different cooling systems by
pregnant sows in experimental pen”. A description of the experimental trials can be found in
that paper.
The goal of the present work is to try to define reference temperatures, useful to set cooling
systems inside the buildings.
B) The second experimental trial was carried out in a dynamic group of sows (average number
192 pregnant sows, with a minimum of 169 and a maximum of 211), where two automatic
showering cages were installed (figure 1). In both cooling stations, placed in the external
feeding-dunging area, the water distribution time was regulated by an electronic controller: the
shower time was fixed in 6 s, while the water consumption was on average 3.6 l/shower.
The sows clearly appreciated the cooling system with shower, based on the free access to the
cage and on the possibility of taking showers all day. The aim of the study was to relate the use
of the showering cages to the thermal conditions.
25
Solution “a” Solution “b”

• shower starting with pressure on the plate • shower starting with photocell
• system appreciated by the sows, particularly • system appreciated by the sows, particularly
pluriparous young ones (after first farrowing)
• suitable system for dinamic groups of sows • suitable system for dynamic groups of sows
Figure 1.Ttwo different individual stations for showering of sows, examined in the trials.
C) A third experimental trial was executed in a farrowing room, with two different cooling
systems:
- drip cooling system (drip nozzles, able to supply 2 l/h per sow); it started four times a day and
worked for 30 min.
- Snout cooling system, coupled with drip cooling system. The air was directed close to the
head of the sow, under the trough, with a flow of 88 m3/h and a constant velocity of 7.2 m/s.
The air could flow over the lying sows, towards either the snout or the neck.
The farrowing crate used in the trials is long enough to allow the sows to lie in advanced or
rear position. In such a way the animals freely choose to profit from cooling systems provided
or not. THI (Ingram, 1965) was used to evaluate the behaviour of the sows during the trials.
Figure 2. Different cooling systems. From left: pipe of the drip cooling system; full steel sheet
under the head of the sow; pipes of snout system; air coming out from the hole.
Results
A) The graph of figure 3 clearly shows the changes in the behaviour of the four sows kept in
the static pen, when the temperature drops.
Over 20°C, when the temperature rises, the use of cooled area with air stream (zone B)
constantly increases. At 26-27°C the frequency rate of this area reaches the maximum value,
with the 50% of preference. Then the rate starts to decrease. At the temperature of 30°C, the
use of zone B drops below the value of zone C.
At the temperature of 26-27°C, the presence of the animals in untreated areas (A and F)
declines under the 20%. Over the value of 30°C the use of these areas drops under the value of
10% of presence.
A further consideration is needed for zone C: how the graph clearly remarks, the use of cooled
area with the coupled system (air and water) is practically nought and it progressively increases
in an almost linear way up to the temperature of 26-27°C. Above such temperature the
frequency rate goes on increasing and soars above 30°C.
26
The graph clearly shows that the use of zone C has an opposite trend in comparison with zone
A. The lines of the two areas crosses at the temperature of 26-27°C, that is an important
reference temperature of the behavioural changes of the sows. Another value to take into
consideration is 30°C: above this temperature the use of the cooled area with the coupled
system becomes predominant. At 31°C the sows remain for the 92.6% of the time in the three
cooled areas.
0,7
0,6
0,5
A
Frequency rate
0,4 B
C
0,3 D
F
0,2
0,1
0
20 21 22 23 24 25 26 27 28 29 30 31
Temperature °C
Figure 3. Use of different areas in relation to air temperature.
B) The results of the use of the two automatic showering cages are very positive. Figure 4
shows the results of the percentage of use of solutions “a” and “b” by the dynamic group of
sows during 26 examined days. During the trial the average temperature was 23.6°C, with daily
maximum values of 29.1°C. The maximum temperature in the whole period was not very high
(33.9°C), in relation to the temperatures of previous years.
The percentage use of the automatic showering cages, that is the number of sows taking at least
a shower during the day on the total number of sows in the group, was 50.4% on average (±
15.89). On the hottest day (i.e. 23rd July) the number of sows taking showers considerably
increased up to 76.9%. However on the coldest day (i.e. 12th July) the percentage of sows
taking showers decreased to 23.7%. Therefore it was possible to show the positive correlation
(r = 0.86) between daily maximum temperature and the percentage of animals making use of
the cooling stations.
Further considerations concerning the employment of the stations are the following: the total
number of showers in the hottest day was 788; in six consecutive hot days the percentage of
sows using the automatic cages reached 86.3%. A single shower had an average length of 59 s,
but this value fluctuated in relation to outside temperature, arriving at a daily value of 80 s on
average.
In the warmest day of the period the distribution of the total number of showers, when 140
sows (76.9% of sows) went under the shower at least once, was: 22 sows took just one shower,
61 took from 2 to 5, and 57 took 6 or over. With lower daily maximum temperatures the
number of sows which went to the automatic showering station at least 6 times was strongly
reduced: on the 11th of July only 2 animals (on a total of 62) took more than 6 showers.
27
Figure 4. Use of the automatic showering cages (black unbroken line) during the 26 days.
The data on the use of the showers from the sows were processed also to verify the changes in
relation to the outside temperature. For this kind of processing 12 hours of the day were
considered, chosen in the daylight period, including the coldest hours (5-7 a.m.) and the hottest
ones (1-8 p.m.). The night time period was excluded from the processing, due to the different
behavioural patterns of the sows: during the night the use of the individual showering cages is
reduced, owing to the sleeping time, and then not strictly related to thermal conditions.
The analysis of the period of 12 daily hours has shown a high positive correlation between air
temperature and the use of the showers from the sows (r = 0.731). The graph of figure 5 clearly
shows how the use increases with temperature. The rise is particularly important above the
temperature of 26°C. At this temperature the use of the showers increases very quickly, in a
linear way.
100%
90%
80%
70%
Use of showers %
60%
50%
40%
30%
20%
10%
0%
14 16 18 20 22 24 26 28 30 32 34 36
Temperature °C
Figure 5. Use of the automatic showering cages in relation to temperatures.
C) The system based on the drip cooling alone does not provide satisfactory results. The sows
seem to appreciate the air flow provided by snout cooling ducts in the front area of the
farrowing crate. The sows remained for long periods with the snout or the neck near the air
outlet hole. With long enough farrowing crates (2.50 m) the sows could freely move back,
lying in such a position as to optimize the effects of the cooling systems.
The graphs of figure 6 clearly shows the changes in the behaviour of the sow during days with
different thermal conditions. The same sow during a warm day (graph on the top) remains for
long times lying ahead. It is clear how during the central hours of the day (mean value from 5
to 7 p.m. 30.16 °C), THI moves from comfort to alert and dangerous zones and the sow
28
chooses to lie ahead: in this situation the fresh air coming out from the pipes is a valid help
against the heat stress of the sows in farrowing room.
In the graph on the bottom temperatures were not very high during the day for a sudden storm
(max 24.07°C; min 21.78°C). Anyway the sow chooses to lie ahead in the cage practically for
all the time, except the central hours of the day. From 3 to 7 p.m. when the temperatures fall
down, the sow changes position inside the crate, avoiding the air on the snout.
In conclusion it is possible to remark the considerable use of the system by the animals which
can choose to profit or not from the air flow, according to the microclimatic conditions inside
the farrowing room.
Alert Dangerous Alert
100 33
Warm 90 31
80
day 29
Ratio of activity %
70
Temperature °C
27
60
25
50
23
40
21
30
20 19
10 17
0 15
8.00
9.00
10.00
11.00
12.00
13.00
14.00
15.00
16.00
17.00
18.00
19.00
20.00
21.00
22.00
23.00
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
Time
Lying Ahead Lying Back Other Activities Temperature
100 25
Cold day
80 23
Ratio of activity %
Temperature °C
60 21
40 19
20 17
0 15
8.00
9.00
10.00
11.00
12.00
13.00
14.00
15.00
16.00
17.00
18.00
19.00
20.00
21.00
22.00
23.00
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
Time
Figure 6. Behaviour of a sow related to temperature and THI: “hot day”(above); “cold day”
(bottom).
Conclusions
The experimental trials shown in the present work aim to provide some suggestions useful in
the planning of cooling systems. The sows always need systems of heat protection, both in
pregnancy and in farrowing phase.
The temperature of 27°C seems to be a first critical threshold to take into consideration, as the
behavioural changes of the sows can prove. When the temperature rises above 30°C the sows
strongly uses the cooling systems available in the pen. In addition to the temperature value,
relative humidity is another critic parameter. So further studies can be useful to define in a
better way the relations between the behaviour of the sows and comfort indexes, such as THI.
References
Barbari M., 2006. Evaluation of Individual Systems for Cooling Pregnant Sows in Collective
Pens. Ageng2006, Bonn, 3-7 September.
29
Use of different cooling systems by pregnant sows in experimental pen
M. Barbari, L. Conti and S. Simonini

University of Firenze, Department of Agricultural and Forest Engineering, Via San
Bonaventura 13, 50145 Firenze, Italy
Keywords: cooling systems, pregnant sows, heat stress, behaviour.
Introduction
Sows are able to decide on the use of cooling systems adopted in the pen. The choice is
obviously affected by thermal conditions.
This study aims to confirm this hypothesis and to identify the temperatures that drive the sows
to find comfort areas.

The experimental campaign was carried out in an intensive pig farm located in the Po Valley.
In a pen for pregnant sows (5.42 m x 4.10 m) three different areas (B, C, D) were arranged
with cooling systems; a fourth area, used as a feeding area with no cooling system, was placed
next to them (Zone A). The four areas were physically divided by metallic partitions. There
was also a middle area (Zone F), that led to an outside dunging alley (Zone E).
Briefly, the following cooling systems were present in the three studied areas:
B. stream of air at high velocity, made of two pipes, whose exit diameter was 0.08 m, which
blew air at high velocity towards the floor (12.5 m/sec), for a total air flow rate of 1810
m³/h.
C. Stream of air at high velocity (as zone B) and supply of water on the floor by a drip cooling
system made using 25 drip nozzles, able to supply 3.5 l/h each, working 4 times a day for
3 minutes. The total amount of water was 36 l/day.
D. Supply of water on the floor, made of 25 drip nozzles, with the same working way
described for the zone C.
As we can see from the low water consumption, the purpose of the system with drip nozzles
was to keep the floor wet and cool instead of cooling directly the bodies of the sows.
The tests in the pen were carried out from the 27th of June to the 24th of August 2005. Six
cycles of observations were conducted, lasting from 7 to 15 days each. Four sows were
observed in each cycle. Thermal-hygrometer data were collected from a data logger, equipped
with special probes. The behaviour of the sows was continuously monitored by means of a
close-circuit television system with infrared cameras.
For each monitored day, the data related to the coldest and the hottest time of the day were
analyzed: respectively, the period between 2 and 7 a.m. and the period between 2 and 7 p.m.
The microclimatic data were acquired every 15 minutes, whereas the behavioural data of the
sows in each separated area were collected every 5 minutes. In order to make those data
sources homogeneous and comparable it was necessary to carry out four repetitions of 15
minutes for each hour of observation, and to calculate the average presence of the animals in
each area at the specific time of the repetition. The next step was to identify 4 classes of
temperature ( T<22°C, 22≤ T <26°C, 26≤ T<30°C e T≥30°C). In this way it was possible to
organize the behavioural data according to the aforementioned classes, analysing the
repetitions corresponding to the class examined.
The methodology followed to arrange the monitored data, both thermal and behavioural was
necessary in order to conduct a statistical analysis (ANOVA and test of Bonferroni).
Furthermore, using χ2-test it was possible to compare the distribution of the observed presences
with expected ones in casual distribution. We supposed as H0 that animals had no preferences
30
choosing the area and as H1 that the choice of animals was affected by temperature. We fixed
high significance P-value = 0.01.
Figure 1. Layout of the pen with the different areas (A: feeding area; B: air-cooled area; C:
air-water cooled area; D: water-cooled area; E: dunging alley; F: passageway).
Results
The first results of the study concern the different percentages of attendance of the sows in the
different areas, for the 6 cycles of observations, in relation to temperatures. Zone B, where the
animals were able to benefit from the air cooling system, was the most frequented area: on
average about 41.3% of the total time, during the 1st, 2nd, 3rd and 4th cycle.
During the 5th and 6th cycle the aforementioned behaviour was reversed, since zone A was the
most frequented area (area without any cooling system, occupied 46% of the time). Such
behaviour could be caused by the temperature decrease, few degrees lower in the last two
cycles: the average temperature was about 20°C, whereas it was about 24°C in the first 4
cycles.
From this analysis it also emerged that the sows preferred, as second choice, the combined air-
water system when temperatures reached or passed 24°C (1st, 3rd and 4th cycle).
The results of the statistical analysis confirmed what was supposed during the elaborations.
The variance analysis allowed to identify significant differences about the average number of
sows in relation to the source of variation identified as the “cooling area” and the one due to
the interaction between the “cooling area” and the “classes of temperature” (Figure 2).
31
T°C<22 22<T°C<26 26<T°C<30 T°C>30

3
a
Average number sows/Time repetition 2,5
2
b
b
1,5 c
c
cd
d
1
e
e
0,5
f
fg fg fg fg fg
fg
g fg
g g g g gg
0
A B C D E F
Zone
Figure 2. Distribution of sows in relation to the zones and classes of temperature (A: feeding
area; B: air-cooled area; C: air-water cooled area; D: water-cooled area; E: dunging alley;
F: passageway).
The figure 2 shows that the presence of the animals in the feeding area, in the air-cooled area
and in the combined air-water cooled area did not occur by chance, but it was influenced by the
factors “cooled area” and “temperature class”, whose highly significant interaction produces a
substantial change in their behaviour. One particular point stands out, that is the different
attendance of the sows in the feeding area, which was quite substantial when temperatures
were under 22°C, but was rather insignificant when temperatures rose up to 30°C and more.
During the highest temperature intervals the animals definitely moved towards the air-cooled
area (with 26≤ T<30°C), and towards the combined one (with T≥30°C). On the other hand we
did not find any statistically significant differences as far as the other 3 areas of the pen are
concerned: this may suggest that the sow behaviour did not depend on the area itself.
The graph of figure 3 is obtained with a different way of analysis (χ2-test). In any case it clearly
shows that the use of cooled areas becomes more and more important with the increase of
temperature. The following considerations can be done:
- below a temperature of 22°C A and B are the favourite areas.
- The number of sows present in zone A regularly decreases with the increase of temperature.
- The presence of the sows in zone B tends to increase up to a temperature of 30°C, when the
more frequented area becomes the C one.
- The use of zone D lightly increases with temperatures, especially with high values. However
the temperatures during the examined period hardly ever reached high values (more than
33°C).
- In zone F the changes are not influenced from the temperatures.
32
100%
90%
80%
70%
D
60%
C
50%
B
40%
A
30%
F
20%
10%
0%
<22 22-26 26-30 >30
Figure 3. Use of different areas from the sows
Conclusions
According to the results obtained, we were able to identify the areas in the pen where the sows
preferred to stay. Besides, through the analysis of the behavioural differences shown by the
animals, combined with the thermal variations, we were able to prove that such behaviour was
not a merely instinctive one, but was actually related to the environmental changes occurring in
the pen.
The system based on the realization of a stream of air at high velocity coupled with a system to
wet the floor was particularly appreciated from the sows during the hottest periods of
experimental trials.
References
Barbari M., 1998. Water sprinkling systems for cooling of sows. Proc. XIII CIGR Word
Congress, Rabat, 2-6 February, 245-252.
Barbari M., 2006. Evaluation of Individual Systems for Cooling Pregnant Sows in Collective
Pens. Ageng2006, Bonn, 3-7 September.
Bull, R.P., P.C. Harrison, G.L. Riskowski and H.W. Gonyou, 1997. Preference among cooling
systems by gilts under heat stress. Journal of Animal Science 75, 2078-2083.
33
Thermoregulatory responses to high ambient temperature in growing pigs:

effects of temperature level and breed.
D. Renaudeau, J.L., Gourdine, C. Anaïs.
INRA Station de Recherches Zootechniques, Domaine de Duclos, 97170 Petit Bourg.

Contact : David.Renaudeau@antilles.inra.fr
Introduction
The thermal environment is the major factor that can negatively affect pigs’ performance in
tropical area especially in pigs of high genetic merit. Within the thermoneutral zone (TNZ),
pigs are able to maintain their body temperature requiring minimal effort. Under this TNZ, the
maximum productivity is normally achieved. When temperature increases above the upper
limit of the thermoneutral zone, heat losses are insufficient to prevent an increase of body
temperature. According to Quiniou et al. (2000), when ambient temperature is above 25°C, the
average daily feed intake (ADFI) is reduced with negative subsequent consequences on growth
performance. The decrease of ADFI is considered as an adaptation process to reduce metabolic
heat production. The effect of temperature is well documented in the literature (Le Dividich et
al., 1998). Most of the studies dealing with the effect of heat stress on growth performance
were performed in heat acclimated animals. Previous studies reported a significant reduction of
ADFI with the first 24 h of exposure to high ambient temperature and thereafter remained
constant or slightly increase with the period of acclimation. Over the same period, the rectal
temperature (RT) shows a rapid increase within the first 24h of exposure and decline thereafter
over the successive days of exposure. Variations in the short-term (a few hours) response to
heat stress among breeds are described in halothane-positive and halothane-negative boars
(Tauson et al., 1998) or in high- and low-producing genotypes (Nienaber et al., 1997), but little
is known about the effect of breed on acclimation to elevated ambient temperature. In the
French West Indies, a local Caribbean breed (Creole pig; CR) is known for its good adaptation
to the harsh tropical environment (Renaudeau, 2005; Renaudeau et al., 2007). For this purpose
it was introduced in our experimental facilities to study the genetic variability of heat tolerance
in pigs. Moreover, as far as we know, the effect of temperature level on heat acclimation is
poorly described in the literature. For these reasons two studies were designed to quantify the
effect of breed and temperature level on medium term responses of growing pigs to high
temperature.
Materials and methods
A total of 120 barrows were used in 2 experiments conducted on replicates of 12 animals at the
experimental facilities of INRA in Guadeloupe (F.W.I., 16° Latitude N., 61° Longitude W.).
Within a replicate, pigs were housed in a climatic controlled room for 40 d including 10 d for
adaptation and 30 d for the experiment. The experimental room contained 12 individual metal-
slatted pens (0.85 × 1.50 m). Each pen was equipped with a feed dispenser and a nipple drinker
designed to avoid water spillage. During the first experiment, the effect of breed was studied in
two replicates of 6 Large White (LW) and 6 Creole (CR) pigs. According to the large
difference in average daily gain (ADG) between CR and LW pigs, this first experiment was
designed to compare both breeds at a same BW range (i.e., 52.0 kg BW at d0). Pigs were kept
at 24°C for 10 d (d-10 to d-1) and thereafter at a constant temperature of 31°C for 20 d (d 1 to
d 20) (Figure 1). In the second experiment eight replicates of 12 LW pigs were used. Pigs were
kept at 24°C for 10 d and thereafter at 24, 28, 32 or 36°C for 20 d (Figure 1). On d 0 the
temperature gradually changed from 24 to 31°Cand from 24 to 28, 32, or 36°C in exp. 1 and 2,
respectively within 4 h starting at 0800. The relative humidity (RH) was kept at 80% over the
total duration of experiment. The pigs were offered ad libitum a diet formulated with maize,
34
wheat middlings, and soybean meal. All pigs were weighed before and after 24-h fasting period
at d-10, d0, d10 and d20. Every morning, feed refusals were manually collected between 0700
and 0800, weighed and sampled for DM determination. Rectal (RT) and cutaneous (CT) body
temperatures and respiratory rate (RR) were measured 3 times daily (i.e., at 0700, 1200, and
1800) every 2 to 3 days (d-10, -7, -5, -3, -1, 0, 1, 2, 4, 7, 9, 11, 11, 14, 16, 18 and 20 of
experiment).
40 100 40 100
90 90
36 36
Ambient temperature, °C
Ambient temperature, °C
80 80
Relative humidity, %
Relative humidity, %
32 70 32 70
28 60 28 60
50 50
24 40 24 40
20 30 20 30
20 20
16 16
10 10
12 0 12 0
-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Day of experiment (d1=first day at 31°C) Day of experiment (d1=first day at the experimental temperature)
Figure 1: Variations in ambient temperature and relative humidity in exp. 1 and exp. 2
For each pig, the daily feed intake (ADFI), ADG, respiratory rate (RR), rectal temperature
(RT) and cutaneous temperature (CT) were averaged over three sub-periods of 10 d each
corresponding to d-10 to d0, d1 to d10 and d11 to d20, respectively. These data were analysed
with breed, replicate, and period or temperature level, replicate and period as mains effect in
exp. 1 and 2, respectively. The repeated measurement option of MIXED procedure of SAS
(2000) was used with an unstructured covariance structure.
Results and discussion
2500 A B B A C D 2500 A B B A B B A B C A B B
2000 2000
ADFI, g/d
d-10-d0
ADFI, g/d
1500 1500 d-10-d0

d1-d10 d1-d10
1000 d11-d20 1000 d11-d20
500 500
0 0
24°C 24°C 24°C 28°C 24°C 32°C 24°C36°C
Creole Large White 24 28 32 36
1200 A B C C D E 1200 A B B A B C A B C A B C
1000 1000
800 800
ADG, g/d
ADG, g/d
d-10-d0 d-10-d0
600 d1-d10 600 d1-d10
d11-d20 d11-d20
400 400
200 200
0 0
24°C 24°C 24°C 28°C 24°C 32°C 36°C
24°C
Figure 2: Effect of breed (Creole vs. Large White; exp. 1) and temperature level (24, 28, 32,
36°C; exp. 2) on the average daily feed intake (ADFI, g/d) and average daily gain (ADG, g/d)
growing pigs over the acclimation period. In exp.1, least square means with a same letter are
not significantly different (P < 0.05). In exp. 2, least square means with a different letter are
affected (P < 0.05) by the duration of exposure.
35
In Exp. 1., according to the breed difference in growth potential, the ADG at 24°C (i.e., from
d-10 to d0) was lower in CR than in LW pigs (675 vs. 1007 g/d, P < 0.001) whereas ADFI was
not influenced by breed (2194 g/d on average) (Figure 2) . Irrespective to the breed, the ADFI
was reduced during the first 10-d exposure to 31°C in order to reduce metabolic heat
production due to metabolic processes of nutrient utilization; this decline was more accentuated
in CR than in LW pigs (- 604 vs. -373 g/d; P < 0.05). The ADFI numerically increased between
d11-d20 and d1 to d10 for both breeds but this increase is significant only for LW pigs (+ 108
g/d; P = 0.021). As shown for ADFI, ADG decreased between d10-d1 and d-10-d0 (-350 g/d,
on average) and slightly increased (+133 g/d on average) during the last 10-d of exposure to
31°C. These variations were not affected by breed. The effect of exposure to 31°C on RT, CT
and RR were presented in Figure 3.
41.0 41.0
A B C D E C B B C A B C A B C A B C
40.5 40.5
Rectal temperature, °C
Rectal temperature,°C
40.0 d-10-d0 40.0 d-10-d0
d1-d10 d1-d10
39.5 d11-d20 39.5 d11-d20
39.0 39.0
38.5 38.5
24°C 24°C 24°C 28°C 24°C 32°C 24°C 36°C
B B B A B C A B C A B C
39.0 A B C D B C 39.0
Cutaneous temperature, °C
Cutaneous tempertaure,°C
38.5 38.5
38.0 38.0
d-10-d0 d-10-d0
37.5 d1-d10 37.5 d1-d10
d11-d20 d11-d20
37.0 37.0
36.5 36.5
36.0 36.0
24°C 24°C 24°C 28°C 24°C 32°C 24°C 36°C
B B B A B B A B B A B C
120 A B C A B C 120
100 100
Respiratory rate, bpm
Respiratory rate, bpm
80 80
d-10-d0 d-10-d0
60 d1-d10 60 d1-d10
d11-d20 d11-d20
40 40
20 20
0 0
24°C 24°C 24°C 28°C 24°C 32°C 24°C 36°C
Figure 3: Effect of breed (Creole vs. Large White; exp. 1) and temperature level (24, 28, 32,
36°C; exp. 2) on the average respiratory rate (breaths per min, bpm), rectal and cutaneous
temperature (°C) in growing pigs over the acclimation period. In exp.1, least square means
with a same letter are not significantly different (P < 0.05). In exp. 2, least square means with
a different letter are affected (P < 0.05) by the duration of exposure
At 24°C, CT and RT were lower in CR than in LW pigs (39.2 vs. 39.4°C and 36.5 vs. 37.1°C;
P < 0.001) whereas RR was not influenced by breed (36 bpm on average). Between d1 to d10
and d-10 to d0, CT increase was higher in CR than in LW pigs (+1.34 vs. 1.15°C; P =0.027).
However, the rises in RT and RR was not influenced by breed (+0.50°C and + 48 bpm, on
average). The elevation of CT is explained by an increase of blood volume in skin vessel to
promote non evaporative heat loss. As the gradient between CT an ambient temperature is
36
reduced at 31°C, water evaporation from the lungs is the more efficient pathway to increase
heat loss. However, the rise in RT suggests that mechanisms implicated in body temperature
regulation (decrease of heat production and/or increase of heat losses) are saturated or not
enough to prevent hyperthermia.
The increase of growth performance between d11 to d20 and d1 to d10 supports the finding of
Morrison and Mount (1971) in 60-kg growing pigs kept at 33°C for 28 d. This result could be
related to a long term acclimation to heat stress. As RT indicates the animal’s efficiency in
maintaining homeothermy, its decline between d11 to d20 and d1 to d10 supports the
hypothesis of an acclimation to elevated temperature. During the same period, RR and CT were
also reduced which suggests that the decrease of heat loss pathways would be considered as a
consequence rather than a cause of the acclimation to heat stress. Finally, it can be
hypothesized that this acclimation is related to a decrease of heat production which is
consistent with numerous reports demonstrating a reduction of metabolic heat production
during acclimation in man (Bianca, 1959). Such an explanation would be consistent with the
decrease of O2 consumption reported by Giles and Black (1991) in pigs kept at 31°C for 11
days. According to our results the long term acclimation response was not affected by breed. In
fact the higher RT in LW pigs at 31°C was related to a delayed threshold time for the onset of
acclimation mechanisms (Renaudeau et al., 2007).
In exp 2., the temperature level significantly affected the ADFI (Figure 2). However, this
variation was not linear. Between d20 to d1 and d-10 to d0, ADFI decrease was more
important between 24 and 36°C than between 24 and 28°C (78 vs. 38 g/d/°C). In other words,
the extent to which ambient temperature affect ADFI depends on the level of temperature.
Comparing the last and first 10-d of exposure to the experimental temperature (24, 28, 32, and
36°C), the ADFI increase was significant only at 32°C (+ 162 g/d, P < 0.001). Similarly, the
ADG decrease was linearly affected by temperature level (-50 g/j/°C on average). From 28°C,
the ADG increase during the duration of exposure to the experimental temperature was
significant (P < 0.05) and it depended on the temperature level (+69, 172, 445 g/j at 28, 32 and
36°C respectively). As shown in exp1., RT, RR and CT increased with the elevation of ambient
temperature from 24 to 28, 32 or 36°C (Figure 3). Moreover a significant effect of duration of
exposure (i.e., d11 to d20 vs. d1 to d10) was reported for RT and CT from 28°C and for RR
only at 36°C.
In conclusion these experiments demonstrate the faculty of pigs to undergo some physiological
changes that make continued exposure to heat stress more endurable. Even if a decrease in heat
production might play a part in this observed acclimation, further studies are required to
understand the mechanism underlying the implicated physiological responses. According to
our works, the acclimation to heat in pigs is affected by breed and the temperature level.
References
Bianca, W. 1959. Acclimatization of calves to hot dry environment. Journal of Agricultural
Science 52:296-304.
Giles, L. R. and J. L. Black. 1991. Voluntary food intake in growing pigs at ambient
temperatures above the zone of thermal comfort. In: E. S. Batterham (Ed.) Manipulating Pig
Production III. pp. 162-166.
Le Dividich, J., J. Noblet, P. Herpin, J. van Milgen, and N. Quiniou. 1998. Thermoregulation.
In: J. Wiseman, M. A. Varley, and J. P. Chadwick (Eds.) Progress in Pig Science. pp. 229-263.
Nottingham University Press, Nottingham.
37
Morrison, S. R. and L. E. Mount. 1971. Adaptation of growing pigs to changes in

environmental temperature. Anim. Prod. 13:51-57.
Nienaber, J. A., G. L. Hahn, R. A. Eigenberg, R. L. Korthals, J. T. Yen, and D. L. Harris.
Genetic and heat stress interaction effects on finishing swine. Bottcher, R. W. and Hoff, S. J.
2[Proccedings of the 13th International Livestock Environment Symposium], 1017-1023.
1997. Bloomington, Minnesota, American Society of Agricultural Engineers. Livestock
Environment.
Quiniou, N., J. Noblet, J. van Milgen, and S. Dubois. 2000. Modelling heat production and
energy balance in group-housed growing pigs exposed to low or high ambient temperatures.
Br. J. Nutr. 84:97-106.
Renaudeau, D. 2005. Effects of short-term exposure to high ambient temperature and relative
humidity on thermoregulatory responses of European (Large White) and Carribbean (Creole)
restrictively fed growing pigs. Animal Research 54:81-93.
Renaudeau, D., E. Huc, and J. Noblet. 2007. Acclimation to high ambient temperature in Large
White and Caribbean Creole growing pigs. J. Anim. Sci. 85:779-790.
SAS. 2000. SAS/STAT User's Guide (version 8.1.). SAS Inst. Inc. cary, NC.
Tauson, A. H., A. Chwalibog, J. Ludviqsen, K. Jakobsen, and G. Thorbek. 1998. Effect of
Short-term exposure to high ambient temperatures on gas exchange and heat production in
boars of different breeds. Anim. Sci. 66:431-440.
38
HEAT STRESS IN DAIRY COWS: A REVIEW OF THE HEAT LOAD

