Published November 25, 2014
Effects of reduced nocturnal temperature on pig
performance and energy consumption in swine nursery rooms1,2,3
L. J. Johnston,*4 M. C. Brumm,† S. J. Moeller,‡ S. Pohl,§ M. C. Shannon,# and R. C. Thaler§§
*West Central Research and Outreach Center, University of Minnesota, Morris 56267;
†Haskell Agricultural Laboratory, University of Nebraska, Concord 68728; ‡Department of
Animal Sciences, The Ohio State University, Columbus 43210; §Department of Agricultural and Biosystems
Engineering, South Dakota State University, Brookings 57007; §§Department of Animal and Range Sciences, South
Dakota State University, Brookings 57007; and #Department of Animal Sciences, University of Missouri, Columbia 65211
ABSTRACT: The objective of this investigation was
to determine the effect of a reduced nocturnal temperature (RNT) regimen on performance of weaned pigs and
energy consumption during the nursery phase of production. The age of weaned pigs assigned to experiments
ranged from 16 to 22 d. In Exp. 1, 3 stations conducted 2
trials under a common protocol that provided data from
6 control rooms (CON; 820 pigs) and 6 RNT rooms
(818 pigs). Two mirror-image nursery rooms were used
at each station. Temperature in the CON room was set
to 30°C for the first 7 d, then reduced by 2°C per week
through the remainder of the experiment. Room temperature settings were held constant throughout the day and
night. The temperature setting in the RNT room was the
same as CON during the first 7 d, but beginning on the
night of d 7, the room temperature setting was reduced
6°C from the daytime temperature from 1900 to 0700 h.
The use of heating fuel and electricity were measured
weekly in each room. Overall, ADG (0.43 kg), ADFI
(0.62 kg), and G:F (0.69) were identical for CON and
RNT rooms. Consumption of heating fuel [9,658 vs.
7,958 British thermal units (Btu)·pig-1·d-1] and electricity (0.138 vs. 0.125 kilowatt-hour (kWh)·pig-1·d-1]
were not statistically different for CON and RNT rooms,
respectively. In Exp. 2, 4 stations conducted at least 2
trials that provided data from 9 CON rooms (2,122 pigs)
and 10 RNT rooms (2,176 pigs). Experimental treatments and protocols were the same as Exp. 1, except
that the RNT regimen was imposed on the night of d 5
and the targeted nighttime temperature reduction was
8.3°C. Neither final pig BW (21.8 vs. 21.5 kg; SE =
0.64), ADG (0.45 vs. 0.44 kg; SE = 0.016), ADFI (0.61
vs. 0.60 kg; SE = 0.019), nor G:F (0.75 vs. 0.75; SE =
0.012) were different for pigs housed in CON or RNT
rooms, respectively. Consumption of heating fuel and
electricity was consistently reduced in RNT rooms for
all 4 stations. Consumption of heating fuel (10,019 vs.
7,061 Btu·pig-1·d-1; SE = 1,467) and electricity (0.026
vs. 0.021 kWh·pig-1·d-1; SE = 0.004) were lower (P <
0.05) in the RNT rooms compared with CON rooms.
This represents a 30% reduction in heating fuel use and
a 20% reduction in electrical use with no differences in
pig growth performance or health. From these experiments, we conclude that imposing a RNT regimen from
1900 to 0700 h is effective in reducing energy costs in
the nursery without compromising pig performance,
which will reduce production costs and decrease emissions of greenhouse gases.
Key words: energy use, pigs, reduced nocturnal temperature, room temperature
© 2013 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2013.91:3429–3435
doi:10.2527/jas2012-5824
INTRODUCTION
1Financial support was provided by the National Pork Board (Des
Moines, IA) and the Minnesota Pork Board (Mankato, MN).
2The authors appreciate the technical assistance of A. M. Hilbrands
in completion of this project.
3The authors acknowledge the assistance of the NCERA-219
Committee on Swine Production Management to Enhance Animal
Welfare in developing and designing these experiments.
4Corresponding author: [email protected]
Received September 6, 2012.
Accepted April 4, 2013.
