Higher light intensity and mat temperature attract piglets to creep areas in farrowing pens

Abstract

Loose farrowing pens have been considered as alternatives to crates to enhance sow welfare. A major concern with pen systems is often higher piglet pre-weaning mortality, especially due to crushing by the sow. An optimal management of light and mat surface temperature may promote greater piglet use of the creep, which has been associated with reduced piglet crushing. A total of 108 sows and their piglets were studied in sow welfare and piglet protection pens on a commercial piggery, across two replicates. Sows were randomly assigned to pens arranged within two creep treatments (bright creep: 300 lx v. dark creep: 4 lx), considering mat temperature as a covariate. Twelve sows and their litters in each treatment (24 in total) had their behaviour continuously recorded for 72-h postpartum (pp), and four focal piglets per litter were weighed on the first and third days pp. In situ behaviour observations were performed daily (from 0800 to 1700 h) on all sows and their litters, every 15 min over 72-h pp to record piglet time spent in the creep, latency to enter the creep for the first time, latency for the litter to remain in the creep for at least 10 min, and piglet and sow use of pen areas immediately in front of (A2) and farthest from the creep (A3). Piglets with access to bright creeps spent on average 7.2% more time (P

Full text

Higher light intensity and mat temperature attract piglets to creep
areas in farrowing pens
G. M. Morello
1†a
J. N. Marchant-Forde
2
, G. M. Cronin
3
, R. S. Morrison
4
and J.-L. Rault
1b
1
Faculty of Veterinary and Agricultural Sciences, Animal Welfare Science Centre, University of Melbourne, Parkville, VIC 3010, Australia;
2
Livestock Behavior
Research Unit, USDA-ARS, 270 S. Russell Street, West Lafayette, IN 47907, USA
c
;
3
School of Life and Environmental Sciences, Faculty of Science, University of
Sydney, Camden, NSW 2570, Australia;
4
Rivalea Australia, Corowa, NSW 2646, Australia
(Received 20 March 2018; Accepted 13 November 2018; First published online 3 January 2019)
Loose farrowing pens have been considered as alternatives to crates to enhance sow welfare. A major concern with pen systems is
often higher piglet pre-weaning mortality, especially due to crushing by the sow. An optimal management of light and mat surface
temperature may promote greater piglet use of the creep, which has been associated with reduced piglet crushing. A total of 108
sows and their piglets were studied in sow welfare and piglet protection pens on a commercial piggery, across two replicates.
Sows were randomly assigned to pens arranged within two creep treatments (bright creep: 300 lx
v
. dark creep: 4 lx), considering
mat temperature as a covariate. Twelve sows and their litters in each treatment (24 in total) had their behaviour continuously
recorded for 72-h
postpartum
(pp), and four focal piglets per litter were weighed on the first and third days pp.
In situ
behaviour
observations were performed daily (from 0800 to 1700 h) on all sows and their litters, every 15 min over 72-h pp to record piglet
time spent in the creep, latency to enter the creep for the first time, latency for the litter to remain in the creep for at least 10 min,
and piglet and sow use of pen areas immediately in front of (A2) and farthest from the creep (A3). Piglets with access to bright
creeps spent on average 7.2% more time (
P<
0.01) in the creeps than piglets in pens with Dark creeps. In addition, for each
degree increase in mat temperature, piglets spent on average 2.1% more time (
P
<0.01) in the creep. Piglets in pens with bright
creeps spent less time in A2 (
P
=0.04) and the least time in A3 (
P
=0.01). Light or mat temperature did not affect sow use of pen
areas or piglet weight gain. Piglets with bright creeps tended (
P
=0.06) to take longer to enter the creep for the first time after
birth, but the latency for 30.0% of the litter to remain clustered for 10 min tended (
P
=0.08) to be shorter in bright compared to
dark creeps. Overall, piglet use of the creep increased with warm mat temperatures and brightness, which should be further
investigated as potential strategies to promote piglet safety and reduce crushing in pen farrowing systems.
Keywords: welfare, crushing, loose farrowing, luminosity, thermal environment
Implications
Attracting piglets to creep areas in loose farrowing pen sys-
tems may be a practical strategy to reduce piglet crushing,
which is the main cause of pre-wean mortality. Piglets spent
more time in creep areas which were brightly lit or had
warmer mats than piglets with access to dark creeps or with
lower mat temperature. Light may act as a cue for the
warmth of creep areas, which led piglets away from pen
areas shared with the sow. Further investigation of appro-
priate light characteristics and potential cues which can act
as attractants to creep areas is warranted.
Introduction
The search for alternative farrowing systems without a crate is
becoming particularly important as the world pork industry
strives for opportunities to improve sow welfare. Piglets are at
risk of being overlain and crushed by the sow in pens, natural
or semi-natural farrowing and lactation systems, while fatal
crushing of piglets by the sow may account for two-thirds of
losses. Reports of pre-weaning mortality were as high as
32.0%, 27.0% and 22.0% in farrowing pen systems
(McGlone and Morrow-tesch, 1990; Blackshaw
et al
., 1994;
†
E-mail: gabriela.morello@i3s.up.pt
a
Present address: Institute for Research and Innovation in Health, University of
Porto, 208 Alfredo Allen St., Porto 4700-135, Portugal.
b
Present address: Institute of Animal Husbandry and Animal Welfare, University
of Veterinary Medicine, A-1210 Vienna, Austria.
c
Mention of any trade name, proprietary product or specific equipment does not
constitute a guarantee or warranty by USDA-ARS and does not imply its
approval to the exclusion of other products that may also be suitable. The USDA-
ARS is an equal opportunity and affirmative action employer and all agency
services are available without discrimination.