EFFECTS ON THE ANIMAL RESPONSE AND RELIEF STRATEGIES
P. Zappavigna1, E. Maltz2, S. D’Archivio1
1 Dipartimento DIPROVAL Università di Bologna, via Fratelli Rosselli, 107, 42100 Reggio
Emilia, Italy
2 Department of Growing Production and Environmental Engineering Institute of Agricultural
Engineering Agricultural Research Organization - The Volcani Center, P.O. Box 6, Bet Degan,
50250 Israel
Key words: heat stress, dairy cows, response, relief strategies
Introduction
Dairy cows as more productive as more are sensible to the heat stress. The following graph
(fig.1), given by Kadzere et al. (2002) represents the increase in milk and heat production of a
cow in the last 50 years. The heat production required to increase the milk production has been
continuously increasing and this makes the cows more sensible to the stressful conditions.
The first part of this work aims to investigate the relationships of the physiological and
productive response of dairy cows to the microclimatic parameters in a confined environment,
using the experimental data available in the bibliography. The second part will point out the the
80
milk energy
70
heat production
60
50
MJ/day
40
30
20
10
0
1930 1940 1950 1960 1970 1980 1990 2000
year
Fig. 1: Energy in milk and heat production increase for dairy cows in the
United States from 1940 to 1995 (Kadzere et al., 2002).
effect of animal related physiological and physical variables, such as behaviour, body weight
and insulation, and the way they influence the capacity and pattern to resist heat stress. Finally
practical aspects of housing, cooling systems and management to provide the best condition for
the animal to exercise optimally its thermoregulatory capacity and new approaches regarding
the evaluation of facilities and management efficiency in reaching this goal are described.
Fig. 2: Effect of the environmental

temperature on milk production of
Holstein and Jersey cows, with air
39
Effects of the climatic parameters on the productive response

Air temperature
The fundamental work of Yeck and Stewart (1959) which is on the basis of the ASAE
5
0
-5
-10
production variation %
-15
-20
-25
-30 Yeck-Stewart (laboratory)
Berry (laboratory)
-35
Baeta (laboratory)
-40
McDowell (farm)
-45
Igono (farm)
-50
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
temperature °C
Fig. 3: Effect of the environmental temperature on milk production according to

various authors.
day
0 1 2 3 4
-30
-28 Spiers T=29 °C
-26
-24 Broucek T=34 °C
-22 Baeta T=34 °C
% noitai ra v noit cudo rp
-20
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
Fig. 4: Effect of the exposure to high temperatures on milk production according

to some authors.
standards appears valid until now in conformity with more recent studies. The graph of fig. 2
points out that the decrease in milk production, for a Holstein cow, begins at about 21 °C, and
becomes significant from 24 °C. If we compare this result to the results obtained in later
investigations (Berry et al., 1964; Baeta et al.,1987; McDowell et al. taken from Reinemann et
al., 1992; Igono et al.,1992) we can see (fig. 3) that the agreement appears very good. This
despite the adaptations required to compare all the data (i.e. we assumed the Igono’s data as
related to the minimum daily temperature, instead of the average, because a drop in the
nocturnal values can neutralize the effect of high diurnal peaks; furthermore the
40
5
18
0 20
-5 22
-10
24
-15
26
-20
28
-25
30
-30
-35
32
-40
34
-45
40 45 50 55 60 65 70 75 80 85 90
relative humidity %
Fig. 5: Production variation (%) in relationship with relative humidity and air
temperature, at an air speed of 0.5 m/s (Baeta et al., 1987).
McDowell data are probably related to the highest daily temperature, but this was not specified
in his paper). For a proper comparison consider also that the first two studies were carried on in
climatic chambers, whilst the last two have been carried out in real barns.
Yeck e Stewart Baeta et al.

0
-3
-5
-10 -7
-15
-20
-25
-25
-30
-31
-35
40 45 50 55 60 65 70 75 80 85 90
relative humidity %
Fig. 6: Production decrease increasing the relative humidity from 44% to 90%, at an
air temperature T=29.5°C: comparison between the data measured by Yeck
and Stewart and those estimated by the Baeta’s model
41
0
-2
-4
-6
-8
produc tion v ariation %
-10
-12
-14
-16
-18 Chiappini, 1983
-20
-22 Igono et al., 1992
-24 Calamari et al., 1994
-26
-28 Bouraoui et al., 2002
-30
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
THI daily average
Fig. 7: Variation in milk production in relationship with the daily average THI:
comparison of the results obtained by various authors in trials carried out in
farming conditions
To analyse the real influence of the temperature on production we must take in addition into
consideration the duration of the exposure to heat stress. This effect has been investigated in
two ways: at an even temperature (as in most of the laboratory tests) and at a variable
temperature (as in real housing conditions and in few laboratory tests). The first type of
investigations gave the results reported in figure 4, where data were taken from Broucek et al.
(1998), Baeta et al. (1987), Spiers et al. (2004). The Spiers’ worst results are due to the higher
production level of cows (35 kg/d) compared to Brouceck (21) and Baeta (24).
On the other side Yeck e Stewart (1959) found that in a cyclic temperature the effect on
production is due to the average temperature: so a decrease in the lower daily temperature can
help to tolerate higher diurnal peaks. In fact Broucek et al. (1998) simulating a temperature
fluctuation from 23 °C to 34 °C (average 28 °C) found no significant effect on the production.
The fact that lower nocturnal temperatures can help to tolerate diurnal peaks has been
confirmed by investigations made by Igono et al. (1992) and Frazzi et al. (2003). The first one
found that a daily period of 3-6 hours with temperatures below 21 °C can minimize the effect
of high diurnal values; the second found that if the temperature falls below 18°C the production
shows no decrease until diurnal values don’t exceed 33 °C.
Air Humidity
The effect of the air humidity itself has been not much investigated. The most complete work
is that one of Baeta et al. (1987) made by means of laboratory tests. The main results are
synthesized in fig. 5 assuming an air speed of 0.5 m/s.
42
These results are partially confirmed by the previous work of Yeck e Stewart (1959). In fact
the comparison of both the studies, possible for a temperature of 29.5 °C, shows a very good
agreement (fig. 6).
-2
-4
-6
air speed
-8
0.5 1 1.5 2
-10
2.5 3 3.5 4
-12
4.5 5 5.5
-14
18 20 22 24 26 28 30 32 34
temperature °C
Fig. 8. Milk production variation in relationship with air temperature and speed, at
RH =40% (Baeta et al., 1987).
Temperature Humidity Index (THI)

The combined effect of the air temperature and humidity has been investigated by various
authors.
The U.S. National Weather Service (1970) using the formula THI = Td + 0.36Tw + 41.2
proposed a classification for different levels of stress, starting from the value of 74.
Using the formula given by Kelly e Bond (1971), THI = Td-(0.55-0.55RH)( Td-58) ,
Wiersma (1990) assuming 72 as a threshold gave the following classification:
72≤THI≤79 mild stress;
79<THI<90 distress;
90<THI<99 severe stress.
1.4
rectal tem perature inc reas e °C
25 kg/day
1.2
30 kg/day
1 35 kg/day
0.8
0.6
0.4
0.2
0
26 27 28 29 30 31 32 33 34 35 36
air temperature °C
Fig. 9: Increase of the rectal temperature as a function of the

environmental temperature and milk production for forced
ventilated dairy cows (Berman et al. 1985)
43
More authors investigated the correlation between THI index and production decrease (Berry et
al. (1964), Chiappini (1983), Igono et al. (1992), Calamari et al. (1994), Bouraoui et al.
(2002)). If we consider the daily average values we can compare all the different works as
reported in fig. 7. The majority of the authors agrees with a threshold level around 72 – 73,
though Berry and Bouraoui found lower values (68-69). Overall the trend of production
decrease with THI appears very similar for all the sources, quantifiable around a 2% less milk
per THI unit over the threshold.
Also for this parameter it has been considered the influence of the exposure to heat stress. In
this case Bouraoui, Chiappini and West et al. (2003) found a better correlation between the
daily production decrease and the average daily THI of the 2nd –3rd previous day.
Air speed in combination with other parameters

The parameter of air speed has been mainly investigated in combination with other parameters.
One of the most complete works is the already mentioned study of Baeta et al (1987) where the
combined effect of air speed, air temperature and air humidity was analysed in laboratory tests.
The results can be reported in graphs like we did in fig. 8.
Overall the main conclusions are that:
a) The basic level ( full production) corresponds to a temperature of 18 °C, RH of 50% and air
speed of 0.5 m/s;
1.8
natural ventilation Berman
rectal temperature increase °C
1.6
natural ventilation Calamari
1.4
1.2 forced ventilation Berman
1 forced ventilation Calamari
0.8
0.6
0.4
0.2
0
26 27 28 29 30 31 32 33 34 35 36
air temperature °C
Fig. 10: Increase of the rectal temperature with the variation of the environmental
temperature as estimated by the Berman’s equations (for cows producing28
kg/day compared to the data measured by Calamari et al (1994).
b) if RH is less than 40%, the production starts decreasing over 24 °C, but by increasing the air
speed until 5.5 m/s it is possible to maintain the same production level until 34 °C;
c) when RH is over 60 %, the production starts decreasing over 22 °C and the air speed
increase can keep even the production just until 26 °C;
d) when RH is over 90 % the production starts decreasing over 20 °C and it can be kept
constant by an increase of the air speed just until 24 °C.
Calamari et al. (1994) through tests carried out in a house with tied up cows found that with an
increase of the air speed up to 0.5 – 0.8 m/s the decrease of production was reduced (in days
with temperatures varying from 23 to 30 °C) of 3.2% instead of the 10% showed by the control
cows subjected to an air speed of 0.1 m/s.
44
Frazzi et al. (2000) working in a farm with loose housing and temperatures up to 32 °C found
that a forced ventilation with an air speed of about 0.5 m/s reduced the loss of production of
3% with respect to the not ventilated control cows (loss of 12.4% instead of 15 %).
100
natural ventilation Berman
95
respiratoty rythm (breaths/min)
natural ventilation Calamari

90
forced ventilation Berman
85
forced ventilation Calamari
80
75
70
65
60
55
50
45
21 22 23 24 25 26 27 28 29 30 31
air temperature °C
Fig. 11: Respiratory rhythm average values as measured at different air

temperatures on cows exposed to natural and forced ventilation
(Berman, 1985 and Calamari et al., 1994).
Effects of the climatic parameters on physiological indicators (rectal
temperature and respiratory rhythm).
Berman et al. (1985) found that air temperatures between 10°C and 24 °C have no influence on
the rectal temperature; instead this indicator varies with production, increasing by 0.02 °C per
kg milk over a production of 24 kg per day. Over the threshold of 26 °C and until 36 °C a
Fig. 12: Effective temperature as a function of the air speed and relative humidity on the
calculated by the model of Frazzi et al., 1998.
45
forced ventilation with an air speed between 1.5 and 3.0 m/s reduced the rectal temperature of
the ventilated cows to the half of the not ventilated (graph of fig. 9).
Berman and coll. investigated also the influence on the respiratory rhythm and their findings
32.0
31.5
31.0
30.5
Effective temperature °C
30.0
air speed
29.5
29.0 0.1 m/s
28.5 0.6 m/s
28.0 1.1 m/s
27.5 1.4 m/s
27.0
26.5
26.0
25.5
25.0
40 45 50 55 60 65 70 75 80
relative umidity %
Fig. 13: Effect of the relative humidity and air speed on the effective temperature (Tair = 30°C)
can be compared to a similar work carried out by Calamari et al. (1994).

Putting the data in the same graph as in fig. 10 we can appreciate a good agreement of the two
works. The same for the respiratory rhythm (fig. 11).
39.4
39.3
rectal temperature °C
39.2
39.1
39
38.9
38.8 Baeta
38.7
Frazzi
38.6
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
air speed m/s
Fig. 14: Influence of the air speed on the rectal temperature

according to Baeta et al. (1987) and Frazzi et al. (1998), at
Tair=30°C and RH=55%.
A more complete work was carried out by Frazzi et al. (1998) through field measurements,
aimed at obtaining an “effective temperature”, defined as the temperature capable of
determining itself the same effects on the rectal temperature and the respiratory rhythm of
various different combinations of air temperature, air speed and relative humidity.
The mathematical model is presented in figure 12.
According to this model and assuming a fixed air temperature of 30 °C we can see in graph of
(fig.13) the effect of the air speed and of the relative humidity on the effective temperature. It
appears from it the prevailing influence of the air speed at least until values of about 1 m/s (at
46
the animal level). The model reveals a decreasing effect the air speed being practically null
over a value of 1.5 m/s (fig. 13).
This result seems different (much lower) from other findings before mentioned, but it can be
explained by the fact that a value of 1.5 m/s of air speed at the animal level (as assumed in this
work) corresponds to a much higher value at the building level.
Finally it is interesting to compare the works of the different authors. First we compared the
Frazzi’s results with Baeta’s. By equalizing the air temperature (30 °C) and the relative
humidity (55%) we obtain the graph of fig 14.
It appears from it a strong difference between the two investigations. A possible explanation
can be the fact that Baeta worked in laboratory chambers, whilst Frazzi worked in a farm,
where the cows took advantage from the nocturnal relief.
The validity of the Frazzi’s results, in a practical farming condition, is confirmed by the
agreement of his results with those of Calamari et al. As far as it is possible to compare the two
works, as we did in table 1, the agreement looks very good.
Tab. 1 Comparison of the average data obtained by Calamari et al. (1994) and the estimates of
the model of Frazzi et al. (1998) at the same microclimatic conditions.
Author and air speed Rectal temperature Respiratory Rhytm

Calamari 0.5-0.8 m/s 38.7 50.4
Frazzi 0.5 m/s 38.65 56.9
Frazzi 0.8 m/s 38.53 55.7
Among the two indicators, the respiratory rhythm has revealed to be a more immediate sign of
1
0
-1
-2
-3
-4
-5
-6
-7 Spiers regression
-8 "Calamari data"
-9
38.5 39 39.5 40 40.5 41
rectal temperature daily average °C
Fig. 15: Milk production variation as a function of the average rectal

temperature: comparison of the Spiers et al (2004) regression
equation and Calamari et al. (1994) collected data.
heat stress, but the rectal temperature has generally been found to be more related to the
productive variations.
This was the conclusion of a study made by Spiers et al. (2004) in climatic chambers where
Freisian cows were exposed to a constant stressful condition (around 29 °C of temperature and
50% of RH). On the other hand the field investigation of Calamari et al. (1994) showed that the
rectal temperature is more related to the production variation (r = 0.72) than the environmental
47
temperature (r = 0.6). It is important to observe that the conclusions of the two studies agree
quite well, as we can see in fig. 15 where the minor decrease in production found by Calamari
et al. can be explained by the cyclic variation of the daily air temperature giving a nocturnal
relief to the animals.
Furthermore Spiers found that the decrease in production is more correlated to the temperature
of the previous day (r2 = 0.89 instead of 0.82).
Since the respiratory rhythm is considered as we said the first sign of heat stress it is generally
preferred as an indicator in the theoretical models aimed at evaluating the animal response to
the climatic variations.
This was the choice made by Berman (2005). Adapting a previous model proposed by
McGovern and Bruce (2000), he set up a model for the calculation of the thresholds of air
temperature, speed and relative humidity over which the heat stress arises in dairy cows
(corresponding to an increase of the respiratory rhythm to the half of the maximum). The
model is quite complex and it is not worth to detail it in this report. However it is interesting to
observe that its conclusions agree very well with the findings of Frazzi’s model, previously
described.
This comparison was made possible by introducing in the Frazzi’s model (for the
determination of the rectal temperature and respiratory rhythm) the values of air temperature,
relative humidity and air speed found by Berman’s calculations to be thresholds of arising
stress. In this comparison we assumed from the Berman’s model the case of the animal body
surface exposed to a forced air stream by the 75% of the total (more realistic than 100%).
It is quite surprising to see that the climatic values obtained by Berman as thresholds of stress
produce, in the Frazzi’s model (tab. 2), a physiological response of typical stressful conditions,
at least according to the literature (i.e. 39.3 °C and 80 beats/min in Bray and Bucklin, 1994).
Tab. 2 Respiratory rhythm and rectal temperature values as obtained by the

model of Frazzi et al. (1998) using the microclimatic parameters
reported in the Berman’s (2005) model.
Berman model data (input) Frazzi model (output)

RH Air speed Temperature Resp. rhytm Rectal
% m/s °C beats/min temp.
°C
75 0.85 31 78.1 38.96
65 1.5 33 81.6 38.94
The influence of animal variables on patterns of coping with heat load

(behaviour, body weight, production, external insulation, tissue insulation)
The animal related variables that influence the capacity and pattern to resist heat stress can be
divided into three: behavioural, physical and physiological. All three are relevant regarding
practical aspects of housing, cooling systems and management, but the first two are dominant.
Behavioural
The first response to thermal discomfort is behavioural. The cow will situate itself in the most
climatically favourable location in the housing facility. When all cows in the group are doing
48
the same, then a crowding problem occurs which reflects of the design, suitability and cooling
efficiency of the housing facility (see below). The cow tends to lie down most of the time
avoid moving and eating which may direct us to feeding routines that can ease down heat load
(see below). Some researchers (Hillman et al. 2005) found that when the body temperature
reached 38.9 ± 0,05°C the cows stood up in order to increase the heat exchange through the
body surface.
Physical
Large animals have a physical advantage in hot climate being able to store some of the heat
generated by metabolism and production by a small elevation of body temperature during the
hot hours of the day and dissipate it passively to the environment during the cold hours without
having to invest energy and water to do it actively (Schmidt-Nielsen 1964). Indeed, the body
temperature of the dairy cow increases during lactation (Shalit et al. 1990) and fluctuates
during the day rising significantly by about 0.8 0C from morning to afternoon (Aharini et al.
2005). It is yet not clear whether body (rectal) temperature rises before, after or in parallel to
production decline. Although the decline in production correlates inversly with rectal
temperature rather than with respiratory rhythm (which is, as we said, a more immediate
symptom of thermal discomfort) it is quite possible that production may drop as the first
response to heat load without a significant increase in rectal temperature. This possibility is
supported by the dramatic decline of the lower critical point in parallel to production increase
(NRC 2000, Berman 2004). The lower critical point may vary within a range of about 25 0C
(depending on wind velocity) between production of zero and 45 kg milk per day.
Insulation is another physical quality that affects the response of the animal to heat load
(Berman 2004). Because the levels of insulation determine the animal’s resistance to heat load,
together with body weight and production, external and tissue insulations determine the ability
and physiological limits of the animal to cope with heat load, thus dictating the practical ways
that we can provide optimal conditions to exercise this ability by adequate design of housing
and cooling facilities as well as management routines.
Physiological
The Physiological mechanisms that cope with heat load are quite similar in all ruminants and
the dairy cow is no exception, However, the physiological load of milk production, which
through aggressive genetic selection increased it far above physiological needs of the offspring,
restricts the ability to cope with heat load without dropping milk production which is one of the
physiological mechanisms activated to reduce heat load. Indeed, the aggressive selection
turned milk production into a strong physiological driving force accompanied by variety of
adaptations that are also related to coping with heat load such as water, electrolytes, nitrogen
and energy metabolism (Shalit et al, 1991, Maltz et al.1996, Silanikove et al. 1997, Maltz et al.
1997) that enables to some extend to maintain production while operating physiological
mechanisms to cope with heat load such as vasodilatation, panting and sweating, but this is
quite limited. Without external interference milk production as well as other physiological
functions such as reproduction will be damaged under conditions of heat load. The external
interference includes a combination of housing conditions, cooling systems and management
routines providing the animal the conditions to exercise optimally the physiological
mechanisms to cope with heat load on one hand and actively cool the cows on the other.
The influence of housing cooling and management on resistance to heat load

Housing (design and materials), cooling systems and management routines help livestock in
general and the dairy cow in particular to cope with environmental heat load. Each of these
measures has its significance and one can not replace the other.
49
Housing
The first aspect of dairy housing facilities is the location and orientation obviously subjected to
local constrains. Recently a survey performed in Israel were the climate is dominated by
rainless very hot summers (Shoshani et al. 2006) evaluated the efficiency of a housing facility
by calculating the threshold temperature in which a cow begins to increase the respiratory rate
by using the heat stress model developed by Berman (2004, 2005). This model takes into
account wind velocity, relative humidity, ambient temperature and physiological characteristics
of a dairy cow which are: milk yield (45 kg/d, 3.5% fat) and 3 mm fur depth. The results
showed that the optimal barn for high yielding cows is as follows: Fully roofed loose house
(vs. free stalls), oriented perpendicular to the dominant wind, with opened roof (sliding or
shatters), opened ridge, roof margins of about 5 m, roof slope 19-22%, and barn width 30-50
m. Barns oriented parallel to dominant wind require higher roofs and margins and narrower
width.
Other aspects as roof insulation or emissivty will be considerated in another paper.
Cooling systems
It was already described that air velocity has a significant effect on the animals ability to
dissipate excessive heat. Therefore fans are very popular in dairy barns either fixed or moving
to ventilate a section of the barn. However, their efficiency is limited only to favourable wind
direction conditions or no wind and it depends also on the capability of air to evaporate sweat
from body surface, that is higher in dry than in wet climate. Therefore, the forced evaporative
cooling (cycles of water sparkling and fans ventilation) becomes popular in hot environments.
It was not until the eighties of the previous century that the combination of sprinkling and
forced ventilation was developed as a cooling system for dairy cows in Israel (Flamenbaum et
al. 1986), which was gradually adopted by the Israeli dairy industry with excellent results
(Israeli Milk Board, 2004). This system is currently implemented in two segments: forced
evaporative cooling in the milking parlor waiting yard and in the feeding alley, and voluntary
evaporative cooling in the feeding alley and lying area. The forced evaporative cooling system
comprises of restricting the cows to an area where the cows are sprinkled and ventilated
successively for about a 1/2-1 hour. This takes place in the milking parlor waiting area before
and between milkings, and in the feeding alley were the cows are yoke-locked when reaching
for the freshly distributed food. The superior effect is reported by the farmers to be the forced
cooling in the milking parlor waiting area. However, the time that this procedure takes each
time it is performed should be limited in order not to impair lying time which is a significant
factor for high producing dairy cows (Fregonesi, and Leaver, 2001; Horning et al. 2001).
Recently it was measured that walking the cows twice daily (between morning and noon and
noon and night milkings) a distance of 50 m to the milking parlor waiting area is not reducing
diurnal and between milkings lying time (Maltz 2006). Also the walking distance between the
shade and the milking parlor is an important factor. It will be a waist of water, energy and time
(on both parts, the farmer and the cow) if the cooling effect will be wasted by walking under a
burning sun to and from the cooling procedure.
The voluntary cooling is operated by timing the sprinkling and ventilation in each shade
according to diurnal heat load and expected presence of cows in accordance to milking and
feeding time. The waist of energy and water of voluntary cooling is much greater because in
the absence of an efficient control system, the system often wets and ventilates areas empty of
cows (see below).
Both, forced and voluntary cooling involve a significant water waist (an important resource in
many hot environments) which makes this cooling system also an environmental concern.
50
Cooling control
Active cooling management should take place after the possibilities offered by housing and
behavioural thermoregulation were exhausted. In the dairy, the object that has to be cooled is
the cow, but only when heat load exceeds the point when energy has to be invested in order to
prevent a positive heat balance – beyond the thermal upper critical point. The means to cool
the animal are water and energy (ventilation), both valuable resources, and wasted water in
the dairy environment becomes an environmental and manure handling problem. All this
suggests that the core of an efficient and effective livestock cooling system is its control. The
control issue can be divided into two: the animal related and the resources related. Regarding
the animal, the cooling system should be operated only when it is needed physiologically and
only in the presence of the "consumer" – the cow. Regarding the resources, the water should
be exploited only for cooling, either by wetting the cow or cooling the air without loosing
water to the manure. Energy should be invested only for ventilating the air to evaporate water
from the wet cow and circulating cooled air in the barn before it is driven out by the wind. This
means that the control system should be operated responding not only to the environmental
but also to the animal conditions normally all the systems operate automatically by means of
timing regarding expected environmental conditions. The dairy cow manifests its
thermoregulatory status by panting that should be used through proper sensors for control
purposes. Presence of cows in a ventilated area in addition to the available sensors for
temperature (dry and wet bulb) wind velocity and direction, can improve cooling efficiency
without waist of resources.
On-line control that includes sensors for environmental conditions and cow's presence and
scattering around the barn is required for the cooling system to operate only when needed and
affecting the animals.
Recently, there is an attempt in Israel to adopt the fogging system that is only cooling the
environment without wetting the cows or the manure and adding to it an air movement control
that combines moving screens on one longitude side of the barn and circulating the cooled air
in the barn space to fully exploit its cooling capacity (Arbel 2006)
Feeding management
Summer feeding has two aspects: one is feeding timing ant the other nutritional. Aharoni et al.
(2005) concluded that the transfer of part of the intake of high-yielding dairy cows in summer
conditions from day to night hours resulted in a decline of energy expenditure during the hot
daytime hours, with no compensation for this decline during night hours. These cows reduced
their feed intake but not their milk yield, compared with day-fed cows, and their decline in
milk yield with time was smaller than that of day-fed cows. The energy expenditure of night-
fed cows in the summer was lower than that of day-fed cows and, as a result, their efficiency of
energy utilization for milk production was higher.
Adin et al. (2006) and Miron et al. (2006), demonstrated the superiority of feeding cows with a
prime cell wall reach diet during the summer on physiological parameters and performance of
high yielding dairy cows.
Integrated relative thermal-comfort index