Global climate change has brought increased
scrutiny on greenhouse gas (GHG) emissions resulting from livestock production (Food and Agriculture
Organization, 2006; de Vries and de Boer, 2010). As a
result, the swine industry has evaluated the production
of GHG in modern production systems with a focus on
3429
3430
Johnston et al.
reducing emissions (Thoma et al., 2011). One approach
to reducing GHG emissions is to reduce consumption of
fossil fuels in pork production systems. Reducing consumption of fossil fuels can result in an economic benefit to pork producers as long as pig performance does
not deteriorate. An obvious way to reduce consumption
of fossil fuel is to reduce ambient temperature within
buildings that house pigs. This approach will likely not
have a dramatic effect on reducing emissions of GHG,
but it will contribute to incremental improvements toward reducing environmental effects of pork production
(Thoma et al., 2011).
In the 1980s and early 1990s, researchers demonstrated that reducing room temperature during nighttime
hours (reduced nocturnal temperature, RNT) significantly reduced consumption of heating fuel (Swinkels et al.,
1988; Brumm and Shelton, 1991) with variable effects on
growth performance of nursery pigs (Brumm et al., 1985,
1995; Swinkels et al., 1988). The commercial swine industry has not adopted RNT regimens for various reasons.
Commercial swine production facilities and equipment
and genetic potential of modern pigs have changed dramatically in the 15 yr since RNT was last studied. The
increased focus on environmental sensitivity of pork production systems and the constant drive to reduce production costs dictates a reevaluation of RNT regimens under
modern conditions. Therefore, 2 studies were conducted
to determine the effects of a RNT regimen on consumption of fossil fuel and pig performance with current genetic lines of pigs in modern facilities. We hypothesized
that RNT would reduce consumption of fossil fuel with
few or no detrimental effects on pig performance.
MATERIALS AND METHODS
The experiments reported herein were conducted
in research facilities owned and operated by cooperating universities or in commercial nursery facilities with
oversight from university faculty. In all instances, research protocols were approved before initiation of the
experiment by the Institutional Animal Care and Use
Committees of each respective university.
Experiment 1
This initial study was conducted as a collaborative
effort among researchers at the University of Minnesota,
University of Missouri, University of Nebraska, and
South Dakota State University. Unanticipated problems
with facility design at South Dakota State University necessitated deleting data from this site, so data reported
are from 3 participating universities. At each station, 2
nursery rooms in the same building were used simultaneously for each replicate of the experiment. Pigs were
assigned randomly to rooms on the day of weaning. The
average age of pigs at weaning ranged from 16 to 22 d.
Each room was assigned randomly to 1 of 2 temperature treatments. Target temperature in the control room
(CON) was set to 30°C for the first 7 d, then reduced
by 2°C per week through the remainder of the experiment. Room temperature settings were held constant
throughout the day and night. The temperature setting in
the RNT room was the same as CON during the first 7
d of the experiment. Beginning on the night of d 7, the
room temperature setting was reduced 6°C from the daytime temperature from 1900 to 0700 h. Rooms assigned
to the RNT treatment were allowed to gradually cool to
the new temperature set point by changing the activation
temperature of the heating furnace. Ventilation settings
were not altered, and no attempt was made to rapidly
cool rooms to the nighttime temperature. At 0700 h, the
room temperature setting was returned to the same as the
CON room. Daytime temperature settings in the RNT
room were reduced 2°C per week similar to the CON
room. Each station conducted 2 replicate trials with room
temperature treatments switched between rooms for the
second trial (Table 1). With the exception of room temperature settings, pigs in CON and RNT were fed and
managed similarly as per the standard operating procedures of each station. All pigs were provided ad libitum
access to feed and water throughout the experiment.
Total weight and number of pigs delivered to each
room were recorded as was the total weight and number
of pigs completing the experiment. Similarly, total weight
of feed consumed in each room was recorded during the
experiment. From these data, ADG, ADFI, and G:F were
calculated over the entire nursery period. The experimental period lasted 35 d in Nebraska and 42 d in Missouri
and Minnesota. Heating fuel (propane or natural gas)
consumption was measured directly with independent gas
meters for each room or was calculated from recorded operation time for each furnace. One liter of propane was assumed to contain 24,186 British thermal units (Btu) and
1 m3 of natural gas was assumed to contain 35,316 Btu for
the calculation of heating fuel use (Midwest Plan Service,
1990). Use of electricity was recorded by a commercial
meter installed for each room. Pig mortality and incidence
of pig treatments were recorded for each room.