Animal
(2019), 13:8, pp 1696–1703 © The Animal Consortium 2018. This is a work of the
U.S. Government and is not subject to protection in the United States.
doi:10.1017/S1751731118003300
animal
1696

Hales
et al
., 2015), 25.0% and 23.0% in pen systems with
communal areas (Marchant
et al
., 2001; Li
et al
., 2009) and
generally lower in farrowing crate systems (Lou and Hurnik,
1994, 15.0% total pre-wean; Gu
et al
., 2011, 10.8% crushing
pre-wean; Kilbride
et al
., 2012, 11.7% total pre-wean).
Attracting piglets to a protected creep area, in between
nursing bouts, may be a solution for reducing piglet mor-
tality, as increased time spent spread around the sow
increases the risks of piglets being crushed (Marchant
et al
.,
2001). The sow welfare and piglet protection (SWAP) far-
rowing pens (Jyden, Vemb, Denmark) have enclosed creeps
exclusive for the piglets and have been studied as alter-
natives to provide greater freedom of movement for the
sows, while allowing for competitive piglet survivability
(Hales
et al
., 2015 and 2016). Yet, more investigation is
needed to optimise piglet use of the creep in SWAP pens.
One strategy to attract piglets to the heated areas is to
increase the temperature gradient between heated and pen
areas (Xin and Zhang, 1999; Schormann and Hoy, 2006; Burri
et al
., 2009), while considering the piglets’sageandtypeof
heating (Xin and Zhang, 1999). However, increasing tempera-
ture gradient between the warm and pen areas in farms
equipped with heat mats must be done with caution, as piglets
tolerated a maximum contact temperature of 46.2°C (Zhang
and Xin, 2000a) with heat mats. Yet, little is known regarding
the ideal mat temperature to promote piglet use of the mat.
In addition, very little has been done to explore other
environmental aspects, such as creep illumination, which
may function as attractants to piglets. Furthermore, the
design of creep areas varies substantially on commercial
farms, from enclosed to fully opened, which affects light. The
reported effects of light intensity on piglet behaviour are
somewhat controversial, as some authors have reported
piglet preference for darkness (Parfet and Gonyou, 1991;
Larsen and Pedersen, 2015), while others reported fear of
darkness (Tanida
et al
., 1996). However, most of the studies
on light have been conducted with distinct light sources,
within distinct contexts and light levels. Still, it is known that
light affects the reproduction, welfare and behaviour of older
pigs (Stevenson
et al
.,1983; McGlone
et al
., 1988; Prunier
et al
., 1994). Thus, light could be used to alter piglet beha-
viour, in an effort to enhance piglet survivability.
The present research, therefore, hypothesised that chan-
ges in the light intensity and mat temperature inside piglet
creep areas alter piglet creep use and consequently their
chances of survival. The current experiment aimed at evalu-
ating the effects of two light intensities (300
v
. 4 lx) without
any source of radiant heat in the creep area on piglet beha-
viour and survivability in farrowing pens, considering the
effects of the mat surface temperature as a covariate.
Material and methods
Facilities
The study was conducted in two farrowing rooms on a
commercial farm located in Central Queensland during Sep-
tember (replicate 1) and November (replicate 2), 2016. The
environment in the farrowing rooms was automatically
controlled using ventilation fans, attic inlets and evaporative
cooling in warmer conditions (not operated during the pre-
sent study). Artificial lighting of ~80 ± 30 lx (ranging from 29
to 186 lx) was provided from 0800 to 1959 h, interrupted by
a 12-h darkness phase (0 lx). The referred light levels were
measured at the sow’s head level when in a standing pos-
ture, averaged for all pens in both farrowing rooms, and was
not statistically different between rooms and replicates.
Farrowing rooms were equipped with four rows of nine
2.0 ×2.8 m SWAP farrowing pens. Each SWAP pen had a
trapezoidal creep area of 1.1 m (length) ×1.0 m (wider
width) ×0.5 m (narrower width) with a trapezoidal metal
heated mat (Lamapor S.A., Navarra, Spain) as the only sup-
plemental heat source for the piglets. The pens had a slatted
floor area for the sow, which the piglets could access. Creep
areas had three sides fully closed, with the open side facing
the pen. Creeps were covered with a black wooden lid, which
Figure 1 (a) Top view of a sow welfare and piglet protection (SWAP) pen with bright creep. (b) Schematic plan view of a SWAP pen, depicting the
trapezoidal covered creep, the areas immediately in front of (A2) and farthest from the creep (A3) distinguished by the dashed line.
Managing creep environment to attract piglets
1697

could be lifted to check on the piglets (Figure 1). Sows were
free to move and were never crated before or after farrowing.