Better understanding of the effects of integrated environmental conditions as well as animal
related variables on the thermoregulatory capacity of the dairy cow on one hand and newly
developed technologies and sensors on the other, call to replace the traditional temperature
humidity index (THI) by a more adequate index. The model suggested by Berman (2004)
indicates that local environmental conditions and animal characteristics such as fur depth and
production level, strongly affect the condition in which a dairy cow starts thermoregulation.
The integration of local environmental and physiological condition in which the animal starts
51
to thermoregulate under given facility conditions, should become the index which evaluates the
efficiency of a housing facility to support the animal’s ability to fully exploit its
thermoregulation capacity. On the other hand, measuring all the relevant variables
(environmental as well as animal) can be integrated into a set point which starts the active
evaporative cooling system. There will be an upper and lower index point, similar to upper and
lower critical temperature point.
Integrated Relative Thermal-Comfort Index (at the stage the animal starts thermoregulation)
(IRTCI) = f(Ta) + f(RH) + f(AC) + f(R)+ f(TI) + f(EI) + f(BW) + f(Prod)
Were:
Ta – Ambient temperature (0C)
RH – Relative humidity (%)
AC – Airflow circulation: pattern of air movement in a livestock facility; direction (an angle
relative to the long axis of the facility if there is any) and velocity (m/sec)
R – Radiation (W/m2)
TI – Tissue insulation 0C/Mcal/m2/day
EI – External insulation (Mcal/m2/0C/day)
BW – Body weight (kg)
Prod – Production (kg/day)
References
Adin, G., E. Yosef, A. Zino, M. Nikbahat, A. Shamai, A. Brosh, I. Halachmi, R. Solomon, E.
Shoshani, I. Flamenbaum, S. Mavgish, J. Miron. (2006).Effect of feeding prime cell wall and
cows cooling in the summer over physiological and energetical aspects. The Annual 18th
Conference on Cattle Sciences, 6-8 June 2006, Jerusalem, Israel (in Hebrew).
Aharoni, Y., Brosh, A., and Harari, Y. (2005). Night feeding for high-yielding dairy cows in
hot weather: Effects on intake, milk yield and energy expenditures. Livestock Production
Science 92: 207-219
Arbel, A., (2006). Cooling livestock buildings by high preasure fogging system with air
ventilation and circulation systems: Israel's solution. In: Animal Housing In Hot Climates. A
multidisciplinary view. Workshop of CIGR section II. Ed: Irenilza de Alencar Nääs, Co-
operating. ed: Daniella Jorge de Moura. Campinas, SP – Brazil, pp 23-27
Baeta F.C., Meador N.F., Shanklin M.D., Johnson H.D. (1987). Equivalent Temperature Index
above the thermoneutral for lactating dairy cows, ASAE, Paper 87-4015, Proc. Am. Soc.
Agric. Eng., St. Joseph, Mich.: ASAE.
Berman, A. (2005). Estimates of heat stress relief needs for Holstein dairy cows, J. Anim. Sci.
83, 1377-1384.
Berman, A., Folman Y., Kalm M., Mamen M., Herz Z., Wolfenson D., Ariell A., Graber Y.
(1985). Upper critical temperatures and forced ventilation effects for high-yielding dairy cows
in a subtropical environment, J. Dairy Sci. 68, 1488-1495.
Berry, I.L., Shanklin M.D., Johnson H.D. (1964). Dairy shelter design based on milk
production decline as affected by temperature and humidity, Transactions of the ASAE, 7(3),
329-331.
Bouraoui, R., Lahmar M., Majdoub A., Djemali M., Belyea R. (2002). The relationship of
temperature-humidity index with milk production of dairy cows in a Mediterranean climate,
Anim. Res. 51, 479-491.
Bray D.R., Bucklin R.A., Montoya R., Giesy R. (1994). Cooling methods for dairy housing in the
Southeastern United States, ASAE Paper No. 94-4501. St. Joseph, Mich.: ASAE.
Broucek, J., Uhrincat’ M., Kovalciková M., Arave C.W. (1998). Effects of heat environment
on performance, behavior and physiological responses of dairy cows, Proceedings of the 4th
International Dairy Housing Conference, ASAE, 217-223.
52
Calamari, L., Maianti M.G., Cappa V., Frazzi E. (1994). The influence of the air speed on yield
and milk characteristics in dairy cows during summer, Proceedings International Conference
on Agricultural Engineering, Milano, 29 Agosto - 1 Settembre, 1, 101-108.
Flamenbaum, I., D. Wolfenson, M. Maman, and A. Berman. 1986. Cooling dairy cattle by
combination of sprinkling and forced ventilation and its implementation in the shelter system.
J. Dairy Sci. 69:314
Frazzi, E., Calamari L., Calegari F. (1998). Dairy cows heat stress index including air speed
parameter, Rivista di Ingegneria Agraria 29(2), 91–96.
Frazzi, E., Calamari L., Calegari F. (2003). Assessment of a thermal comfort index to estimate
the reduction of milk production caused by heat stress in dairy cow herds, Proceedings of the
5th Dairy Housing Conference, ASAE 1, 269-276.
Frazzi, E., Calamari L., Calegari F., Stefanini L. (2000). Behavior of dairy cows in
Fregonesi, J. A., and J. D. Leaver. 2001. Behaviour, performance and health indicators of
welfare for dairy cows housed in strawyard or cubicle systems. Livest. Prod. Sci. 68:205-216.
Hillman P.E., Lee C.N., Willard S.T. (2005). Thermoregulatory Responses Associated With
Lying and Standing in Heat Stressed Dairy Cows. Transactions of the ASAE 48(2), 795-801.
Horning, B., C. Zeitlmann, and J. Tost. 2001. Differences in the behaviour of dairy cows in the
lying area of 40 loose houses. KTBL-Schrift 403:153-162.
Igono, M.O., G. Bjotvedt, and H.T. Sanford-Crane. 1992. Environmental profile and critical
temperature effects on milk production of Holstein cows in desert climate. Int. J. Biometerol.
36,77-87.
Israel Milk Board. 2004. Year Book 2004. Israeli Milk Board Publication, Rishon Le'Zion,
Israel (in Hebrew).
Kadzere, C.T., Murphy M.R., Silanikove N., Maltz E. (2002). Heat stress in lactating dairy
cows: a review. Liv. Pro. Sci. 77, 59-91.
Kelly, C.F., Bond T.E. (1971). Bioclimatic factors and their measurements, In: A guide to
environmental research in animals. National Academy of Sciences, Washington, D.C., p.77.
Maltz, E., N. Silanikove. (1996). Kidney function and nitrogen balance of high - yielding dairy
cows at the onset of lactation. J. Dairy Sci. 79: 1621-1626.
Maltz, E., Devir, S., J.H.M. Metz, H. Hogeveen. (1997). The body weight of the dairy cow: I.
Introductory study into body weight changes in dairy cows as a management aid. Livestock
Production Science. 48:175-186
Maltz, E., (2006). Behaviour sensor for welfare assessment and physiological status of the
dairy cow. Forth International Workshop Smart Sensors in Livestock Monitoring, September
22-23, Gargnano (BRESCIA), Italy.
McDowell, R.E., Hooven N.W., Camoens J.K. (1976). Effects of climate on performance of
Holsteins in first lactation, J. Dairy Sci. 59, 965–973.
Miron, J., E. Yosef, A. Zino, M. Nikbahat, A. Shamai, A. Brosh, I. Halachmi, R. Solomon, E.
Shoshani, I. Flamenbaum, S. Mavgish, G. Adin. (2006). Effect of feeding prime cell wall and
cows cooling in the summer over feed intake and milk yield and composition. The Annual 18th
Conference on Cattle Sciences, 6-8 June 2006, Jerusalem, Israel (in Hebrew).
Reinemann, D.J., Smith T.R., Timmons M.B., Meyers A.P. (1992). Cumulative effects of heat
stress on milk production in Holstein herds, Paper No. 924027, Written for presentation at the
1992 International Summer Meeting sponsored by ASAE, Charolette, North Carolina, June 21-
24.
Schmidt-Nielsen, K., (1964) Desert animals: Physiological problems of heat and water.
Oxford: Oxford University Press.
Shalit, U., E. Maltz, N. Silanikove, A. Berman. (1991). Water, Na, K and Cl metabolism of
dairy cows at onset of lactation in hot weather. J. Dairy Sci. 74:1874-1883
53
Shoshani, E., A. Hetzroni, A. Levi, and R. Brikman (2006). Characteristics of barn design for
dairy cows under harsh conditions derived by a new stress model. EAAP, Book of abstracts
No. 12 (2006), Antalya, Turky, 17-20 September
Silanikove, N., E. Maltz, A. Halevi, D. Shinder (1997). Metabolism of water, sodium,
potassium, and chlorine by high yielding dairy cows at onset of lactation. J. Dairy Sci. 80:949-
956.
Spiers, D.E., Spain J.N., Sampson J.D., Rhoads R.P. (2004). Use of physiological parameters
to predict milk yield and feed intake in heat-stressed dairy cow, J. Thermal Biol. 29, 759-764.
USDC-ESSA (1970). Livestock hot weather stress, Central Regional Operations Manual Letter
70-28. Environmental Sciences Services Admin., U.S. Dept. Commerce, Kansas City, MO.
West, J.W., Mullinix B.G., Bernard J.K. (2003). Effect of hot, humid weather on milk
temperature, dry matter intake, and milk yield of lactating dairy cow, J. Dairy Sci. 86, 232-242.
Wiersma, F. (1990). Temperature-humidity index table for dairy producer to estimate heat
stress for dairy cows, Department of Agricultural Engineering, The University of Arizona,
Tucson.
Yeck, R.G., Stewart R.E. (1959). A ten-year summary of the psychroenergetic laboratory dairy
cattle research at the University of Missouri, ASAE Trans., 2(1), 71-77.
54
Method to evaluate and optimise climate control strategies in livestock

buildings taking into account dynamic behaviour of animal heat production
D. Berckmans, E. Vranken, Division M3-Biores, Catholic University Leuven, Belgium
Introduction
Although a whole range of new technology has become available for ventilation, heating and
cooling of livestock buildings, little progress has been made in the development of adequate
control algorithms. To obtain more benefit from these new technologies, more knowledge
about the interaction between the animal responses (e.g. heat production) and the control
actions has to be integrated into the applied control algorithms. Although several authors tried
to give some general guidelines about the use of ventilation control equipment for livestock
buildings (Van ‘t Klooster et al., 1989; Van Ouwerkerk, 1995 CIGR, 1989; Albright, 1990),
there is no internationally accepted procedure and there is no agreement between the guidelines
for tuning ventilation controllers in livestock buildings. Consequently, much ignorance exists,
resulting very often in a sub-optimal indoor climate control.
An appropriate technique to develop and optimise control algorithms is the use of simulation
models. In the last decades, different authors have described mathematical models to simulate
the indoor climate in livestock buildings (Surbrook et al., 1979; Mitchell, 1993; Overhults et
al., 1994; Zhang et al., 1995; Van Ouwerkerk, 1995; Timmons et al., 1995;), but most of them
assume constant steady state heat, moisture and gas balances applied to a perfectly mixed
ventilated space. By combining steady state equations with the equations that describe the
controller actions, the resulting indoor climatic conditions at different outdoor temperatures can
be calculated, but they do not take into account the dynamic responses of animal heat
production. Such approach is only acceptable for evaluating the indoor temperature if the
temperature difference between the indoor and the outdoor temperature is high and the daily
fluctuations of the outdoor temperature is relatively low (Berckmans, 1986).
In the evaluation of the efficiency of ventilation and cooling strategies, a more realistic
dynamic modelling approach is necessary. This means that, due to the dynamic changes of
inputs (animal heat production, ventilation rate, cooling rate, etc.) and disturbing factors
(outdoor climate); the calculated indoor conditions rapidly change over time.
In order to describe the dynamics of such a complex process like the temperature and humidity
in a livestock building, a set of differential equations can be written, which mostly cannot be
solved analytically and often numerical solutions are used. Examples of such dynamic models
were given by Axaopoulos et al. (1992) who used four terms in the thermal equation: animal
sensible heat production, heat exchange through the walls, heat exchange through the floor and
ventilation losses. Chao et al. (1992) incorporated the effect of the climate controller in his
model. The weak point in these models is the prediction of the dynamic behaviour of the
animal sensible and latent heat production. The most referred to work is the model of Bruce
and Clark (1979) for fattening pigs and Reece & Lott (1982) for chickens, but they are only
valid under steady state conditions. Wagemans & van Ouwerkerk (1995) used a combination
of a model that calculates the thermal comfort zone of pigs (Sterrenburg & van Ouwerkerk,
1986) and a dynamic model of the climate and energy balance of a pig house and a reference
year to calculate the effect of ventilation control techniques on the energy consumption and the
ammonia emission of pig houses. The same combination of models was used by Van ‘t
klooster et al. (1989) to calculate climate standards for pigs. Zhang et al. (1992) developed a
dynamic model of the thermal environment that has proven useful to analyse alternative on/off
control strategies in a perfect mixed air space.
Most of the authors never incorporated the dynamic responses of animal heat and moisture
production into their model analysis and did not perform validation works under real field
55
conditions and with living animals. Consequently the reliability of such simulation models
remains questionable, especially when these equations are used in combination with other sub
models. This is often done in simulations of the total system to evaluate control algorithms.
This paper explains how dynamic animal heat and moisture production can be used in a global
simulation model that uses dynamic energy and mass balances to calculate the fast fluctuating
changes of indoor temperature and humidity for optimisation of control actions for heating,
ventilation and cooling of livestock buildings.
Methods
The global climate simulation model, adapted from Berckmans et al., 1992, consists of
different sub models, as presented in Fig. 1. These parts describe by a set of mathematical
equations the dynamic behaviour of different sub-systems of a mechanically ventilated pig
house, such as the (1) controller, (2) the heating system, (3) the fan, (4) the process of heat and
mass exchange within the ventilated structure and (5) the temperature sensor. The inputs of the
model are the control settings as a function of time. Disturbance variables are the outdoor
climatic data of temperature and humidity, originating from a reference year (Dogniaux et al.,
1980).
Outdoor climatic data

Control settings
Xo To
2
Temperature Qs
Heating system Xi
controller
Temperature
Ti
set point
Φa Φa
Fan
1 3
Process
4
5
Temperature sensor
Figure 1: The different sub models in the climate dynamic simulation model’s, heat supply of
heating element; Ti , indoor temperature; To ,external temperature; Xi, indoor absolute
humidity; Xo, outdoor absolute humidity; Φa, air flow
The mechanistic model simulates the dynamic behaviour of outside temperature and humidity,
the sensible and latent heat production of the animals, and hence calculates the dynamic
behaviour of the resulting inside temperature and humidity and the corresponding energy use
for heating and ventilation.
Inside temperature was derived from inside enthalpy and humidity, with (Berckmans et al.,
1992):
H −εi ⋅ Xi
Ti = i (1)
c a + cv ⋅ X i
Moisture and enthalpy balance in the building was calculated according to Eqn. 2 and Eqn. 3,
respectively:
56
dX i IL Xio + φl ε i + X e
= (2)
dt V ⋅γ
dH i ILHio + SL + φl + φ s + φh
= (3)
dt V ⋅γ
Input variables in the different sub-model parts were outside weather conditions, number and
weight of animals, data on their feed intake in relation to maintenance requirements, building
thermal characteristics, as well as control settings for ventilation and cooling.
Building characteristics included wall and floor heat transfer coefficients and their thermal
capacities, as well as the dimensions of the compartments.
The dynamic animal heat production
The well established model of Bruce and Clark is used to determine the sensible and latent heat
production of the animals below the critical temperature and within the thermo neutral zone
(Bruce & Clark, 1979). The model variables are: indoor air temperature, air velocity, floor type,
live weight, and group size.
As a first input variable, the animal weight M has to be known as a function of time. Traditionally,
the fattening period start with animals of 20 kg and grow to 105 kg. The length of each fattening
period is 140 days, but different growth curves can be used as well. During simulation, the animal
weight is calculated each time step from a pre-set growth curve.
Then a whole set of variables has to be derived to finally calculate the upper and lower critical
temperatures:
- Tissue thermal resistance: Rt = 0.02 M 0.33 (4)
- Minimum evaporative heat loss per m2 of pig area: E LCT = 8 + 0.07 M (5)
- Total area per animal: A = 0.09 M 0.67 (6)
- Contact area ratio between pigs: Ac = 0.15 N pen − 1 (7)

A N pen
- Maintenance energy requirement: m = 5.09 M 0.75 (8)
- Metabolizable energy intake in feed: F = mlf (9)

lf is the nutrition level: it is kept in a user definable data file as a function of live weight and
obtained by interpolation for the present weight.
- Efficiency of utilization of metabolizable energy for growth:
f = 0.625 + 0.00141M (animals < 100kg)
f = 0.75 (animals > 100kg) (11)
- Thermo neutral heat production (the sum of sensible and latent heat production is considered
to be constant in the thermo neutral zone):
Q n = m + (1 − f )( F − m ) (12)
57
- External thermal resistance at skin exposed to air:
1
Ra = 0.6
(13)
5.3 + 15.7 va0.13
M
va is taken as 0.2 m/s
- Effective thermal resistance of the floor:
0.33
M   A f  0.5
R f = R f 45    0.2 A  N pen (14)
 45   
But: if Rf45=0 : Rf = Ra (15)
Rf45 is the Rf resistance for an animal of 45 kg and varies with the type of floor between 0 (metal
mesh) and 0.23 (wet straw)
- Contact area ratio animal - floor:
Af
When R f ≥ Ra : = 0. 2 (16)
A
Af
When R f < Ra : = 0.1 and R f = R f (17)
A 2
- Latent heat production at lower critical temperature:
H LLCT = E LCT A (18)
- Latent heat production at upper critical temperature:
H LUCT = EUCT A (19)
EUCT is an experimental extension of the model of Bruce and Clark: 37 W/m2
- Sensible heat production at lower critical temperature:
H SLCT = Q n − H LLCT (20)
- Sensible heat production at lower upper temperature:
H SUCT = Q n − H LUCT (21)

- Lower critical temperature:
Q n ( Ra + R f ) − E LCT A Ra
T LC = T b − (22)
 A R − R f  Ac
A 1 + f a −
 A Rt + R f  A
58
- Upper critical temperature; is this temperature region it is accepted that (extension of the
model):
A f
= 0.2 (23)
A
0.33
M 
if Rf45 < > 0 : R f = R f 45   (24)
 45 
and further: Rt = 0 and Ac/A = 0
Q n Ra − EUCT A Ra
T UC = T b − (25)
 A R − R f 
A 1 + f a
 A R f 
Once the sensible and latent heat productions at both upper and lower critical temperature are
known, they can be derived for any other temperature by interpolation (neutral zone) or
extrapolation (otherwise).
For the thermo neutral zone, the interpolation equations are:
H Sst = H SLCT + H SUCT H SLCT (T i − T LC )

−
(26)
T UC − T LC
H Lst = H LLCT + H LUCT H LLCT (T i − T LC )

−
(27)
T UC − T LC
Below TLC
To extrapolate, a “Bruce & Clark Extrapolation factor” is defined (all the variables in this
equation are those defined to calculate the lower critical temperature):
 A f Ra − R f Ac 
A 1 + − 
 A Rt + R f A  dH Set
EPF BC = = (28)
(R a + R t ) dT i
This is the rate of change of sensible heat production with inside temperature below TLC. The
extrapolation equations are:
H Set = H SLCT + EPF BC (T LC − T i ) (29)
H Lst = H LLCT (30)
Above TUC, the sensible heat starts dropping and will become zero for Ti = Tb.
59
H SUCT ( −
H Set = H SUCT − T i T UC ) (31)
T b − T UC
H Lst = H LUCT (32)
The Bruce and Clark model is a steady state model. To calculate the heat production as a function
of time, the animal is assumed to act as a first order system to temperature variations. This is
based on the idea that the animal will not waste energy by reacting as a higher order system. Until
more data are available, this assumption is used and an estimated value for τSH and τLH is applied.
To obtain the dynamic heat production as a function of time, a first order system was applied to
the static acquired animal heat production by the equations:
dH s = 1 (
H Sst − H S ) (33)
dt τ SH
dH L 1
= (H Lst − H L ) (34)
dt τ LH
When these derivates are known, the heat production at each time step can be calculated.
Figure 2: Calculation of the dynamic animal heat production
Results
The global simulation model calculates the dynamic changes of the indoor temperature and air
humidity in a livestock building with a time step of 3 seconds over a time period of one year by
60
using outdoor temperature and humidity data obtained from a dynamic reference year. The
time step was chosen at 3 seconds, corresponding to the smallest time constant of the system,
in this case the fan.
The model uses a number of input data, such as the time constants of the different sub systems,
the dimensions and thermal characteristics of the building structure, the simulation period; the
animal growth data, the heat and moisture production and finally the control settings.
A typical simulation output for a typical Belgian fattening pig house compartment of 80 pigs is
shown in figure 2 (Vranken et al., 1997).
Simulation on yearly basis

80 fattening pigs
Fattening period 1 Fattening period 2 Fattening period 3
30
25
20
Temperature (°C)
15
10
5
0
-5
-10
0 30 60 90 120 150 180 210 240 270 300 330 360
Time (day number)
inside temp outside temp lower comfort level upper comfort level
Figure 2: Simulation output of indoor temperature on yearly basis for a typical Belgian pig
house compartment.
The figure clearly shows the critical periods in relation to the thermal comfort zone of the
animal. During the first 10 days (January), the indoor temperatures are just below the lower
comfort level and in the second half of fattening period 2, the temperature are above upper
critical values during hot summer days.
The described method allows to do dynamic analyses to evaluate the effect of control systems
on the dynamic behaviour of indoor climate. Moreover it gives a solution to do analyses taking
into account the dynamic behaviour of animal heat and moisture production.
This type of analysis is very helpful for design of ventilation control equipment, such as fans,
heating systems or cooling equipment. The model was used to design and evaluate the dynamic
effect of evaporative cooling systems, such as fogging on indoor climate in pig facilities
(Haeussermann et al., 2005a,b).
Simulation of evaporative cooling for a Belgium reference year showed a reduction of the
maximum indoor temperature by 2.3°C , while for a reference year from the Hohenheim region
in Southern Germany a similar cooling system reduced the maximum indoor temperature by
about 4.8°C. The difference in the reduction of the maximum temperature was mainly caused
by the outside humidity, which averaged at 85 % r. h. in Belgium and at 76 % r. h. in Southern
Germany, giving a higher potential to use adiabatic cooling at the latter region (Haeussermann
et al., 2005a).
Conclusions
Although modern controllers for ventilation and cooling of livestock buildings use advanced
hardware technologies, they have the disadvantage that the basic control algorithms do not
contain any biological process information. Consequently many setpoints have to be tuned by
61
the end user, who in most cases is no ventilation expert. In this paper a simulation model is
presented to evaluate the tuning of ventilation controller setpoints taking into account the
dynamic heat and moisture production of pigs.
As an example, the model was used to design and evaluate the dynamic effect of evaporative
cooling systems, such as fogging on indoor climate in pig facilities.
This example showed that the added value of modern ventilation control systems must be
found in the use of more optimal control strategies and the method described in this chapter can
be used to develop these.
References
Albright, L. D. (1990). Environment control for animals and plants. Michigan,USA: The
American Society of Agricultural Engineers, ISBN 0929355083
Axaopoulos, P., Panagakis, P., & Kyritsis, S. (1992). Computer simulation assessment of the
thermal micro-environment of growing pigs under summer conditions. Transactions of
the ASAE, Vol 35(3), pp.1005-1009.
Berckmans, D. (1986a). Analyse van de klimaatbeheersing in dierlijke produktie-eenheden ter

optimalisering van de regeling - Analysis of climate control in animal houses for optimal
control purposes . Ph.D-thesis 146 K.U.Leuven - Fac. Agric. Sciences, 374 p.
Berckmans, D., Van Pee, M., & Goedseels, V. (1992). Evaluation of Livestock environment by
simulation technique. International Summermeeting ASAE, Charlotte, North Carolina,
1992, June 21 (Paper no. 92-4055).
Bruce, J. M., & Clark, J. J. (1979). Models of heat production and critical temperature for
growing pigs. Animal Production, vol.28. pp.353-369.
Chao, K. L., Gates, R. S., & Chi, H. (1992). Diagnostic hardware/ software system for
environmental controllers. ASAE-paper 92- 3560.
CIGR. (1989). 2nd report of CIGR working group on Climatization of animal houses. SFBIU,
Aberdeen
Dogniaux, A., Lemone, N., & Sneyers E.R.S. (1980). Années types moyennes pour le
traitement des problèmes des charges thermiques de bâtiments. I.R.M. M.I.C. Serie B,
nr.45.
Haeussermann, A., Vranken, E., Aerts, J.M., Hartung, E., Jungbluth, T., and D. Berckmans
2005a. Evaluation Method for Improved Control of Adiabatic Air Cooling in Pig
Facilities. ASAE Annual International Meeting, July 17-20, 2005, Tampa, Florida, USA,
paper No. 054019, 12 p.
Haeussermann, A., Vranken, E., Aerts, J.M., Hartung, E., Jungbluth, T., and D. Berckmans
2005b. Process control of evaporative indoor air cooling with a combined data based and
mechanistic model. In: Book of abstracts 2nd European Conference on Precision
Livestock Farming, Uppsala, Sweden, 2 p.
Surbrook, T. C., Esmay, M. L., & Bickert, W. G. (1979). Control effectiveness and energy
comparison for animal housing ventilation and heating systems. ASAE paper no 79-
4526, 12 p.
62
Mitchell, B. W. (1993). Process Control System for Poultry House Environment. Transactions
of the ASAE, Vol 36(6), pp.1881-1886.
Overhults D.G., & Gates R.S. (1994). Energy use in tunnel ventilated broiler housing with
different controls. 1994 ASAE International Winter Meeting, Atlanta Hilton & Towers,
Atlanta, Georgia, 1994, December 13, 1994
Reece, F. N., & Lott, B. D. (1982). The effect of environmental temperature on sensible and
latent heat production of broiler chickens. Poultry Science, vol.61. pp.1590-1593.
Sterrenburg, PP., & Van Ouwerkerk, E. N. J. (1986). Rekenmodel voor de bepaling van de
thermische behaaglijkheidszone van varkens (BEZOVA). Rapport 78, Instituut voor
Mechanisatie,Arbeid en Gebouwen, IMAG, Wageningen, (In Dutch)
Timmons, M. B., & Gates, R. S. (1985). Optimizing broiler production with an expert system.
Winter Meeting of the ASAE, Chicago, 1985, December 17, 1985 paper no.85-4547.
Van Ouwerkerk, E. N. J., & Aarnink, A. J. A. (1995). Gasproduktie in vleesvarkensstallen (gas

production in pig houses). Imag-DLO rapport, 94-32. (In Dutch)
Van 't Klooster, C. E. et al. (1989). Klimaatsnormen voor varkens. (climate standards for pigs)
Proefverslag Proefstation voor Varkenshouderij, P.1.43, Rosmalen (NL). (In Dutch)
Vranken, E., Berckmans, D., & Goedseels, V. (1997). Analysis of Livestock environment by
simulation technique and field data. Proceedings of the CLIMA 2000 conference,
Brussels, 1902, August 30, 1997
Wagemans, M. J. M., & van Ouwerkerk, E. N. J. (1995). Effekt van koel- en
ventilatiemethoden in varkensstallen op het energieverbruik en de ammoniakemissie.
IMAG-DLO nota V95-12, 54 p. In Dutch
Zhang, Y., Barber, E. M., & Sokhansanj, S. (1992). A model of the dynamic Thermal
Environment in Livestock Buildings. J. Agric. Eng. Res., vol 53, pp.103-122.
Zhang Y., & Barber E.M. (1995). An evaluation of heating anf ventilation control strategies
for livestock buildings. Journal of Agric. Engineering Research, vol.60, pp. 217-225.
63
Symbols
A Animal area [m2]
Ac Contact area between two pigs [m2]
Af Contact area pig – floor [m2]
ca heat capacity of air [J kg-1 K]
cv heat capacity of vapor [J kg-1 K]
ELCT evaporative heat loss at lower critical temperature [W m-1]
EUCT evaporative heat loss at upper critical temperature [W m-1]
EBFBC Bruce & Clark extrapolation factor [W °C-1]
f efficiency of utilisation of metabolisable energy for growth
F metabolisable energy intake in feed [W]
Hi inside enthalpy [J kg-1]
HL latent heat production [W]
HLST statical latent heat production [W]
HLLCT latent heat production at lower critical temperature [W]
HLUCT latent heat production at upper critical temperature [W]
HSLCT sensible heat production at lower critical temperature [W]
HSUCT sensible heat production at upper critical temperature [W]
Hs sensible heat production [W]
HSst statical sensible heat production [W]
ILXio humidity losses through ventilation [g kg-1 dry air]
ILHio enthalpy losses through ventilation [J kg-1]
lf nutrition level
m maintenance energy requirement [W]
M animal weight [kg]
64
Npen number of animals per pen
Qn thermoneutral heat production [W]
SL heat losses through structure [W]
Ra external thermal resistance at skin exposed to air [°C m2 W-1]
Rf effective thermal resitance of floor [°C m2 W-1]
Rf45 effective thermal resitance of floor for pig of 45 kg [°C m2 W-1]
Rt tissue thermal reistance [°C m2 W-1]
Tb deep body temperature [°C]
Ti inside temperature [°C]
TLC lower critical temperature [°C]
TUC upper critical temperature [°C]
va air speed [m s-1]
V air volume [m³]
Xi inside humidity [g kg-1 dry air]
γ density of the air [kg m-³]
εi evaporation heat of water at inside temperature [J g-1 H2O]
Φl latent heat supply through animals [W]
Φs sensible heat supply through animals [W]
Φh heat supply heating system [W]
65
Is it convenient to condition the resting area in dairy cows barn?
F. Calegari, P. D’Alessio and E. Frazzi
Università Cattolica del Sacro Cuore – Facoltà di Agraria

Via Emilia Parmense, 84 - 29100 PC - Italy
Keyword: dairy cows, resting area, cooling
Introduction
The problems of heat stress are important in a southern European country, such as Italy, in
relation to the conditions of summer, when high temperatures are coupled with high humidity.
It is documented that heat stress has negative effects on the performances of dairy cows. The
high temperature during the summer season is a severe problem and the animals exposed to
heat stress show a decline in productivity. We also observed a decrease in cheese making
properties (Bernabucci and Calamari 1998; Calamari and Mariani 1998; Frazzi el al., 2002).
The high production levels reached in our farms have changed also the microclimatic cattle’s
needs, so the farmer is forced to adjust the structure of the barn in a way which relieves the
negative effects of heat stress. During a warm summer animal behavior modifications have
been observed; during the night cattle tend to go outside and they spend a considerable part of
the day standing rather than lying down. The percentage of cows resting or ruminating in the
standing position has been observed to increase linearly as temperatures increased.
Research on dairy cows under heat stress and environmental modification of these situations
have demonstrated the positive effect of microclimatic intervention such as forced ventilation,
ventilation with misting and ventilation with sprinkling on cow performance (Bucklin el al,
1991; Calegari el al., 2000; Calegari et al., 2003; Frazzi et al., 2000; Lin et al., 1998; Turner et
al., 1992).
To verify the effect of the forced ventilation on the resting area, we used two pens, one with
climatization in the feeding and in the resting area, one with climatization only upon feeding
area.
This trial was conducted to compare the systems and to evaluate their effects on the
animals (behavior and milk yield), the organizational problems and operating of the
systems.
Material and methods