Experiment 2
This study was conducted as a collaborative effort among researchers at the University of Minnesota,
University of Missouri, The Ohio State University, and
South Dakota State University. Each participating station
completed at least 2 replicate trials (Table 2). Each trial
lasted 28 to 42 d depending on the pig production flow at
the research location for each station. Pig age at the start
3431
Reduced nocturnal temperature for nursery pigs
Table 1. Participating stations, time periods for each
replicate, and size of each replicate (Exp. 1)
1
Start
End
No. of Number of pigs/room
date
date
pigs
CON
RNT
December 2004 January 2005 118
59
59
November 2005 January 2006 120
60
60
Missouri
February 2006 March 2006 240
120
120
March 2006
April 2006 240
120
120
Minnesota
June 2007
July 2007
415
208
207
February 2008 March 2008 505
253
252
1CON = control room; RNT = reduced nocturnal temperature room.
Station
Nebraska
of the trial ranged from 16 to 22 d. Data from 1 CON
room in South Dakota were excluded from all analysis
because of a malfunction of the temperature controller. A
computer malfunction rendered fuel use data inaccessible
in the first 2 replicate trials conducted in Minnesota, but
data on pig performance and room temperature were of
good quality and were included in the analysis. Both heating fuel and electrical use data were recorded for the last
2 replicate trials in Minnesota. The commercial nursery
used in Ohio recorded the use of heating fuel but was not
equipped with electric meters for each room, so electrical
use data were not recorded in Ohio.
One of 2 temperature treatments was imposed randomly on 2 identical nursery rooms at each participating station. Temperature treatments were CON and RNT.
In the CON rooms, temperature controllers were set at
30°C for the first week and reduced by 2°C per week
throughout the remainder of the trial. Daytime and nighttime temperature settings were the same in CON rooms.
In the RNT rooms, temperature controllers were set at
30°C for the first 4 d after the arrival of pigs. Beginning
on d 5, daytime temperatures were set to equal the CON
room, but temperature controllers were set to reduce
nighttime temperature by 8.3°C from the daytime temperature from 1900 to 0700 h. Daytime temperature in
the RNT room was also reduced 2°C per week during
the remainder of the study. Consumption of heating fuel
(propane or natural gas) and electricity in each room
was recorded weekly as described for Exp. 1. Operation
time for lights was kept constant between CON and RNT
rooms as were supplemental heat sources (heat lamps
or mats) if used. Room temperatures were recorded at
pig level in 2 or 3 locations in each room throughout
the experiment using automated data recording devices
(Hobo Pro V II, Onset Computer Corp., Bourne, MA).
Temperatures were recorded every 5 min throughout
the experiment. Recorded temperatures were averaged
across each recorder at each time, and 1 average temperature for the daytime (0700 to 1900 h) and nighttime
(1900 to 0700 h) was used in data analysis.
Pigs were assigned randomly to experimental rooms.
Temperature treatments in the first replicate were assigned randomly to rooms. Room temperature treatments were switched between rooms for the second
trial at each location. Within rooms, pigs were assigned
randomly to pens based on the normal operating procedures of the unit. At each station, total number and
weight of pigs entering and leaving each room were recorded. In addition, total consumption of feed by pigs in
each room was recorded. At the University of Missouri
and South Dakota State University, individual pig BW
was recorded on all pigs at entry to the nursery room
and weekly throughout the experiment. Similarly, feed
disappearance on a pen basis was recorded weekly. At
the University of Minnesota, pig BW and feed disappearance on a pen basis were recorded on a random
subsample of pigs (15 pens of 9 pigs/pen in each room)
weekly throughout the experiment. Pig BW, but not feed
intake, were recorded on a pen basis weekly throughout
the experiment at the Ohio station. The number of pigs
that died, pigs that received individual treatments for ill
health, and total number of days pigs received individual health treatments were recorded for each room. Pigs
were fed nutritionally adequate diets as per normal oper-
Table 2. Participating stations, time periods for each replicate, and size of each replicate (Exp. 2)
Start date
End date
No. of pigs1
January 2010
February 2010
700
March 2010
May 2010
720
South Dakota
January 2010
February 2010
50
February 2011
March 2011
100
Missouri
March 2010
April 2010
180
December 2010
January 2011
180
Minnesota
December 2009
January 2010
536 (270)
February 2010
March 2010
595 (270)
November 2010
December 2010
607 (270)
January 2011
February 2011
630 (286)
1Number in parentheses indicates number of pigs included in weekly performance measures.