A total of 108 multiparous (Large White ×Landrace) sows
(64 sows in replicate 1 and 44 in replicate 2) were randomly
allocated into the 72 farrowing pens available across the two
studied farrowing rooms, ~3 days before their farrowing due
date. Farm practices were performed following their stan-
dard operation procedures, with minimal external inter-
ference from the research team, hence the unbalanced
distribution of sows across the two replicates. As part of the
farm’s routine, one or two employees were available cover-
ing 24-h per day to constantly assist with births when
necessary, and care for the welfare of sows and their litters.
Experimental design
The experiment was performed in a generalised randomised
block design, with two blocks or replicates (September and
November) and a one-way arrangement of treatments
(bright creep: 300 lx
v
. dark creep: 4 lx), with mat surface
temperature treated as a covariate.
The bright creep treatment was achieved by illuminating
the creep areas with cool white light-emitting diode (LED)
light strips. Creeps assigned with the dark creep treatment
did not have any source of artificial light inside them. Two of
the four 15-pen rows inside each farrowing room were
assigned with the bright creep treatment. The 15-pen rows
with the bright creep treatment were staggered with the
rows assigned with the dark creep treatment. Creep light
levels were measured at floor level, before piglets were born,
and did not add any extra heat to the creep environment.
Mat temperature was set for automatic control at room
level for 30.0°C in farrowing room 1
v
. 35.0°C in farrowing
room 2 to promote temperature variation among mats of
distinct rooms. Mat set-point was later swapped between
rooms 1 and 2 for the second experiment replicate, in an
effort to prevent the effects of the mat temperature to be
biased towards possible existent unknown effects of the
farrowing rooms. Although mat temperature was controlled
automatically, mat temperature varied greatly among dis-
tinct pens and over the studied days, hence its consideration
as a covariate for the statistical analysis.
Pregnant sows, due to farrow in 3 to 5 days, were ran-
domly assigned to pens of distinct treatments and farrowing
rooms after balancing for their parities (first
v
. multiparous
sows), as they were brought from gestation pens into the
farrowing rooms, following the farm’s schedule. Sow and
piglet behaviour was studied for 72-h
postpartum
(pp),
which is the period with the greatest risk for piglet crushing
by the sow (Marchant
et al
., 2001; Baxter
et al
., 2009;
Kilbride
et al
., 2012).
Data collection
Behaviour
. Six focal sows and their litters (three in replicate 1
and three in replicate 2) from each of the two possible
treatments in each of the two farrowing rooms (total of 24
focal sows and litters) had their behaviour continuously
recorded, with night view cameras (model QC-3694; SigNet
Technologies Inc., Beltsville, MD, USA). The remaining 84 study
sows and their litters (52 in replicate 1 and 32 in replicate 2)
were scan sampled through live observation once every 15 min
(from 0800 to 1700 h) for their location. Piglets’use of the
creep, as well as piglets’and sows’use of pen areas immedi-
ately in front of the creep (A2) and farthest from the creep (A3)
were recorded (Figure 1). Sows were considered to be in a
certain area if at least 70.0% of their body was in this area,
otherwise sows were marked as being in the middle of the
pen. Latency for piglets to enter the creep for the first time
after birth, and latency for 30.0% and 75% of the litter to
remain in the creep area for a continuous period of 10 min for
the first time after birth (adapted from Burri
et al
., 2009),
were coded from the recorded videos.
Light and temperature measurement
. Light and temperature
levels were automatically registered hourly inside creep areas
of the 24 focal pens, with the use of light and temperature
loggers (Hobo, U12; Onset Computer Corporation, Bourne,
MA, USA). Ambient temperature and light levels were mea-
sured once daily at 1200 h, in the centre of each study pen, at
the sow’s head level when in a standing position, using a heat
stress meter (model 800036; Sper Scientific Ltd, Scottsdale,
AZ, USA). Ambient light levels during darkness were measured
in the centre of each study pen, once before each experiment
replicate. Room temperature was also recorded from the
rooms’control systems at 0800, 1200 and 1700-h daily. Mat
surface temperature was automatically registered using a
temperature coin logger (model DS1922LThermocron; OnSo-
lution Pty Ltd, NSW, Australia), attached directly to the mat,
0.5 m from the longest (1.1 m) edge of the mat and 0.3m
away from the second longest (1.0 m) edge. The coin logger
placement allowed for it to be in contact with the actual
heated area of the mat, while still close to the edges, to avoid
attracting the piglets’attention to it. The coin loggers were
covered with 0.5 cm thick extruded polystyrene foam and
insulation tape, except for the face in contact with the mat, to
avoid measuring air instead of mat surface temperature. This
method of measuring mat surface temperature was tested and
validated before the study. Radiant temperature was mea-
sured inside the creep areas once before each replicate, to
ensure that light treatment did not add extra heat to the
creep’s thermal environment.
Piglet weight and mortality
. Piglet mortality was assessed
hourly during the 72-h pp period, for all pens in the rooms.
Death cause was recorded and classified into death due to
crush, weak and small piglets (without any signs of crush-
ing), and other/unknown. Dead piglets were removed from
the pen. Four piglets (the lightest- and the heaviest-born
female and male) from each of the 24 focal litters were
marked soon after birth and weighed both in the first and
third days of life (72-h pp), to evaluate treatment effects on
piglet weight gain. Average weaning weights for the focal
litters were retrieved from the farm’s records.
