Content uploaded by Karen Schwartzkopf-Genswein
Author content
All content in this area was uploaded by Karen Schwartzkopf-Genswein on Sep 23, 2015
Content may be subject to copyright.
3630
ABSTRACT: The objective of the present study
was to identify and quantify several factors affecting
shrink in cattle during commercial long-haul transport
(≥400 km; n = 6,152 journeys). Surveys were designed
and delivered to transport carriers to collect relevant
information regarding the characteristics of animals, time
of loading, origin and destination, and loaded weight
before and after transport. In contrast to fat cattle, feeder
cattle exhibited greater shrink (4.9 vs. 7.9 ± 0.2% of BW,
respectively; P < 0.01), and experienced longer total
transport durations (12.4 vs. 14.9 ± 0.99, respectively;
P < 0.01) due to border crossing protocols which require
mandatory animal inspection. Shrink was greater (P <
0.001) for feeder cattle loaded at ranches/farms and feed
yards compared with those loaded at auction markets.
Cattle loaded during the afternoon and evening shrank
more than those loaded during the night and morning
(P < 0.05). Shrinkage was less in cattle transported
by truck drivers having 6 or more years of experience
hauling livestock compared with those with 5 yr or less
(P < 0.05). Shrink increased with both midpoint ambient
temperature (% of BW/ºC; P < 0.001) and time on truck
(% of BW/h; P < 0.001). Temperature and time on truck
had a multiplicative effect on each other because shrink
increased most rapidly in cattle transported for both
longer durations and at higher ambient temperatures
(P < 0.001). The rate of shrink over time (% of BW/h)
was greatest in cull cattle, intermediate in calves and
feeder cattle, and slowest in fat cattle (P < 0.05) but such
differences disappeared when the effects of place of
origin, loading time, and experience of truck drivers were
included in the model. Cull cattle, calves and feeder cattle
appear to be more affected by transport compared with
fat cattle going to slaughter because of greater shrink.
Several factors should be considered when developing
guidelines to reduce cattle transport stress and shrink
including type of cattle, ambient temperature, transport
duration, driving quality, and time and origin of loading.
Key words: commercial transport, livestock, shrinkage
Factors affecting body weight loss during
commercial long haul transport of cattle in North America
1
L. A. González,*†
2,3
K. S. Schwartzkopf-Genswein,* M. Bryan,* R. Silasi,* and F. Brown*
*Agriculture and Agri-Food Canada, Lethbridge Research Centre, Lethbridge, T1J 4B1 AB, Canada;
and †University of Manitoba, Department of Animal Science, Winnipeg, R3T 2N2 MB, Canada
© 2012 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2012.90:3630–3639
doi:10.2527/jas2011-4786
INTRODUCTION
The amount of BW loss or shrink animals
experience during transport is directly related to their
level of hydration and body and carcass weight (Jones
et al., 1990; Warriss, 1990; Schaefer et al., 1992). In
addition, greater shrink has been associated with reduced
performance and greater morbidity (e.g., shipping fever)
after transport (Camp et al., 1981, 1983), reduced meat
quantity and quality after slaughter (Jones et al., 1990),
and greater incidence of lame, non-ambulatory, and
dead animals on board (González et al., 2012c). Thus,
shrink has detrimental impacts on animal well-being and
profi tability, and it is therefore essential to identify and
quantify those factors affecting it to reduce its impact.
Shrink is related to the length of time of feed and
water deprivation, environmental conditions, body
condition of the animals, time of the day when loading
occurs, composition of the diet, driving quality, and
management and handling procedures before and
during transport (Warriss, 1990; Coffey et al., 2001).
1
Financial support from Alberta Beef Producers, Alberta Livestock
Industry Development Fund, Alberta Farm Animal Care Association,
and Agriculture and Agri-Food Canada is greatly acknowledged. The
fi rst author received an NSERC postdoctoral fellowship. Authors are
very thankful to all dispatchers, trucking companies and drivers for their
honesty in fi lling out the surveys and dedication to the cattle industry
and welfare of animals. A special thanks to Rick Sincennes and Tracey
Greer for their invaluable assistance.
2
Present address: CSIRO Livestock Industries, Australian Tropical
Sciences and Innovation Precinct, Townsville, 4810 QLD, Australia.
3
Corresponding author: LucianoAdrian.Gonzalez@gmail.com
Received October 4, 2011.
Accepted May 3, 2012.
Published January 20, 2015
Shrink of cattle during transport
3631
Stress, physical activity, and environmental conditions
during transport increase energy and water demands,
and excretion of feces and urine (Warriss, 1990; Parker
et al., 2003). Meanwhile, animals are typically deprived
of feed and water within North America as cattle are
not provided with feed or water during transport and
are not off loaded for such purposes (González et al.,
2012a). Body reserves must therefore be mobilized to
supply extra energy and water (Schaefer et al., 1992)
required to cope with transport but eventual depletion
could cause dehydration and threaten survival.
Shrink is easy to obtain and could therefore be used
as an indicator of the conditions during commercial
transport as the majority of cattle liners are routinely
weighed before and after transport. Such data have
been collected by our group along with information
about transport practices and conditions (González et
al., 2012a,b). The objective of the present study was to
identify and quantify those factors affecting cattle shrink
during commercial long hauls of cattle in North America.
MATERIALS AND METHODS
The present study collected information about prac-
tices employed during the commercial transport of cattle.
Care and handling of animals during transportation was
therefore not supervised or controlled by the research team.
Description of the Survey
Field surveys were designed to collect data
regarding the characteristics of cattle transported during
long hauls departing from and arriving to the province
of Alberta, Canada. Surveys were considered if the
transport distance was equal to or longer than 400 km
between the place of origin where cattle were loaded
and the place of fi nal destination where cattle were
unloaded from the trailer. Surveys consisted of a set of
questions separated into 5 sections which were designed
to gather information regarding the livestock, driver and
equipment, animal loading, conditions during transport,
and unloading, respectively. A detailed description of
the surveys and calculations were presented elsewhere
(González et al., 2012a,b,c).