2CON = control room; RNT = reduced nocturnal temperature room.
3Data for this room were lost due to an errant temperature controller.
Station
Ohio
Number of pigs/room1,2
CON
RNT
350
350
360
360
—3
50
50
50
90
90
90
90
268 (135)
268 (135)
297 (135)
298 (135)
303 (135)
304 (135)
314 (143)
316 (143)
3432
Johnston et al.
Table 3. Effect of reduced nocturnal temperature (RNT)
on nursery pig performance and energy use in nursery
rooms (Exp. 1)
Table 4. Effect of reduced nocturnal temperature (RNT)
on nursery pig performance and energy use in nursery
rooms (Exp. 2)
Trait1
CON2
RNT2
SE3
No. of rooms
6
6
—
Initial pig BW, kg
6.2
6.2
0.13
Final pig BW, kg
23.7
23.7
1.04
ADG, kg
0.43
0.43
0.022
ADFI, kg
0.62
0.62
0.025
G:F
0.69
0.69
0.011
Heating fuel used, Btu·pig-1·d-1 9,658
7,958
1,662
Electricity used, kWh·pig-1·d-1
0.138
0.125
0.014
1Btu = British thermal unit; kWh = kilowatt-hour.
2CON = control room; RNT = reduced nocturnal temperature room.
3Pooled SE.
Trait
CON1
RNT1
SE2
No. of rooms
9
10
—
Initial pig BW, kg
6.2
6.2
0.03
CV of initial pig BW, %
19.0
18.7
0.18
Final pig BW, kg
21.8
21.5
0.64
CV of final pig BW, %
16.2
16.0
0.18
ADG, kg
0.45
0.44
0.016
ADFI, kg
0.61
0.60
0.019
G:F
0.75
0.75
0.012
Heating fuel used,3 Btu·pig-1·d-1 10,019a
7,061b
1,467
Electricity used,4 kWh·pig-1·d-1
0.026a
0.021b
0.004
a,bWithin a row, means with a different superscripts differ (P < 0.05).
1CON = control room; RNT = reduced nocturnal temperature room.
2Pooled SE.
3No. of observations were 7 and 8 for control (CON) and RNT, respectively. Btu = British thermal units.
4No. of observations were 5 and 6 for CON and RNT, respectively.
kWh = kilowatt-hour.
ating procedures at the research site. Pigs had ad libitum
access to feed and water throughout the experiment.
Statistical Analysis
Statistical analysis of the data was accomplished
using the PROC MIXED procedure with repeated measures in time where appropriate (SAS Inst. Inc., Cary,
NC). The initial statistical model included the fixed effects of temperature treatments (CON or RNT), station,
and their interaction. When the interaction was not significant, it was removed from the statistical model for
the final analysis. Replicate nested within station was
included in all statistical models as a random effect.
Room served as the experimental unit in the statistical
analysis. Separation of means was accomplished using
the PDIFF option of SAS with the Tukey adjustment
for multiple comparisons. Enumeration data such as
incidence of mortality and morbidity collected in Exp.
2 were analyzed using χ2. The level of statistical significance was set at P < 0.05, with P < 0.10 indicating
a significant trend.
RESULTS
Experiment 1
Significant effects of station were observed for
many of the response criteria in this experiment. Station
effects were not the focus of this experiment and are not
presented. More importantly, there were no significant
station by treatment interactions for any of the response
criteria, which indicates a consistent response to room
temperature treatments across all participating stations.