Morello, Marchant-Forde, Cronin, Morrison and Rault
1698

Statistical analysis
The procedure GLMSELECT was performed on SAS (SAS
Institute, Cary, NC, USA) through a stepwise method to find
the best model explaining the pig behavioural parameters,
after data were checked for normality and homo-
scedasticity. The response variables tested were the pro-
portion of the 72-h pp spent by piglets inside the creep, A2
and A3. The parameters considered as possible indepen-
dent variables were creep light treatments (bright and dark
creep), farrowing room (room 1 or room 2), mat tempera-
ture, sow parity, time spent by the sow in A2 and A3,
replicate (September or November), ambient temperature,
ambient light, number of piglets in the litter and all possible
interactions among these variables. The proportion of the
72-h spent by the sows in A2 and A3 was also considered as
responsible variables as functions of the remaining referred
independent variables. After the best models were selected,
analyses of variance were performed through procedure
GLM on SAS (SAS Institute) and least square means were
compared. Individual sows and their litters within a pen
were considered as experimental units. Means are pre-
sented along with their standard errors.
Results
Effects of the light treatment
The interaction between creep light treatment and mat
temperature was not significant (
F
7,101
=0.04,
P
=0.83),
and therefore results are presented separately for light
intensity and mat temperature. Light intensity significantly
affected piglet use of the creep area (
F
7,101
=8.72,
P
<0.01).
Piglets with access to the bright creep spent 7.2% more
(
P
<0.01) time in the creep than piglets with access to dark
creeps (Figure 2).
Piglets with bright creeps tended (
F
5,19
=3.90,
P
=0.06)
to take longer (104 ± 37 min) to enter the creep for the first
time, compared to piglets with dark creeps (24 ± 6 min).
Conversely, the latency for 30.0% of the litter to be in the
creep together and remain in the creep for a minimum period
of 10 consecutive min for the first time tended to be shorter
(
P
=0.08) for piglets of bright creeps, but no significant dif-
ferences (
P
=0.49) were found for the latency of 75% of the
litter to enter and remain in the creep for a minimum period
of 10 consecutive min. Piglets in bright creeps spent 5.0%
less time of the total 72-h pp period in A3 compared to
piglets in dark creeps (
P
=0.01), and 2.9% less time in A2
(
P
=0.04, Figure 3).
Effects of mat temperature
There was a substantial variation in temperature measured
at the surface of the mats among distinct pens and over time,
even within the same farrowing room. Thus, mat tempera-
ture was not significantly different across the two desired set
points (32.9 ± 2.9°C
v
. 34.3 ± 7.1°C,
P
=0.24). Mat tem-
perature averaged 33.0 ± 6.2°C (ranging from 20.3°C to
42.2°C) and 34.2 ± 5.3°C (ranging from 34.2°C to 41.7°C)
for bright and dark creeps, respectively, and did not statis-
tically differ between light treatments (
P
=0.40).
Mat temperature was related to piglet use of the creep
(
F
7,101
=31.0,
P
<0.01), as well as A2 and A3. For each
increase of 1.0°C in mat temperature, there was an approx-
imate 2.1% increase (
P
<0.01) in piglet use of the creep, and
conversely a 0.7% decrease in piglet time spent in A2
(
P
=0.02) and a 1.6% decrease in A3 (
P
<0.01).
Effects of ambient temperature
Ambient temperature was significantly higher (
P
<0.01) in
replicate 2 (27.3 ± 0.1°C) compared to replicate 1 (23.8 ± 0.1°C).
Ambient temperature strongly affected the piglets’use of the
creep (
F
7,101
=28.8,
P
<0.01) and remaining pen areas. Gen-
erally, for each 1.0°C increase in ambient temperature, there
was an approximate 4.8% reduction (
P
<0.01) in piglet use of
the creep, and conversely a 1.2% increase in piglet time spent
in A2 (
P
=0.02) and 3.4% increase in A3 (
P
<0.01).
Sow location
The proportion of time spent by sows in the areas A2 and A3
was not affected by light (A2:
F
6,102
=0.70,
P
=0.40; A3:
F
6,102
=0.33,
P
=0.57) or mat temperature (A2:
F
6,102
=0.05,
P
=0.83; A3:
F
6,102
=2.29,
P
=0.13). Sow location within the
pen was related (
P
<0.01) to piglet location within A2 and
A3. For each increase of 1.0% of time spent by the sow in A2
**
0
10
20
30
40
50
Dark Bright
Mean time spent in the creep, %
Figure 2 Mean time spent by the piglets inside the dark and bright
creep areas, as a percentage of the 72-h
postpartum
period. **
P<
0.01.
*
**
0
10
20
30
40
50
A2 A3
Mean time spent in the pen
areas, %
Dark
Bright
Figure 3 Mean time spent by the piglets in the areas immediately in
front of (A2, Figure 1) and farthest from the creep (A3, Figure 1), as a
percentage of the 72-h
postpartum
period. *
P<
0.05 and **
P<
0.01.
Managing creep environment to attract piglets
1699

and A3, there was an approximate 20.0% increase of time
spent by the piglets in the same area as the sow.