The fi rst section of the survey was designed to gather
animal information including number, average BW, and
category. Cattle category was defi ned according to BW as
well as their place of origin and transport destination into
fat, feeder, calf, and cull cattle as described in González
et al. (2012a). Possible origin and destination selections
included farm/ranch, feed yard, auction market, and
slaughter. Animals were considered fat cattle when
loaded at feed yards and unloaded at slaughterhouses.
Feeder was used to designate animals loaded at feed
yard or auction market and their BW was between 275
and 500 kg. Animals were considered calves when
unloaded at farms, auction markets or feed yards and
weighed <275 kg. Cull cattle were cows and bulls going
from auction market or farm to feed yard or slaughter.
The second section was designed to gather information
regarding the experience of the driver hauling cattle (0
to 2, 3 to 5, 6 to 10, and more than 10 yr). Maximum and
minimum ambient temperatures within the journey were
recorded by the truck drivers from sensors usually located
in the side mirrors of the cabin or in the bumper of the
truck. Midpoint temperature during the journey was the
average between maximum and minimum temperature.
Start and end time and date of loading and unloading
of the animals was also requested in the surveys. Time
on truck was calculated as the length of time cattle spent
in the trailer from the start time of loading at the place
of origin until the end time of unloading at the fi nal
destination. Loading time was classifi ed as ‘night’ (2100
to 0600 h), ‘morning’ (0600 to 1200 h), ‘afternoon’ (1200
to 1600 h), and ‘evening’ (1600 to 210 0h). Surveys were
collected between June 1, 2007, and December 1, 2008.
However, the monthly distribution of cattle transport
throughout the year, as a percentage of all journeys over
the year, was analyzed using only data between June 1,
2007, and May 31, 2008, to avoid duplication of some
months in the data (e.g., November). In addition, the
year was also divided into 4 seasons: spring (March 21
to June 21), summer (June 22 to September 21), fall
(September 22 to December 21), and winter (December
22 to March 20).
The reason and length of delays experienced during
transport were also included in the survey with possible
reasons being border crossing, waiting to unload animals
at destination, driver rest stops, mechanical breakdown,
traffi c, poor road conditions due to inclement weather
or traffi c problems, and ‘other’. Delays were assigned a
duration of 0 for the purpose of analysis if no delay type
and length was recorded by the drivers. Loaded weight of
the animals before and after transport was also required
in the surveys. Drivers typically recorded the weight of
the group of animals being loaded from ground scales
at the place of origin before loading and at the place of
destination after unloading animals. Alternately, drivers
determined loaded weights as the difference between
loaded (animals present) minus tare weight (i.e., no
animals present). Average BW of cattle was calculated
as scale weight before transport divided by the number
of animals loaded. Body weight loss (% shrink) was
calculated as [1 – (scale weight after transport/scale
weight before transport)] × 100.
González et al.
3632
Statistical Analysis
Mixed-effects regression models were used in the
MIXED procedure (SAS Inc., Cary, NC). The best
model was chosen according to fi tting statistics with
the least Bayesian Information Criterion and Akaike’s
Information Criterion. Vehicle ID number within com-
pany and company were random effects. Different vari-
ances among cattle categories were observed and mod-
eled using the repeated statement grouped by cattle
category. Analysis of variance and covariance were used
including fi xed categorical effects (e.g., cattle category,
driver experience) and covariates being the continuous
variables (e.g., time on truck and temperature), and all
possible interactions. Time of day of animal loading was
used as linear, quadratic, and cubic covariates to assess
the change in shrink according to hour of day of animal
loading ranging from 0000 to 2400 h. Many factors af-
fected shrink of cattle during commercial transport in the
present study, and therefore an analysis was performed
to determine the relative importance of each factor when
all were included in the same mixed-effects linear re-
gression model. All signifi cant dependent variables
were thus placed into the same model to obtain multiple
regression equations. Multiple regression models were
constructed using a manual backward method consisting
of adding all factors into the model and progressively
eliminating the one with greatest P-value until only sig-
nifi cant factors were left (P < 0.10). Denominator de-
grees of freedom of mixed models were calculated us-
ing the Kenward-Roger method. Means were calculated
through the least square method and multiple compari-
sons adjusted by the Bonferroni’s test. Interactions were
further investigated at P ≤ 0.10, signifi cance declared
at P ≤ 0.05, and tendencies discussed at P ≤ 0.10 un-
less otherwise noted. All data were checked for outliers
(eliminated if necessary) and normality of the residuals.
Transformations were carried out on the data to normal-
ize distribution of residuals such as log-transformation
of delay lengths (length + 1 because there were zero val-
ues) but values are presented in the original units.
RESULTS
Monthly Distribution and Temperature
During Cattle Transport
Each cattle category showed slightly different
monthly frequency distribution patterns. Calves were
transported most frequently in November (32%), feeder
cattle in October (22%), and fat cattle were transported
in equal proportions from August through November
(approximately 14% each month; Figure 1). As a result,
fat cattle experienced greater (P < 0.05) temperature
ranges within the journey compared with feeder cattle
and greater (P < 0.05) maximum temperature compared
with calves. However, midpoint temperature did not dif-
fer among cattle categories (P > 0.10; Table 1).
Delays and Time on Truck
Feeder cattle experienced the longest delays at the
border, fat cattle intermediate, and calves and cull cattle
the shortest (P < 0.01; Table 2). Delays due to rest stops
were longer in feeder cattle compared with fat and cull
cattle, and intermediate in calves (P < 0.05). In contrast
to border and rest stop delays, delays due to waiting to
unload cattle at their destination were longest in cull, in-
termediate in fat, and shortest in feeder cattle and calves
(
P < 0.001; Table 2). Unloading delay length calculated
as the time from arrival until the start of unloading at
Figure 1. Monthly distribution of cattle transport (% of all hauls during
the year) for fat cattle (n = 4,032), feeder cattle (n = 707), calves (n = 63), and
cull cattle (n = 43) during long hauls originating from, or shipped to Alberta,
Canada (>400 km).
