- Email: [email protected]
Effects of under- and overstocking freestalls on dairy cattle behaviour
G Model
ARTICLE IN PRESS
APPLAN-4085; No. of Pages 6
Applied Animal Behaviour Science xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim
Effects of under- and overstocking freestalls on dairy cattle behaviour Christoph Winckler a,∗ , Cassandra B. Tucker b , Daniel M. Weary c a Department of Sustainable Agricultural Systems, Division of Livestock Sciences, University of Natural Resources and Life Sciences, Vienna, Gregor-Mendel-Strasse 33, A-1180 Vienna, Austria b Department of Animal Science, University of California, Davis, One Shields Avenue, Davis, CA 95616, USA c Animal Welfare Program, University of British Columbia Vancouver, 2357 Main Mall, Vancouver, BC V6T 1Z4, Canada
a r t i c l e
i n f o
Article history: Received 28 November 2014 Received in revised form 10 June 2015 Accepted 15 June 2015 Available online xxx Keywords: Cubicle Animal welfare Stocking density Housing Agonistic interactions
a b s t r a c t Freestall availability affects cattle behaviour and most studies in this area have focused on overstocking. We studied the effects of three levels of stall availability, including both over- and understocking on the time budgets and agonistic interactions in 36 dairy cattle in four stable groups. Using a switch-back design, with treatment order balanced, groups of nine cows were given access to 6, 9, and 12 stalls for 1 week each, allowing for a within-cow test of stocking density of 150, 100 and 75% (cows/stalls). After 5 days of acclimatization at each density, time budgets and displacements from stalls were measured during the last 48 h of each treatment period using continuous video recording and direct observation. When animals had access to fewer stalls, they spent less time lying down (11.6, 12.6, 12.8 h/24 h in 150, 100 and 75% treatments, respectively; SE: 0.31, 0.31, 0.28 h/24 h), particularly at night (6.6, 7.5, 7.6 h; SE: 0.20, 0.20, 0.17 h). Lying behaviour was also more synchronous when more freestalls were available (Kappa coefficient of agreement 0.00, 0.13, 0.17 for lying time in 150, 100 and 75% treatments, respectively). Cows spent more time standing in the alleyways when overstocked (between two rows of stalls: 1.8, 0.8, 0.6 h/24 h in 150, 100 and 75%, respectively; SE: 0.09, 0.09, 0.06 h/24 h; between feeder and stalls: 1.5, 1.3, 1.3 h/24 h in 150, 100 and 75%, respectively; SE: 0.13, 0.13, 0.11 h/24 h), but did not alter the time they spent feeding. Moreover, cows were more likely to displace one another from stalls at greater stocking densities (2.9, 1.1, 0.6 displacements per cow/24 h in 150, 100 and 75% treatments, respectively; SE: 0.16, 0.16, 0.11 displacements per cow/24 h). Cows that were less successful at displacing others spent a higher proportion of their time lying during the day when overstocked, indicating that lying during this time is less preferred. For all variables, the magnitude of response was most affected by overstocking; this practice reduced lying time, especially at night, synchrony of lying behaviour and increased competition for stalls. Understocking provided benefits, but the degree of behavioural change was smaller than when stalls were limited. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Recent surveys suggest that in four out of every 10 USA freestall barns there are more cows than stalls available (USDA, 2010). Dairy producers overstock to save building costs, or because herd growth is not matched by expansion of barns. Epidemiological evidence indicates that this overstocking explains a significant portion of the non-dietary variation in milk production, with a lower milk yield per cow in overstocked farms (Bach et al., 2008). The mechanism associated with lower productivity is not understood, but could be related to increased lameness within the herd (e.g. Leonard et al., 1996), as lame cows or those with hoof injuries produce less milk
∗ Corresponding author. Tel.: +43 1 47 654 3261; fax: +43 1 47 654 3254. E-mail address: [email protected] (C. Winckler).