Room temperature treatments had no influence on final
pig BW or growth performance of pigs over the study
(Table 3). Pig mortality (CON = 1.1%; RNT = 1.0%)
was not affected by temperature treatments at the 2 sta-
tions that reported mortality data. Implementing the
RNT regimen reduced consumption of heating fuel by
18% and electricity by 9%. However, these differences
were not statistically significant.
Experiment 2
Similar to Exp. 1, significant station effects were
observed for many response variables, but these were
considered of minimal interest and are not presented.
No significant treatment by station interactions existed,
so only the main effects of room temperature treatments are presented. Initial and final BW of pigs were
not different between CON and RNT rooms (Table 4).
Likewise, within-pen variation of pig BW measured by
the coefficient of variation was not different between
room temperature treatments. Pig performance over the
entire experimental period was nearly identical across
CON and RNT rooms. Neither BW of pigs (Fig. 1) nor
ADFI of pigs (Fig. 2) were different during any measurement period of the experiment. Consequently, ADG
and G:F were not different for pigs housed in CON or
RNT rooms during any period of the experiment (data
not shown). The number of sick pigs (41 vs. 49; χ2 =
10.3, df = 12; P > 0.50), the number of days sick pigs
were treated (98 vs. 97; χ2 = 12.9, df = 13; P > 0.40),
and the number of pigs that died (32 vs. 26; χ2 = 4.2, df =
8; P > 0.80) were not different in CON and RNT rooms,
respectively.
Effects of station and room temperature treatments
interacted (P < 0.01) to influence temperatures recorded
in rooms (data not shown). In all instances, these were
Reduced nocturnal temperature for nursery pigs
3433
Figure 1. Effect of control (CON) and reduced nocturnal temperature
(RNT) regimens on changes in pig BW throughout Exp. 2. See online version
for figures in color.
Figure 2. Effect of control (CON) and reduced nocturnal temperature
(RNT) regimens on ADFI of pigs throughout Exp. 2. See online version for
figures in color.
interactions of magnitude, not direction, which suggests
that desired differences between treatments within station
were achieved as designed in the study. Room temperatures recorded during both days and nights were not different across CON and RNT rooms during the first 4 d
of the experiment (Fig. 3). Throughout the ensuing 31 d,
temperature in RNT rooms at night was lower (P < 0.01)
than daytime temperatures in CON and RNT rooms and
lower than nighttime temperatures in CON rooms. The
magnitude of difference between RNT and CON nighttime temperatures seemed to decline as the study progressed. Daytime temperatures in RNT rooms were
cooler (P < 0.05) from d 15 to 21 and tended (P < 0.10)
to be cooler during the other measurement periods than
daytime temperatures in the CON rooms. The cooler temperatures in RNT rooms were associated with a reduction
(P < 0.05) in the quantity of heating fuel and electricity
consumed in RNT compared with CON rooms (Table 4).
were noted for any response criteria. The absence of this
interaction suggests responses to the RNT regimen were
consistent in direction and magnitude across stations. In
both studies at every station, pig performance was not affected by implementation of the RNT regimen. Similarly,
numerical reductions in consumption of fossil fuel were
observed in RNT rooms at every station except Nebraska
in Exp. 1, where there was no difference. This consistency
of findings lends confidence to the results of these studies.
Pig performance and health were unaffected by the
RNT regimens imposed in both studies reported herein.