Table 1. Body weight and temperature for fat, feeder,
calves, and cull cattle transported during long hauls
(>400 km) from and to Alberta, Canada [mean ± SEM
(N)]
1
Item
Cattle category
P-value
Fat cattle Feeder cattle Calves Cull cattle
BW, kg 655 ± 4.4
x
388 ± 4.8
y
230 ± 9.0
z
758 ± 9.0
w
<0.001
Temperature, ºC
Minimum 5.5 ± 1.25
(4,350)
3.5 ± 1.36
(672)
−1.1 ± 2.44
(61)
0.7 ± 2.40
(37)
0.081
Midpoint
2
11.1 ± 1.29 10.4 ± 1.33 9.1 ± 1.64 8.2 ± 1.89 0.073
Maximum 16.9 ± 1.64
x
(4,657)
15.2 ± 1.67
xy
(710)
13.1 ± 2.04
y
(63)
14.2 ± 2.11
xy
(38)
<0.001
Range
3
13.3 ± 0.92
x
10.6 ± 0.97
y
9.7 ± 1.52
xy
13.1 ± 1.51
xy
0.013
w-z
Within a row, means without a common superscript differ (P ≤ 0.05).
1
The number of observations was the same for minimum, mean and range
of temperatures.
2
Calculated as the average between the maximum and minimum ambient
temperatures within a journey.
3
Calculated as maximum minus minimum temperatures within each journey.
Shrink of cattle during transport
3633
fi nal destination was longer in fat compared with feeder
cattle and calves (P < 0.01) with cull cattle experiencing
the numerically longest unloading delays but differing
from calves only (P < 0.05; Table 2). The length of all
delays (total delay) was longest for feeder cattle com-
pared with the rest of the categories (P < 0.001; Table 2).
Total time animals spent on truck was longer in feeder
compared with fat cattle and calves (P < 0.05), despite
the fact that feeder cattle were transported for the same
distance compared with the rest of categories (P > 0.10;
Table 2). However, calves were transported for short-
er distances than fat cattle (P < 0.01) because of their
Canadian origin and destination (data not shown).
Place of Origin and Loading Time
Fat cattle were almost exclusively loaded on trucks
at feed yards whereas a greater proportion of feeder cattle
were loaded at auction markets and farms/ranches. In
contrast, the greatest proportion of calves and cull cattle
were loaded at auction markets (χ
2
; P < 0.001; Table
3). Loading of fat cattle occurred almost exclusively
during the night and morning (Table 3) as 95% of them
were loaded between 0500 and 0900 h (data not shown).
However, loading time of the other cattle categories
occurred with a more even distribution throughout the
day (χ
2
; P < 0.001; Table 3).
Body Weight Loss during Transport
Mean shrink across the dataset was 5.3 ± 1.79%
of BW (mean ± SD); however, several factors affected
the amount of shrink experienced by cattle during
commercial transport. Cattle transported by truck drivers
with less than 3 (5.09 ± 0.12% of BW) and 3 to 5 (5.11 ±
0.13% of BW) years of experience hauling cattle had
greater shrinkage than those transported by drivers with 6
to 10 (4.79 ± 0.13% of BW) and more than 10 yr (4.86 ±
0.12% of BW; P ≤ 0.01; data not shown). Animals loaded
during the afternoon and evening experienced greater
shrink than those loaded during the night and morning
(P < 0.05; Figure 2A). In addition, a cubic relationship
between shrink and time of the day when animals were
loaded was also observed when a continuous model was
fi tted because shrink increased from 0700 until 1800 h
and decreased thereafter (Figure 2B). Animal origin also
infl uenced the extent of shrink observed (P < 0.001) as
cattle loaded at auction markets (4.4 ± 0.88% of BW)
shrank less compared with those loaded at farms (6.5 ±
0.78% of BW) and feed yards (7.2 ± 0.73% of BW; data
not shown). However, this was only observed for feeder
cattle (P < 0.001) as there were no differences (P > 0.10)
within fat cattle or cull cattle (Table 4).
Body weight loss during transport differed among
cattle categories with feeder cattle experiencing 62%
greater shrink than fat cattle (P < 0.001; Table 5). These
differences among means were maintained even after
correcting the means by the linear effects of time on
truck and midpoint ambient temperature during transport
(P < 0.001; Table 5). In these linear regression models,
shrink increased linearly with Midpoint Temperature (β =
Table 3. Proportion of journeys that loaded cattle
according to origin, loading time, and season for each
cattle category during long haul transport (≥ 400 km)
1
Item
Cattle category
Fat cattle Feeder cattle Calves Cull cattle
Origin
Auction market 0.1 16.1 45.6 78.0
Farm/ranch 1.3 21.8 27.8 22.0
Feed yard 98.6 62.1 26.6 0.0
Loading time
2
Night 16.1 19.6 11.1 23.1
Morning 80.9 24.0 33.3 53.8
Afternoon 1.5 19.6 16.7 19.2
Evening 1.5 36.8 38.9 3.9
Season
Summer 37.7 26.9 14.1 36.6
Fall 36.5 40.4 70.6 43.9
Winter 17.6 15.8 9.8 17.1
Spring 8.2 16.9 5.5 2.4
1
Values reported are proportion of all surveys within each cattle category
(within a column) over the total number of trailers for that cattle category.
2
Time of loading were 0600 to 1200 h for morning, 1200 to 1600 h for
afternoon, 1600 to 2100 h for evening, and 2100 to 0600 h for night.