(Amory et al., 2008; Archer et al., 2010; Green et al., 2002). Indeed, the cow:stall ratio was eligible for inclusion in the final model evaluating lameness prevalence in two recent epidemiological studies (Dippel et al., 2009; Espejo and Endres, 2007). Increased lameness in overstocked barns is likely mediated by changes in behaviour. Cows spend less time lying down (Fregonesi et al., 2007; Hill et al., 2009; Krawczel et al., 2008, 2012) and less time standing partially in the stall (Hill et al., 2009; Lombard et al., 2010) when fewer freestalls are available. These changes in time budgets, particularly lying time, are robust; they are apparent regardless of the method used to experimentally overstocking freestalls, by either blocking stalls or adding new individuals to the group (Krawczel et al., 2012) or the method used to record stall usage, either through continuous monitoring over 24-h periods (Fregonesi et al., 2007; Hill et al., 2009; Krawczel et al., 2012) or less accurate single measures within a day
http://dx.doi.org/10.1016/j.applanim.2015.06.003 0168-1591/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model APPLAN-4085; No. of Pages 6
ARTICLE IN PRESS C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
2
(Krawczel et al., 2008; Ito et al., 2009). However, lying time is not consistently affected by stocking density in on-farm research. Some studies have failed to find a relation between stocking density and lying behaviour (Charlton et al., 2014; Ito et al., 2014; Lombard et al., 2010). Understocking has been included in the epidemiological comparisons (e.g. median stocking density was 96% in Charlton et al., 2014; range was 71–197% in Ito et al., 2014), but has been rarely included in experimental studies. Thirty-nine percent of US freestall farms understock (USDA, 2010) and recent research suggests that this practice could increase lying times beyond those seen when 1 stall is offered for every cow (Telezhenko et al., 2012). However, little is known about the other effects of this practice. For example, cows kept in overstocked conditions are more likely to both compete directly for freestalls and adopt non-competitive strategies, such as using the freestalls more at non-peak times, and thus reducing synchrony of lying behaviour (Fregonesi et al., 2007). It is unclear how understocking would affect competition and stall use. Our objective was to better understand how both under- and overstocking freestalls affects time budgets, social behaviour and synchrony of dairy cattle in stable groups.
2. Materials and methods 2.1. Animals and treatments The experiment was conducted at the University of British Columbia Dairy Education and Research Centre in Agassiz, British Columbia. Thirty-six dairy cows were randomly assigned to four groups of nine animals averaging (mean ± SD) 40.2 ± 8.07 kg/day milk production, 146 ± 25.6 days in milk, 2.4 ± 1.19 parity, 679 ± 70.2 kg body weight and 47 ± 16.3 months in age. Each experimental pen (width = 7.5 m, length = 13.5 m) contained 12 freestalls configured in three rows. In two rows, the stalls were open at the front (‘head-to-head’; two cows facing one another) and had a bed length of 2.4 m. The third row of freestalls faced a cement wall, and these stalls were 0.3 m longer to allow more space for the cow to lunge forward when getting up and lying down. Freestalls were separated by Dutch comfort style partitions and measured 1.2 m wide centre to centre and the neck rail was 1.14 m from the stall surface. Stalls were deep-bedded with 0.4 m of sand. The flooring throughout the pens was grooved concrete. The alley closest to the feed bunk measured 3.5 m and the floor of half of this alley was inlaid with a 2.5 cm-thick durable rubber mat. The rubber surface was level with the concrete flooring and had grooving similar to the concrete flooring. Alleys were cleaned six times per day with automatic scrapers. Each pen had 7.5 m of feed bunk space available through a post and rail barrier. Animals were fed for ad libitum consumption with a total mixed ration of corn silage, grass silage, barley, canola meal and soybean meal. Fresh feed was provided twice daily (at 06:00 h and 15:30 h) and feed was pushed up three times per day. Water was freely available from a self-filling trough. Cows were milked twice daily (approximately 06:00 h and 17:00 h) and spent 1.3 h/day (0.50 SD) away from the pen. We manipulated stocking density by providing 6, 9, or 12 freestalls to groups of nine animals, thus creating stocking densities of 150, 100, and 75% (cows/stalls), respectively. Using a switch-back design with treatment order balanced, all groups were exposed to all treatments for 1 week each. Each group returned to the 75% stocking density treatment after exposure to either the 150% or the 100% treatment. Stalls along one side of the pen were blocked to create the 100% and 150% treatments. For the 100% treatment, the three stalls along a wall and closest to the crossover/water trough were blocked. To create the 150% treatment, cows were denied