Other researchers using a similar protocol have imposed
a 10°C nighttime reduction for 12 h (Swinkels et al., 1988;
Brumm and Shelton, 1991) or a 6°C nighttime reduction
for 16 h (Brumm et al., 1995) and reported no differences
in ADG, ADFI, G:F, or health of nursery pigs similar to the
8.3°C reduction imposed in Exp. 2. In contrast, reduced
nighttime temperatures of 6°C (Brumm and Shelton, 1988)
or 10°C (Brumm et al., 1985) have elicited increased vol-
DISCUSSION
Experiment 1 was designed as a validation study to
evaluate the feasibility of a RNT regimen with a 6°C temperature reduction imposed 7 d after pigs arrived in the
nursery. This validation study was conducted to determine if responses to RNT documented in the late 1980s
were repeatable with younger weaning ages, improved
genetics, and modern nursery diets. Because there were
no detrimental effects on pig performance and apparent
health of pigs was not affected by the RNT regimen imposed, a second experiment with a more aggressive RNT
regimen was conducted. In Exp. 2, the RNT protocol was
imposed on the fifth day after pigs arrived in the nursery,
and target temperature at night was reduced 8.3°C from
the daytime temperature setting. In both experiments, significant station effects were observed for most response
variables, but these effects are not presented because
they were considered of minimal interest. More importantly, no station by room temperature treatment effects
Figure 3. Average daytime and nighttime temperatures in control (CON)
and reduced nocturnal temperature (RNT) rooms over the course of Exp. 2.
Temperature data were truncated at d 35, which was the last day all stations
had data to report. Within time period, aRNT-Night differs from all others (P <
0.001), RNT-Day tends to differ from CON-Day (P < 0.10); bRNT-Night differs from all others (P < 0.001), RNT-Day differs from CON-Day (P < 0.05);
cRNT-Night differs from all others (P < 0.001), RNT-Day tends to differ from
CON-Day (P < 0.10); dRNT-Night differs from all others (P < 0.001), RNTDay tends to differ from CON-Day (P < 0.10). Pooled SE = 0.14. See online
version for figures in color.
3434
Johnston et al.
untary feed intake and, consequently, improved ADG of
nursery pigs with no improvements in efficiency of BW
gain. When Brumm et al. (1985) imposed nighttime temperature reductions immediately on entry of the pigs to the
nursery (5°C for 3 d followed by 10°C for remaining days),
ADFI and ADG of pigs increased but so did the occurrence
of pig removals due to death and unthriftiness. In the current study, the RNT regimen was imposed 7 d (Exp. 1) or 4
d (Exp. 2) after pigs entered the nursery with no evidence
of compromised pig health. These observations suggest
that pigs may benefit from a brief acclimation period of 4
to 7 d in the nursery to recover from weaning and transport stresses and establish consistent feed intake before a
RNT regimen can be imposed successfully. Although not
studied directly in the current experiments, it seems prudent to let the nursery room cool gradually to the reduced
nighttime temperature rather than increasing ventilation
rates to rapidly drive room temperature down to the lower
nighttime target. Gradual temperature reductions will allow pigs ample time to adjust to temperature changes and
will not waste heat energy that may have been recently introduced into the room by furnaces.
The similarity in pig performance between CON and
RNT regimens implemented in both experiments is not
surprising when one considers temperature preferences of
pigs. Using operant conditioning approaches, researchers
(Baldwin and Ingram, 1968; Baldwin, 1979; Morrison
et al., 1987) have demonstrated that nursery pigs prefer
cooler room temperature at night compared with daytime.
Bench and Gonyou (2007) observed that pigs as young
as 12 to 14 d controlled room temperature when provided with controllers in the pen and chose a nighttime
temperature that was about 2.5°C cooler than daytime
temperatures. Pigs generate less heat at night, which can
be partially explained by the decrease in activity of pigs
during night (van der Hel et al., 1984; Verstegen et al.,
1986). Because pigs prefer cooler nights and generate less
heat during the night, it seems logical to operate nursery
rooms cooler at night than during daytime.
Average daytime and nighttime temperatures were
similar in both CON and RNT rooms during the first
4 d of EXP. 2 as outlined in the experimental protocol
imposed. These first 4 d in the nursery were important
for pigs to become acclimated to their new surroundings
and establish a reliable pattern of feed intake. During
the remainder of the experiment, average temperature in
the RNT rooms at night was significantly cooler than the
CON rooms or RNT rooms during daytime, which indicates successful implementation of the intended experimental treatments. Average daytime temperature in the
RNT room was supposed to be similar to CON rooms
but actually tended to be slightly cooler. This cooler
temperature is most likely due to the lag time required
each morning beginning at 0700 h for the RNT room to
warm up. During the first 1 to 2 h of each day, temperatures in RNT rooms were lower than CON rooms as they
warmed up to the new daytime target. Actual room temperature during this period is considered in the daytime
average room temperature for RNT rooms. Average
room temperature in the CON rooms was not different
between day and nighttime as expected.