Table 2. Length of delays and time on truck experienced
by fat, feeder, calves, and cull cattle transported dur-
ing long hauls (>400 km) from and to Alberta, Canada
[mean ± SEM (N)]
Item
Cattle category
P-value
Fat cattle Feeder cattle Calves Cull cattle
Delay, h
Border
1
0.70 ± 0.062
y
2.03 ± 0.136
x
0.23 ± 0.093
z
0.18 ± 0.079
z
< 0.001
Rest
1
0.31 ± 0.134
y
1.64 ± 0.177
x
1.12 ± 0.323
xz
0.40 ± 0.287
yz
< 0.001
Unload
1
0.38 ± 0.018
x
0.05 ± 0.018
y
0.01 ± 0.024
y
1.10 ± 0.380
x
< 0.001
Unload
2
0.51 ± 0.024
x
(4,213)
0.14 ± 0.024
y
(664)
0.10 ± 0.029
y
(58)
0.69 ± 0.255
xy
(30)
< 0.001
Total
3
1.98 ± 0.226
z
(5,044)
5.56 ± 0.281
x
(959)
3.26 ± 0.558
y
(95)
1.70 ± 0.461
z
(43)
< 0.001
Time on
truck, h
12.4 ± 0.96
y
(3,610)
14.9 ± 0.99
x
(710)
11.6 ± 1.39
y
(63)
12.9 ± 1.45
xy
(29)
< 0.001
Distance,
km
956 ± 45
x
(4,705)
890 ± 46
xy
(902)
839 ± 55
y
(94)
866 ± 64
xy
(42)
< 0.001
x-z
Within a row, means without a common superscript differ (P ≤ 0.05).
1
Values reported by truck drivers in the surveys and log-transformed for
analysis. The number of observations was the same as for total delays.
2
Values calculated as time period since arrival at destination until start of
unloading animals off the trailer.
3
Values calculated with time of loading and unloading for unloading delay
and waiting to depart.
González et al.
3634
0.0142 ± 0.00436%/ºC; P < 0.001) whereas the Time
on truck × Midpoint Temperature interaction indicated
that rate of BW loss with increasing time on truck was
faster at high temperatures (β
2
= 0.0023 ± 0.000266%/
(h × ºC); P < 0.001). This linear relationship affected all
cattle categories similarly (data not shown). However,
increasing time on truck affected the shrink observed by
cattle categories to different extents (cattle category ×
Time P = 0.07). This interaction indicated that longer
time on truck resulted in faster shrink (% of BW/h) of cull,
followed by calves, then feeder cattle for cattle having
the slowest rate of BW loss with longer time on truck
(Table 5). Therefore, the regression equations explaining
shrink had different intercepts and slopes for each cattle
category.
All ‘single’ signifi cant factors from previous
analyses were still signifi cant when placed together in
the same multiple regression model (P < 0.05; i.e., cattle
category, time on truck, ambient temperature, loading
time, and origin). However, the cattle category × Time
on truck interaction was no longer signifi cant (P = 0.32;
data not shown) and the main effect of time on truck
became quadratic (P = 0.002) with shrink increasing at
a decreasing rate as time on truck increased reaching
maximum values at approximately 40 h on truck (data
not shown). This indicates that time on truck had a similar
effect on shrink for all cattle categories even though
different intercepts and mean values were observed for
each category (Table 5). In the complete model, both
feeder cattle and cull cows showed the greatest shrink,
calves intermediate, and fat cattle the least (P < 0.001;
Table 5). Most of the variation in shrink can be explained
by cattle category and time on truck as shown by the
F-values (Table 6). The equation explaining shrink
responded to (coeffi cient [± SEM]):
Shrink = Intercept + 0.154 [± 0.0197] Time
on truck – 0.00164 [± 0.0005] Time on truck
2
+ 0.0258 [± 0.0046] Temperature + 0.00146
[± 0.0003] Time on truck × Temperature
Where Shrink is % of BW lost during transport; the
intercept is different for each level or group of cattle
category, loading time, origin of cattle, and driver
experience; time on truck is defi ned as the elapsed
time from start of loading to fi nish of unloading (h);
and temperature is midpoint ambient temperature (ºC).
A graph of this equation is presented for fat cattle in
Figure 3 and to which values of 1.95, 0.58, and 5.69%
of BW should be added to the intercept for feeder cattle,
calves, and cull cattle, respectively.
DISCUSSION
It is well documented that the long haul commercial
transport of cattle results in shrink values averaging 2 to
14% of BW depending of several factors (Warriss, 1990;
Tarrant and Grandin, 2000). With the mean shrink (7.9%
of BW) and BW (388 kg) values reported in the present
study for feeder cattle, a loss of revenue to the industry
of approximately CAN$55 per animal is estimated
(based on average BW price of CAN$1.78 /kg in 2010).
However, feeder cattle are normally marketed with a 5%
Figure 2. Shrink of cattle according to the time of the day when animals
were loaded in a categorical (A) and continuous (B) model during long haul
transport in Alberta, Canada. Data in between 2200 and 0400 h is not included
in the later because less than 0.4% of truck loads were reported within this time
period.
x, y
means without a common superscript differ (P < 0.05).
Shrink of cattle during transport
3635
pencil shrink on the BW before loading for transport or
on live BW after unloading at fi nal destination. Fat cattle
are normally paid according to carcass weight and meat
quality attributes both of which are reduced with greater
shrink (Jones et al., 1990). For instance, 430-kg steers
that lost 10.7% of their BW had a carcass weight that was
5% or 14 kg lighter compared with counterparts that lost
3.1% of their BW (Jones et al., 1990). Additional costs of
shrink may include veterinary treatments because greater
shrink has also been associated with increased morbidity
(Camp et al., 1981, 1983), and with greater incidence
of lame, non-ambulatory and dead animals (González
et al., 2012c). Therefore, better understanding of the
factors affecting shrink could help to develop strategies
to reduce the negative impacts associated with it.
Cattle experience BW loss when held off feed and
water but to a greater extent if they are also transported
in addition to withholding them from feed and water
(Phillips et al., 1991; Cole et al., 1986) as a result of the
adrenocorticoid stress response triggered by transport
(Parker et al., 2003, 2007). It is important to notice that
only 0.04% of all cattle loads assessed in the present
study received both feed and water, and 0.05% received
only water during stops where cattle were unloaded
(González et al., 2012a). Also, North American-style
livestock trailers participating in the present study are
not equipped to provide feed or water within the trailer
as is the case of some European trailers. Two types of
shrink can be experienced by cattle, fi ll shrink and tissue
(or carcass) shrink (Coffey et al., 2001). Fill shrink is the
result of the loss of contents from the gastrointestinal
tract in the form of manure and from the bladder as urine.