Light treatments and mat temperature did not affect sow
posture (
P
>0.10). However, most of sow standing (76.0%)
and sitting (74.5%) events were observed in A3, excluding
the situations where sows had <70.0% of their bodies within
a specific pen area (middle of the pen). Percentage of lying
events (both lateral and sternal recumbence) were similar in
A2 (52.9%) and A3 (47.1%).
Piglet weight, mortality and number of piglets in the pen
Average piglet weight at weaning was 7.3 ± 0.2 kg, for an
average birth to wean period of 27.8 ± 0.8 days, not sig-
nificantly different (
F
4,15
=1.11,
P
=0.39) between repli-
cates, light treatments and not affected by mat temperature.
Piglet weight gain within the first 72-h pp was also not
affected (
F
3,16
=0.83,
P
=0.49) by light intensity or mat
temperature.
Average piglet mortality for the 72-h pp was 8.7 ± 9.2%
and 8.8± 15.7% for the bright and dark treatments, respec-
tively. Piglet mortality was not affected by light treatment
(
F
1,108
=1.20,
P
=0.28) or mat temperature (
F
1,108
=0.09,
P
=0.76). However, the risk of piglet death within 72-h pp
tended to reduce with increased ambient temperature
(
F
1,106
=3.50,
P
=0.06), and tended to be reduced with
increased piglet birth weight (
F
1,108
=3.37,
P
=0.07).
The number of piglets in the pen had a small, but sig-
nificant effect on piglet use of pen areas, as the increase of
one piglet in the litter led to a 1.2% decrease in time pig-
lets spent in the creep (
P
=0.03) and conversely a 0.7%
increase in time piglets spent in A2 (
P
=0.01), but no effect
of number of piglets in the pen was found for piglet use of
A3 (
P
=0.25).
Discussion
Illuminating creep areas in farrowing pens led to higher use
of the creep by piglets within 72-h pp on a commercial
piggery. Consequentially, piglets with bright creeps spent
less time in the pen areas shared with the sow, and the
reduction was greater in the area furthest from the creep
(A3) where sows spent most time standing. Piglet use of the
creep area also increased with increased mat surface tem-
perature and decreased ambient temperature, and a lower
number of piglets per pen. Piglet weight gain, mortality and
sow use of pen areas were not affected by light intensity or
mat temperature.
Light treatment
The positive effect of bright creeps on piglet creep use con-
trasts with the findings of Larsen and Pedersen (2015), in
which piglets did not change their use of lit (130 lx) creep
areas compared to piglets with access to dark creeps (lx not
reported). Larsen and Pedersen (2015) also reported a pre-
ference of piglets to sleep in the darkness, which was not
seen in the present research.
The higher use of bright creep areas seen in the present
study may not necessarily mean piglets preferred the lit
(300 lx) over the dark environment (4 lx), especially because
there was a trend that piglets took more time to enter bright
creeps for the first time, compared to dark creeps. Instead,
the light could have acted as an extra cue for the warmth of
the creeps, helping piglets to associate the creep area with a
warmer environment. This would explain why, although
taking longer for individual piglets to enter bright creeps for
the first time, once piglets began entering the creep, it ten-
ded to take less time for piglets to remain in the bright creeps
with their litter mates (30.0% of litter together) for 10 con-
secutive min or more, as compared to dark creeps. The ten-
dency for a greater latency for piglets to enter the bright
creep for the first time could also be the result of neonates
having to adjust from the intra-uterine (dark) environment to
the extra bright environment of the bright creeps, thus not
necessarily an aversion from piglets to the bright creeps.
Another possible explanation for the contrasting results
between the present and Larsen and Pedersen’s (2015) stu-
dies relates to the type of heat source in the creep. In Larsen
and Pedersen’s study (2015), creep heat was delivered solely
through radiation, whereas in the present study heat source
was conductive, through heating mats. It is possible that
type of heating affects how rapidly piglets perceive heat from
distinct heat transfer mechanisms (i.e. radiation
v
. conduc-
tion), or how effectively piglets associate the creep areas
with a warm zone. Piglets have been reported previously to
spend more time under radiant heat for the first 2 days of life
than on conductive heat mats (Zhang and Xin, 2001). It is
possible that piglets perceive radiant heat more readily on
their skin (head, neck, sensitive areas) compared to con-
ductive heat (from the mats), where piglets need to lay down
and be in direct contact to exchange heat with heat mats.
Radiation may also help neonates to get dry, thus decrease
heat loss soon after birth (Zhang and Xin, 2001), which is
relevant for their survival in the very first few hours after birth
(Vasdal
et al
., 2011). Thus, in Larsen and Pedersen’s study
(2015), which used solely radiant heat, brightness may have
been less important to attract piglets to the creep area, as
compared to the current study, which used mats as the
additional heat source for the piglets in the creeps. Therefore,
in the latter case, brightness could have added an extra
dimension to the cues of the warm and safe environment of
creeps, hence the current results indicating a greater use of
bright creeps by the piglets.
Moreover, creep light levels in the bright treatment were
constantly above pen light levels (300
v
.80lx,respectively)
in the present study, thus there was a constant light gra-
dient in which the higher light intensity was inside the
creep. Conversely, in Larsen and Pedersen’s study (2015)
creep light intensity was below the room light level during
the day (130
v
. 300 lx), which may be an indication that
light gradient has a greater influence on neonatal beha-
viour than illumination
per se
.