The magnitude of difference in temperatures between RNT and CON rooms at night waned as the experiment progressed. After about the first 10 d of the RNT
regimen, we could not achieve the desired temperature
differential of 8.3°C. This decreased differential is most
likely a result of the increasing BW of pigs. According to
Harmon et al. (1997), sensible heat production increases
from about 60 Btu/h for pigs weighing 7 kg to 137 Btu/h
for pigs weighing 11 kg. The research of Brown-Brandl
et al. (2004) suggests sensible heat production increases
further to 217 Btu/h as pigs reach a BW of 22 kg. These
estimates could be reasonably increased 10% to account
for genetic improvements in pig growth rate and leanness
over the last 10 yr after Harmon and Brown-Brandl conducted their work. This suggests that sensible heat production increases from about 65 Btu/h when pigs weigh
7 kg to 240 Btu/h when pigs weigh 22 kg. This nearly
4-fold increase in heat production by pigs as they grew
surely contributed to the heat supply for the room and
likely explains why we could not achieve the desired differential in nighttime room temperatures between CON
and RNT. In Minnesota, after about 14 d on the RNT
regimen, furnaces in the RNT rooms did not run at all
from 1900 to 0700 h, suggesting that no supplemental
heat was necessary to maintain target room temperatures.
With no heater running, the only source of heat could be
the pigs themselves. Obviously, the ability to achieve this
temperature depression is a function of building design,
wall and ceiling insulation, ventilation rates, and pig size
and will vary tremendously among nursery barns.
Numerical reductions in the consumption of heating fuel and electricity were observed in RNT rooms
for 2 of 3 stations in Exp. 1, and statistically significant
reductions were observed in RNT rooms of all stations
that participated in Exp. 2. The high variation in energy
consumption data and the limited number of replications
in Exp. 1 limited our ability to declare statistical significance of the differences. However, the RNT regimen
reduced heating fuel consumption by 18% and electrical
use by 9%. In Exp. 2, with a more aggressive temperature reduction protocol, we used 30% less heating fuel
and 20% less electricity in the RNT rooms. The magnitude of savings observed in the current studies compares
favorably with energy savings for RNT regimens reported in the literature ranging from 15.5% (Brumm and
Shelton, 1988) to 35% (Swinkels et al., 1988). These
Reduced nocturnal temperature for nursery pigs
savings in energy consumption represent an economic
benefit through reduced production costs.
The reduction in energy consumption also results in
environmental benefits. In Exp. 2, the RNT regimen reduced consumption of heating fuel 2,958 Btu·pig-1·d-1.
Assuming a nursery period of 35 d and that liquid propane heating fuel contains 24,168 Btu/L, the RNT regimen saved 4.28 L of propane and 6.9 kg CO2 equivalents (CO2-e; 1.62 kg per CO2-e per liter of propane
combusted; Deru and Torcellini, 2007) per pig leaving
the nursery. Similarly, RNT reduced electrical use by
0.005 kilowatt-hour (kWh)·pig-1·d-1, which can result
in an estimated reduction of 0.175 kWh per pig leaving the nursery. Assuming 1 kWh of electricity generated in the Midwest region of the United States causes
0.78 kg CO2-e of emissions (Environmental Protection
Agency, 2007), the savings in electricity realized in RNT
rooms could reduce GHG emissions 0.136 kg CO2-e per
pig leaving the nursery. Combined, the RNT regimen
could reduce CO2-e by 7.0 kg per pig leaving the nursery, which is a small fraction of the total CO2-e emitted in the production of consumer-ready pork. However,
Thoma et al. (2011) suggested that improvements in
the carbon footprint of consumer-ready pork will result
from incremental improvements in a suite of production
practices and technologies rather than major advances in
a specific aspect of production.