Fill shrink has been estimated to be approximately 3.2%
of BW and occurs at a rate of 1%/h during the fi rst 3 to
4 h of transport (Coffey et al., 2001). Thus, all animals
transported in the present study should have experienced
most of the fi ll shrink because all journeys were longer
than 3.25 h (González et al., 2012a). Tissue or carcass
shrink is the result of cellular losses of fl uids and body
reserves of energy and nutrients through sweating,
oxidation and respiration which may account for over
half of the total shrink (Self and Gay, 1972; Jones et
al., 1990). In contrast to fi ll shrink, carcass shrink is
experienced after long periods without feed and water
and cattle require more time to physiologically recover
the lost BW (Self and Gay, 1972; Phillips et al., 1991).
Average shrink of feeder cattle in the present study
is comparable with previous studies that reported
approximately 8% of BW loss despite the fact that
animals had been transported between 22 and 34 h
(Self and Gay, 1972; Camp et al., 1981; Arthington et
al., 2008) or held off feed and water from 12 to 16 h
(Aiken and Tabler, 2004; Phillips et al., 2006). Therefore,
rate of shrink (%/h) seems to be highly variable among
studies which could be explained by differences in cattle
Table 5. Body weight loss experienced by cattle, and intercept and regression coeffi cient transport time, during
long haul transport from and to Alberta, Canada (> 400 km)
Item
Cattle category
P-value
Fat cattle Feeder cattle Calves Cull cattle
N 3,935 497 17 11 −
Shrink,
1
%BW 4.90 ± 0.115
y
7.94 ± 0.157
x
6.13 ± 0.989
xy
6.60 ± 1.325
xy
< 0.001
Corrected shrink 1,
2
%BW 4.89 ± 0.079
y
7.11 ± 0.1812
x
4.22 ± 1.384
y
7.09 ± 1.320
x
< 0.001
Intercept 1 3.06 ± 0.09 4.42 ± 0.44 -0.15 ± 2.91 -2.01 ± 4.33 0.002
P-value
3
< 0.001 < 0.001 0.66 0.97 −
Time on truck, %BW/h 0.089 ± 0.0029
z
0.127 ± 0.0185
y
0.240 ± 0.1104
xy
0.533 ± 0.2348
x
0.07
3
P-value
4
< 0.001 < 0.001 0.05 0.06 −
Corrected shrink 2,
5
%BW 4.36 ± 0.251
y
6.31 ± 0.255
x
4.94 ± 1.146
xy
10.06 ± 1.615
x
< 0.001
Intercept 2,
5
%BW 2.31 ± 0.129
z
4.26 ± 0.244
y
2.89 ± 1.147
yz
8.00 ± 1.687
x
< 0.001
x-z
Within a row, means without a common superscript differ (P ≤ 0.05).
1
Not corrected by linear effects of time on truck or temperature.
2
All means are corrected for the linear effects of time on truck for each cattle category, and overall Midpoint temperature (β = 0.0142 ± 0.00436% of BW / ºC;
P < 0.001) and Transport time × Temperature interaction (β = 0.0023 ± 0.000266% of BW / (h × ºC); P < 0.001).
3
P-value of the Cattle Category × Time on Truck interaction.
4
P-value for the difference from zero of the regression coeffi cient and intercept.
5
Corrected for all terms in the model containing the main effects of origin, time of day of loading of the animals, driver experience, time on truck, temperature,
and time × temperature.
Table 4. Body weight loss experienced by each cattle
category according to origin during long haul transport
from and to Alberta, Canada (>400 km)
Origin
Cattle category
Fat cattle Feeder cattle Calves Cull cattle
Shrink,
1
%BW
Auction 6.53 ± 0.971 3.87 ± 0.562
b
1.90 ± 3.284 7.62 ± 1.564
Farm 5.28 ± 0.207
y
7.19 ± 0.362
ax
7.46 ± 2.682
xy
13.46 ± 4.931
xy
Feedyard 4.90 ± 0.112
y
8.05 ± 0.166
ax
8.19 ± 1.472
xy
−
x-z
Within a row, means without a common superscript differ (P ≤ 0.05).
a-c
Within a column, means without a common superscript differ (P ≤ 0.05).
1
Not corrected by linear effects of time on truck or temperature.
González et al.
3636
management and environmental conditions, although
such factors were rarely considered and not reported
in many studies. Surprisingly, few published transport
studies have assessed the extent of shrink in fat cattle
and cull cows during long haul transport without access
to water between weighing, however shrinkages in the
present study are close to those reported for cattle that
have been feed and water deprived. For example, Jones et
al. (1990) found that shrink of slaughter cattle transported
for 5 km to the abattoir increased from 3.1 to 10.6%
of BW as time off feed and water while held in pens
increased from approximately 4 to 48 h before slaughter
but a plateau was reached at 36 h. Similarly, Warriss et
al. (1995) reported that 12- to 18-mo old steers shrank
4.6, 6.5, and 7.0% of BW when transported for 5, 10, and
15 h, respectively. When lactating beef cows were fasted
for 8 to 22 h shrink increased from approximately 5.2 to
10% of BW (Heitschmidt, 1982). The amount of shrink
reported for 160-kg calves that were transported between
30 and 44 h was between 7.5 and 9.5% of BW (Lofgreen
et al., 1975) and between 6.6 and 10% of BW for 150-d
old calves transported for 15 h (Schwartzkopf-Genswein
et al., 2007), which are in agreement with results of the
present study. Also in agreement with the present study,
Self and Gay (1972) observed no difference in shrink
between calves and feeder cattle. However, Phillips
et al. (1991) reported that 16-mo-old steers shrank less
than 9-mo-old steers. There are several potential reasons
for reduced shrink reported in fat compared with feeder
cattle in the present study. First, slaughter weight cattle
may have better body condition to cope with the stress
and conditions of transport. Second, most of the fat cattle
in the present study were loaded at feed yards where high-
concentrate diets are normally fed whereas more feeder
cattle were loaded at farms where rations typically contain
greater proportions of roughage. Third, fat cattle were
more frequently loaded in the morning (0600 to 1200 h;
data not shown) which has been associated with less gut
fi ll and therefore shrink as found in the present study and
supported by Coffey et al. (1997). Fourth, fat cattle were
transported for shorter durations compared with feeder
cattle although at similar ambient temperatures. However,
shrink of fat cattle was still less after accounting for
all these effects (i.e., origin, loading time, and driver
experience) but differences in rate of shrinkage with
increasing time on truck disappeared (% of BW/h). This
may indicate that factors other than those considered in
the present study may lead to greater shrink of feeder
compared with fat cattle in addition to the longer transport
duration of the later. One speculation is that feeder cattle
are more stressed during transport as a result of no
previous experience with transport and human handling
because of more extensive rearing, and most likely recent
weaning. Stressful conditions are associated with greater
BW loss because of mobilization of body reserves, and
greater defecation and urination (Warriss, 1990; Phillips
et al., 1991; Parker et al., 2003).