The trial was carried out in the summer season (from June to September) 2006 with 30
Italian Friesian cows raised in a free stall barn located in the Po Valley. The largest side
(exposed to the west) was completely open to an unshaded paddock, while the other was
half closed by a masonry wall. The cows were fed total mixed ration (TMR) distribuited
ad libitum at 08.00 h.
The animals were subdivided in 2 groups of 15 cows each, homogeneous for days in milk,
calving number and milk yield. The groups were raised in two boxes with free stalls bedded
with straw and both boxes were equipped with a cooling by surface system in the feeding area
(heavy drop sprinklers and fans). One of the two groups was used as control (C), the animals
of the other one (group W) were raised in a box equipped with a cooling system also in the
resting area (forced ventilation on the free stalls) (fig. 1).
The feeding area of C and W were equipped with axial flow fans (90 cm diameter; 22500 m3/h
airflow rate), height approximately 2.5 m, angled downward at about 10 degrees from vertical.
66
The fans were thermostatically controlled and were switched on at 25.5° C. In this area, along
the feeding passage, a set of showers (sprinklers) was placed perpendicular to the air flow of
the fans. The set of showers (4 sprinkler for every box) was made of a polyethylene pipe with a
diameter of 5.0 cm. Every sprinkler had a delivery rate of 5.5 L/min and a pressure of 150-200
kPa. Also this system was thermostatically controlled and the sprinklers started to work at a
temperature of 28 °C: 50 seconds of showering and ventilation followed by 10 minutes of only
ventilation, the time of ventilation decreased of 20 second every degree over 28°C.
fans
sprinklers sprinklers
Figure. 1. Layout of the experimental barn
The resting area of W was equipped with axial flow fans (70 cm diameter; 15000 m3/h airflow
rate) one fan for one row of cubicles. Also these fans were fixed at a height approximately 2.5
m and angled downward at about 10 degrees from vertical, thermostatically controlled and
switched on at 25.5° C.
During the trial these data were collected:
- microclimatic parameters (temperature, relative humidity) inside the barn. (by electronic
probes located in different areas inside the barn);
- breathing rate were measured every ten days in the afternoon (h 15:30);
- daily individual milk yield;
- cow behaviour (by video cameras placed in the barn)
The breathing rate and milk yield were statistically analyzed with the MIXED procedure of
SAS (1999).
All informations of cows behaviour were analyzed using the Pearson (X2) squared.
Results
The weather during the summer 2006 was normally hot, with maximum air temperatures about
35-36°C during the hotter hours of the day. Figure 2 shows the graph of the maximum
temperature and of the minimum relative humidity recorded inside the barn during the trial.
The hottest period was between 18 - 31 July and 04 - 08 September, with maximum intensity
between 22 and 27 July. As is usual in the area under consideration, high temperatures were
associated with high relative humidity and limited air movement.
67
40 100
Max temperature Min relative humidity
90
35
80
Relative humidity (%)

Temperature (°C)
70
30
60
25
50
40
20
30
15 20
/6
/6
/7
/7
/8
/8
/8
/9
/9
7
8
2/
1/
12
22
12
22
11
21
31
10
20
Figure 2. Maximum daily air temperature, and minimum daily relative humidity, during the
trial.
Examining the milk yield from the two groups of cows, it is possible notice that, in the phase
characterized by elevated temperature, the milk yield was higher (+1,0/kg/cow/day) in the W
cows raised in the pen equipped with fans in the resting area. (fig. 3)
35
34
33
Milk yield (kg/d)
32
31
30
29
W C
28
27
n
ug
ug
ug
ug
ep
l
l
Ju
/Ju
/Ju
/Ju
/Ju
Se
Se
/S
/A
/A
/A
A
7/
14
21
28
1/
8/
30
4/
15
11
18
25
Figure 3. Behavior of milk yield, in all group, during the trial
The results show that the breathing rate values were lower in W group than C group especially
during the hotter periods. The behaviour of these values was related to the variation of the
microclimatic parameters during the trial in both groups. The maximum average value of the
breathing rate was observed in C group (fig. 4).
68
70
65
Breathing rate (n/min)

60
*
55
*
50
C W
45
40
/6
/7
/7
/7
/8
/8
/8
/9
7
9
7/
4/
1/
8/
30
14
21
28
11
18
25
15
Figure 4. Behavior of breathing rate, during the trial. (*: P<0.05)
The data referred to the cows lying in the free stalls show a presence in the hottest hours of the
day and in the morning time period that was, more or less, the same in both in the C and W
pens; in the evening and during the night, the cows (W) used the resting area a little more than
the cows of the control group (C). (fig. 5)
100
90
80
70
60
Animal (%)
50
40
*
30
20
C W
10 * * * *
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hours
Figure 5. Cows lying in the free stalls during the hottest period (18 - 31 July), (*: P<0.05)
Conclusion
We could conclude that the treatment of ventilation on the resting area (with cooling by surface
in the feeding area) has positive effects on the thermal balance of the cow that, in this way, it is
able to dissipate an higher rate of heat. The results seem to indicate that, in a free stall barn
with cooling by surface in the feeding area and straw litter in the free stall, a treatment of
forced ventilation in the resting area is the better system to maintain an high production level
under heat stress conditions.
69
References
Bernabucci, U., Calamari L. (1998). Milk production in hot environment. Zoot. Nutr. Anim.,
24, 247-258.
Bucklin, R.A., Turner L.W., Beede D.K., Bray D.R., Hemken R.W. (1991). Methods to relieve
heat stress for dairy cows in hot, humid climates. Applied Engineering in Agriculture 7(2):241-
247.
Calamari, L., Mariani P. (1998). Effects of the hot environment conditions on the main milk
cheesemaking properties. Zoot. Nutr. Anim., 24, 259-272
Calegari, F., L. Calamari and E. Frazzi. (2000). Effects of housing system on the behaviour and
welfare of dairy cows during hot periods and under environment conditioning. In Proc. XIV
memorial CIGR World Congress 2000, Tsukuba, (Japan), 1317-1322.
Calegari F., Calamari L., Frazzi E. 2003 - Effects of ventilation and misting on behaviour of
dairy cattle in the hot season in South Italy. In Proc. “Fifth international dairy housing
conference”, Forth Worth, Texas, 29-31 January, 303-311.
Frazzi, E., Calamari L., Calegari F., Stefanini L. (2000). Behavior of dairy cows in response to
different barn cooling systems. Transaction of the ASAE, 43(2):387-394.
Frazzi, E., L. Calamari and F. Calegari. 2002. Productive response of dairy cows to different
barn cooling systems. Transaction of the ASAE, 45(2):395-405.
Lin, J.C., Moss B.R., Koon J.L, Flood C.A., Smith R.C., Cummins K.A., Coleman D.A.
(1998). Comparison of various fan, sprinkler, and mister systems in reducing heat stress in
dairy cows. Applied Engineering in Agriculture 14(2):177-182.
Turner, L.W., Chastain J.P., Hemken R.W., Gates R.S., Crist W.L. (1992). Reducing heat
stress in dairy cows through sprinkler and fan cooling. Applied Engineering in Agriculture
8(2):251-256.
70
Influence of the thermal environment on the physiological responses of dairy

goats breeding on deep bedding systems.
Santos, C.R; Souza, C.F; Tinôco I. F.; Cruz, V. F.; Pereira, V.N; Acevedo, R.R.; Mendonça,
H.V.
Cláudia Ribeiro dos Santos – crsantos74@yahoo.com.br
Abstract
The objective of this study was to evaluate the influence of thermal comfort, by means
of thermal index, calculated from the Black Globe, dry and wet bulb temperatures, on the
physiological responses of dairy goats raised in the deep bedding system. The experiment
was carried out during the spring at the goat farming sector of the Federal University of
Viçosa in Minas Gerais State, Brazil. The study was performed with 48 Brown Alpine and
Saanen goats, at their eighth week of lactation, divided in three homogeneous groups, with
mean production of 2.0kg/goat and initial weight of 36.0kg. Each group of sixteen animals
were housed in three stalls, in which the following bedding treatments were applied: rice
straw (T1), wood shaving (T2) and grass (T3). To characterize the thermal environment,
black globe, dry and wet bulb temperatures were recorded both inside and outside the
facility in alternate days, every two hours. Thermal comfort was evaluated with basis on the
physiological responses (breathing frequency and rectal temperature). Considering the
conditions in which the experiment was conducted and the obtained results, it can be
concluded that, on average, the thermal comfort of the goats, in the three treatments – rice
straw, wood shaving and grass, evaluated by Black Globe Temperature and Humidity Index
(BGHTI), was considered satisfactory, in other words, the materials did not affect the
thermal comfort of the animals. The resulting physiological responses (breathing frequency
and rectal temperature) were similar to the other records for goats in termoneutral condition.
Key Words: goats, thermal environment, deep bedding
Introduction
From the bioclimatic point of view, even though the goat is considered to
be a rustic animal, the association between high temperatures and a high relative
humidity of air may promote behavioral and physiological alterations, such as
increase in rectal temperature and breathing frequency, reduction of food intake
and therefore decrease in productivity (LU, 1989). This study was done with the
objective of evaluating the effects of thermal comfort, by means of the Black
Globe Temperature and Humidity Index (BGHTI) on physiological responses of
lactating goats breeding in deep bedding systems.
Methodology
The experiment was developed during the spring, in the Goat Production Sector of the
Department of Animal Sciences of the Federal University of Viçosa in Minas Gerais State,
Brazil. The city is located in the latitude of 20 45' 45” and longitude of 42 52’ 04”, with
altitude of 657m. The climate of the region, according to the Köppen classification, is Cwa
(hot, rainy temperate, with dry season in the winter and hot summer). The study was performed
with 48 Brown Alpine and Saanen goats, at their eighth week of lactation, divided in three
homogeneous groups, with mean production of 2.0kg/goat and initial weight of 36.0kg. Each
group, formed from sixteen animals, was housed in three stalls, in which the following bedding
treatments were applied: rice straw (T1), wood shaving (T2) and grass (T3). Each one bed type,
71
with a height of 0.50m, constituted one treatment. The long axis of the building run north and
south; ceiling height of 3.45m; roof of asbestos tiles and eaves close to 1.50m. The walls, with
a height of 1.0m, above the bed surface were built with iron grating. To characterize the
thermal environment, in alternate days, every two hours, from 8:00am until 6:00pm, Black
Globe, dry and wet bulbs temperatures, inside each division as well as in the exterior of the
facility were recorded, throughout a period of three months. The environmental thermal
comfort was evaluated by means of the Black Globe Temperature and Humidity Index
(BGHTI) (Buffington et al., 1981), for each treatment, as well as the physiological responses
such as breathing frequency (BF) and rectal temperature (RT) of the goats.
Results and Discussions
The general mean, did not have a significant difference among the values of the thermal
environment indexes for the bedding treatments: rice straw, wood shaving and grass (p>0,05),
which was expected once the environments interacted, since the stalls were open. The
maximum values of the Black Globe Temperature and Humidity Index (BGHTI) occurred from
12:00pm to 2pm, being, on average, around 73 for all of the treatments, in the interior of the
facility and 81 in the exterior. According to the National Weather Service (1976), values of this
index up to 74 represents a safe environment for animal production. This indicates that well
ventilated and open facilities are better suited to raise animals, in deep bedding systems, in
regions of hot climate.
The treatments had no influence (p>0,05) in Breathing Frequency (BF) nor in the Rectal
Temperature (RT) of the goats, because there was no difference between the treatments for
these parameters. The largest value of RT observed in the afternoon was still within the
normality for goats. According to Tavares (1989), these animals keep the rectal temperature in
a daily average of 39ºC, with a variation from 37.5 to 40.5ºC. The BF values was on average
36 movements/minute in the morning and 52 movements/minute in the afternoon , which were
very different than those mentioned by BRASIL et al. (2000), where they found a variation in
BF of 80 movements/minute in the morning , and 174 movements/minute in the afternoon in
animals under thermal stress. This indicates that the animals in the study were not under
thermal stress.
Conclusion
Considering the way this investigation was conducted and the results obtained, it can
be concluded that the thermal comfort of the goats, in the three treatments: rice straw, wood
shavings and grass bedding, evaluated with basis on the BGHTI values was considered
satisfactory, that is, the materials had not intervened with the thermal comfort of the animals.
The resulting physiological responses (breathing frequency and rectal temperature) proved the
viability of the use of deep bedding systems for goat breeding in hot climate conditions.
References
BRASIL, L.H.A; WECHESLER, S.W.; BACCARI, W.J.; GONÇALVES, H.C.; BONASSI,
I.A. 2000. Efeitos do Estresse Térmico Sobre a Produção, Composição Química do Leite e
Respostas Termorreguladoras de Cabras da Raça Alpina. Revista Brasileira de Zootecnia. 29,
(6).
BUFFINGTON, C. S., COLLAZO-AROCHO, A., CANTON, G. H., PITT, D., THATCHER,
W.W., COLLIER, R. J. 1981. Black globe humidity index (BGHI) as comfort equation for
dairy cows. Transaction of the ASAE, 24 (3): 711-714.
LU, C. D. 1986. Heat stress and goat production. In: SIMPÓSIO INTERNACIONAL DE
BIOCLIMATOLOGIA ANIMAL, 1, Fortaleza, CNPC – EMBRAPA,
72
NATIONAL WEATHER SERVICE - Central Region. Livestock hot weather stress. Letter C -
31-76, 1976.
TAVARES, S. L. S., Reações fisiológicas e produção de cabras leiteiras, sob quatro

temperaturas aparentes, em câmaras climáticas. Dissertação (Mestrado em Engenharia
Agrícola) – Universidade Federal de Viçosa, Viçosa – MG, 60p 1989.
73
Dairy cows thermal comfort evaluation in hot climates using

temperature humidity index
Maurício Perissinotto 1); Vasco Fitas da Cruz 2); Daniella Jorge de Moura3)
1)
Universidade de São Paulo/Escola Superior de Agricultura “Luiz de Queiroz”, Dep de Engenharia Rural. Av.
Pádua Dias, 11 CP. 09 CEP: 13418-900, Piracicaba, São Paulo e ICAM – Instituto de Ciências Agrárias
Mediterrânicas, mperissi@esalq.usp.br
2)
Universidade de Évora, Departamento de Engenharia Rural e ICAM – Instituto de Ciências Agrárias
Mediterrânicas, Apartado 94, 7002-554, Évora, Portugal, vfc@uevora.pt
3)
Universidade Estadual de Campinas, Faculdade de Engenharia Agrícola. Cidade Universitária “Zefferino Vaz”,
Distrito de Barão Geraldo, CP: 6011, Cep: 13083970, Campinas, São Paulo,
daniella.moura@agr.unicamp.br
Introduction
Thermal environment influences animal performance affecting heat transfer mechanisms and
thermal balance regulation.
In order to characterize or quantify thermal comfort zones according with animal species
thermal comfort indexes were developed. Those indexes joint in one variable all the elements
that characterized the thermal environmental associated to the animal. One of the most used
index is the THI (Temperature Humidity Index) developed by Thom (1959). Animal inside one
adequate range of THI produce according his genetic potential. The limits of THI classes
(ranges) vary with the zone and animal type, so in some cases it is necessary to make
adaptations in the index.
Animal adaptation to hot climates is also evaluated by physiological parameters like rectal
temperature and respiratory rate. Using data mining from data of animal physiological
parameters and performance and from data of weather stations there is the possibility of
evaluate the thermal stress impact on animal and to modelling it.
The aim of this paper is to determine the critical intervals of THI in dairy cows breeding in
Alentejo, south of Portugal through data mining of meteorological and zootechnical data.

The study was carried out at Mitra dairy farm (Èvora University) located in Alentejo, Southeast
of Portugal. Climatic data were collected from an automatic weather station. With those data
THI index was calculated according with Thom (1959) equation:
THI = Tbs + 0,36 Tpo + 41,5, where Tbs is the dry bulb temperature (°C) and Tpo is the dew
point temperature (°C).
Rectal temperature and respiratory rate of 6 middle lactation Holstein Frisian cows with
average live weight of 650 Kg were collected during 20 days of August and September of
74
2005. The housing systems was the free stall system and animal were ad libidum fed. They
were milking two times per day.
The computational program Weka® made data mining. 3 classes of rectal temperature (lower
than 38,8ºC, between 38,8 and 39,2ºC, and higher than 39,2ºC) and of respiratory rate (lower
than 56, between 56 and 64, and higher tahn 64 mov.min-1). Thermal comfort was classified
as high, medium and low.
Results
Results did not show significant differences in critical THI values according with milk
production level. Good conditions of thermal comfort were achieved when THI register lower
values than 77. For THI values higher than 80 dairy cows’ thermal conditions were very poor.
Data analysis provide one model of decision tree with a total precision of 0,94.
Figure 1 shown the critical values of higher comfort (without stress), medium comfort
(moderate stress ) e lower comfort (high stress) founded in function of physiological
parameters (rectal temperature and respiratory rate).
Figure 1. Decision tree for THI
According with critical values obtained by the figure 1 THI classes for this kind of animals in
Alentejo zone, were established in function of air temperature and relative (Figure 2).
Dry bulb temperature (ºC)
HR (%) 26 27 28 29 30 31 32 33 34 35 36
20 68,0 69,3 70,6 71,9 73,2 74,5 75,7 77,0 78,3 79,6 80,9
30 70,1 71,4 72,7 74,0 75,3 76,6 77,9 79,2 80,5 81,8 83,1
40 71,6 72,9 74,2 75,6 76,9 78,2 79,5 80,8 82,2 83,5 84,8
50 72,8 74,1 75,5 76,8 78,1 79,5 80,8 82,1 83,5 84,8 86,1
60 73,8 75,2 76,5 77,9 79,2 80,5 81,9 83,2 84,5 85,9 87,2
70 74,7 76,1 77,4 78,8 80,1 81,5 82,8 84,1 85,5 86,8 88,2
80 75,5 76,9 78,2 79,6 80,9 82,3 83,6 85,0 High Comfort
90 76,2 77,6 78,9 80,3 81,6 83,0 84,4 85,7 Average Comfort
100 76,9 78,2 79,6 80,9 82,3 83,7 85,0 86,4 Low Comfort
Figure 2. Classes of Temperature and Humidity Index
Figure 2 shows the importance of relative humidity in thermal comfort. When relative humidity
increases lower are the values of temperature which provide thermal comfort for the animal.
75
Pennington et al. (2005), refer that moderate thermal stress signs appear when temperature
range 27 e 32ºC only with relative humidity is higher than 50%. When temperature range is
between 32 e 38ºC and relative humidity is higher than 50% cows show evident signs of
thermal stress.
Conclusions
Rectal temperature and respiratory rate of housing dairy cows were related with THI.
The results of this experiment shown that is possible to adopt THI values in each
region and according with animal type to evaluate animal thermal comfort. However the
validation of this methodology in large size farms it is necessary.
References
BACCARI JUNIOR, F. Adaptação de sistemas de manejo na produção de leite em clima quente. In: Simpósio
brasileiro de ambiência na produção de leite, Piracicaba, 1998. Anais. Piracicaba: FEALQ, 1998. p.24 – 67.
BERMAN, A.; FOLMAN, Y. M.; KAIM, M.; et al. Upper critical temperature and forced ventilation effects of high
yielding dairy cows in a tropical climate. Journal of Dairy Science, v.67, p. 488-495, 1985.
DU PREEZ, J.D.; GIESECKE, W.H.; HATTINGH, P.J.; EISENBERG, B.E. Heat stress in dairy cattle and other
livestock under Southern African conditions. II Identification of areas of potential heat stress during summer
by means of observes true and predicted temperature-humidity index values. Onderstepoort Journal Vet.
Res., v 57, p. 183-187, 1990.
FUQUAY, J. W. Heat stress and it affects animal production. Livestock Environment, v.2, p.1133-1137, 1997.
HAHN, G. L. Management and housing of farm animals in hot environment. In: In:YOUSEF, M. K. Stress
physiology in livestock. v.2, 1985. p.151-174.
HUBER, J. T. Alimentação de vacas de alta produção sob condições de estresse térmico. In: Bovinocultura leiteira.
Piracicaba: FEALQ. 1990. p.33-48.
IGONO, M. O.; JOHNSON, H. D. Physiologic stress index of lactating dairy cows based on diurnal pattern of rectal
temperature. Journal of Interdisciplinary Cycle Research., v. 21, p.303-320, 1992.
MARTELLO, L. S. Diferentes recursos de climatização e sua influência na produção de leite, na termorregulação
dos animais e no investimento das instalações. Pirassununga, 2002. Dissertação (Mestrado) - Faculdade de
Zootecnia e Engenharia de Alimentos, Universidade de São Paulo.
PENNINGTON, J. A.; VANDEVENDER, K. Heat Stress in Dairy Cattle. Agriculture and Natural Resources.
University of Arkansas, United States Department of Agriculture, and County Governments Cooperating.
http://www.uaex.edu. Acesso em 18/11/2005.
REZENDE, S. O.; PUGLIESI, J. B.; MELANDA, E. A.; DE PAULA, M. F. 2005. Mineração de Dados. Sistemas
Inteligentes: fundamentos e aplicações. São Paulo. Ed. Monole, pp.307-336.
THOM, E.C. 1959. Cooling degrees - days air conditioning, heating, and ventilating. Transactions of the ASAE,
55:7:65-72.
76
A VIRTUAL ANIMAL FOR CLIMATE CONTROL DESIGN:

STATIC AND DYNAMIC SIMULATIONS OF HEAT LOSSES
J.-M. Aerts, D. Berckmans
Division M3-BIORES, Department of Biosystems, Catholic University Leuven,

Kasteelpark Arenberg 30, 3001 Leuven, Belgium;
Email: Daniel.Berckmans@biw.kuleuven.be
ABSTRACT. The simulation of the effects of different control strategies on the interior climate in animal houses
has been a frequent focus of research work for the last 30 years. Only in a few casesrealistic computer
simulations of animal occupants have been used in the development of new climate control strategies. The aim of
this research was to develop a virtual chicken (VirChick) for computer-aided design and engineering of climate
controllers for poultry houses. The objective was to develop a dynamic simulation model of the energy and mass
transfer between the chicken and its thermal environment as a first step in the making of a VirChick. A static as
well as dynamic model of heat loss for VirChick was developed. Two experiments were performed to generate
data for the evaluation of the static heat loss model. It was demonstrated that the modeling results for the heat
loss components were in agreement with data found in the literature. Furthermore, on the basis of three dynamic
experiments, it was demonstrated that dynamic responses of total heat loss to step variations in temperature
(ranging from 18°C to 35°C) could be modeled with a correlation coefficient between measured and simulated
total heat loss of 0.83 to 0.96. In the future, such a virtual chicken can be equipped with many more properties,
such as realistic locomotion, thermoregulatory and other behaviors, artificial intelligence, etc.
Keywords. Chicken, Climate, Design, Modeling, Virtual.
The simulation of the effects of different control strategies on the environmental variables in
animal houses has been the aim of a lot of research work in the last 30 years (Bjerg et al.,
2000). Computational fluid dynamics (CFD) is one of the most popular modeling tools used to
simulate fluid behavior for this purpose (e.g., Worley and Manbeck, 1995; Predicala and
Maghirang, 2003). The use of CFD in relation to livestock houses, so far, has been mainly
concentrated on the development of CFD simulations that have been compared with
measurements in full-scale or scale-model test rooms (e.g., Harral and Boon, 1997). In most of
these applications, the climate simulations are calculated without animals in the room (Harral
and Boon, 1997; Sun et al., 2002) or with static physical animal models (Zhang et al., 1999;
Bjerg et al., 2000). However, in order to make realistic simulations of climatic variables inside
animal houses, the dynamic responses of the animals should also be taken into account because
the complex geometry of the animals and the energy and mass transfer between the animal and
the environment influence the resulting climate (Bjerg et al., 2000). Only in a few studies were
more realistic simulations of animal or human occupants used (e.g., cows: Wu and
Gebremedhin, 2001; humans: Murakami et al., 2000). To the authors' knowledge, for broiler
chickens no such advanced simulation models exist that can be used in computer-aided design
and engineering (CAD/CAE) applications.
Figure 1 presents the different parts needed to develop a virtual chicken (VirChick) that can be
used in CAD/CAE applications for designing climate controllers for poultry houses. In this
research, the aim was to develop the first part of VirChick. More specifically, the objective was
to develop a dynamic simulation model of the energy and mass transfer between the animal
and its thermal environment.
77
VirChick
Energy and 3D geometry Artificial

mass transfer and locomotion intelligence
Figure 1. Schematic presentation of the different parts of VirChick.