In summary, imposing a RNT regimen from 1900 to
0700 h does not have detrimental effects on pig health or
performance provided pigs are allowed a brief acclimation period to establish normal feed consumption. This
temperature management strategy will reduce consumption of heating fuel and electricity by as much as 30%
and 20%, respectively. These reductions in fossil fuel
consumption will reduce production costs and decrease
emission of greenhouse gases, which will improve the
carbon footprint of modern pork production.
LITERATURE CITED
Baldwin, B. A. 1979. Operant studies on the behavior of pigs and
sheep in relation to the physical environment. J. Anim. Sci.
49:1125–1134.
Baldwin, B. A., and D. L. Ingram. 1968. Factors influencing behavioral thermoregulation in the pig. Physiol. Behav. 3:409–415.
3435
Bench, C. J., and H. W. Gonyou. 2007. Temperature preference
in piglets weaned at 12 – 14 days of age. Can. J. Anim. Sci.
87:299–302.
Brown-Brandl, T. M., J. A. Nienaber, H. Xin, and R. S. Gates. 2004.
A literature review of swine heat production. Trans. ASAE
47:259–270.
Brumm, M. C., and D. P. Shelton. 1988. A modified reduced nocturnal temperature regimen for early-weaned pigs. J. Anim. Sci.
66:1067–1072.
Brumm, M. C., and D. P. Shelton. 1991. Two reduced nocturnal
temperature regimens for early-weaned pigs. J. Anim. Sci.
69:1379–1388.
Brumm, M. C., D. P. Shelton, and J. M. Dahlquist. 1995. Interaction
of diet composition and a reduced nocturnal temperature regimen in weanling pigs. J. Anim. Sci. 73:2518–2523.
Brumm, M. C., D. P. Shelton, and R. K. Johnson. 1985. Reduced
nocturnal temperatures for early weaned pigs. J. Anim. Sci.
61:552–558.
Deru, M., and P. Torcellini. 2007. Source energy and emission factors for energy use in buildings. Tech. Rep. No. NREL/TP-55038617. Natl. Renewable Energy Lab, Golden, CO.
de Vries, M., and I. J. M. de Boer. 2010. Comparing environmental
impacts for livestock products: A review of life cycle assessments. Livest. Sci. 128:1–11.
Environmental Protection Agency. 2007. Year 2007 summary tables:
EGRID2010 Vers. 1.1. www.epa.gov/cleanenergy/energy-resources/egrid/index.html. (Accessed 16 April 2012.)
Food and Agriculture Organization. 2006. Livestock’s long shadow:
Environmental issues and options. Food Agric. Organ. of the
United Nations, Rome.
Harmon, J. D., H. Xin, and J. Shao. 1997. Energetics of segregated
early weaned pigs. Trans. ASAE 40:1693–1698.
Midwest Plan Service. 1990. Mechanical ventilating systems for
livestock housing. Rep. No. MWPS-32. Midwest Plan Serv.,
Ames, IA.
Morrison, W. D., L. A. Bate, I. McMillan, and E. Amyot. 1987.
Operant heat demand of piglets on four different floors. Can. J.
Anim. Sci. 67:337–341.
Swinkels, J. W. G. M., E. T. Kornegay, and M. W. A. Verstegen. 1988.
The effect of reduced nocturnal air temperature and feed additives on the performance, immune response and scouring index
of weanling pigs. J. Anim. Physiol. Anim. Nutr. 60:137–145
Thoma, G., R. E. Martin, D. Nutter, R. Ulrich, C. Maxwell, J.
Frank, and C. East. 2011. National life cycle carbon footprint
study for production of US swine. http://www.pork.org/filelibrary/NPB%20Scan%20Final%20-%20May%202011.pdf.
(Accessed 15 April 2012)
van der Hel, W., M. W. A. Verstegen, W. Baltussen, and H. Brandsma.
1984. The effect of ambient temperature on diurnal rhythm
in heat production and activity in pigs kept in groups. Int. J.
Biometeorol. 28:303–315.
Verstegen, M. W. A., W. van der Hel, R. Duijghuisen, and R. Geers.
1986. Diurnal variation on the thermal demand of growing pigs.
J. Therm. Biol. 11:131–135.