There may be age-related changes in attractiveness of
different light intensities. While Larsen and Pedersen (2015)
Morello, Marchant-Forde, Cronin, Morrison and Rault
1700

did not find an overall illuminance preference of pre-weaning
piglets, a previous study by Taylor
et al
. (2006) demonstrated
that 7 to 11 week old piglets preferred dimmer conditions
(2.4 lx) over bright (400.0 lx) conditions. However, it is
important to point out that in Taylor
et al.
’s study (2006),
illumination was provided by incandescent bulbs, which are
highly inefficient in converting energy into visible light; only
5.0% or less of the energy used by incandescent bulbs is
converted to visible light, while the remaining 95.0% or more
energy is converted into heat. Taylor
et al
. (2006) did control
their experiment for ambient temperature, but there were no
reports on the direct measurement of radiant heat, which has
a direct impact on the piglet skin temperature, thus heat
exchange with the surroundings. The authors mentioned
keeping lights energised on the darkest treatment with a
layer of black card between lights and chamber to maintain a
radiant load on this treatment, but the magnitude of the
achieved radiant load across treatments was not reported. It
is possible that at higher illumination treatments, in which
there were less gel filters between the lights and the pigs
compared to dim treatments, there was also a greater radi-
ant heat load perceived directly by pigs on their surface.
Therefore, the preference found by older piglets for spending
time in dimmer chamber in Taylor
et al.
’s study (2006) could
be somewhat confounded with preference for less radiation,
especially as older piglets require reduced heat load.
Based on the results of the present study, light can be
potentially used as an additional cue in creeps to increase
their use by newborn piglets. However, given the longer
latency to enter bright as compared to dark creeps, the light
configurations (cool white LED, 300 lx) used in this experi-
ment may not be the ideal ones for increasing piglet use of
the creep area. Lights of different intensity and frequencies,
within the pigs’visible spectrum, should be further studied as
potential cues or attractants to creep areas. Furthermore,
other sensory stimuli can be explored in creeps to increase
piglet use of these areas, as piglets have been previously
reported to be attracted to a variety of olfactory, auditory
and tactile stimuli (Morrow-Tesch and McGlone, 1990; Parfet
and Gonyou, 1991).
Mat temperature
The increased use of the creep area as mat temperature
increased from 20.3°C to 42.2°C could be a result of the
increased temperature gradient between the mat and the
pen environment, as mat temperature increased. Generally,
the greater the difference between the pen and heated areas,
the more piglets will seek the heated areas (Xin and Zhang,
1999). Although 30.0°C to 35.0°C are common mat set
points in farrowing facilities, mat temperatures above this
range, measured at the mat’s surface, did not lead to
reduced use of the creep area in this study. This result was in
agreement with the one reported by Zhang and Xin (2000a),
in which piglets tolerated up to 46.2°C of contact tempera-
ture with the heated mat, without leaving the warm area.
The variation observed among pens and over time in mat
temperature, despite the central automatic control at room
level, highlights the variation that is commonly observed on
commercial farms, and demonstrates the importance of sys-
tematically and regularly checking mat temperatures, not
just thermostat settings. In addition, part of the mat tem-
perature variation found in this work could be a result of the
extra heat load from the piglets on the mat’s surface as
piglets used the mat. Zhang and Xin (2000b) demonstrated
that differences in temperature between the occupied and
unoccupied mat areas by piglets could be as high as 7.0°C to
12.0°C. Thus, the substantial differences in temperature
readings from mat surfaces in the present research could be a
reflection of how piglets actually used heat mats, which
could explain, at least partially, why piglets did not leave
heat mats at high temperatures.
Ambient temperature
Ambient temperature was higher in replicate 2 compared to
replicate 1 due to the seasonal variation between the two
replicates (September, end of winter
v
. November, end of
spring in Australia, respectively). The increase in creep use
with the reduction in ambient temperature found in the
present research may also have been a result of the increased
temperature gradient between the heated creep and the pen
area as ambient temperature decreased. Vasdal
et al
. (2010)
reported that piglets showed clear thermal preferences when
the thermal gradient between choices was 8.0°C to 16.0°C,
whereas this disappeared when the gradient was 4.0°C to
8.0°C. The lower ambient temperature in replicate 1 led to an
average temperature gradient of ~10.0°C between the mat
surface and pen environment, whereas this gradient was
about 5.0°C in replicate 2. Thus, the consequently decrease
in temperature gradient with higher ambient temperatures
could partly explain the lower creep use in such ambient
temperatures, in agreement with Vasdal
et al
. (2010).
Sow location
Despite all the environmental effects found on piglet creep
use, sow location was the main factor associated with
piglet location within the pen, when outside the creep.
Early access to colostrum is vital for piglet survivability
(Hoy
et al
., 1995) and piglets will tend to remain near the
sow while nursing or seeking to nurse within the first few
hours of life (Hrupka
et al
., 1998). It is also possible that in
pen farrowing systems the sow herself tends to seek and
remain near her piglets. Yet, the use of the creep as
affected by the ambient and mat temperatures may be an
indication of a shift in the piglets’motivation from willing
to stay near their mother to stay warm, in an enclosed area,
depending on the thermal context.