access to the three stalls adjacent to the stalls blocked for the 100% stocking density treatment. 2.2. Sampling and measurement Behaviour was recorded using eight Panasonic WV 330 cameras, positioned approximately 10 m above the experimental pens for 48 h during each of the 5 weeks (i.e. days 6 and 7) at 3 frames/s. The cameras were attached to a Panasonic video multiplexer (WV-FS416) and time-lapse recorder (AG-6540p; Panasonic; Mississauga, Ontario, Canada). Red lights (100 W) were hung approximately 10 m above the pens to facilitate video recording at night. Cows were marked with unique symbols using hair dye to identify individuals. Behaviour was recorded in two ways (Table 1). Time budgets were scored from video using instantaneous scan sampling once every 10 min. At each scan we scored if the cow was in the freestall (lying, standing with two or four legs in the stall), standing between the feeder and stalls, in the crossover, between the two rows of stalls, or at the feeder (Table 1). The number of displacements from the stalls was also recorded for each cow during 48 h of continuous observation from video, resulting in 240 h of observations per cow. Finally, we recorded social interactions at the feeder with continuous live observation in the 2 h after morning and afternoon milking. Each pen was observed for 30 min twice a day for 2 days during each week, for a total of 10 h observation/pen. The order of the 30-min observation period was balanced across pens and days. Both aggressive and positive interactions were recorded (Table 1). Displacements from the stalls were used Table 1 Definitions of behaviours used to evaluate the effects of freestall availability on dairy cattle. Behaviour In the stall Lyinga Standing with front two legsa Standing with four legsa Displacementa In the alley Standing between feeder and stallsa Standing in crossovera Standing between two rows of stallsa At the feeder Feedinga Displacementsb
Head buttingb
Pushingb
Total agonistic interactionsb Positive interactionsb
a b
Definition Flank in contact with ground Two front legs in contact with the stall surface, weight on legs Four legs in contact with the stall surface, weight on legs Contact (butt or push) immediately followed by leaving the stall Four legs in contact with the alley surface between the feeder and the row of stalls closest to the feeder, weight on legs, head not above the feeder Four legs in contact with the alley surface on the crossover between the alleys, weight on legs Four legs in contact with the alley surface on the alley between the two rows of stalls, weight on legs Presence at the feeder with the head above the feed bunk Contact (butt or push) immediately followed by leaving the feeding place by at least a cow width or half a cow length Contact with the forehead or horn base using a forceful movement, receiving animal does not give up its present position Applying steady force with any part of the head, receiving animal does not give up its present position Sum of displacements, head butting and pushing Sum of social licking (tactile oral contact directed to the body except the anal region, udder or claws) and horning (rubbing of foreheads or horn bases against the head or neck of another animal)
Scored from video recordings (48 h/week). Scored with live observation (2 h/week).
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model APPLAN-4085; No. of Pages 6
ARTICLE IN PRESS C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
to calculate an index of social success (Mendl et al., 1992; Galindo and Broom, 2000). This index of social success could vary from 0 to 1 and equalled the number of times the cow displaced other cows divided by the total number of times she displaced other cows and was displaced herself.
3
affected by the 100% and 150% treatments. Pearson correlation coefficients were calculated between the time spent lying in stalls blocked to create the 150% and 100% treatments during the periods with 75% stocking density (%) and how much the cow’s lying time was influenced by the treatments (lying time in 75% treatmentlying time in 150% treatment).