The capacity of the gut and the amount of feed
contained within it depends on the quality and quantity
of feed consumed before loading and may therefore
determine the extent of shrink cattle experience. The
origin of cattle may be related to both the quality and
quantity of the diets consumed. For instance, cattle
loaded at auction markets shrank less than those loaded
in feed yards and farms during the present study. This
may be due in part to the fact that cattle already lost
some gut fi ll while being transported to the markets in
the present study as well as the fact that the stressful
Table 6. Numerator degrees of freedom, value of F-ratio
test, and probability value for each factor left in the fi nal
model affecting body weight loss during long haul trans-
port of cattle (>400 km)
Effect Numerator df
F-value P-value
Cattle category
1
3 74.51 <0.001
Place of origin
2
2 8.8 <0.001
Driver experience
3
3 3.8 0.01
Loading time
4
3 29.2 <0.001
Time on truck, linear
5
1 61.0 <0.001
Time on truck, quadratic
5
1 11.8 <0.001
Midpoint ambient temperature
6
1 31.8 <0.001
Time on truck × Midpoint temperature 1 27.0 <0.001
1
Cattle category was calves, feeder cattle, fat slaughter cattle, or cull cattle.
2
Place of origin was feedlot, farm/ranch, or auction market.
3
Driver experience was less than 2, 3 to 5, 6 to 10, and >10 yr hauling cattle.
4
Loading time was morning, afternoon, evening, and nighttimes.
5
Time since start of loading to fi nish of unloading animals.
6
Midpoint ambient temperature was the average of maximum and
minimum temperature registered within each journey.
Figure 3. Effect of time spent on truck and average ambient temperature
during the journey on shrink of fat cattle during commercial long haul
transport in North America (> 400 km). Add 1.56% of BW for feeder cattle,
2.60 for calves, and 3.56 for cull cattle to the value from the any point in
the fi gure. Ambient temperature was the midpoint between the minimum and
maximum values reported within each journey.
Shrink of cattle during transport
3637
environment in auction markets may not encourage
animals to consume feed and water even if available.
This was more evident in feeder cattle as a greater
proportion of them were loaded at markets in contrast
to fat cattle which were loaded at feed yards. In addition,
the place of origin of cattle may also be related to gut-fi ll
as a result of the quality of the diet consumed (Phillips et
al., 2006) or whether they have been preconditioned or
not (Schwartzkopf-Genswein et al., 2007). For instance,
slaughter cattle loaded at feed yards are commonly fed
high quality, concentrate diets which are associated with
less gut-fi ll and therefore less shrink. In contrast, cattle
fed diets with greater forage content have greater gut
volume or fi ll which leads to greater shrinkage during
fasting (Dinius and Cross, 1978; Phillips et al., 2006),
greater water excretion through urine and feces, and
shrink at a faster rate (Cole et al., 1986). Schwartzkopf-
Genswein et al. (2007) reported a marked difference
in BW loss between conditioned and nonconditioned
calves (23.6 and 14.6 kg, respectively) during long-
haul transport. This was attributed to the conditioned
calves being weaned and consuming long-hay and grain
before transport thereby having more gut-fi ll compared
with nonconditioned calves that were still suckling
until the day of transport. In contrast to the results of
our study, Self and Gay (1972) reported a shrink of 7.2
and 9.1% of BW for feeder cattle loaded at ranches and
sale yards, respectively; however, the reasons for such
discrepancies between studies is not evident. Finally,
the extent of gut fi ll, and therefore shrink, is affected
by the amount and time of the last voluntary feed and
water consumed before loading. Cattle normally have a
major feeding bout after sunrise and gut-fi ll increases
throughout the day as shown by Coffey et al. (1997)
who reported that the amount of BW lost (kg) and shrink
(% of BW) recorded after a 16-h fast increased when
animals were allowed to graze for longer after sunrise.
In agreement with this, the present study found that
animals loaded in the afternoon and evening shrank
more compared with those loaded at night or in the
morning as 90% of ‘morning’ haul cattle were loaded
between 0600 and 0800 h (data not shown). In addition,
we previously reported that only 6% of all truck loads
assessed had no access to feed and water for more than
30 min before loading. This is due to the fact that it is
common industry practice to allow cattle access to both
feed and water until the time of loading (González et al.,
2012a).
Quantifi cation of the effect of time on truck using
large data sets and considering other factors such as
ambient temperature, animal type and management
have not been previously reported. The intercept for
shrinkage observed in the multiple regressions for fat
and feeder cattle corresponds well with previously
reported fi ll-shrink (Self and Gay, 1972; Coffey et al.,
2001). This fi ll shrink also agrees with the rapid BW loss
of approximately 1% of BW/h within 3 to 4 h after feed
and water withdrawal reported by Coffey et al. (1997).
Aiken and Tabler (2004) also reported similar values up
until 4 h, however shrinkage continued increasing by
approximately 0.5% of BW/h up until 10 h after fasting.