Therefore, a possible strategic solution to reduce crushing
in pen systems may be to manage the environment to attract
piglets to protected creep areas, away from the sow between
suckling episodes. Attracting piglets to the creeps should be
done within the first 24-h to 72-h pp, which are the most
critical for the risk of piglet crushing by the sow (Marchant
et al
., 2001; Andersen
et al
., 2005).
Managing creep environment to attract piglets
1701

Piglet weight, mortality and number of piglets in the pen
Piglet weight gain within 72-h pp and between birth to
weaning was not affected by light and mat treatments,
thus the higher use of bright and warmer creeps possibly
did not affect piglet nursing behaviour, although further
analysis would be needed to confirm this hypothesis. It is
important that any strategy which promotes a greater use
of the creep by the piglets, and consequently less time near
the sow, allows for efficient milk intake during the nursing
bouts. Missing milk ejection events can lead to a sub-
stantial decrease in milk intake, which can reduce piglet
weight gain and vitality, while lighter piglets have more
chance of dying due to starvation or crushing by the sow.
Reduced milk intake can be even more aggravated with
increasing litter size, where sibling competition can be
higher and access to a teat more difficult (Andersen
et al
.,
2011; Bozděchová
et al
., 2014).
Although mortality was not affected by creep light and
mat treatments, this study was not designed to detect
mortality differences among sows, as this would require
studying a substantially greater number of litters. Still,
previous research has demonstrated that the risk of a piglet
being crushed is significantly higher if piglets are spread
around within 0.5 m from the sow as she lies down
(Marchant
et al
., 2001), as opposed to piglets being safely
clustered together.
The decrease in the use of the creep area with the increase
in number of piglets in the pen may be partly due to the
clustering dynamics among piglets of larger litters or due to
increased heat generation among greater number of piglets.
In support of the social dynamics hypothesis, increased
incidence of crushing of at least three piglets by the same
sow was found in litters of over 13 piglets, as compared with
litters smaller than 10 piglets (Morello, 2015). Hence, it is
possible that in larger litters piglets are less likely to remain
close together in one single cluster, which has been
demonstrated to be safer for piglets as opposed to being
spread around the sow (Marchant
et al
., 2001). In support of
the heat generation hypothesis, 2-day-old piglets generate
~5 W/kg of heat (Zhang and Xin, 2000a; Brown-Brandl
et al
.,
2004), which could be enough heat added up in larger litters
to reduce the need of piglets to seek for warmer areas.
The present research demonstrated that provision of light
and increased mat temperature in creep areas increased use
of the creep by newborn piglets in a commercial setting.
Whether the increase in creep use translates to lower piglet
crushing and total reduction in pre-wean mortality should be
evaluated on a larger sample size. Furthermore, other sen-
sory stimuli can be explored as attractants to creep areas in
an effort to improve piglet survivability, while providing
better welfare to sows housed in pen systems.
Acknowledgements
The authors thank Bettapork Pty Ltd for their collaboration with
the present study, the Australian Pork Cooperative Research
Centre (project 1A-116) for funding and support of the present
research, as well as David Murden, Igor Moreira Lopes and
Seungyeol Ko for their help with data collection and analysis.
Declaration of interest
The authors declare that there are no conflicts of interest
involved with this research.
Ethics statement
The present research was approved by the Animal Ethics
Committee from the University of Melbourne, following the
National Health and Medical Research Council’sAustralian
Code for the Care and Use of Animals for Scientific Purposes
(eighth edition, 2013).
Software and data repository resources
None of the data were deposited in an official repository.
References
Andersen IL, Berg S and Bøe KE 2005. Crushing of piglets by the mother sow
(
Sus scrofa
)–purely accidental or a poor mother? Applied Animal Behaviour
Science 93, 229–243.
Andersen IL, Nævdal E and Bøe KE 2011. Maternal investment, sibling compe-
tition, and offspring survival with increasing litter size and parity in pigs
(
Sus scrofa
). Behavioral Ecology and Sociobiology 65, 1159–1167.
Baxter EM, Jarvis S, Sherwood L, Robson SK, Ormandy E, Farish M, Smurthwaite
KM, Roehe R, Lawrence AB and Edwards SA 2009. Indicators of piglet survival in
an outdoor farrowing system. Livestock Science 124, 266–276.
Blackshaw JK, Blackshaw AW, Thomas FJ and Newman FW 1994. Comparison of
behaviour patterns of sows and litters in a farrowing crate and a farrowing pen.
Applied Animal Behaviour Science 39, 281–295.
Bozděchová B, Illmann G, Andersen IL, Haman J and Ehrlenbruch R 2014. Litter
competition during nursings and its effect on sow response on day 2 post-
partum. Applied Animal Behaviour Science 150, 9–16.
Brown-Brandl TM, Nienaber JA, Xin H and Gates RS 2004. A literature review of
swine heat production. Transactions of the ASAE 47, 259–270.
Burri M, Wechsler B, Gygax L and Weber R 2009. Influence of straw length, sow
behaviour and room temperature on the incidence of dangerous situations for
piglets in a loose farrowing system. Applied Animal Behaviour Science 117,
181–189.