2.3. Statistical analyses 3. Results Several additional dependent variables were calculated: time spent lying during the day (time between am and pm milking), proportion of lying time occurring during the day (time between am and pm milking) and time to lie down after return from milking. For all variables, data were averaged to create single values for each 48 h period, i.e. days 6 and 7 of the 5 weeks of treatment. A linear mixed model (procedure MIXED; SAS) was used to analyze differences between the three levels of freestall availability (2 df) for all response variables as well as the interaction between treatment and cows’ ability to displace other cows from the stalls (1 df). The model included week as a repeated measure and cow within pen as subject, using a compound symmetry covariance structure. If not significant, the interaction term was removed from the model. Specified contrasts were used to test the effects of stocking density (i.e. 150% vs. 100% and 100% vs. 75%; df 1). Distribution of residuals was graphically evaluated using normal-qq-plots to verify model assumptions. As a measure of synchrony (Asher and Collins, 2012), we also calculated generalized Kappa statistics for lying, standing and feeding in the three treatments. Using the data from instantaneous scan sampling, the %MAGREE macro in SAS computes estimates and tests of agreement with individual cows in the four groups treated as multiple raters and the behavioural categories lying, standing and feeding as responses. Effects of treatment (df 2) were tested using a general linear model (procedure GLM; SAS) that included pen. Contrasts (df 1) were specified as described above. In addition to the primary objective of the study, we also explored whether cows that used the blocked stalls during the 75% stocking density periods (12 stalls available) were differentially
Cows spent less time lying in the freestalls when fewer stalls were available, with a 1.2 h difference between the 75% and 150% treatments (Table 2). The change in lying time was more apparent at night (6.6, 7.5 and 7.6 h lying at night in 150%, 100%, and 75% treatments, SE: 0.20, 0.20 and 0.17 h, P < 0.01, Fig. 1), than during the day (5.0, 5.1, and 5.2 h lying during day in 150%, 100%, and 75% treatments, SE: 0.15, 0.15, 0.13 h, P = 0.04). The proportion of lying time during the day increased with overstocking, and time to lie down after return from milking was shorter (Table 2). Freestall availability also affected standing behaviour, with cows changing where they stood within the pen. Cows spent less time standing with only front legs in the stall when fewer freestalls were provided, spending on average 0.9 h/24 h at a stocking density of 150% compared with 1.3 h when stocking density was 100%. At higher densities cows spent more time standing around the stalls, both between the feeder and the stalls (P = 0.04) and in the area between two rows of freestalls (P < 0.01) as well as in the crossover area (P < 0.01, Table 2). There was no effect of density on time spent standing at the feeder. Cows were five times more likely to displace others from the stall when only six freestalls were available (150% stocking density), compared with the 75% stocking density treatment (Table 2). In total, we observed 426 displacements from the stall. Of these, 46%, 43% and 12% occurred when the displaced cow was lying, standing with the front two legs or all four legs in the stall, respectively. Social success, as measured with displacements from the stall, influenced the response to treatments in several situations. Cows that were less successful displacing others from the stall
Table 2 Mean and S.E. for durations and frequencies of behaviours at stocking densities of 150%, 100%, and 75% (i.e. 6, 9, and 12 stalls for 9 cows/group; n = 4 groups). P-values are for the test of stocking density and the interaction between density and a cow’s success at displacing other cows from the stall as well as for contrasts comparing specific densities. Variable
Stocking density 150% S.E.
In the stall Lying (h/24 h) Proportion of lying time during day (%) Latency to lie upon return from milking (min) Standing with front two legs (h/24 h) Standing with four legs (h/24 h) Displacements (no./24 h)
P
100%
75% S.E.
Stocking density
Intx with social success
S.E.
Specified contrasts 150% vs. 100%
100% vs. 75%
11.6 43.2 59 0.9 0.2 2.9
0.31 0.77 5.4 0.17 0.05 0.16
12.6 40.6 61 1.3 0.2 1.1
0.31 0.77 5.4 0.17 0.05 0.16
12.8 40.7 62 1.4 0.2 0.6
0.28 0.52 4.2 0.15 0.04 0.11
<0.01 <0.01 0.04 <0.01 <0.07 <0.01
excl.a <0.01 0.04 <0.01 excl.a n/ab
<0.01 <0.01 0.04 0.01 – <0.01
0.36 0.91 0.23 0.69 – 0.04
In the alley Standing between feeder and stalls (h/24 h) Standing in crossover (h/24 h) Standing between two rows of stalls (h/24 h)
1.5 1.3 1.8
0.13 0.11 0.09
1.3 1.1 0.8
0.13 0.11 0.09
1.2 1.1 0.6
0.11 0.09 0.06
0.04 <0.01 <0.01
excl.a 0.02 excl.a
0.17 0.20 <0.01
0.55 0.17 0.03
At the feeder Feeding (h/24 h) Displacements (no./2 h) Head butting (no./2 h) Pushing (no./2 h) Total agonistic interactions (no./2 h) Sociopositive interactions (no./2 h)
5.4 2.0 5.0 2.7 9.8 0.9
0.13 0.53 1.07 0.52 1.65 0.26
5.4 2.2 5.3 2.4 10.0 0.7
0.13 0.53 1.07 0.52 1.65 0.26
5.5 2.6 7.1 2.2 12.0 1.0
0.12 0.44 0.84 0.37 1.38 0.17
0.67 0.32 0.04 0.57 0.13 0.58
excl.a n/ac n/ac n/ac n/ac n/ac
– – 0.78 – – –
– – 0.11 – – –
a b c
Excluded from final model as not significant. Displacements were used to calculate social success. Not applicable to research question.