Modelling the Energy and Mass Transfer
The processes of heat exchange between an animal and its surroundings can be described by
appropriate transfer equations. To model the energy and mass transfer between VirChick and
its environment, the following assumptions were made: the deep body temperature of VirChick
was a constant 41°C (Wathes and Clark, 1981b); the sensible heat was mainly lost by radiation
and convection; heat transfer by conduction could be neglected (Bouchillon et al., 1970;
Wathes and Clark, 1981b; McArthur, 1987); heat loss equaled heat production at every
moment; VirChick was living indoors and did not receive solar radiation; and the convective
heat transfer occurred in a mixed regime (a fair assumption in practice, Wathes and Clark,
1981a).
First, a static mechanistic model was developed based on existing knowledge describing the
responses of heat losses due to convection, radiation, and evaporation to the environmental
variables air temperature and air velocity. The sensible heat loss by convection (C) was
calculated as (McArthur, 1987):
ρc p
C= (Tc − Ta ) (1)
raH
where C is the convective heat loss (W m-2), ρcp is the volumetric specific heat of air (J m-3 K-
1
), raH is the thermal resistance of the boundary layer (s m-1), Ta is the air temperature (°C), and
Tc is the surface temperature of the feathers (°C).
The thermal resistance of the boundary layer (raH) can be calculated as (McArthur, 1987):
ρc p d t
raH = (2)
k a Nu
where ρcp is the volumetric specific heat of air (J m-3 K-1), dt is the characteristic dimension of
the chicken (m), ka is the thermal conductivity of air (W m-1 K-1), and Nu is the Nusselt
number.
According to Mitchell (1930), the characteristic dimension (dt) for a chicken can be determined
by the following equation:
d t = 0.131 ⋅ W 0.33 (3)
where W is the body mass of the chicken (kg).
The Nusselt number (Nu) can be calculated based on the Reynolds number assuming a
convective mixed regime as (Wathes and Clark, 1981a):
Nu = 2 + 0.79. Re 0.48 (4)
where Re is the Reynolds number, which can be calculated as:
vd t ρ
Re = (5)
µ
where v is the air velocity (m s-1), dt is the characteristic dimension of the bird (m), ρ is the air
density (kg m-3), and µ is the air viscosity (kg m-1 s-1).
78
The radiant heat loss, Ln (W m-2), was calculated as (McArthur, 1987):

ρc p
Ln = (Tc − Te ) (6)
rR
where ρcp is the volumetric specific heat of air (J m-3 K-1), rR is the radiative resistance (s m-1),
Te is the radiant temperature of the environment (°C), and Tc is the surface temperature of the
feathers (°C). The surface was assumed to be a black body (McArthur, 1987).
The radiative resistance of the boundary layer (rR) was calculated as follows (McArthur, 1987):
ρc p
rR = (7)
4σTce3
where ρcp is the volumetric specific heat of air (J m-3 K-1), σ is the Stefan-Boltzmann constant
(W m-2 K-4), and Tce is the average temperature of Tc and Te (K).
The evaporative heat loss, Ge (W m-2), of broilers can be calculated as a function of air
temperature (Ta, for Ta ranging between 20°C and 28°C and assuming a relative humidity of
60%) by the equation of Chwalibog et al. (1985):
 1 
(
Ge =   (0.0116) − 202 + 471.W 0.64 + 6.Ta ) (8)
 As 
where W is the bird body mass (kg), Ta is the air temperature (°C), and As is the skin surface
area (m2). The equation describing the skin surface area (As) for chickens as a function of body
mass was derived by Mitchell (1930) based on the general Meeh equation:
As = 0.000819 ⋅ W 0.705 (9)
2
where As is the skin surface (m ), and W is the chicken body mass (g). According to Schmidt-
Nielsen (1984), errors in the coefficients of the Meeh equation are in the range of 10% to 20%.
Therefore, we assumed that the coefficients of the equation, although it originates from 1930
and broilers have evolved a lot since then, can still be used today.
The total heat loss, Gt (W m-2), is the sum of these partial heat losses and is calculated as:
A 
Gt =  c  (C + Ln ) + Ge (10)
 As 
where Ac is the surface area of the feathers (m2) and can be calculated by the following
equation (Walsberg and King, 1978):
Ac = 0.081.W 0.667 (11)
where W is the body mass of the chicken (kg). Because the quantities C and Ln are expressed
per unit area of coat surface, the ratio (Ac/As) allows Gt to be expressed per unit area of skin
(Turnpenny et al., 2000).
Secondly, the static mechanistic model was extended to a dynamic model by introducing the
time constants of the dynamic response of total heat loss of broiler chickens. These time
constants were determined in previous work (Aerts et al., 2000). In order to do so, the
following assumptions were made: the changes in temperature were in discrete time and
regarded as a sequence of step changes; the response of Gt to step changes in air temperature
was described by first-order dynamics (Aerts et al., 2000); the time constant of the step
response was not a function of age; and the time constant was a function of the direction of the
change (on-transient versus off-transient) (Aerts et al., 2000):
 ∆k
 −   Ac
Gt (k ) = Gt (k − 1) + 1 − e τ    A (C (k ) + Ln (k ) )
 
  s (12)
+ Ge (k )  − G (k − 1) 
 t 
 
where Gt(k) is the total heat loss at time instant k (W m-2), ∆k is the considered time interval, τ
is the time constant of the dynamic response (min); C(k) is the convective heat loss at time
79
instant k (W m-2), Ln(k) is the radiant heat loss at time instant k (W m-2), and Ge(k) is the
evaporative heat loss at time instant k (W m-2).
The time constant for total heat loss (Gt) was assumed to be 8.6 min for the on-transient and
12.8 min for the off-transient (Aerts et al., 2000).
Experimental Data
Growth Experiments under Semi-Practical Conditions
Two growth experiments (Exp1 and Exp2) were performed with groups of 2900 broilers (Ross
308, mixed sex) during a production period of 42 days in a broiler compartment measuring 12
× 16 m. The compartment was equipped with four axial fans (two in each side wall, 45 cm
diameter, Fancom Exavent) pulling the air through the ridge in the compartment. The
compartment was heated by a central heating system with hot water pipes under the ridge. The
testing facility was situated in a broiler house of the Agricultural Research Centre - Ghent in
Belgium.
Air temperature (Ta) was measured in the ridge and at two positions near the side walls (about
1 m above the floor) by means of thermocouples (0.5°C accuracy). For the simulations with
VirChick, the mean air temperature of the three sensors was used. Mean air temperature was
set to 30°C during the first three days. From day 3, room temperature was set at 28°C and then
decreased 1° every three days until a constant temperature of 21°C was reached. Air humidity
was not measured. A conventional lighting schedule of 23 h of light and 1 h of darkness was
used.
The radiant temperature (Te) of the building envelope was determined based on images of an
infrared camera (AGEMA Thermovision 570, 0.1°C nominal sensitivity, 7.5 to 13 µm infrared
spectrum, 24° lens, thermal emissivity set to 0.95). The radiant temperature used in the
calculations was the weighed average of the measured average temperatures of the walls, the
roof, and the litter (weighing coefficients for the litter, side walls, back wall, and roof were
0.41, 0.13, 0.43, and 0.03, respectively). The temperatures of the different surfaces were taken
from the middle of the house by directing the camera to the middle of each surface. The
temperature of the feathers (Tc) was measured by means of the same infrared camera. More
specifically, the feather temperature was calculated as the average of the back side
temperatures of five randomly chosen freely moving broilers. The back side temperatures were
measured with the camera at a distance of 1.5 m.
Air velocity (v) was measured with a unidirectional air velocity sensor (TSI 8455, accuracy =
2% of reading). Air velocity was measured at 0.3 m above the litter, and the sensor was
positioned to measure the airflow along the longitudinal axis of the house. Every measuring
day, measurements were made at five different positions in the first half of the house. For the
calculations of VirChick, the average value of the five measured air velocities was used.
The compartment was equipped with four scales (Fancom B.V., 0.34 m diameter, 1 g accuracy)
connected to a computer (Fancom F747, version A1) for measuring the average body mass (W)
of the flock every 24 h. Feed consumption was recorded daily.
During the growing period, measurements of Tc, Te, and v were performed on day 3, 8, 16, 23,
30, 37, and 42 for Exp1 and on day 3, 10, 15, 22, 30, 37, and 42 for Exp2. The values of air
temperature (Ta) and average body mass (W) of the chickens were registered daily from the
process controllers.
Dynamic Experiments under Laboratory Conditions
In addition, data of three experiments (Exp3, Exp4, and Exp5) of dynamic heat loss responses
to step variations in air temperature (measurement interval of 4.5 min) were used. The
experiments were carried out in a respiratory chamber with inside dimensions of 0.30 × 0.55 ×
80
0.50 m that was placed in a temperature-controlled cell (10°C to 40°C). The chamber was
made of stainless steel and was only partly insulated so it could easily exchange heat with the
cell. Humidity was not controlled. The temperature in the respiration chamber was measured
with a platinum resistance temperature detector (Pt-100, 0.2°C accuracy), placed at the outlet.
Light was provided by a fluorescent lamp (60 W). Light intensity in the chamber was 35 lux
(light on). Total heat loss was calculated based on O2 consumption and CO2 production.
Samples were taken with a difference-measuring CO2 analyzer and a difference-measuring O2
analyzer (difference between animal chamber and fresh air).
Depending on the age of the birds, four to eight broilers were placed in the chambers.
Measurement interval (i.e., time between subsequent samples) was 4.5 min. Full calibration of
the respiration unit was performed with ethanol burning tests. The chamber recoveries
averaged 94% ±2% (SE). The O2 and CO2 analyzers were calibrated before each experiment.
An error analysis of the measurement system indicated a relative error of 5.6%. In each
experiment, a step up and a step down in air temperature was generated. Every step up was
followed, after reaching steady state, by a step down of the same magnitude. After a step
change in air temperature, the new steady-state level of air temperature was kept constant for at
least 2 h in order to let the animals reach a new steady-state level of heat loss. The magnitude
of step changes in air temperature was fixed at 17°C. Light intensity was kept constant at 35
lux. The animals were 6 days old in Exp3, 14 days old in Exp4, and 21 days old in Exp5. For
more details about the experiments and the measurement of the heat loss by means of indirect
calorimetry, refer to Aerts et al. (2000).
Results and Discussion
Growth Experiments under Semi-Practical Conditions
Figures 2 and 3 show results of the simulations of the static model of VirChick with input data
from the 6-week growing experiments (Exp1 and Exp2, respectively). In figures 2a and 3a, the
measured air temperature (Ta), radiant temperature of the environment (Te), and coat
temperature (Tc) are shown. The coat temperature of birds in clusters was on average 1°C to
2°C higher than the coat temperature of freely moving birds. This observation was in
agreement with the measurements of Wathes and Clark (1981c).
81
Figure 2. Measured and simulated data for the growth experiment Exp1: (a) measured air temperature Ta, radiant temperature of the
environment Te, and coat temperature Tc; (b) measured air velocity v and calculated Nusselt number Nu; (c) simulated convective heat
loss C, radiant heat loss Ln, and total sensible heat loss Gc; (d) simulated sensible heat loss Gc, evaporative heat loss Ge, and total heat
loss
Gt.
Figure 3. Measured and simulated data for the growth experiment Exp2: (a) measured air temperature Ta, radiant temperature of the
environment Te, and coat temperature Tc; (b) measured air velocity v and calculated Nusselt number Nu; (c) simulated convective heat
loss C, radiant heat loss Ln, and total sensible heat loss Gc; (d) simulated sensible heat loss Gc, evaporative heat loss Ge, and total heat
loss Gt.
82
The measured air velocity and calculated Nusselt number are shown in figures 2b and 3b. The
measured temperatures and air velocities in Exp1 are very similar to the field measurements in
a broiler house described by Wathes and Clark (1981c). These authors found Nusselt numbers
varying between 15 to 30, compared to between 15 to 35 in this research. In Exp2, the
measured air velocities and calculated Nusselt numbers were higher compared to Exp1 (up to
0.5 m s-1 and 50 for air velocity and Nusselt number, respectively). This could be explained by
the fact that during the last two weeks of Exp2, the outside air temperature was high (30°C),
resulting in an increased ventilation rate and air velocity in the house (see fig. 3b). As
mentioned before, each air velocity value is the mean of the air velocities measured at five
different positions within the house. For Exp1, the difference between the minimum and
maximum spatially measured air velocity ranged from 0.05 to 0.1 m s-1 depending on the
measurement day. For Exp2, the difference between the minimum and maximum spatially
measured air velocity ranged from 0.05 to 0.2 m s-1. The spatial variations were especially
large during the last two weeks, when outside temperature was high.
Figure 2c shows the course of the convective (C), radiant (Ln), and total sensible heat losses
(Gc) during the growing process. The amount of the total sensible heat losses was characterized
by a sharp increase from day 1 up to day 8, a plateau from day 9 to day 21, and finally a
decrease between day 22 and day 42. The decrease after day 22 was less pronounced in Exp2
than in Exp1. This could be explained by the fact that, due to the higher inside air temperatures
during the last two weeks, the gradient between the coat temperature (Tc) and the
environmental air temperature (Ta) and radiant temperature (Te) was higher, as can be seen in
figure 3a. Similar patterns in total sensible heat losses were found by Wathes and Clark
(1981c) for Ross1 broilers. The absolute values in this work were higher compared with their
results. This can probably be explained by the fact that the measured feather temperature in this
research was higher than in Wathes and Clark (1981c). In addition, the proportion of the
convective and radiant heat losses to the total sensible heat losses were higher compared with
the results of Wathes and Clark (1981c) and Mitchell (1985) (70% vs. 50%). This is probably
due to the fact that the radiant heat losses are underestimated. In the growth experiments, it was
assumed that the birds stayed in clusters 66% of the time, resulting in a reduction of radiant
heat losses due to the surrounding birds with the same coat temperature. This assumption was
made based on the work of Wathes and Clark (1981c), who quantified the clustering behavior
of broilers during the production process under similar conditions of air temperature and air
velocity. Compared with isolated individual animals, the calculated radiant heat loss was only
33% of the total sensible heat loss.
Graphs of the simulated total heat losses (Gt), the sensible heat losses (Gc), and the evaporative
heat losses (Ge) are shown in figure 2d and figure 3d. In Exp1 the proportions of Gc and Ge to
Gt were 60% to 40% for Gc and 40% to 60% for Ge. In Exp2, the proportions for Gc ranged
between 60% and 55% and for Ge between 40% and 45%. This is in agreement with numerical
values found in the literature. Based on field tests, Reece and Lott (1982) determined that the
proportion of Gc and Ge to Gt for broilers was 60% and 40%, respectively. The data of Curtis
(1983) indicated that the proportion of Gc was 60% to 30% for laying hens when air
temperature varied from 25°C to 35°C. Chwalibog et al. (1985) measured proportions for Gc
from 30% to 60% for growing broilers from 1 to 37 days old. Finally, the results of Xin et al.,
(1996) indicated that the proportion of Gc to Gt was 30% to 50% for broilers of 6.5 weeks old.
Dynamic Experiments under Laboratory Conditions
The following variables were used as input in the dynamic model of VirChick: air temperature
(Ta), air velocity (v), and average bird body mass (W). Radiant temperature of the environment
(Te) and coat temperature (Tc) were not measured. It was assumed that the radiant temperature
83
of the environment (Te) equaled the air temperature (Ta). The coat temperature (Tc, °C) was
calculated as a function of Ta by the following equation:
Tc = Ta + 0.7(41 − Ta ) (13)
This relationship was theoretically derived based on the equations of sensible heat transfer in
the bird and from the bird to the environment. The sensible heat transfer (Gc) from the deep
body of the bird to the coat can be expressed by the following equation (Wathes and Clark,
1981c):
ρc p
Gc = (Tb − Tc ) (14)
rb + rc
where Gc is the sensible heat loss (W m-2), ρcp is the volumetric specific heat of air (J m-3 K-1),
Tb is the chicken body temperature (°C), Tc is the chicken coat temperature (°C), rb is the
thermal resistance of chicken tissue (s m-1), and rc is the thermal resistance of chicken feathers
(s m-1).
When it is assumed that Te equals Ta, the sensible heat transfer (Gc) from the coat to the
environment can also be written as (McArthur, 1987):
 ρc p ρc p 
Gc =  +  (Tc − Ta )
 (15)
 raH rR 
where Gc is the sensible heat loss (W m-2), ρcp is the volumetric specific heat of air (J m-3 K-1),
Tc is the chicken coat temperature (°C), Ta is the ambient air temperature (°C), raH is the
thermal resistance of the boundary layer (s m-1, see eq. 2), and rR is the radiative resistance of
chicken feathers (s m-1, see eq. 7).
By combining equations 14 and 15, the following relation between Tc and Ta can be derived:
raH rR
Tc = Ta + (Tb − Ta ) (16)
raH rR + (raH + rR )(rb + rc )
The values for raH and rR were estimated by applying equations 2 and 7 to the measured data of
Ta, Te, Tc, and v of the growth experiments. The value for rb+rc was set to 150 s m-1, based on
the estimations of Wathes and Clark (1981c) for individual birds. By using the estimated
thermal resistances in equation 16 and by assuming body temperature (Tb) to be 41°C, we got
equation 13.
The dynamic model was tested on three data sets for animals of different ages (6, 14, and 21
days) and for step up as well as step down responses. Figure 4 shows the simulation results for
one example. Figure 4a shows the measured total heat loss and air temperature. In figure 4b,
the measured total heat loss is compared with the simulated total heat loss. Figure 4c shows the
simulated total heat loss and the different simulated partial heat losses (convective heat loss,
radiant heat loss, total sensible heat loss, and evaporative heat loss). Finally, in figure 4d, the
proportions of the different partial heat losses to the total heat loss are shown. Table 1
summarizes the modeling accuracies for the different experiments in terms of the correlation
coefficient (r) between the measured and simulated total heat loss response.
84
Figure 4. An example of measured and simulated dynamic heat losses: (a) measured air temperature and total heat loss; (b) measured
versus simulated total heat loss; (c) simulated responses of total sensible heat loss, convective heat loss, radiant heat loss, and
evaporative heat loss to a step down in air temperature expressed in absolute values and (d) as a percentage of total heat loss.
Table 1. Modeling results of the dynamic simulation model expresses as correlation coefficient (r) between the measured and simulated
total heat losses.
Correlation Coefficient (r)

For Step Up For Step Down
Experiment Response Response
Exp3 0.96 0.85
Exp4 0.89 0.85
Exp5 0.88 0.83
Based on the graphical results as well as the results in table 1, it is seen that the model for
energy and mass transfer between VirChick and the thermal environment succeeds in
describing the static as well as the dynamic response of total heat loss (r varying between 0.83
and 0.96). Furthermore, the courses of the partial heat loss responses were realistic. In all the
simulated heat loss responses, an increase in air temperature (Ta) resulted in a decrease in total
heat loss (Gt), a decrease in total sensible heat losses (Gc), and an increase in evaporative heat
loss (Ge). Alternatively, a decrease in air temperature resulted in an increase in the simulated
total heat loss and total sensible heat losses and a decrease in the evaporative heat loss. The
opposite proportional relation between Gt on one hand and Ta and Te on the other hand is
confirmed by, among others, Chwalibog and Eggum (1989). The opposite proportional relation
between Gc on one hand and Ta and Te on the other hand is confirmed by Richards (1976) and
Curtis (1983).
85
The results of Wathes and Clark (1981c) showed that for broilers of 10 days old (in a cluster),
with an air temperature of 25°C and a radiant temperature of the environment of 24°C, the total
sensible heat loss was between 40 and 50 W/m2. For similar conditions of Ta and Te and age of
birds (6 days old), the simulated total sensible heat loss of VirChick was nearly 60 W/m2. An
increase in Ta caused a decrease in the proportion of convective heat losses (C) to Gc due to the
diminishing difference between Ta and Tc (Chwalibog and Eggum, 1989). This has also been
observed in the simulations of VirChick. The calculations of Wathes and Clark (1989c)
showed that the convective heat losses (C) of broilers of 8 days old (Ta and Te of 24°C) were
50% of Gc. For similar conditions of Ta and Te and age of birds (6 days old), the proportion of
the simulated convective heat loss of VirChick was also 50%.
The simulated evaporative heat losses (Ge) of VirChick were 33% to 35% of Gt for Ta varying
between 20°C and 25°C. These values are in accordance with the results of Richards (1976),
who demonstrated that Ge was 30% of Gt when Ta was 26°C.
In the case of high values of Ta and Te (35°C to 38°C), the proportion of the simulated
evaporative heat losses increased strongly. In Exp4 and Exp5, the proportion of Ge in Gt
increased to 80%. However, in Exp3, the proportion of Ge in Gt increased up to 95%. This can
be explained mainly by the fact that the air temperature during the first half of the experiment
(40°C) was nearly as high as the body temperature of the birds (fig. 4a), so the birds had to lose
almost all their heat by evaporation.
Other authors have modeled homeostatic responses of broiler chickens to their thermal
environment. Bouchillon et al. (1970) modeled the effect of air temperature and air humidity
on the evaporative and sensible heat losses of chickens. The concept of the model was
described, and the steady-state simulations were evaluated qualitatively, but not quantitatively.
Kettlewell and Moran (1992) made a model describing heat losses of broilers as a function of
the thermal environment with the aim at designing ventilation systems for broiler transporters.
Although some of the model parameters were not well defined, the steady-state simulations
were in good agreement with data from the literature. In contrast with the models described in
this work, the models of these authors did not describe dynamic responses of heat losses to
variations in the thermal environment.
In the future, additional efforts should be made to further improve the simulation accuracy of
VirChick. A first improvement could be made by taking into account the (relative or absolute)
air humidity as a variable in the calculations of the evaporative heat losses (Ge). The equation
that was used in this work (eq. 8) was derived for air temperature ranging between 20°C and
28°C and for a relative humidity of 60%. In addition, using more recent equations describing
the relationships between chicken body mass on the one hand and characteristic dimension and
chicken skin surface on the other hand might improve the simulation results, since the
relationships described by Mitchell (1930) are likely to be outdated after 75 years.
Conclusion
This research was a first attempt in the development of a virtual chicken (VirChick) that can be
used for CAD/CAE applications for designing climate controllers. It was demonstrated that the
static as well as dynamic energy and mass transfer between VirChick and the thermal
environment is modeled in an accurate and realistic way. The development of virtual organisms
offers opportunities for design purposes in general and for the design of climate control
strategies in particular. In the future, such a virtual chicken can be equipped with many more
properties, such as a realistic three-dimensional shape, locomotion, thermoregulatory system,
feeding and social behaviors, learning abilities (artificial intelligence), etc. In this way, the
development of virtual chickens becomes a very challenging interdisciplinary research work.
86
An additional advantage of developing virtual organisms is that they reduce the need for using
live birds in research trials.
Acknowledgements
This research was supported by the Katholieke Universiteit Leuven (Grant No. PDM/01/125).
The authors are very appreciative for this support.
References
Aerts, J.-M., D. Berckmans, P. Saevels, and E. Decuypere. 2000. Quantification of dynamic
and static responses of total heat production of broiler chickens to temperature and
light intensity. Trans. ASAE 43(6): 1835-1841.
Bjerg, B., K. Svidt, G. Zhang, and S. Morsing. 2000. The effects of pen partitions and thermal
pig simulators on airflow in a livestock test room. J. Agric. Eng. Res. 77(3): 317-326.
Bouchillon, C. W., F. N. Reece, and J. W. Deaton. 1970. Mathematical modeling of thermal
homeostasis in chicken. Trans. ASAE 13(5): 648-652.
Chwalibog, A., and B. O. Eggum. 1989. Effect of temperature on performance, heat
production, evaporative heat loss, and body composition in chickens. Arch. Geflug. 53:
179-184.
Chwalibog, A., J. Pedersen, and B. O. Eggum. 1985. Evaporative and sensible heat loss from
chickens kept at different temperatures. Arch. Geflug. 49: 50-54.
Curtis, S. E. 1983. Environmental Management in Animal Agriculture. Ames, Iowa: Iowa State
University Press.
Harral, B. B., and C. R. Boon. 1997. Comparison of predicted and measured airflow patterns
in a mechanically ventilated livestock building without animals. J. Agric. Eng. Res.
66(3): 221-228.
Kettlewell, P. J., and P. Moran. 1992. A study of heat production and heat loss in crated
broiler chicken: A mathematical model for a single bird. British Poultry Sci. 33: 239-
252.
McArthur, A. J. 1987. Thermal interaction between animal and microclimate: A comprehensive
model. J. Therm. Biol. 126: 203-238.
Mitchell, H. H. 1930. The surface area of single comb white leghorn chickens. J. Nutr. 2: 443-
449.
Mitchell, M. A. 1985. Effects of air velocity on convective and radiant heat transfer from
domestic fowls at environmental temperatures of 20° and 30°C. British Poultry. Sci. 26:
413-423.
Murakami, S., S. Kato, and J. Zeng. 2000. Combined simulation of airflow, radiation, and
moisture transport for heat release from a human body. Build. Environ. 35(6): 489-500.
Predicala, B. Z., and R. G. Maghirang. 2003. Numerical simulation of particulate matter
emissions from mechanically ventilated swine barns. Trans. ASAE 46(6): 1685-1694.
Reece, F. N., and B. D. Lott. 1982. The effect of environmental temperature on sensible and
latent heat production of broiler chickens. Poultry Sci. 61: 1590-1593.
Richards, S. A. 1976. Evaporative water loss in domestic fowls and its partition in relation to
ambient temperature. J. Agric. Sci. 87: 527-532.
Schmidt-Nielsen, K. 1984. Scaling: Why is Animal Size so Important? London, U.K.:
Cambridge University Press.
Sun, H., R. R. Stowell, H. M. Keener, and F. C. Michel, Jr. 2002. Two-dimensional
computational fluid dynamics (CFD) modeling of air velocity and ammonia distribution
in a High-Rise hog building. Trans. ASAE 45(5): 1559-1568.
Turnpenny, J. R., A. J. McArthur, J. A. Clark, and C. M. Wathes. 2000. Thermal balance of
livestock: 1. A parsimonious model. Agric. Forest Meteorol. 101(1): 15-27.
87
Walsberg, E. G., and King, J. R. 1978. The relationship of the external surface area of birds to
skin surface area and body mass. J. Exp. Biol. 76: 185-189.
Wathes, C. M., and J. A. Clark. 1981a. Sensible heat transfer from the fowl: Boundary-layer
resistance of a model fowl. British Poultry Sci. 22: 161-173.
Wathes, C. M., and J. A. Clark. 1981b. Sensible heat transfer from the fowl: Thermal
resistance of the pelt. British Poultry Sci. 22: 175-183.
Wathes, C. M., and J. A. Clark. 1981c. Sensible heat transfer from the fowl: Radiative and
convective heat losses from a flock of broiler chickens. British Poultry Sci. 22: 185-196.
Worley, M. S., and H. B. Manbeck. 1995. Modeling particle-transport and airflow in ceiling inlet
ventilation systems. Trans. ASAE 38(1): 231-239.
Wu, B., and K. G. Gebremedhin. 2001. CFD development and simulation of flow fields in
ventilated spaces with multiple occupants. Trans. ASAE 44(6): 1839-1850.
Xin, H., J. L. Sell, and D. U. Ahn. 1996. Effects of light and darkness on heat and moisture
production of broilers. Trans. ASAE 39(6): 2255-2258.
Zhang, G., K. Svidt, B. Bjerg, and S. Morsing. 1999. Buoyant flow generated by thermal
convection of a simulated pig. Trans. ASAE 42(4): 1113-1120.
NOMENCLATURE
Ac = chicken coat surface (m2)
As = chicken skin surface (m2)
C = convective heat loss (W m-2)
dt = characteristic dimension (m)
Gc = sensible heat loss through the coat (W m-2)
Ge = evaporative heat loss (W m-2)
Gt = total heat loss (W m-2)
ka = thermal conductivity of air (0.02492 W m-1 K-1 at 10°C; 0.02700 W m-1 K-1 at
37.8°C)
Ln = radiative heat loss (W m-2)
Nu = Nusselt number
Re = Reynolds number
raH = thermal resistance of the boundary layer (s m-1)
rb = thermal resistance of chicken tissue (s m-1)
rc = thermal resistance of chicken feathers (s m-1)
rR = radiative resistance (s m-1)
Ta = air temperature (°C)
Tb = deep body temperature (°C)
Tc = coat surface temperature (°C)
Tce = average temperature of Tc and Te (K)
Te = mean radiative temperature of the environment (°C)
Ts = chicken skin temperature (°C)
v = air velocity (m s-1)
W = chicken body mass (kg)
µ = air viscosity (0.000018 kg m-1 s-1 at 101.325 kPa and 20°C)
ρcp = volumetric specific heat of air (1220 J m-3 K-1 at 20°C and standard pressure)
ρ = air density (1.15 kg m-3 at 101.325 kPa and 20°C)
σ = Stefan-Boltzmann constant (5.67 10-8 W m-2 K-4)
τ = time constant (min)
88
An Approach to Evaluate the Suitability of an Evaporative Pad Cooling

System in Animal Housing to the Summer Climate in China
Baoming Li*, Chaoyuan Wang, Wei Cao and Zhengxiang Shi
Department of Agricultural Structure and Bio-environmental Engineering, China Agricultural
University, Beijing 100083, China.
*
Corresponding author: Prof. Baoming Li; phone: +86 10 62736904; fax: +86 10 62737570;
e-mail: libm@cau.edu.cn.
Abstract
The cooling effect of an evaporative pad cooling system in animal housing is very
significant in those regions characterized by hot and dry climates in summer, but it will be
dramatically decreased under the humid climate conditions. The climate in China varies
considerably from the North to the South. Thus, a theoretical analysis of the suitability of the
evaporative pad cooling system to different climate regions is very helpful in guiding the
environmental design of animal housing in China. In this paper, the suitability of an
evaporative pad cooling system in animal housing to the summer climate is evaluated by a
fuzzy mathematic method through analyzing 20-year’s weather data of nine representative
cities with varied climate features in China. The results show that the inside air temperature of
the animal housing located in the Yellow River basin and to its northern area can be basically
lowered below 30 oC in summer, and it can be dropped to 32 oC in Yangtze River valley as well
as the south coastal area by applying the evaporative fan-pad cooling system. The result
indicates that the system can meet the cooling requirement of the farm buildings in those
regions and provides a scientific basis for the extension and application of the evaporative pad
cooling system in China.
Keywords: evaporative pad cooling system, animal housing, cooling, climate
1. Introduction
During the summertime, the evaporative fan-pad cooling system is widely used in
greenhouse and animal housing for its high cooling efficiency in China (Jin et al., 1987; Li et
al., 1992; Liu and Chen, 2003; Liu et al., 2006; Zhou, 1988; Zhang et al., 2006). In the
facilities applied with the system, the outside air cooled down by the wetted pad takes the
excessive sensible heat when passing through the buildings and is exhausted via the fan, but
increases the relative humidity level at the same time. The system is very efficient in the
regions with hot and dry climates, but it will be dramatically decreased in humid conditions.
Meanwhile, the climates in China vary greatly from the north to the south. Thus, it is highly
needed to evaluate the suitability of the evaporative fan-pad cooling system to different climate
conditions in China theoretically in order to guide the environmental design of animal housing
and greenhouses.
In this paper, a fuzzy mathematic method was developed to evaluate the suitability of the
evaporative fan-pad cooling system to the different summer climates by analyzing 20-year’s
weather data of nine representative cities across the northern to southern China with varied
climate features.
2. Materials and methods
2.1 Cooling efficiency of the evaporative fan-pan cooling system
The cooling efficiencyη c of the evaporative pad cooling system is defined by the ASHRAE
as follows (ASHRAE, 1983):
89
Tdb ,o − Ti
ηc = (1)
Tdb,o − Twb ,o
where Tdb,o and Twb ,o are the dry and wet bulb temperatures outside the agricultural facilities
equipped with the evaporative pad cooling systems in oC, respectively; and Ti is dry bulb
temperature of the cooled air inside the facilities in oC. Equation (1) works well for the
evaporative fan-pad cooling system because the cooling process occurs nearly at a constant wet
bulb temperature of the outside air. In the hot and dry summer climates in China, the
evaporative pad-fan cooling system is very efficient, and the cooling efficiency of the system
could be as high as or above 80% when the air velocity through the wetted pad reaches 1.0~1.5
m/s (Zhou, 1988).
Rearranging equation (1) gives the air temperature of cooled air inside the operations and the
temperature difference between the outside and inside air as follows:
Ti = Tdb,o − η c (Tdb,o − Twb ,o ) = (1 − η c )Tdb,o + η cTwb ,o (2)
∆T = Tdb,o − Ti = η c (Tdb,o − Twb ,o ) (3)
where ∆T is the temperature difference between the inside and outside air of the agricultural
operations with this evaporative cooling system in oC. Both Ti and ∆T are the functions of dry
and wet bulb temperatures of the outside air. Therefore, the cooled air temperature inside the
facilities and the temperature difference can be calculated through the equations (2) and (3) by
using the weather data (outside dry and wet bulb temperatures) of any region under a constant
cooling efficiency condition (Li, 1991). Thus, the cooling effect of the fan-pad cooling system
can be determined by analyzing the weather data ( Tdb,o and Twb ,o ), and the suitability of the
system to different climate conditions can be also evaluated.
2.2 Establishment of a fuzzy judge model
The cooled air temperature and the temperature difference of the operations are two most
important parameters of an evaporative fan-pad cooling system, and are normally used to
evaluate the cooling effect of the system, and the cooled air temperature (inside dry bulb
temperature) is even more crucial in judging whether the system can meet the temperature need
of inside environment for the animals. However, it is difficult to evaluate precisely the cooling
effect of an evaporative pad cooling system by general mathematics. Through
calculating Ti and ∆T , the suitability of the evaporative pad cooling system to the different
summer climates is comprehensively judged by a fuzzy mathematical method in analyzing 20-
year’s (1981~2000) weather data of nine representative cities in China. The region for the
judgment is:
U = {∆T , Ti } (4)
As twenty years’ weather data (1981-2000) are used, Ti and ∆T sets are:
∆T = {∆T1 , ∆T2 , ∆T3 ,....∆T20 }
 (5)
Ti = {T1 , T2 , T3 ,....T20 }
Therefore, the judge mode is a two stages fuzzy synthetical judge mode (Wang and Song,
1988). And the comment region V for discussion is:
90
V= {very suitable, more suitable, suitable, not suitable} (6)