The present study showed a quadratic increase in shrink
with time on truck; however, shrink reached a plateau at
low but not at high ambient temperature within transport
durations of up to 45 h. Unfortunately, there were no
data points during the period of time where the rate of
shrink was most rapid (<4 h) and a low frequency (4.7%)
of values occurring at greater than 30 h (González et
al., 2012a). Interestingly, the rate of shrink as a result of
increasing time on truck (%/h) was slowest in fat cattle,
greater in feeder cattle and calves, and fastest in cull
cattle. Cattle category, origin and loading time may help
to explain this result as such differences disappeared
after their introduction into the model.
The average rate of shrink with greater transport
duration (%/h) and temperature (%/ºC) increased at high
ranges of both time on truck and ambient temperature
(fastest in the upper right corner than in the lower left
corner of Figure 3). For instance, the rate of shrink during
the fi rst 20 h was approximately 0.12%/h at an ambient
temperature of 0ºC and 0.17%/h at 30ºC. Similarly, the rate
of shrink increased by 0.03%/ºC for animals transported 5
h and at 0.09%/ºC if transported for 40 h. Therefore, the
rate of shrink increases most rapidly as both temperature
and time on truck increase concurrently because of their
combined effects on the animals. Such factors should
be considered together to thoroughly understand and
manage BW loss in cattle. Thus, it seems more benefi cial
and important to reduce transport time under hot weather
conditions to avoid rapid and large increases in BW loss.
However, ambient temperature by itself (main effect) had
a signifi cant but relatively small effect on shrink (i.e., 0.2%
of BW per 10ºC of increase in temperature). Phillips et al.
(1991) reported greater shrinkage at high (9.5% of BW; 18
to 34ºC) compared with low (7.7% of BW; −16 to −6ºC)
ambient temperatures in feeder cattle transported for 48 h.
Interestingly, data from that study suggested that a greater
proportion of the BW lost at high temperatures was from
tissue and not fi ll shrink, which may be more detrimental
to animal welfare because it indicates that body water
reserves are being exhausted and less readily available
to cope with the stress of transport such as evaporative
cooling. The results of our study are diffi cult to compare
with previous research because they did not consider the
combined effects of transport, temperature, time, and the
origin of the animals on shrink. In feeder cattle, Self and
Gay (1972) reported a rate of shrink of 0.38% units per
González et al.
3638
100 km of transport distance although the background of
the animals and transport duration was unknown.
The fact that time on truck, temperature and
management (such as time of the day and origin of animal
loading) may affect shrinkage for each cattle category to
different extents suggests that these factors should be
considered together when assessing factors that affect
shrink. Furthermore, other factors not considered in the
present study may affect the extent of shrinkage during
transport such as relative humidity, which is used to
calculate the temperature humidity index. According to
Randall (1993), humidity should also be considered to
assess the effect of trailer microclimate on animal welfare
outcomes, especially at temperatures above 30ºC. The
design of the vehicles and loading density of animals might
also affect trailer microclimate, and this is a strong reason
to avoid extrapolating the results from the present study to
other conditions. A companion paper of the present study,
indicated that the incidence of cattle becoming lame, non-
ambulatory, and dead onboard increased sharply in loads
where shrink reached values above 10% of BW (González
et al., 2012c). Interestingly, González et al. (2012c) also
showed sharp increases in the incidence of those welfare
outcomes as transport duration increased above 30 h
and ambient temperature above 30ºC. The present study
shows that 10% of BW loss would be experienced by
calves, feeder cattle, and cull cattle at approximately 30 h
of transport and 30ºC but shrink does not reach this value
in fat cattle (Figure 3). Onboard mortality was greater
in feeder compared with fat cattle, whereas calves and
cull cattle showed the greatest incidence of lame, non-
ambulatory, and dead animals, which is similar to the
pattern of shrink values among cattle category presented
herein. These results demonstrate the close relationship that
exists between transport duration, ambient temperature,
shrink, and cattle category with welfare outcomes such
as non-ambulatory and dead animals. Furthermore, this
demonstrates that shrink is cumulative because it increases
with the number and severity of stressors to which cattle
are exposed. Our results also indicate that animals being
transported by drivers with less experience (≤5 yr) shrank
more compared with those transported by drivers with
more experience (≥6 yr). Although such a difference
(0.3% of BW) might not have practical importance, it may
be indicative of transport conditions and better driving
quality and skills of more experienced drivers resulting
in smoother cornering, breaking and shifting gears, and
minimize delays. This could reduce the physical exertion
and energy expenditure cattle require to maintain balance
during transport. Interestingly, animals transported by
drivers with ≤5 yr of experience hauling cattle showed a
greater likelihood of becoming non-ambulatory compared
with those transported by drivers with ≥6 yr of experience
(González et al., 2012c).
In conclusion, shrink during the long haul transport
of beef cattle is an important issue because it reduces
profi tability as it is related to live and carcass weight, meat
quality, performance, morbidity, and mortality during
and after transport. Body water and energy reserves are
essential for animals to cope with stress and environmental
demands during transport. Its usefulness as a monitoring
tool is augmented by the facts that shrink is easy to
measure and is already part of the standard documentation
process of the transport industry. Transport duration
was the single variable having the greatest infl uence on
shrink, especially at high ambient temperatures because
both factors have a multiplicative effect on each other.
Thus, every attempt should be made to reduce transport
duration and shrink such as avoiding unnecessary delays
by thorough planning of the journey and border crossing
inspections for feeder cattle. Transport management should
be more careful during hot weather as it will exacerbate
the effects on shrink. Type of animal or category also
had a great infl uence on shrink because it determines the
physical and physiological status of the cattle at the time
of transport. Thus, fattened cattle going to slaughter were
the most able to cope with transport showing the lowest
shrink compared with feeder and cull cattle. However,
caution must be exercised if using shrink as an indicator
of transport conditions because it can be strongly affected
by numerous factors such as time of loading (morning vs.
afternoon) and place of origin (e.g., auction vs. feedlot).