Gu Z, Gao Y, Lin B, Zhong Z, Liu Z, Wang C and Li B 2011. Impacts of a freedom
farrowing pen design on sow behaviours and performance. Preventive Veter-
inary Medicine 102, 296–303.
Hales J, Moustsen VA, Nielsen MBF and Hansen CF 2015. Temporary confine-
ment of loose-housed hyperprolific sows reduces piglet mortality. Journal of
Animal Science 93, 4079–4088.
Hales J, Moustsen VA, Nielsen MBF and Hansen CF 2016. The effect of tem-
porary confinement of hyperprolific sows in sow welfare and piglet protection
pens on sow behaviour and salivary cortisol concentrations. Applied Animal
Behaviour Science 183, 19–27.
Hoy S, Lutter C, Puppe B and Wähner M 1995. Correlations between the vitality
of newborn piglets, teat order, mortality, and live weight development up to
weaning. Berliner und Munchener Tierarztliche Wochenschrift 108, 224–228.
Hrupka BJ, Leibbrandt VD, Crenshaw TD and Benevenga NJ 1998. The effect of
farrowing crate heat lamp location on sow and pig patterns of lying and pig
survival. Journal of Animal Science 76, 2995–3002.
Kilbride AL, Mendl M, Statham P, Held S, Harris M, Cooper S and Green LE 2012. A
cohort study of preweaning piglet mortality and farrowing accommodation on 112
commercial pig farms in England. Preventive Veterinary Medicine 104, 281–291.
Larsen MLV and Pedersen LJ 2015. Does light attract piglets to the creep area?
Animal 9, 1032–1037.
Morello, Marchant-Forde, Cronin, Morrison and Rault
1702

Li Y, Johnston L and Hilbrands A 2009. Pre-weaning mortality of piglets in a
bedded group-farrowing system. Journal of Swine Health Production 18,
75–80.
Lou Z and Hurnik JF 1994. An ellipsoid farrowing crate: its ergonomical
design and effects on pig productivity. Journal of Animal Science 72,
2610–2616.
Marchant JN, Broom DM and Corning S 2001. The influence of sow behaviour on
piglet mortality due to crushing in an open farrowing system. Animal Science 72,
19–28.
McGlone JJ and Morrow-tesch J 1990. Productivity and behavior of sows in
level vs. sloped farrowing pens and crates. Journal of Animal Science 68,
82–87.
McGlone JJ, Stansbury WF, Tribble LF and Morrow JL 1988. Photoperiod and
heat stress influence on lactating sow performance and photoperiod effects on
nursery pig performance. Journal of Animal Science 66, 1915–1919.
Morello GM 2015. Investigating piglet crushing by the sow: a data mining
approach. PhD thesis, Purdue Univeristy, West Lafayette, IN, USA.
Morrow-tesch J and Mcglone JJ 1990. Sources of maternal odors and the
development of odor preferences in baby pigs. Journal Animal Science 68,
3563–3571.
Parfet KAR and Gonyou HW 1991. Attraction of newborn piglets to audi-
tory, visual, olfactory and tactile stimuli. Journal of Animal Science 69,
125–133.
Prunier A, Dourmad JY and Etienne M 1994. Effect of light regimen under
various ambient temperatures on sow and litter performance. Journal of Animal
Science 72, 1461–1466.
Schormann R and Hoy S 2006. Effects of room and nest temperature on the
preferred lying place of piglets –a brief note. Applied Animal Behaviour Science
101, 369–374.
Stevenson JS, Pollmann DS, Davis DL and Murphy JP 1983. Influence of sup-
plemental light on sow performance during and after lactation. Journal of Ani-
mal Science 56, 1282–1286.
Tanida H, Miura A, Tanaka T and Yoshimoto T 1996. Behavioral responses of
piglets to darkness and shadows. Applied Animal Behaviour Science 49,
173–183.
Taylor N, Prescott N, Perry G, Potter M, Le Suerur C and Wathes C 2006. Preference
of growing pigs for illuminance. Applied Animal Behaviour Science 96, 19–31.
Vasdal G, Møgedal I, Bøe K, Kirkden R, and Andersen IL 2010. Piglet preferencefor
infrared temperature and flooring. Applied Animal Behaviour Science 122, 92–97.
Vasdal G, Østensen I, Melišová M, Bozděchová B, Illmann G and Andersen IL 2011.
Management routines at the time of farrowing –effects on teat success and post-
natal piglet mortality from loose housed sows. Livestock Science 136, 225–231.
Xin H and Zhang Q 1999. Preference for lamp or mat heat by piglets at cool and
warm ambient temperatures with low to high drafts. Applied Engineering in
Agriculture 15, 547–551.
Zhang Q and Xin H 2000a. Modeling heat mat operation for piglet creep heat-
ing. Transactions of the ASAE 43, 1261–1267.
Zhang Q and Xin H 2000b. Static and dynamic temperature distribution of heat
mats for swine farrowing creep heating. Applied Engineering in Agriculture 16,
563–569.
Zhang Q and Xin H 2001. Responses of piglets to creep heat type. Applied
Engineering in Agriculture 17, 515–519.
Managing creep environment to attract piglets
1703