2.3 Standards of fuzzy judge
Although there are many reports on the cooling effect of the evaporative fan-pad cooling
system for agricultural facilities, the publications on the fuzzy judge standard of the system
could rarely be found. According to different practical temperature demands in China, the
appropriate air temperature for an enclosed environment is from 24 to 28 oC in summer (GBJ-
87, 1989). If the highest inside air temperature in summer does not exceed 28 oC, the thermal
environment can normally meet the temperature demands of the agricultural facilities such as
greenhouses, livestock and poultry houses, microbiological ferment houses and so on. The
poultry housing for instance, it is quite economic and suitable for the growth of the birds when
the inside air temperature is below 28 oC in summer, which is within the thermo neutral range
of the poultry. But when the inside air temperature exceeds 30 oC, the body temperature of
birds will become higher, and the production will become lower. Under this condition, the egg
production per birds decreases and average egg weight becomes lighter, and the egg shell
becomes thinner as well the mortality rate becomes higher. When the inside air temperature
reaches 32 oC, heavy heat stress may occur on the birds. If in hot and humid climates, the
harmfulness to birds could be even larger, and the health as well as production can be
significantly affected (Wen, 1981; Li, 2003).
According to the current researches, the suitability of most housed animals to the inside air
temperature ( Ti ) in summer could be described by the following four ranges:
Ti < 28°C (very suitable)
T = 28 ~ 30°C (more suitable)
 i
 (suitable) (7)
Ti = 30 ~ 32°C
Ti ≥ 32°C (not suitable)
In the design and operation of a ventilation system for poultry housing, the temperature
difference between air inlet and outlet (exhaust fan location) is generally controlled at about 3 oC
to achieve a better economic efficiency for both the investment of the ventilating equipments and
the operating cost as well (Shen, 1981). Thus, the relation between the temperature of wetted pad
(Tp) and the inside average temperature (Ti) could be described as:
Tp = Ti - 1.5 (8)
From equations (7) and (8), the cooling effect of the evaporative pad cooling system and the
suitability to the climates can be determined by:
 T p < 26.5°C (very suitable)
26.5°C ≤ T < 28.5°(more
C suitable)
 p
 (suitable) (9)
28.5°C ≤ T p < 30.5°C
 T p ≥ 30.5°C (not suitable)
The cooling effect is normally considered to be very significant when the inside air
temperature could be cooled down by 7 oC or more. It is less significant between 5~7 oC, and it
is still acceptable for 3~5 oC. But if it is less than 3 oC, the cooling system is usually treated as
not suitable for the agricultural facilities. Thus, the suitability of the system can relevantly be
judged based on the temperature difference ( ∆T ) by:
91
 ∆T ≥ 7°C (very suitable)

5°C ≤ ∆T < 7°C(more suitable)

 (suitable) (10)
3°C ≤ ∆T < 5°C
 ∆T < 3°C (not suitable)
Applying the judgment rules (9) and (10), the suitability of wetted pad cooling to the
climatic conditions in China can be evaluated.
2.4 The representative cities selected for evaluating the suitability of the evaporative pad
cooling system
Based on the climatic conditions of China, nine representative cities are chosen, and the
suitability of the evaporative pad cooling system in these nine cities is evaluated. The selected
cities for the judgement are Beijing, Jinan (Shandong province), Xi’an (Shanxi province),
Chongqing (Sichuan province), Wuhan (Hubei province), Nanjing (Jiangsu province),
Changsha (Hunan province), Shanghai, and Guangzhou (Guangdong province), which are
distributed in different climatic regions from the Northern to Southern China. The weather data
at 14:00 o’clock every day from June to September in 1981 to 2000 are used for the
calculation. Considering the practical conditions of the production, only the data of the hot
days with the outside air temperatures higher than 30 oC are chosen to do the analysis.
3. Results and discussion
Taking the judgment of Shanghai in 1981 for example, the process of fuzzy comprehensive
judge (FCJ) is specified. By using the mode developed through the work, the results of
calculating the inside air temperature Ti show that there are 3, 48, 13 and 0 days assigned into
“very suitable”, “more suitable”, “suitable” and “not suitable” mode respectively in 1981. In
total, there were 64 days with the highest outside air temperature higher than 30 oC, and the
ratios of the days fell into the four modes are 0.047, 0.750, 0.203 and 0, respectively.
Therefore, the subordinate degree of the climatic suitability in Shanghai area in 1981 is:
RT1 = (0.047， 0.750， 0.203， 0.000 ) (11)
Similarly, the ∆T subordinate degree of the climatic suitability in Shanghai area in 1981 can
be relevantly gained:
R∆T1 = (0.031， 0.234， 0.672， 0.063) (12)
The subordinate degree of climatic adaptability in other years can be calculated by the same
method. Thereby, the first matrix of fuzzy comprehensive judge can be gotten as follows:
T
0.05 0.07 0.08 0.06 0.07 0.04 0.03 0.29 0.06 0.10 0.14 0.35 0.19 0.16 0.19 0.12 0.11 0.31 0.14 0.06
 
0.75 0.46 0.56 0.53 0.60 0.60 0.60 0.56 0.60 0.65 0.53 0.44 0.67 0.53 0.67 0.79 0.59 0.51 0.64 0.54
RT = 
0.20 0.48 0.32 0.40 0.32 0.33 0.37 0.16 0.29 0.25 0.33 0.21 0.14 0.31 0.14 0.10 0.27 0.14 0.21 0.40
 
0.00 0.00 0.04 0.02 0.02 0.02 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.04 0.00 0.00
Using the method described by Li (Li, 1990), the power set of factors can be gained:
Thereupon, the results of FCJ to each factor can be calculated by equations (11), (12) and
(13) :
T
0.03 0.00 0.00 0.09 0.02 0.16 0.03 0.04 0.02 0.04 0.08 0.03 0.00 0.03 0.02 0.00 0.07 0.07 0.07 0.00
 
0.23 0.14 0.18 0.28 0.18 0.25 0.55 0.33 0.23 0.15 0.44 0.15 0.14 0.19 0.26 0.33 0.18 0.32 0.29 0.20
R∆T = 
0.67 0.59 0.74 0.59 0.71 0.57 0.37 0.56 0.57 0.73 0.47 0.79 0.85 0.72 0.68 0.62 0.62 0.59 0.60 0.63
 
0.06 0.27 0.08 0.04 0.09 0.02 0.05 0.07 0.02 0.08 0.02 0.03 0.02 0.06 0.04 0.05 0.13 0.01 0.05 0.17
BT = A × RT = (0.129， 0.594， 0.265， 0.012 )
A＝(0.064 0.044 0.050 0.068 0.044 0.049 0.065 0.045 0.047 0.052 0.064 0.034 0.052 0.032 0.057 0.042 0.045 0.071 0.042 0.035) （ 13）
B ∆T = A × R ∆T = (0.043， 0.264， 0.625， 0.068)
92
Following the process of FCJ, the suitability of the evaporative pad cooling system to the
summer climate in Shanghai area could be figured out. Concerning the inside air temperature
controlling, the result is:
0.129 0.594 0.265 0.012
BShanghai = + + +
very suitable more suitable suitable not suitable
The result shows that the inside air temperature can be controlled below 30 oC by equipped
with the evaporative pad cooling system for over 70% of the days in which the highest outside
air temperatures are excess 30 oC in summer. For only 26.5% of the days (average 13 days per
year), the inside air temperature reaches 30 to 32 oC, and the temperature is rarely higher than
32 oC. According to the principle of maximum subordination (Wang and Song, 1988), the
result indicates that the evaporative pad cooling system is very suitable for application in the
agricultural facilities in Shanghai region.
For the temperature difference ∆T , the result is:
0.043 0.264 0.625 0.068
BShanghai = + + +
very suitable more suitable suitable not suitable
It means that the inside air temperature could lowered by 3 to 7 oC for about 90% of the hot
days when the evaporative pad cooling system is used in Shanghai.
The calculation results of the subordinate degrees of the evaporative pad cooling system in
nine representative cities are shown in Table 1.
Table 1. Climatic suitability of evaporative fan-pad cooling system in China
Subordinate degree of Subordinate degree of factor
Cities factorΔT (oC) Ti (oC) Suitability
≧7 5~7 3~5 <3 <28 28~30 30~32 ≧32
Beijing 0.519 0.269 0.192 0.020 0.739 0.225 0.036 0
very
Jinan 0.544 0.274 0.167 0.015 0.675 0.240 0.081 0.004
suitable
Xi’an 0.630 0.313 0.056 0.001 0.719 0.231 0.049 0.001
Chongqing 0.436 0.389 0.173 0.002 0.399 0.549 0.051 0.001
Wuhan 0.175 0.444 0.358 0.023 0.204 0.445 0.329 0.022
Nanjing 0.149 0.379 0.432 0.040 0.219 0.489 0.291 0.001
suitable
Changsha 0.289 0.431 0.266 0.014 0.156 0.605 0.235 0.004
Shanghai 0.043 0.264 0.625 0.068 0.129 0.594 0.265 0.012
Guangzhou 0.038 0.281 0.619 0.062 0.101 0.757 0.133 0.001
In Table 1, it shows that the inside air temperature of the agricultural facilities equipped with
evaporative fan-pad cooling systems in the areas of Beijing, Xi’an, Chongqing and Changsha
could be normally cooled down by 5~7 oC in hot days in summer, and in the other areas it is
from 3 to 7 oC. The inside air temperatures in the Yellow River valley and to its north areas
where Beijing, Xi’an, Jinan cities are located, can be controlled below 28 oC with the cooling
system in hot climates, and commonly are not higher than 30 oC. It presents that these areas are
extremely suitable for the application of the evaporative fan-pad cooling system. While in the
other areas including the Yangtze River valley and to its south south costal areas, the inside air
temperature can basically lowered below 30 oC in hot climate days, and is seldom higher than
32 oC. That means the evaporative pad cooling system is quite suitable in these areas also.
The evaporative fan-pad cooling system has been applied to the greenhouse and the animal
housing in the cities such as Beijing, Changsha, Shanghai, Chongqing, and Guangzhou from
1985. The filed measurement of the inside air temperature shows that it is rarely higher than 32
o
C, even the outside air temperature reaches as high as 38 oC and above. And the inside air
temperature in the animal housing applied with the cooling system is usually below 30 oC (Li,
1990). This proves the correction of the above results.
93
4. Conclusions
A fuzzy mathematics method was developed to evaluate the suitability of the evaporative
fan-pad cooling system applied to the agricultural facilities to the summer climates in China.
The results show that the inside air temperatures of the animal housing with the cooling system
in the areas of Yellow River valley and to its north could normally be controlled below 28 oC,
and 5 to 7 oC lower than the outside temperature even in the hottest days in summer. And the
cooling system is very suitable for the application in these areas. While in the Yangtze River
valley and to its south coastal areas, it can basically be dropped under 30 oC and 3 to 7 oC
lower.
By measuring the inside air temperatures and the temperature differences of the animal
housing equipped with the evaporative pad cooling systems, the results calculated by the
developed fuzzy mathematic method are proved. The method is very helpful in guiding the
cooling system design of the enclosed environments for agricultural productions under
different climate conditions in China.
References
ASHRAE, 1983. Evaporative air cooling equipment. Equipment Handbook, ASHRAE
GBJ-87, 1989. Code for Design of Heating, Ventilation and Air Conditioning. Beijing: China
Planning Press
Jin, Y.Z., Z.D. Huang and Y.A. Cui, 1987. Thermal environment of the greenhouse with the
evaporative wet-curtain cooling system. Transactions of the CSAE, 3(1), 47-57
Li, B.M., 1990. Study on the tunnel ventilation in the chicken housing in summer. M.S.
dissertation, Beijing Agr. Eng. Univ., Beijing
Li, B.M., Y.J. Zhou and Y.A. Cui, 1991. Improving the airflow organization of the tunnel
ventilation system for a chicken house. J. Beijing Agr. Eng. Univ., 11(2), 61-66
Li, B.M., Y.J. Zhou and Y.A. Cui, 1992. Study and use of tunnel ventilation system for poultry
houses in summer. Transactions of the CSAE, 8(4), 83-89
Li, R.Z. (editor), 2003. Livestock Environment and Hygiene. China Agriculture Press, Beijing
Liu, J. and J.L. Chen, 2003. Preliminary studies on the effect of wet curtain cooling system
applied to the quality breeder hens. J. Foshan Univ., 21(1), 78-80
Liu, W.D., Z.C. Wang and L. Wang, 2006. Effect of the wet curtain cooling system on the
production of the egg breeding chickens. China Poultry, 28(10), 27-28
Shen, J.S., 1981. Calculation of solar radiative heat of a laying hen house and the air exchange
amount in summer. Transactions of the CSAM, 4, 92-100
Wang, C.H. and L.T. Song (editors), 1988. The Methodology of the Fuzzy Theory. China
Architecture & Building Press, Beijing
Wen, S.Z. (editor), 1981. Animal Environment and Control. Agricultural Press, Beijing
Zhang, S.G., W.T. Song, G.H. Teng, L.H. Gao and Z.D. Huang, 2006. Cooling effect of
different installation height of wet-curtain fan-cooling system. J. Agr. Mach., 37(3): 91-94
Zhou, C.J., 1988. Experimental study on the optimal construction of honeycomb paper pad.
Transactions of the CSAE, 4 (2), 37-47
Zhou, Y.J., 1988. Design and application of the wet-curtain cooling system in chicken housing.
Transactions of the CSAE, 4(4), 38-46
94
Developing a set of strategies, in Portugal, to monitor and prevent damages

in animal housing, due to hot climate conditions
Vasco Fitas da Cruz1, José Carlos Barbosa2, João Santos e Silva3
1
Universidade de Évora. Departamento de Engenharia Rural. vfc@uevora.pt
2
Escola Superior Agrária de Bragança. bb@ipb.pt
3
Ministério Agricultura. joao.silva@draedm.min-agricultura.pt
Keywords: animal production, hot climate, animal houses.
In Portugal, animal production (mainly meat and milk) represents 32% of the Agriculture
Domestic Product and, in some regions, its socio-economic importance is quite relevant (PEN,
2007). Cattle are more common in Alentejo; in the littoral North; and in the Azores island.
From these, dairy cattle can mainly be found in littoral North and the Azores island, whereas
beef cattle is more common in Alentejo. Pigs are mainly raised in the littoral center of the
country and in Alentejo, in the South. Sheep and goat raising is more common in Alentejo, and
in the inland regions of the Center and North.
Table 1. Livestock location of animal domestic species in Portugal (x103) in 2005 (INE, 2005).
Dairy
Region Cattle Cows Swine Sheep Goats
Entre Douro e Minho (see A) 247,4 90,7 95,0 129,8 60,0
Trás-os-Montes (see B) 64,3 11,7 35,0 298,3 64,2
Beira Litoral (see C) 112,5 49,2 399,9 167,8 68,9
Beira Interior (see D) 46,8 9,9 45,0 418,3 97,2
Ribatejo e Oeste (see E) 139,8 21,3 802,4 229,4 47,3
Alentejo (see F) 474,7 17,2 354,0 1 225,8 78,4
Algarve (see G) 9,7 0,3 36,1 56,7 16,0
Açores island 217,0 86,7 48,5 3,6 6,6
Madeira island 3,3 0,3 17,9 3,2 5,2
Portugal is located in Southwestern Europe and it has a mediterranean climate. Winter is cold
and wet. Summer is hot and dry particularly in Alentejo and northeastern regions. Significantly
high temperatures combined with dry air (or even wet air) may bring about serious problems or
damage to livestock and losses to the farmer. In Portugal, it is estimated that 20% of the annual
losses in animal production are due to adverse climatic factors or deficient regulation in indoor
climatization of animal housing. The situation is most problematic in summer when very high
temperatures occur.
According to the HWDI (Heat Wave Duration Index) since 2000, in Portugal three heat waves
have occurred; in August 2003 over a period of 17 days; in May 2005, during 11 days, with
temperatures over 40 degrees C in some places; in June 2005, a heat wave that lasted for 12
days, with temperatures over 40 degrees C in several regions. Figure 1 shows the localization
of the regions most affected by these heat waves. Generally, in all regions, these periods of
very high temperatures are combined with dry air (low relative humidity).
95
Figure 1. Duration (days) of heat waves from July/August, 2003; May 2005; and June 2005.
In most cases, the buildings are not suitable for animal housing under high temperatures. They
lack appropriate equipment to control indoor environmental conditions.
To minimize the effects of these adverse climatic situations on animal production, in Portugal,
we intend to carry out a work project to tackle this problem. For that purpose, we intend to
develop a set of strategies aiming at:
- collecting information about animal breeding in different regions and animal housing
conditions or climatization of animal houses;
- identifying the climatic factors that affect livestock, in all portuguese regions, specially those
related to hot climates;
- identifying the climatic diversity of the several regions and, above all, the factors that can
affect animal housing;
- evaluating the effects of the climatic factors on animal production, in different regions;
- advising farmers about the risk of heat waves or, even, occasional high temperatures;
- studying strategies and methods to help farmers cope with the problem;
- creating a webpage, as a tool to spread information and advice to the farmers.
To carry out this work we have to organize a multidisciplinary team, in order to embrace all
different fields of interest related to this problem.
We also intend to involve several entities, like breeder associations, producer organizations and
public institutions.
References
Ministério da Agricultura (2007) Plano Estratégico Nacional – Plano de Desenvolvimento
Rural 2007-2013. http://www.gppaa.min-agricultura.pt/drural2007-2013/doc/PEN_set06.pdf
(Jan. 2007)
INE (2005) Inquérito à estrutura das explorações agrícolas 2005. INE: Instituto Nacional de
Estatística, Lisboa.
World Meteorological Organization (2001) WCDMP-47: Report on the activities of the
working group on climate change detection and related rapporteurs, WMO-TD nº 1025,
Geneve.
96
Modeling of Technology Implementation and Dairy Farm

Foundation in Hot Climates
M. Samer1, H. Grimm1, P. Epinatjeff 1, M. Hatem 2, R. Doluschitz1, and T. Jungbluth1

1
University of Hohenheim (440), 70593 Stuttgart, Germany
2
Agricultural Engineering Department, Faculty of Agriculture, Cairo University, Giza, Egypt
Keywords: modeling, precision dairy farming, technology
Introduction
Open housing system is used in hot climates, and it is usually a shade structure covering
yard(s) or corral(s) (Lindley and Whitaker, 1996). Linear programming (LP) involves the
planning of activities to obtain optimal solution (Hillier and Lieberman, 2005). The objective is
to develop a model to help in designing dairy farms in hot climates, to assist in implementing
technology, and to calculate the costs.
Material and Methods

A computer-based management information system (CBMIS) was made to collect the
necessary technical data of the new technologies on an ongoing basis. The LP is used to
optimize the technology selection, and then the Excel-Solver is used to solve the mathematical
model of the LP. The MS-Excel is used to develop the model, and to show the results of the
input settings automatically by calculating not only the dimensions of the dairy cow building
and its components, but also the fixed, variable, and total costs.
Results and Discussion

The dairy farm foundation model (DFFM) is divided to 3 sub-models (Fig. 1): (1) design
model (DM) which assists in designing the corrals and the cowsheds and in planning the dairy
farm, (2) technology model (TM) which is a mathematical model using the LP in making
decisions to implement new technologies in the designed dairy cow buildings by the DM and
using the Excel-Solver to derive solutions from the model, (3) costs calculation model (CM)
which is a programmed Excel worksheet that shows the results of the input settings
automatically by calculating the fixed, variable, and total costs. The sub-models have mutual
effects, (e.g.) after making a building design by the DM; the technologies will be implemented
by the TM, and therefore it may be necessary to return to the DM to fit the building dimensions
to the implemented technology.
Design Model
Technology Costs Calculation

Model Model
Figure 1. Dairy Farm Foundation Model.
97
Conclusion
The operations research (OR), especially the LP, can be used successfully in implementing
new technologies in dairy cow farms as a new pattern for the precision dairy farming.
References
Hillier, F. S. and G. J. Lieberman. 2005. Introduction to Operations Research. 8th Edition. The
McGraw-Hill Companies, Inc., New York, USA. 1061 p.
Lindley, J.A. and J.H. Whitaker. 1996. Agricultural Buildings and Structures: Dairy Cattle
Housing, p. 529-564. ASAE, St. Joseph, Michigan, USA.
98
ANALYSIS OF THE EFFECTS OF THE ROOFING DESIGN ON HEAT

STRESS IN DAIRY COW HOUSING
P. Zappavigna, P. Liberati
Dipartimento DIPROVAL Università di Bologna, via Fratelli Rosselli, 107, 42100 Reggio
Emilia, Italy
Key words: roofing design, heat stress, dairy cows, view factor, simulation
Introduction
In animal housing the roof plays a primary role in the determination of the thermal exchanges
of the animals (Liberati and Zappavigna, 2004). In particular, in the hot climate, a high thermal
resistance in daily hours can be helpful in order to reduce the effect of the solar radiation. But
by increasing the thermal resistance it is reduced the possibility to the animals of discharging
heat through the roof in the night hours.
To reduce the diurnal negative effect of the radiative heat load onto the animals the use of
insulation materials is often recommended. But this is an expensive solution the usefulness of
which is not ascertained, depending on various factors: climate, latitude, building geometry and
orientation, constructive solutions, animal physical and spatial parameters.
In dairy housing, many real buildings can be seen all over the world where the roof insulation
is avoided and the reduction of the solar effect simply pursued throughout some geometrical
parameters: slope, height, ridge.
This study aims at investigating the effective influence of the roof constructive parameters on
the animal heat exchange (and welfare) during the whole day, trying to find out the optimal
solutions for different combinations of the more relevant factors: insulation, slope, orientation,
height, roof shape.
The work, carried out by a specific theoretical model, is in this paper referred to a location in
the Northern Italy (Po valley), but it can be equally applied to other different climate and
latitude conditions.

A simulation model determining the heat flow exchange between the animals housed inside a
building and the roof was developed considering various relevant factors: constructive
materials, slope, height, orientation, latitude, external air temperature, solar load, animal
position.
The model was applied with reference to the climatic conditions of the Italian Northern area
and to dairy cows.
Heat exchanged between the cow and the roof was calculated by the following:
σ (T 4 r − T 4 coat )
Q r −cow j = (W ) (1)
1 − εr 1 1 − ε cow
+ +
ε r A r A r ⋅ Fr −cow j ε cow A cow
where Tsi is the internal surface temperature of the roof (calculated separately for each roof
pitch), Tcoat is the surface temperature of the cow fur, Ar is the roof area surface , Acow is the
99
area surface of the cow, εr and εcow are, respectively the emission coefficient for the roof (set at
0.9) and the cow (set at 0.98); σ is the Stefan-Boltzmann constant. Qr-cowj will be positive when
the net balance at cow level is incoming (cow absorbing heat), and negative when the cow is
releasing heat (the roof subtract heat from the cow).
Determination of the inside surface temperature (Tsi)
The developed model to calculate Tsi (described in: Zappavigna e Liberati, 2005) was
dynamic and took into account all aspects affecting the internal environment. The inputs were:
building geometry, building orientation, vent opening (size and position), thermal-inertial
characteristics of construction materials (thermal conductivity, density, specific heat capacity,
heat transfer coefficients), wall material composition, terrestrial coordinates (longitude,
a) b)
R2 R1
R2 R1
A B C A B C
orienta- cow orienta- cow orienta- cow orienta- cow

tion exposition tion exposition tion exposition tion exposition
NS W NS E NS E NS W
EW N EW S EW N EW S
Figure 1. Roof configurations: a) gabled roof; b) multiple shed roof. A, B, and C, cow positions
inside the barn for view factor calculation. Three roof slope (10, 30 and 45 %), and two eaves
height (3.5 m and 4.0m) had been evaluated.
latitude), type of animal, as well as local climatic conditions (direct and diffuse solar sun
radiation, wind (velocity and direction), air temperature and humidity).
Roof contribution was calculated considering the internal surface temperature evaluated by
modelling the thermal behaviour by using a one-dimensional Fourier equation solved through
the finite difference method (FDM) (using Crank-Nicolson formulation).
The boundary conditions at the outside surface consider the solar contribution, both direct and
diffuse radiation. The sun's position in the sky was updated each calculation step; the incident
radiation was calculated in relation to the building orientation also.
Considering also the radiative exchange between wall surface and sky at the outside roof
surface, allowed the radiating surface temperature itself to drop, if the case, below the air
temperature.
View factor calculation

View factor between the roof and an hypothetical cow had been calculated considering
different affecting parameters, as roof sloping, eaves height, roof type (gabled and multi shed
roof). Three position for the cow had been considered, A, B and C (fig.1); longitudinal position
had been set at the middle.
The cow had been simulated as an ellipsoid (x2 / a2 + y2 / b2 + z2 / c2 = 1, where a is the half
cow length, b is the half cow width, c is the half trunk height). Ellipsoid area surface was
similar to that obtained with the well known formula for Holstein cows, weighing 600 kg, that
is of 5.46 m2: As = 0.14 × W0.57 (m2) (Berman, 2003).
100
A computer program had been developed to calculate view factors under different conditions.
After a meshing step regarding the roof and the cow, view factors was calculated using a
discretized formula as follow:
1 cos(α j ) a j ⋅ cos(α i ) a i
π r2
Fr −cow = (2)
Ar
∑∑
j i
where Fr-cow is the view factor between the roof and the cow, aj and ai are the resulting
elementary faces of the meshed surface of the roof and the cow, αj and αi are the angles
between the normal to each face a and the connecting straight line of the faces; r is the distance
between the faces itself.
In this way, for each cow position, the view factor had been calculated with respect to each
pitch of the roof. For example, for a cow in A position, two view factors had been calculated:
FR1-A, and FR2-A. The calculated view factors are reported in fig. 2.
-4 -4
30
View factor - gabled roof (x 10 )
pitch R1 30
View factor - multiple shed roof (x 10 )
pitch R1
25
pitch R2 25 pitch R2
20 20
15 15
10 10
5 5
0 0
cow: A B A B A B A B A B A Bcow: A B A B A B A B A B A B
|------ 10% --------||--------- 30% -----||----------- 45%-- -----| |------ 10% --------||--------- 30% -----||----------- 45%-- -----| |------ 10% --------||--------- 30% -----||----------- 45%-- -----| |------ 10% --------||--------- 30% -----||----------- 45%-- -----|
|--------------------------------- 3.5m ---------------------------||----------------------------------- 4.5m ---------------------------| |--------------------------------- 3.5m ---------------------------||----------------------------------- 4.5m ---------------------------|
Figure 2. View factors (Froof-cow) for cows in A and B position , for gabled and
shed roof shapes.
To calculate Tcoat of the cow fur from internal air temperature, has been considered the work
done by Turnpenny et al., 2000. In particular, a linear function Tcoat = f (Tair) has been found
considering two points: 1) Tcoat = 19°C with Tair =15°C and 2) Tcoat = 30.4°C with Tair = 30°C.
That is the gradient Tcoat - Tair is decreasing with increasing Tair.
Two building orientations has been tested (NS, WE), with the roof without (only fibre-cement)
and with insulation (fibre-cement plus 4 cm of polystyrene). Two roof shape (gabled and multi
shed), with three slopes: 10%, 30% and 45% (fig.1). Then, also two eaves height had been
tested: 3.5 and 4.5m.
The model was applied to a fully open building (simple shelter), but the effect on the animal
heat exchange would be much higher in a closed building since the heat exchanged throughout
the roof influences the animal not only in a radiative form, but also in a convective form,
increasing or decreasing the internal air temperature.
Result and discussion