LITERATURE CITED
Aiken, G. E., and S. F. Tabler. 2004. Technical note: Infl uence of fast-
ing time on body weight shrinkage and average daily gain. Prof.
Anim. Sci. 20:524–527.
Arthington, J. D., X. Qiu, R. F. Cooke, J. M. B. Vendramini, D. B.
Araujo, Jr., C. C. Chase, and S. W. Coleman. 2008. Effects of pre-
shipping management on measures of stress and performance of
beef steers during feedlot receiving. J. Anim. Sci. 86:2016–2023.
Camp, T. H., R. A. Stermer, and D. G. Stevens. 1983. Shrink of feeder
calves as an indicator of incidence of bovine respiratory disease
and time to return to purchase weight. J. Anim. Sci. 57(Suppl 1):66.
Camp, T. H., D. G. Stevens, R. A. Stermer, and J. P. Anthony. 1981.
Transit factors affecting shrink, shipping fever and subsequent
performance of feeder calves. J. Anim. Sci. 52:1219–1224.
Coffey, K. P., F. K. Brazle, J. J. Higgins, and J. L. Moyer. 1997.
Effects of gathering time on weight and shrink of steers grazing
smooth bromegrass pastures. Prof. Anim. Sci. 13:170–175.
Coffey, K. P., W. K. Coblentz, J. B. Humphry, and F. K. Brazle. 2001.
Review: Basic principles and economics of transportation
shrink in beef. Prof. Anim. Sci. 17:247–255.
Cole, N. A., W. A. Phillips, and D. P. Hutcheson. 1986. The effect of
pre-fast diet and transport on nitrogen metabolism of calves. J.
Anim. Sci. 62:1719–1731.
Dinius, D. A., and H. R. Cross. 1978. Feedlot performance, carcass
characteristics and meat palatability of steers fed concentrate
for short periods. J. Anim. Sci. 47:1109–1113.
Shrink of cattle during transport
3639
González, L. A., K. S. Schwartzkopf-Genswein, M. Bryan, R. Silasi,
and F. Brown. 2012a. Benchmarking study of industry prac-
tices during commercial long haul transport of cattle in Alberta,
Canada. J. Anim. Sci. 90:3606–3617.
González, L. A., K. S. Schwartzkopf-Genswein, M. Bryan, R. Silasi,
and F. Brown. 2012b. Space allowance during commercial
long distance transport of cattle in North America. J. Anim. Sci.
90:3618–3629.
González, L. A., K. Schwartzkopf-Genswein, M. Bryan, R. Silasi, and F.
Brown. 2012c. Relationships between transport conditions and wel-
fare outcomes during long haul transport of cattle in North America.
J. Anim. Sci. 90:3640–3651.
Heitschmidt, R. K. 1982. Diurnal variation in weight and rates of
shrink of range cows and calves. J. Range Manage. 35:717–720.
Jones, S. D. M., A. L. Schafer, W. M. Robertson, and B. C. Vincent.
1990. The effects of withholding feed and water on carcass
shrinkage and meat quality in beef cattle. Meat Sci. 28:131–139.
Lofgreen, G. P., J. R. Dunbar, D. G. Addis, and J. G. Clark. 1975.
Energy level in starting rations for calves subjected to market-
ing and shipping stress. J. Anim. Sci. 41:1256–1265.
Parker, A. J., G. P. Dobson, and L. A. Fitzpatrick. 2007. Physiological
and metabolic effects of prophylactic treatment with the osmo-
lytes glycerol and betaine on Bos indicus steers during long du-
ration transportation. J. Anim. Sci. 85:2916–2923.
Parker, A. J., G. P. Hamlin, C. J. Coleman, and L. A. Fitzpatrick.
2003. Dehydration in stressed ruminants may be the result of a
cortisol-induced diuresis. J. Anim. Sci. 81:512–519.
Phillips, W. A., S. W. Coleman, and H. S. Mayeux. 2006. Case study:
Changes in body weight, fi ll, and shrink of calves grazing wheat
pasture in the winter and spring. Prof. Anim. Sci. 22:267–272.
Phillips, W. A., P. E. Juniewicz, and D. L. VonTungeln. 1991. The ef-
fect of fasting, transit plus fasting, and administration of adreno-
corticotropic hormone on the source and amount of weight lost
by feeder steers of different ages. J. Anim. Sci. 69:2342–2348.
Randall, J. 1993. Environmental parameters necessary to defi ne
comfort for pigs, cattle and sheep in livestock transporters.
Anim. Prod. 57:299–307.
Schaefer, A. L., S. D. M. Jones, A. K. W. Tong, B. A. Young, N. L.
Murray, and P. Lepage. 1992. Effects of post-transport electrolyte
supplementation on tissue electrolytes, hematology, urine osmo-
lality and weight loss in beef bulls. Lives. Prod. Sci. 30:333–346.
Schwartzkopf-Genswein, K. S., M. E. Booth-McLean, M. A. Shah,
T. Entz, S. J. Bach, G. J. Mears, A. L. Schaefer, N. Cook, J.
Church, T. A. McAllister. 2007. Effects of pre-haul manage-
ment and transport duration on beef calf performance and wel-
fare. Appl. Anim. Behav. Sci. 108:12–30.
Self, H. L., and N. Gay. 1972. Shrink during shipment of feeder cat-
tle. J. Anim. Sci. 35:489–494.
Tarrant, V., and T. Grandin, 2000.
Cattle transport. Pages 151–173
in
Livestock handling and transport. 2nd ed. T. Grandin,
ed. CABI Publishing, Wallingford, Oxon, UK.
Warriss, P. D. 1990. The handling of cattle pre-slaughter and its effects
on carcass and meat quality. Appl. Anim. Behav. Sci. 28:171–186.
Warriss, P. D., S. N. Brown, T. G. Knowles, S. C. Kestin, J. E.
Edwards, S. K. Dolan, and A. J. Phillips. 1995. Effects on cattle
of transport by road for up to 15 hours. Vet. Rec. 136:319–323.