Instantaneous thermal power exchanged between the cow and the roof
As an example in figures 3 is reported the instantaneous net balance of the radiative thermal
power exchanged by a cow in B position with a gabled roof. The heat gained at the peak value
by the cow with non-insulated roof is 4-5 times than the insulated roof, even if the heat lost by
the cow is greater with the non-insulated roof (but less than one time, in the better
configuration).
101
To evaluate the response of the different configurations and cow position, we refer to figures 4
and 5, reporting the peak values of the instantaneous net power balance at cow surface
(positive and negative, respectively for cow absorbing and releasing heat), with cow in A or B
position).
Comments concerning the thermal roof behaviour will be done with respect to different
affecting factors: roof shape, insulation level, building orientation, eaves height, cow location
inside the barn.
To improve the readability of the observations, the following abbreviations will be adopted:
GR = gabled roof;
SR = shed roof;
A, B = cow in A or B position;
NS, WE = building orientation;
10%, 30%, 45% = roof slope;
PPB, NPB = Positive or Negative thermal Power Balance at the cow level (peak values). For
example, corresponding to the positive and the negative values for GR-B_WE configuration of
fig.3.
So, GR-A_45%, regard a cow located in A position with a gabled roof with a slope of 45%;
A_10%_WE refer to a cow in A position, roof slope 10% and WE oriented.
(24, -32)>INSULATED-4.5m-45% (25, -37)>INSULATED4.5m-30%

(26, -38)>INSULATED4.5m-10% (147, -55)>NON-INSUL.-4.5m-45%
(155, -55)>NON-INSUL.-4.5m-30% (163, -56)>NON-INSUL.-4.5m-10%
(27, -41)>INSULATED3.5m-45% (28, -41)>INSULATED3.5m-30%
(29, -42)>INSULATED3.5m-10% (159, -60)>NON-INSUL.-3.5m-45%
(168, -60)>NON-INSUL.-3.5m-30% (179, -62)>NON-INSUL.-3.5m-10%
Gabled roof, cow position B, WE

orientation
180
150
120
Qroof-cow_B (W)
90
60
30
0
-30
-60
2.00 5.00 8.00 11.00 14.00 17.00 20.00 23.00 2.00 5.00 8.00 11.00
time
Figure 3. Gabled roof. Instantaneous thermal power exchanged between the roof
an the cow in B position, for WE orientation, with different roof configurations. In
legend labels, inside parenthesis, the positive and negative peek values.
Insulation level factor

Clearly, roof insulation is the most important factor to reduce heat load of the cow. But
insulation, in night time can prevent heat loss from the cow itself.
Passing from an insulated roof to another non-insulated we can have:
PPB period
102
180 180
GABLED ROOF
MULTIPLE SHED ROOF gained gained
heat heat
150 150 lost heat byheat

lost the cow
by the cow
exchanged heat (W)

exchanged heat (W)
120 120
90 90
60 60eaves 4.5 m eaves 4.5 m eaves 3.5 m eaves 3.5 m

eaves 4.5 m eaves 4.5 m eaves 3.5 m eaves 3. 5m
30 30
0 0
-30 -30
-60 -60
POSITION
POSITION BA
BA BA BA BA BA BA
BA BA BA BA
BA BA
BA BA
BA BA
BA BA
BA BA
BA BA BA BA
BA BA BA BA
BA BA BA BA
BA BA BA BA
BA BA
BA BA
BA BA
BA BA
BA BA
ORIENT.WE-NS
ORIENT.WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS WE-NS
WE-NS
SLOPE45% 45%30% 30%10% 10% 45% 45%30% 30%10% 10%45% 45%30% 30%10% 10% 45% 45%30% 30%10% 10%
SLOPE
INSULATED
INSULATED NON-INSULATED
NON-INSULATED INSULATED
INSULATED NON-INSULATED
NON-INSULATED
Figure. 5. Multiple shed roof. Thermal power (peak values) exchanged by

Figure. 4. Gabled roof. Thermal power (peak values) exchanged by the
the cow trought the roof (positive, gained by the cow, negative lost), for
cow trought the roof (positive, gained by the cow, negative lost), for
different roof configuration.
different roof configuration.
Whit GR-B_45%, we have a PPB of only 18W, arising up to 98W if we keep out the insulation
(+512%), about the same with cow in A position and a slope of 10%.
With SR, the response is differentiated depending on the orientation: with WE we increase
PPB of 463% (511% for 10% slope), up to 700% for NS orientation (505% for 10% slope).
NPB period
In this context, passing from an insulated to non-insulated roof we improve heat losses from
the cow.
In particular, in GR_45% we pass from 32 to 55W (+72%, +48% in A position) , about the
same for both WE and NS and for slope of 10%.
With SR, improvement for A position is the same than GR, but reduced for B position (46%,
passing from 22 to 48W).
Résumé
With insulation PPB is reduced, but also NPB; slope and orientation are not important in GR,
while in SR, passing from a slope of 10% to 45%, PPB is reduced of 25%. Always with
insulation, A is better than B (PPB reduced of 25% up to 38% with slope of 45%).
Roof shape factor

Roof shape (gabled and shed) is an important aspect, much more as greater is the roof slope.
Passing from GR to SR the cow reduces its PPB, but, in the meantime, also the NPB during the
night time.
PPB period
Insulated roof:
Passing from GR_45% to SR_45%, a cow located in B position reduce PPB about of 25%
(both for NS and WE), passing from 24W to 18W; in A position, there is an ulterior
improvement, depending on building orientation: with WE the reduction is of the 35% (from
17W to 11W), while with NS the reduction is of the 45% (from 17W to 11W).
With a slope of 10%, passing from GR to SR we obtain the following reductions: B_10% = 9%
both for WE and NS; A_10%_WE = 11%, A_10%_NS = 14%.
Non-insulated roof:
Reduction passing from GR to SR_B_45%_WE is 33% (5% for NS), passing from 147W to
98W. With cow in A position, we have a reduction from GR to SR_A_45% of 39/% (25% for
NS); with slope at 10%, the improvement is, for all the configurations, about of 9%.
103
NPB period
Insulated roof:
In B position, the roof do not show particular variation (38W), both for 45% and 10% slope. In
A position, NPB passes from 29W to 21W (-27%, both for NS and WE) with a slope of 45%,
while with roof slope of 10% we have a smaller worsening of only 7%.
Non-insulated roof:
In B position, worsening is of 13% (WE, passing from 55 to 48W of the NPB), worse in NS
orientation (26%), like in A position (from 42 to 31W, both for WE and NS).
So, if the SR improve the situation during daytime (up to a reduction of 45% of PPB), in night
time we have a general worsening (up to -27% of NPB). This behaviour of the roof is due to
the fact that in the SR we have generally a reduced view factor.
Résumé
With insulation, passing from GR to SR, PPB is reduced from 25% to 35% with a slope of
45% (-10% with slope of 10%); also NPB is reduced (of 27% for SR-A_45%, less with slope
of 10%).
Without insulation, PPB is reduced of 33-40% with slope of 45%, 9% with slope of 10%. In
the meantime also NPB go down, more with slope of 45% (-13/26%), less with slope of 10% (-
4/7%).
Building orientation factor

Building orientation seem does not affect the response of insulated roofs.
Non-insulated roof:
PPB period
In GR-B position the orientation is not important, while in SR-B and A WE orientation is
better than NS: reduced PPB of 29% with slope of 45%, 19% with slope of 30)%, and 6 %
with slope of 45%.
NPB period
Building orientation do not affect the response of the roofs.
Résumé
With insulation slope is indifferent (both for PPB and NPB).
Without insulation, in GR-B is not important. In GR-A, PPB is reduced of 13% with slope of
45% (5% with slope of 10%).
In SR non-insulated, both in A and B, WE is better than NS, with PPB reduced of 29% with
slope of 45%, 19% with slope of 30%, and 6% with slope of 10%.
Eaves height factor

Increasing of the eaves height, reduce the view factor; this fact, from a side reduce PPB, on the
other side reduced NPB. So, increasing of eaves height improve daytime situation, worsening
night time condition.
Insulated roof:
PPB period
In particular, passing from 4.5 to 3.5m in B position we increase PPB of 13%, from 24W to
27W (11% in SR), and about the same in 10% of slope and WS or NS orientation.
In A position, we have a better result in WE (+20%, similar in SR; 15% in NS).
NPB period
In GR-B_45%, passing from 4.5 to 3.5m, we have +28% in NPB; in GR-B_10%, we have the
same in NS, less in WE (11%).
In SR, we have a better condition in A position (+19%, both for 45% and 10% of slope), while
in B position only an improvement of 13%, both for WE and NS.
104
Non-insulated roof:
In non-insulated roof, GR and SR have a similar behavior.
PPB period
In B position, (45% and 10% slope), we have an increased PPB of 9% (from 147 to 159W),
while in A position we have 11% in WS, and 16% in NS (the same of 10% slope, both WE and
NS).
In SR-A, we have an increasing of 16% (both fro WE nad NS, and 45% and 10% slope)
NPB period
In this period, we have the same values seen for the PPB period; only for GR-A_45% we have
+14% instead of 11%, both for WE and NS.
Résumé
With insulated GR and SR, PPB and NPB are about reduced of 10%.
Cow location inside the barn (A and B) factor

PPB period
In daytime A location is better than B location, and vice versa in night time, both in GR and in
SR, insulated or non-insulated.
Meanly speaking, GR-A is 20% better than B, SR-A is 27%
NPB period
GR-B is 18 % better than A, while SR-B is 30%.
Résumé
With regard to PPB, A position is better than B; vice versa if considering NPB.
Slope factor
PPB period
Generally, in GR, passing from 10 to 45% PPB is reduced of 11%; in SR-B_WE, passing from
slope of 10% to 45%, we have an improvement of 20-16%. With SR-A_WE, we have 28-
21%.
With NS, we have a reduced improvement, 6% in B, 13% in A.
NPB period
Thermal behavior does not depend on orientation, but only from the view factor, increasing
with decreasing slope. In GR there are not variation. In SR-B, passing from 10% to 30% of
slope we reduce NPB of 7% (4% passing from 30% to 45%)
Résumé
With insulation, in SR, PPB is reduced of 25% passing from 10% to 45% (only of 8% in GR);
NPB, in GR and SR, is reduced of 12% in B (29% in A).
In non-insulated SR-B, WE oriented, passing from 10% to 45% PPB goes down to 16-20%
(21-28% in A). In NS, reduction is about only 10%.
NPB worsens as slope is increased (due to the reduced view factor).
Exchanged energy
Time duration for PPB and NPB are different. In particular, with non-insulated roof the PPB
period start early in the morning (about 8 am, figure 3), before than in the insulated roof (10
am), but terminates lightly before (about one half hour); the NPB period, start before with the
non-insulated roof (just one half hour), but terminate before. So, the PPB period is meanly
9.4 h for the insulated roof (11.7 h for non-insulated), while the NPB period is 14.6 h for
insulated and 12.3 h for non-insulated roof.
This differences in PPB and NPB duration, makes important the analysis of the energy
exchanged in that periods. Figure 6 shows the integral of the thermal energy gained (+) or lost
(-) by the cow during 24 hours.
105
Generally speaking, increasing slope reduce the gained energy, and WE is the best orientation
in cow located in position GR-A, SR-A and SR-B.
gabled roof - cow A gabled roof - cow B
(Wh) (Wh)
1400 1400
1200 1200
1000 lost 1000 lost
gained gained
800 800
600 600
400 400
200 200
0 0
-200 -200
-400 -400
-600 -600
NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE
10% 30% 45% - 10% 30% 45% 10% 30% 45% - 10% 30% 45% 10% 30% 45% - 10% 30% 45% 10% 30% 45% - 10% 30% 45%
|------------- 3.5 m------------|------------- 4.5 m------------------||------------- 3.5 m------------|------------- 4.5 m--------------| |------------- 3.5 m------------|------------- 4.5 m ----------------||------------- 3.5 m------------|------------- 4.5 m--------------|
|------------------ INSULATED ------------ ---||------------------ NON-INSULATED --------------- |------------------ INSULATED -------------------||------------------ NON-INSULATED ---------------|
|
multiple shed roof - cow A multiple shed roof - cow B
(Wh) (Wh)
1400 1400
1200 1200
1000 lost 1000 lost
gained gained
800 800
600 600
400 400
200 200
0 0
-200 -200
-400 -400
-600 -600
NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE NS WE
10% 30% 45% - 10% 30% 45% 10% 30% 45% - 10% 30% 45% 10% 30% 45% - 10% 30% 45% 10% 30% 45% - 10% 30% 45%
|------------- 3.5 m------------|------------- 4.5 m------------------||------------- 3.5 m------------|------------- 4.5 m--------------| |------------- 3.5 m------------|------------- 4.5 m-----------------||------------- 3.5 m------------|------------- 4.5 m--------------|
|------------------ INSULATED ---------------||------------------ NON-INSULATED ---------------| |------------------ INSULATED -------------------||------------------ NON-INSULATED ---------------|
Figure. 6. Thermal energy (Wh) exchanged by the cow and the roof, for different roof
slope, building orientation, eaves height, with and without insulation, during daytime
(positive values, heat gained by the cow) and night time (negative values, heat lost by
the cow).
With respect to the heat lost by the cow, the best solution is with non-insulated roof, 3.5m
eaves height, and a slope of 10%
Normally, a low view factor for a given roof configuration causes a better condition, reducing
the PPB during daytime (considering roofs with the same degree of insulation): view factor
decrease if increasing slope and eaves height; SR shows the lower values, mainly due to the
pitch R1, and more pronounced with the cow in A position (fig. 2). Smaller view factor
produces a smaller NPB.
GR is less affected by parameters changing than SR, and SR is generally better than GR.
Generally speaking, insulated roof shows a similar behavior (both in the GR and SR), although
SR is better than GR.
Insulated roofs are the best solution for the lower net heat transferred to the cow (the best
solution is: SR, 10%, 4.5m, with a positive peak of only 11 W).
Referring to the cow position, A is better than B during the day; in night time, A and B are at
the same level (both for GR and SR), while in considering previous comment about power
exchanged (figures 4 and 5), A was generally better then B.
Conclusions
In this work, we considered only the radiative exchange between the roof and the cow. From
this point of view, the evidence found (obviously) is that we need of insulated roofs if we want
to reduce cow heat load: this assertion is fundamental for closed barn.
But, for open shade shelter, other aspects must be take into account. For example, if the site
where the shelter is located presents a good wind action level; in this case, non-insulated
shelter could be adopted. The non-insulated multi shed roof with a slope of 45% and an eaves
height of 4.5 m oriented in WE, shows a peak of PPB of 98W in B position and a NPB of 48W
(62W and 31W, respectively, in A position); the absorbed energy by the cow in B position
106
during the PPB period is 770 Wh (483Wh in A position). Released heat durin NPB period is
457Wh (60% of that gained in PPB period). This heat load could be not excessive if we
consider also that part of the heat produced by the cow could be removed by the convection
way.
An ulterior aspect to be considered is the location of the cows inside the barn: in A position we
have a reduction of the heat load of about 30%.
References
Berman A. 2003. Effects of Body Surface Area Estimates on Predicted Energy Requirements
and Heat Stress. J. Dairy Sci. 86:3605–3610.
Bray, D.R., Bucklin R. A., Montoya R. and R. Giesy, 1994. Means to reduce environmental
stress on dairy cows in hot, humid climates. Proc. Third International Dairy Housing
Conference, Orlando, Florida, 2-5 February, 589-597.
Bucklin, R.A., R.W.Bottcher, G.L. Van Wicklen and M.Czarick, 1993. Reflective Roof
Coatings for Heat Stress Relief in Livestock and Poultry Housing. Applied Engineering in
Agriculture 9 (1): 123-129.
Buffington, D. E. and T.C. Skinner, 1980. Solar Radiation and Wind effects as Functions of
Building Orientation. Trans. ASAE, 23 (6),1482-1488
Garret, W.N., Bond T.E. and N. Pereira, 1967. Influence of shade height on physiological
responses of cattle during hot weather. Trans. ASAE, 10 (4).
Cascone G., 1980. Influenza della copertura sul microclima degli edifici zootecnici nell'area
mediterranea. Atti del IV Convegno Nazionale A.I.G.R. su: "Ingegneria per lo sviluppo
dell'agricoltura", Porto Conte (Alghero), 4-6 maggio 1988, pp. 265-270.
Liberati P., Zappavigna P., “Performance of ventilated roofs in hot climate”, International
ymposium of the CIGR 2nd Technical Section, Evora, Portogallo, May 2-6, 2004, pp. 1-8, CD
N. FB04_611.
eppsson K.H., and G. Gustafsson, 2001. Solar Heat Load in Uninsulated Livestock Buildings,
J. agric. Engn. Res. 78 (2), 187-197.
Perry, R.L. and E.P. Speck, 1962. Geometric factors for thermal radiation exchange between
cows and their surroundings. Trans. ASAE.
Swierstra, D. and E.N.J. van Ouwerkerk, 1985 A model estimating the effectiveness of shade
structures on the production of dairy cows in hot climates. Seminar of the 2nd Technical
Section of the CIGR Agricultural Buildings in Hot Climate Countries, Catania, Italy,
September.
Stowell, R.R., W.G. Bickert and F.V. Nurnberger, 1998. Radiant Heating and Thermal
Environment of Metal Roofed Dairy Barns. Proc. Fourth International Dairy Housing
Conference, St.Louis, Missouri, 193-200.
Turnpenny J.R., McArthur A.J., Clark J.A., Wathes C.M., 2000. Thermal balance of livestock.
1. A parsimonial model. Agricultural and forest meteorology 101: 15-27.
Zappavigna, P. and Liberati P., 2002. Thermal Behaviour of Animal Houses in Hot Climate:
Experimental Contribution to the Theoretical Approach. Proc. ASAE Annual International
Meeting and CIGR XVth World Congress, Chicago, USA, July 28-31: CD Paper.
Zappavigna P., Liberati P., “Slotted roofs as a tool for improving the housing conditions in hot
climate”, Proceedings of the Seventh International Symposium “Livestock Environment VII”
A.S.A.E., Beijing, China, 18-20 May 2005, pp. 1-6, ISBN 1-892769-48-4.
107
Technical Solutions for Reduction of Heat Stress in Livestock Buildings in Germany
H.-J. Mueller. Leibniz-Institute for Agricultural Engineering, Department of Engineering for

Livestock Management, Max-Eyth-Allee 100, 14469 Potsdam, Germany
Keywords: heat stress, ventilation system, micro climate, emission
Introduction
High air temperatures together with high air humidity lead to increase the heat stress for the
animals. Also in Germany during the summer period high temperatures occur. In this period in
the middle of the day the temperatures increases outside up to 32 °C; 36 °C – 38 °C are
possible. From that high temperatures inside the building a negative influence with respect to
animal welfare and animal performance results. The reduction of heat stress is possible by
suitable design of the building and of the ventilation system. Such measures can be the
reduction of solar radiation, using heat store capacity of the ground, increase of the air
movement in the animal zone, evaporative cooling system and technical cooling systems.
Examples will be given which solutions are used in Germany for cattle, pigs und poultry.
Background
The determinant meteorological factors for the microclimate inside the building are the outside
temperature, the outside humidity, the outside enthalpy, the solar radiation and the wind
velocity. In the time period 1893 – 1952 the maximum temperature in Potsdam was 37.5 °C. In
this period a temperature value 30.0 °C was exceeded 19.3 hours per year and 157 hours per
year a value of 25.0 °C was exceeded. The maximum relative frequency of wind velocity
during the summer period is 12% for 3.5 m/s. Only 2 % of the summer period the wind
velocity is lower than 1 m/s. This value is interesting for naturally ventilated livestock
buildings.
The microclimate in animal houses can be evaluated by balance methods. Necessary values for
heat, moisture and CO2 production can be found for example in the standard DIN 18910.
The definition of thermal comfort for animals is much more difficult than for humans. In the
literature there are different models to evaluate the different influencing factors such as
temperature, humidity and velocity, with a so-called “Thermal comfort index”. A buildup of
different approaches can be found by Naas et al. (2006).
In many cases the increasing of air velocity in the animal zone is advised (see Herkner et al.
2002).
Investigation Methods
Especially the air temperature, the air humidity and the air velocity is measured inside the
building. The air flow pattern are observed by smoke generator and recorded by video camera.
Gas concentrations (mainly CO2 and NH3) are measured by multi-gas-monitor to analyse the
micro climate and to determine the emission mass flows. For measuring the air exchange rates
to calculate the emission streams the ATB has developed special tracer gas methods. The
advantage of using Krypton 85 (radioactive inert gas) is a high resolution in time and place, if
there are complicated flow conditions.
Results
In cattle houses with natural ventilation different solutions to increase the air velocity in the
animal zone are investigated. One example for using ceiling fans is shown in Figure 1. High air
108
velocities are measured beneath the ceiling fan. The higher velocities near the side wall result
from the outside wind.
0,00-0,20 0,20-0,40
0,40-0,60 0,60-0,80
0,80-1,00 1,00-1,20
1,2
12 m
1,0
0,8
8m
0,6
Feed pass side wall
0,4 4m
0,2
0,0 0m
0m 4m 8m 12 m 16 m
Figure 1. Ceiling fan (left) and air velocity field 1.2 m above the floor (04 May 2006). The
velocity data are average values over 2 minutes.
In contrast to the cattle barns the pig houses mostly are forced ventilated. For this kind of
animals cooling systems – like evaporative systems, geothermal heat exchanger using heat
store capacity of the ground and technical cooling – are applied to reduce heat stress.
In poultry houses mostly are used ceiling fans to improve the air movement in the animal zone
and evaporative cooling systems.
Conclusion
Animal welfare and high performance require the compliance with good microclimate
conditions inside the animal houses over the whole year. Especially in the summer period high
temperatures lead to problems. In naturally ventilated cattle houses additional forced
ventilation systems are used to improve the air movement. In pig farming the geothermal heat
exchanger have provide themselves. In connection with biogas production the application of
technical cooling systems in pig production is under development in Germany. The evaporative
cooling systems are successfully applied in poultry production.
References
Herkner, S, Lankow, C, Heidenreich, T and Panzer, K. 2002.
Mindestsommerluftvolumenströme für Hochleistungskühe. Landtechnik. 5/2002, pp 286-287.
Naas, I. et al., 2006. Animal Housing in Hot Climates: A Multidisciplinary Review,
Published by Research Centre Bygholm, Danish Institute of Agricultural Sciences, Schüttesvej
17, 8700 Horsens, Denmark. 105 pages. ISBN 87-88976-94-7. (www.cigr.org)
109
Evaluation of fogging in a mechanically ventilated pig facility

Angelika Haeussermann1, Eberhard Hartung1, Erik Vranken2, Jean-Marie Aerts2, and Daniel
Berckmans2
1
Institute of Agricultural Engineering, Christian-Albrechts-University Kiel,
Max-Eyth-Strasse 6, 24188 Kiel, Germany; ahaeussermann@ilv.uni-kiel.de
2
Division M3-BIORES, Department of Biosystems, Catholic University of Leuven,
Kasteelpark Arenberg 30, 3001 Leuven, Belgium; Daniel.Berckmans@biw.kuleuven.be
Abstract
Evaporation of water to the ambient air is generally a cost effective solution to alleviate heat
stress in animal facilities. In this investigation, a high pressure fogging system (7 MPa) was
tested in a mechanically ventilated research facility for growing-finishing pigs. A positive
effect of fogging was found on the indoor temperature, the temperature-humidity index (alert
situations were reduced from 13% to 1%), and on the weight gain of the animals during hot
summer conditions. The simulated water consumption averaged on 2.5 L pig-1 d-1 if fogging
was used on maximum settings. Main future tasks of an optimized control algorithm are to
optimize indoor conditions in terms of temperature, relative humidity and air flow, to minimize
water consumption and energy use, and to avoid wetting of surfaces.
Keywords
Adiabatic cooling, ventilation, evaporation model
Problem
During hot environmental conditions, pigs regulate body temperature by reducing the time and
amount of feed intake. In consequence, metabolic rate and performance drop (Nienaber et al.,
1996, Quiniou et al., 2000; Huynh 2005). In order to reduce heat stress of animals, heat loss
can be improved by providing showers at separate pen areas and subsequent evaporation of
water from the wetted body surface. Likewise, evaporation of small water droplets
(misting/fogging) is utilised for increasing the latent heat content of the ambient air at large or
complete pen areas, which consequently improves sensible heat loss of the animals. Equally
important, indoor air cooling can counteract an increase of atmospheric emissions, such as NH3
or CH4, during hot summer conditions (Haeussermann et al., 2005; 2006).
High pressure (fogging) systems realise in general a higher evaporative fraction than low
pressure (misting) systems. An accurate control of fogging systems can either be reached by
varying the ventilation rate or by varying the supplied water amount and system pressure,
hence evaporation rate (Gates et al., 1991a; 1991b; Arbel et al., 1999; 2003). Notwithstanding,
the effective evaporation rate depends on several factors and can vary considerably among
different housing systems or outside conditions. In order to improve the control of a fogging
system, information on evaporation characteristics of the specific system are needed.
Objectives
A high pressure fogging system was tested in a mechanically ventilated research facility for
growing-finishing pigs during year-round measurements. Main objectives of the study were to:
(1) investigate effects of the tested adiabatic cooling system on indoor air climate and weight
gain of the animals; (2) model the evaporation process and select the main influencing
variables on the evaporation rate; (3) simulate different control settings for ventilation and
fogging.
110
Methodology
The research facility comprised two equally designed and comparable compartments (54 pigs
each; 30 kg to 105 kg). Both compartments were equipped with: two pens; a slotted concrete
floor with a manure pit underneath each pen; alternatively either with a sensor liquid or a mash
feeding system; a mechanical ventilation system (negative pressure, under floor extraction).
Fresh air was supplied via two porous air inlet ducts per compartment, each arranged centrally
above the animal area (Haeussermann et al., 2004).
Fogging lines were placed one in front of the air inlet ducts (2 nozzles per inlet) and one inside
the compartments (3 nozzles per pen). Water supply was 885 ml min-1 per compartment when
both fogging lines were operated; pressure of the system was fixed to 7 MPa. The fogging-
system duty cycle (4 min on / 3 min off) and operation of fogging was controlled by the
ventilation controller. Fogging started either when the indoor temperature exceeded the set
point temperature by more than 1.5°C or when the indoor relative humidity dropped below
50%. Fogging cycles were interrupted when indoor relative humidity exceeded 80%.
Measurements included: outside and indoor temperature and relative humidity (PT 100 and
capacitive, respectively), and ventilation rate (measuring impellor), recorded continuously at a
measuring frequency of 1 min; water consumption (water meter), recorded once per day; and
the average animal weight per pen, recorded in three-week intervals (weighing scale).
Based on indoor temperature T in °C and indoor relative humidity RH in %, the temperature-
humidity index THI was calculated according to NWSCR (1976), Eqn (1):
THI = [1.8 ⋅ T + 32] − [0.55 ⋅ ( RH 100)] ⋅ [(1.8 ⋅ T + 32) − 58] (1)

Evaporation characteristics, such as the time constant to reach 63% of the evaporation rate at
steady state and the evaporative fraction in percent of the water amount supplied for fogging,
were calculated by considering the evaporation process as a dynamic system described by a
single-input, single-output transfer function (Haeussermann et al., 2007a). For the validation of
the model, this transfer function was integrated as sub-model into a dynamic mechanistic
simulation model developed in Berckmans et al. (1992) (Fig. 1). In this connection, different
control settings for fogging (turning on the system at 1.5°C and at 3°C above set point
temperature, fogging system duty cycle 30 s and continuous) and for ventilation (control range
3°C and 6°C, maximum available ventilation rate 32 and 43 air volume changes per hour) were
simulated (Haeussermann et al., 2007b).
Outside climate
Control Energy-
settings use
XO TO
Heating Q Gi
ventilation system Process
controller
I Xi
Fan
Xe Ti
Fogging Q
system
TS /RHS
temperature / Water
humidity sensor consumption
Figure 1. Dynamic simulation model (modified after Berckmans et al., 1992): T =
temperature, X = humidity, RH = relative humidity, G = gas concentrations, Q =
heat supply, I = ventilation rate, o = outside, i = inside, s = sensor, and e =
evaporation.
111

Possible effects of indoor air cooling were mainly found during days with mean daily outside
temperatures above 14°C. Thereby, temperature peaks in the afternoon were lowered in
average by about 4°C to 5°C comparing ventilation with and without fogging (Fig. 2).
Maximum indoor temperature was reduced by about 7°C. The inter-quartile range of the indoor
relative humidity rose on 64% to 82%. Hence, dry indoor conditions with low relative
humidity (< 40% RH) were mainly avoided.
30 30
Mean indoor temperature, °C
Mean indoor temperature, °C

28 28
26 26
24 24
22 22
20 20
18 18
10:30
12:15
14:00
15:45
17:30
19:15
1:45
10:30
12:15
14:00
15:45
17:30
19:15
0:00
3:30
5:15
7:00
8:45
21:00
22:45
1:45
0:00
3:30
5:15
7:00
8:45
21:00
22:45
(a) Time (b) Time
Figure 2. Mean diurnal course of indoor temperature for ventilation: (a) with fogging (45
days) and (b) without fogging (38 days) at mean daily outside temperatures above
14°C; simulated values for parallel time periods; measured: with fogging;
without fogging
Reference, measured ; simulated: with fogging; without fogging
The positive effect of fogging on indoor climatic conditions was confirmed furthermore by the
temperature-humidity index (THI): without cooling, 15.5% (13%; 567 h) of the measured
(simulated) values were in minimum in an alert range (THI: 75-78), thereof 5% (4%; 189 h)
were dangerous (THI: 79-83) and 3% (2%; 73 h) were categorised as emergency (THI: ≥ 84).
With indoor air cooling, only 0.8% (1%; 34 h) of the values reached an alert category. As the
average outside temperature during the investigation (14°C) was higher than the typical yearly
average in southern Germany, the number of hours in the different classes did correspond well
with the analysis performed by Lucas et al. (2000) for pig production in Portugal using indices
given by NWSCR (1976) and Ingram (1965).
Due to hot climatic outside conditions in summer 2003, the average daily weight gain of the
pigs in a growing-finishing stage between 50 and 90 kg was clearly reduced compared to the
average daily weight gain at comparable growing-finishing stages during four fattening
periods. Thereby, the weight gain during growing-finishing stage 50 to 70 kg was reduced by
about 47 g pig-1 d-1 with cooling and by about 134 g pig-1 d-1 without cooling. Though cooling
was thereafter exchanged between the two compartments, the average weight gain during
growing-finishing stage 70 to 90 kg was 46 g pig-1 d-1 lower without cooling but 38 g pig-1 d-1
higher with cooling, compared to the average weight gain of this growing-finishing stage.
According to investigations of Quiniou et al. (2000) and Huynh (2005), the reduction in
average daily weight gain was presumably caused by a lowered voluntary feed intake during
heat stress periods.
The evaporative fraction and time constant to reach steady state conditions were influenced by
environmental conditions, such as temperature and saturation deficit. The supplied water
amount evaporated completely during warm and dry indoor conditions (28°C, 53% RH), while
112
the evaporative fraction in steady state dropped to 89% for moderate indoor conditions (21°C;
69% RH) and to 65% for cold and humid indoor conditions (13°C, 83% RH). The information
about the evaporative fraction is crucial for the control and the simulation of control settings
for fogging and ventilation. Year-round simulations resulted in a water consumption of
2.5 L pig-1 d-1 if fogging was used on maximum settings. It was lowered by about 25% for an
optimized ventilation setting, namely a reduction of the ventilation capacity by one-third in
combination with an increased maximum temperature for onset of fogging. Thereby,
ventilation and fogging control influenced also the number of hours with indoor temperatures
above an upper limit of 24°C as well as the distribution of hours in specific temperature-
humidity categories.
Conclusions
In conclusion, high pressure fogging of water was an appropriate method for indoor air cooling
in the investigated pig facility. Main future tasks of an optimized control algorithm are to
optimize indoor conditions in terms of temperature, relative humidity and air flow, to minimize
water consumption and energy use, and to avoid wetting of surfaces.
References
Arbel A; Yekutieli O; Barak M (1999). Performance of a fog system for cooling greenhouses.
J. Agric. Eng. Res., 72, 129-136
Arbel A; Yekutieli O; Barak M (2003). Combination of forced ventilation and fogging systems
for cooling greenhouses. Biosystems Engineering, 84(1), 45-55
Berckmans D; Van Pee M; Goedseels V (1992). Evaluation of livestock environment by
simulation technique. ASAE paper No. 92-4055, St. Joseph, MI, USA, 25 p.
Gates R S; Timmons M B; Bottcher R W (1991a). Numerical optimization of evaporative
misting systems. Trans. ASAE, 34(1), 275-280
Gates R S; Usry J L; Nienhaber J A; Turner L W; Bridges T C (1991b). An optimal misting
method for cooling livestock housing. Trans. ASAE, 34(5), 2199-2206
Haeussermann A; Hartung E; Jungbluth T (2004). Development of innovative ventilation
control for pig facilities – method and first results. Agrartechnische Forschung, 10(1), 7-15
Haeussermann A; Hartung E; Jungbluth T (2005). Environmental effects of pig house
ventilation controlled by animal activity and CO2 indoor concentration. In: Precision Livestock
Farming ‘05 (Cox S, ed.), Wageningen Academic Publishers, 57-64
Haeussermann A; Hartung E; Gallmann E; Jungbluth T (2006). Influence of season, ventilation
strategy and slurry removal on methane emissions from pig houses. Agriculture, Ecosystems
and Environment 112, 115-121
Haeussermann A; Vranken E; Aerts J M; Hartung E; Jungbluth T; Berckmans D (2007a).
Cooling Effects and Evaporation Characteristics of Fogging Systems in an Experimental
Piggery. Biosystems Engineering, submitted
Haeussermann A; Hartung E; Jungbluth T; Vranken E; Aerts J M; Berckmans D (2007b).
Evaluation of control strategies for fogging systems in pig facilities. Trans. ASABE, in print
Huynh T T T (2005). Heat stress in growing pigs. PhD diss. Wageningen Institute of Animal
Science, Wageningen University, The Netherlands
Ingram D L (1965). Evaporative cooling in the pig. Nature, 207(4995), 415-416
Lucas, E M; Randall J M; Meneses J F (2000). Potential for evaporative cooling during heat
stress periods in pig production in Portugal (Alentejo). J. Agric. Eng. Res., 76(4), 363-371
113
Nienaber J A; Hahn G L; McDonald T P; Korthals R L (1996). Feeding patterns and swine

performance in hot environments. Trans. ASAE 39(1), 195-202
NWSCR (1976). Livestock Hot Weather Stress. Regional operations manual letter C-31-76.
National Weather Service Central Region, USA
Quiniou N; Dubois S; Noblet J (2000). Voluntary feed intake and feeding behaviour of group-
housed growing pigs are affected by ambient temperature and body weight. Livestock
Production Science, 63(3), 245-253
114
NEW TRENDS IN ANIMAL HOUSING IN GREECE:

GREENHOUSE TYPE LIVESTOCK BUILDINGS
C. Nikita – Martzopoulou1
1
Laboratory of Agricultural Structures and Equipment
Aristotle University of Thessaloniki, 54006 Thessaloniki, Greece, nikita@agro.auth.gr
Introduction
The lack of sufficiency in animal products is well recognized in Greece, although a lot of
effort was put, up today, in order to improve the livestock production indices. As a result of
this situation, the capital spent to import animal products is remarkably high (i.e. about 1.0
billion € in 1980 and 1.5 billion € in 2000).
However, the factors affecting the livestock development and productivity are the
genotype, the nutrition and the environment. The latter is the factor, which can be obviously
improved by replacing the old fashion livestock structures with economically and
environmentally acceptable modern buildings.
Greenhouse type livestock buildings

The livestock buildings in Greece are mainly traditional heavy structures with a high
initial cost of investment. Another problem for some livestock branches in Greece is the
pastoral raise. For example sheep and goats production for all Mediterranean countries is very
profitable. The last years in Greece sheep/goats production is declined due, probably, to the
fact that new farmers are not susceptible of the pastoral raise and they prefer housing the
livestock.
Substantial contribution to overcome the above problems is the use of light structures
such as the greenhouse type buildings.
These structures have the following advantages:
●Light structures ●Prefabricated, easily transported with minimum time of installation ●Low
cost of installation and maintenance ●Easy replacement of spoiled parts ●Better natural
ventilation and lighting ●Adapted insulation ●Safe sanitary conditions ●Easy cleaning
●Esthetic asset ●For their installation no permission from the urban planning is required.
Nowadays, the livestock buildings of greenhouse type tend to replace the traditional
heavy structures for the most species reared in Greece, sheep/goats, poultry, swine, dairy cattle
and mink animals.
A greenhouse type sheep house was firstly used in England during the winter 1979. It
was an experimental tunnel type greenhouse, with galvanized steel frame and polythene cover.
To improve the ventilation, the lower part of sides, in a height of 1m, was replaced by a
material called tensar. This is a polyethylene perforated permitting the fresh air to enter the
structure in the animal level. Along the roof of the tunnel a continuous opening, of 30cm
width, was used for the air outlet. The fronts were covered also by tensar to reduce the air
draughts. The dimensions of the structure were 10 x 30 m and 100 sheeps were confined.
During the winter the inside temperature was 1 – 2oC higher than the outside while the air
velocity inside the tunnel was 30% lower than outside. The humidity was the same, except the
days with no wind where the inside humidity was higher. The only problem was the roof
opening allowing the rain to wet the corridor of the structure. Generally, the function of this
structure was considered satisfactory and in nowadays is used extensively in Great Britain, for
sheep production.
These structures started to be installed also in Greece the last 7 years.
115
Construction materials
In the arch type structures the frame is of galvanized steel while in the single span type,
mainly used for mink confinement, could be also of wood.
The covering is usually a sandwich. The outer layer is made of PVC canvas
(550 – 600g/m2) of 10 years life, or of green PE (250µ thickness). The inner layer is made of
black-white PE (160 – 200µ thickness) to protect the structure from the water vapour. In
between a layer of glass wool (8cm thickness) assures insulation and amelioration of the inside
environment. As covering material a thin galvanized steel, corrugated, with a good insulator
(i.e. polyurethane) underneath may also be used.
The sides, in a height of 1m above the floor, are covered by sandwich panels. The panels
generally should meet the requirements of livestock buildings in fire protection, insulation,
mechanical resistance, water and air tightness, quick erection and economy.
The fronts are covered by PVC or panels.
Ventilation system
Natural or dynamic ventilation is used as in greenhouses. The natural ventilation is
achieved by the following openings: a. Continuous side openings of 1m width equipped by
PVC curtain and protection net. b.Continuous roof opening of 30cm minimum width for the air
outlet. To protect the inside space from the rainfall a cap over this opening is necessary. In
dynamic ventilation fans are used which sometimes are combined with cooling pads.
Technical specifications – Certification

In Greece, the legal condition for a greenhouse type livestock building installation is to
be provided by the certificate of quality from the Center of Agricultural Structures Control.
This Center has been authorized by the Ministry of Agriculture and it belongs to the Lab of
Agricultural Structures and Equipment of Aristotle University of Thessaloniki.
To certify such a structure the technical specifications described above must be kept and
the static study should be harmonized with the following eurocodes:
Eurocode 1- Basis of design and actions on structures.
Eurocode 3 – Design of steel structures.
Eurocode 5 – Design of wood structures.
Draft prENG 13031-1.1997. Greenhouses: Design and References.
Fig.1. Greenhouse type sheep house in GR Fig.2. Greenhouse type poultry house in GR
116
Diagnosis of the Air Quality for Matrices Raised in Collective Stalls and Individual Cages
Campos, J.A., Tinoco, I.F.F., Silva,J.N., Baêta, F.C., Cruz, V.F., Mauri, A.L.
Abstract
The quantification of the gaseous emissions in the agro-industrial systems has been,
nowadays, a world-wide concern, since they can affect the environment, the health of the
employees that works inside production complexes, and yet able to diminish animal’s
performance, by causing discomfort and illnesses. The main goal was the diagnoses of the air
quality (concentration of CO2, NH3 and CO), in two environments of gestation for swine
matrices. In one of them the matrices were kept in individual cages and another in collective
stalls, being both types of installations used in Brazil for production. The experiment was
carried out under winter conditions, season in which the concentration of gases inside the
installations is usually elevated due fact that the unit is more often closed. The gestation
building was composed of 50 cages, each one with one animal. The gestation in collective
stalls was composed of 3 jails, each one with 6 pregnant swine. Measurements of instantaneous
concentrations of ammonia (ppm), carbon monoxide (ppm) and carbon dioxide were done. The
gases were sampled four different times (09:00 am, 12:00 noon, 15:00 pm and 18:00 pm)
during daylight inside the hangars, and this happened during all the experimental period at the
height of the animals head. Experimental were data had been submitted to the regression
analysis. In both evaluated gestations, were found CO2 and NH3 concentrations levels below
the ones capable of affecting animals’ health. In numerical terms, the environment inside
collective stalls presented higher concentration of gases as compared to the one with individual
cages. Finally no CO was found inside both unities.
Key Words: concentration of gases, gestation, swine
Introduction
Surrounding air, which is source of oxygen for the animal metabolism, is, also, vehicle
of waste of the excess heat, water vapor and gases emitted by the animals and originating from
the decomposition of dejections, and dust. These factors can act polluting and modifying the
ideal air characteristics, causing an increase of the susceptibility to respiratory diseases of the
animals with consequent damage in the productive process (MACARI & FURLAN, 2001).
High ammonia concentrations in closed installations for pigs (>10ppm), may carises irritation
of the eyes and of respiratory membranes. It is also a chronic stressor that can affect the course
of infectious disease and to influence directly the growth of young animals (CURTIS, 1983).
Carbon dioxide (CO2) is a gas without odor and normally, it’s in air in a concentration near to
300 ppm (NI, 1998). In accordance with NADER et al. (2002), a 3000 ppm or higher
concentration the gas carbonic causes, to animals, an increase of the respiratory rhythm and
deeper breaths. In concentration of 40000 ppm or higher it can causes anxiety, followed by
vertigo and possible death. The CO2 produced by animals is directly related with its heat
production, and is function of its corporal weight, as well as the thermal environment. Studies
carried through with carbon dioxide (CO2) and carbon monoxide (CO) gives evidences that
from certain limits of concentration, say about: 3000 ppm of CO2 and 10 ppm CO these gases
affect considerably the health of swine (NADER et al, 2002). Therefore the concentration of
those gases in the environment is of vital importance because they do affect the animal health
directly (poisoning for gases) or indirectly (causing stress and consequently affecting the
immunity, propitiating the infection beside the point disease).
117

This work was developed during the period winter of 2005, inside of an industrial pork
production plant. The gestation building was composed of 50 cages, each on them with one
animal. The gestation in collective stalls was composed of 3 jails, each one with 6 pregnant
swine. Measurements of instantaneous ammonia concentrations (ppm), carbon monoxide
(ppm) and carbon dioxide (ppm) were done. Those gases were collected in four different times
(09:00 am, 12:00 noon, 03:00 pm and 06:00 pm) during daylight inside the hangars. This
happened during all the experimental period at the height of head of the animals.
Results and Discussions

In this study of gestations in collective stalls and cages, the average levels of CO2
concentration had been statistically different within the treatments only at 12:00 noon and
06:00 pm, being these values higher in collective stalls. In the gestations in cages a bigger
circulation of air exists, however it does not happen in the gestations in bay where the walls
turns difficult the ventilation, leading to a higher concentration of gases in this last one. The
levels of concentration CO2 had not exceeded the concentration of 800 ppm, values these
below the limit that can affect the health of the animal (3000 ppm), as cited by NADER et al.
(2002). About the evaluation of ammonia, the concentration had differed (P<0,05) within time
at 15:00 and 18:00 hours only. The concentration of NH3 had been higher in the collective
stalls. This is explained due to the fact that this installation happen to have an environment
surrounded by walls and the animals are free inside of it, leading to a bigger dispersion of
excrements and piss, that consequently facilitates to the volatilization of ammonia increasing
its concentration in air, as AARNINK, 1997. A research done by CURTIS (1983), NIOSH
(2005) and HAMILTON (1996), indicates that these values of concentration of ammonia do
not affect the health of the animal, being below the acceptable maximum levels (20 ppm).
Measurable values of CO not found in both evaluated gestations systems.
Table 1 - Average values of CO2 and NH3, in ppm, for the respective combinations of time (9
am,12noon,3 pm and 6 pm) and local (GE1: individual cages and GE2: collective stalls).
Gás Local H=9 am H=12 noon H=3 pm H=6 pm
GE1 646,8 A 510,4 B 663,7 A 735,6 B
CO2
GE2 655,6 A 684,2 A 684,6 A 853,6 A
GE1 1,1 A 1,5 A 2,0 B 0,3 B
NH3
GE2 2,4 A 2,1 A 4,6 A 1,7 A
The averages followed by the same letter on the column do not differ at a 5% level of probability for the Tukey
test.
Conclusions
In both evaluated gestations, were found CO2 and NH3 in concentrations below the
ones capable of affecting the animals’ health; In numerical terms, the environment inside
collective stalls presented higher concentration of gases as compared to the one with individual
cages. Finally no CO was found inside both unities.
References:
CAMPOS, J.A. Qualidade do ar, ambiente térmico e desempenho animal em dois tipos de
suinoculturas. Disssertação de tese de mestrado. Departamento de engenharia agrícola. Viçosa:
UFV, 2006. 69p.
CURTIS, S.E. Environmental management in animal agriculture. Ames. The Iowa State
University, 1993. 409p.
118
HAMILTON, T.D.C.; ROE J.M.; WEBSTER A.F.. The synergistic role of gaseous ammonia
in the aetiology of pasterella multocida induced atrophic rhinitis in swine. J Clin Microbiol 43,
1996. p 01-33.
MACARI, M.; FURLAN, R.L. Ambiência na produção de aves em clima tropical. Editado por
Iran José Oliveira da Silva, Piracicaba, SP; 2001. p.31-87.
NADER, A. S.; BARACHO, M. S.; NAAS I. A; SAMPAIO, C. A. P. Avaliação da qualidade
do ar em creche de suínos. IN: SEMINÁRIO: POLUENTES AÉREOS RUÍDOS EM
INSTALAÇÕES PARA PRODUÇÃO DE ANIMAIS. Campinas, São Paulo, Setembro de
2002. p 49 - p 56.
NI, J. Q.. Emmission of carbon dioxide and ammonia from mechanically ventilated pig house.
Ph.D. Thesis. Department of Agriculture Engineering, Catholic University of Leuven. Leuven,
Belgium, 1998. 227 p.
119
Diagnosis of the CO and CO2 concentration in the broiler chicken production in half-
acclimatized poultry facilities
Menegali, I.; Tinôco, I.F.F.; Baêta, F.C.; Guimarães, M.C.C.; Cordeiro,

M.B.
Abstract
For the conditions of the Brazilian poultry houses (opened) one becomes impracticable
to calculate the gas emission, a time that this has direct relation with the air speed, which is
very changeable in natural conditions. Thus the aim of this project is to diagnose the air quality
based on the concentrations of monoxide and carbon dioxide found in the environment,
associated to the surrounding thermal comfort in the production of broiled chicken. Two
different systems of ventilation had been used (with negative pressure/SVN and positive
pressure/SVP), under winter conditions, in the southern of Brazil. For evaluation of the thermal
environment, randomized blocks was used, in split-plot design, with 2 replications. The
experiment was developed from July to September, 2004, in integrated commercial poultry
facilities in the company “Perdigão Agroindustrial S.A.”, available for light female breeding.
The top gas concentrations were detected during the morning for the both treatments (SVN and
SVP), with concentrations levels of CO and CO2 still in acceptable limits for the good
development of the herd.
Key Words: Thermal comfort, carbon monoxide, carbon dioxide
Introduction
The volumes of reached production and exportation had detached the Brazilian poultry
keeping, as much for the generation of income to the agricultural way, how much for it offers
to the population a protein of high quality and health (UBA, 2006). The supply of heat for the
birds is essential in the initial phases of life, when risks of stress for cold exists (Tinôco, 1996).
On the other hand, the ventilation presents important function, mainly for sanitary reasons,
making the renewal of air, to prevent undesirable concentrations of gases inside of poultry
houses (Silva & Sevegnani, 2001).When airborne pollutants alter the ideal characteristics of the
air, it increases susceptibility of poultry breathing diseases and damages in the production
process (Tinôco, 2004). The carbon dioxide is a gas without odor present in the atmosphere in
a concentration next to 300 ppm (Macari & Furlan, 2001). Its concentration can be increased in
the interior of the installation in environments badly ventilated; therefore this gas is deriving,
mainly, of the breath of the animals and heaters where combustion occurs. It also can be set
free for the decomposition of excretas. The Carbon Monoxide lightly less dense than air, is
odorless , and generally has its concentration increased in an animal installation from the
incomplete combustion of a fuel, which had an imperfection in the adjustment of the heater,
together with inadequate ventilation of the system. For poultry installations, Wathes (1999)
recommends the limit of 3.000 ppm for CO2 and 10 ppm for CO as the maximum for
continuous exposition of the birds in the installations. The rising of the conditions of the air
quality in each one of the ventilation systems adopted by the Brazilian poultry keeping, for
each different area and climatic regions of the country, is an imperative and urgent need of
animal production.
Methodology
The experiment was developed from July to September, 2004, in integrated commercial
poultry facilities in the company “Perdigão Agroindustrial S.A.”. This experiment was
conducted in sheds in the same area of production, positioned side by side, orientation east-
west, in the dimensions of 100 x 12 m, a height of 2.80 m. Instantaneous measurements of
concentrations of carbon dioxide (during all the productive period of the birds) and carbon
120
monoxide had been carried through (in the two first weeks - heating phase), in ppm, to the
level of the birds (0,30 m, that is, approximately the height of breath of the same ones), in the
schedules of 9:00 AM and 3:00 PM.
The values of UR had been between 60 and 70% (average for first week 64,7 in SVN and
65.8% in the SVP) with the lesser values occurring around 2:00 PM. For the ITGU, in all the
schedules the values had been bigger in treatment SVN in the two first weeks of life of the
birds. Such values are adjusted to the production of the birds for the first week, in the second
week, treatment SVN presented superior values than 85%, in the majority of the schedules,
what it indicates discomfort for the birds, as the values found in research carried through for
Teixeira (1983). The concentrations of found carbon dioxide in the two treatments are inside of
the acceptable limits for poultry facilities, once recommends the limit of 3.000 ppm as the
maximum for continuous exposition of the animals in the installations (Nader et al., 2002). For
the concentrations of carbon monoxide, the values found in two environments had been inside
of a considered band of security, not representing risks to the lodged birds. In table 1 the
average values of the concentration of gases are presented.
TABLE 1. Averages of CO and CO2 concentrations, in ppm.
CO concentration (ppm) CO2 concentration (ppm)

Treatments
9:00 AM 3:00 PM 9:00 AM 3:00 PM
SVN 8,79 aA 3,93 aA 1897,2 aA 1400,8 bA
SVP 2,65 aB 1,87 aA 2062,3 aA 1693,1 bA
The averages following by at least a same small letter in the line and capital letter in the column doesn't
differ amongst themselves at the level of 5% of probability for the test of Tukey
Conclusion
The values of thermal comfort surrounding relative humidity of air and ITGU in the interior of
the installations had presented similar behavior for two treatments SVN and SVP, with
satisfactory levels of relative humidity for a good development of the breeding. The
concentrations of monoxide and carbon dioxide, had not presented harmful values for the two
systems that is, remained inside of the acceptable limits of quality of air, as much in the period
of the morning even in the afternoon.
References:
Macari, M., Furlan, R.L. Ambiência na produção de aves em clima tropical. Editado por
Iran José Oliveira da Silva – Piracicaba – SP: 2001. 31-87p.
Nader, A.S.; Baracho, M.S.; Nääs, I.A ; Sampaio, C.A.P. In: Seminário: Poluentes Aéreos e
ruídos em instalações para produção de animais. Campinas, São Paulo. 2002. 49-56p.
Silva, I.O., Sevegnani, K.B. Ambiência na produção de aves em clima Tropical. Editado por
Iran Jose Oliveira da Silva. FUNEP, 2001. v.2. 185p.
Teixeira, V.H. Estudo dos índices de conforto em duas instalações de frango
de corte para as regiões de Viçosa e Visconde do Rio Branco - MG. Viçosa,
1983. 62p. Dissertação (Mestrado) – Universidade Federal de Viçosa.
Tinôco, I.F.F. Conforto ambiental para aves/ponto de vista do engenheiro. In. Simpósio Goiano
de Avicultura (2: 1995: Goiânia-Goiás). Anais do II Simpósio Goiano de Avicultura –
Goiânia, 1996. 47-56p.
121
Tinôco, I.F.F. A granja de frangos de corte. Produção de frangos de corte / editado por Ariel
Antônio Mendes, Irenilza de Alencar Nääs, Marcos Macari – Campinas: FACTA, 2004.
356p.
UBA. Relatório anual 2005/2006. http://www.uba.org.br/ubanews_files/rel_uba_2005_06.pdf.
10 Jan. 2007.
Wathes, C.M. World Poultry, v.15, n.3, p.17-19, 1999.
122

Animal Housing in Hot Climates PDF

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Animal Housing in Hot Climates PDF

Diunggah oleh

Hak Cipta:

Format Tersedia

CIGR Workshop “Animal Housing in Hot Climate”, Cairo, Egypt, April 1-4, 2007

Workshop of CIGR Section II

Animal Housing in Hot Climates

Dr. Daniel Berckmans Belgium daniel.berckmans@biw.kuleuven.be

Dr. Mohamed Hatem Egypt hatem@cu.edu.eg

Dr. Panos Panagakis Greece ppap@aua.gr

Dr. Vasco Fitas da Cruz Portugal vfc@uevora.pt

Dr. Paolo Zappavigna Italy paolo.zappavigna@unibo.it

Issues related to livestock housing under hot climatic conditions including

Heat and moisture production

Animal heat production, W

Figure 2 Total heat production for pigs, CIGR 2002

Voluntary feed intake, g pig d

Start to Increase Start to Increase Increase Increase

Animal response to heat stress

Consequences of high temperatures

Traditional ways of alleviating heat stress

It is essential (and economically justifiable) to install well-controlled sprinkling systems in

Improved housing design and management practices

Innovative technologies to alleviate heat stress

Feeding 2 maintenance 1.5 maintenance

Flooring Straw Concrate

Air speed Still Draught Still Draught

Stocking Group Single Group Single

Modeling buildings and climate control systems

Outdoor climatic data

Summary and conclusions

Behavioural changes related to air temperature in sows kept in different

M. Barbari and M. Bianchi

Keywords: cooling systems, pregnant sows, heat stress, behaviour.

Materials and Methods

Solution “a” Solution “b”

Figure 3. Use of different areas in relation to air temperature.

Figure 5. Use of the automatic showering cages in relation to temperatures.

Lying Ahead Lying Back Other Activities Temperature

Use of different cooling systems by pregnant sows in experimental pen

M. Barbari, L. Conti and S. Simonini

Keywords: cooling systems, pregnant sows, heat stress, behaviour.

Materials and Methods

T°C<22 22<T°C<26 26<T°C<30 T°C>30

Figure 3. Use of different areas from the sows

Thermoregulatory responses to high ambient temperature in growing pigs:

D. Renaudeau, J.L., Gourdine, C. Anaïs.

INRA Station de Recherches Zootechniques, Domaine de Duclos, 97170 Petit Bourg.

1500 1500 d-10-d0

Respiratory rate, bpm

Morrison, S. R. and L. E. Mount. 1971. Adaptation of growing pigs to changes in

HEAT STRESS IN DAIRY COWS: A REVIEW OF THE HEAT LOAD

Fig. 2: Effect of the environmental

Effects of the climatic parameters on the productive response

Fig. 3: Effect of the environmental temperature on milk production according to

Fig. 4: Effect of the exposure to high temperatures on milk production according

Yeck e Stewart Baeta et al.

Temperature Humidity Index (THI)

Fig. 9: Increase of the rectal temperature as a function of the

Air speed in combination with other parameters

1 forced ventilation Calamari

natural ventilation Calamari

Fig. 11: Respiratory rhythm average values as measured at different air

can be compared to a similar work carried out by Calamari et al. (1994).

Fig. 14: Influence of the air speed on the rectal temperature

Author and air speed Rectal temperature Respiratory Rhytm

Fig. 15: Milk production variation as a function of the average rectal