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model
ARTICLE IN PRESS
APPLAN-4085; No. of Pages 6
C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
4
100 90 80
% cows lying
70 60 75%
50
100%
40
150%
30 20
4:00
2:00
0:00
22:00
20:00
18:00
16:00
14:00
12:00
10:00
8:00
0
6:00
10
Fig. 1. Diurnal variation in lying time at stocking densities of 150%, 100% and 75% (i.e. 6, 9, and 12 stalls for 9 cows/group; n = 4 groups). Each point shows the average for all 36 cows at 10-min intervals.
A) Proportion of lying time during the day 70
- - - 75% 60
___100% … 150%
% 50 40 30
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
B) Standing time with two front legs in stall
Synchrony of lying, feeding and overall synchrony (when considering the behaviour categories lying, feeding and standing) was lower with greater stocking density (all P < 0.01, Table 3). The stalls blocked to create the 150% and 100% treatments were less popular during the 75% treatment periods compared with stalls available during the treatments with higher stocking density. The three stalls blocked to create the 100% treatment were used for 20.2 ± 18.4% (mean ± SD) of lying during the 75% treatment (expected: 25%), and stalls blocked to create the 150% treatment accounted for 46.1 ± 20.6% of lying during the 75% treatment (expected: 50%). However, cows that used the blocked stalls during the 75% treatment were not differentially affected by reduced stall availability (Fig. 3; rpearson ≤ 0.20).
h/24h
6 5
4. Discussion
4
Lying time was reduced when cows were overstocked. Cattle spent more time standing in alleyways when fewer stalls were provided, but feeding time was not affected by stall availability. Overstocking also reduced synchrony of lying behaviour, resulted in less lying at night and increased competition for stalls. Cows that were less successful at displacing others from stalls were more affected by overstocking than were socially dominant animals. Lying time was reduced by 1 h/day when stalls were overstocked from 100 to 150%; understocking provided relatively little benefit in terms of lying time (increasing just 0.2 h/24 h). Experimental studies of freestall availability report that lying time is always reduced when fewer stalls are provided. Most studies have focused on overstocking (Falk et al., 2012; Fregonesi et al., 2007; Hill et al., 2009; Krawczel et al., 2008, 2012). In agreement with the present study, Telezhenko et al. (2012) found a similar small increase in lying with understocking. Epidemiological studies, with both under- and overstocked farms included, have often not identified stocking density as an important risk factor for total lying time (Charlton et al., 2014; Ito et al., 2014; Lombard et al., 2010). The current finding that understocking has a much smaller effect on lying time may explain some of the discrepancy between these two types of evidence, as a focus solely on overstocking may make it easier to detect a clear pattern in cattle responses. Other factors, such as the individual variation of lying time and the power of within-animal tests (all experimental work to date uses this approach), also likely play a
3 2 1 0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Index of social success at freestalls
0.8
Fig. 2. Proportion of time lying during the day (Panel A) and standing with two front legs in the stalls (Panel B) in relation to the social success index as regards displacements from the stalls. Three stocking densities at the freestalls were compared: 150%, 100%, and 75% (i.e. 6, 9, or 12 freestalls for groups of nine cows).
spent more time lying during the day in the 150% treatment (Pintx social success = 0.01) and spent a higher proportion of their lying time during the day in the 150% treatment (Pintx social success < 0.01, Fig. 2A). Similarly, cows that were less successful changed their standing behaviour in the stall when there was less competition: they spent more time standing with two legs in the stall in the 75% and 100% treatments (Pintx social success < 0.01, Fig. 2B). A lower social success index was also associated with more time spent standing in the crossover in the 100% and 150% treatments (Pintx social success = 0.02).
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model
ARTICLE IN PRESS
APPLAN-4085; No. of Pages 6
C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
5
Table 3 Mean Kappa coefficients and SE for synchrony of lying, feeding, standing and overall synchrony at stocking densities of 150%, 100%, and 75% (i.e. 6, 9, and 12 stalls for 9 cows/group, n = 4 groups). P-values are for the overall test of density and for contrasts comparing specific densities. Behaviour
Lying Feeding Standing Overall
Stocking density
P
150%
100%
75%
SE
Stocking density
0.00 0.15 0.02 0.05
0.13 0.19 0.08 0.12
0.17 0.23 0.04 0.15
0.01 0.01 0.03 0.01
<0.01 0.01 0.34 <0.01
Specified contrasts 150% vs. 100%
100% vs. 75%
<0.01 0.04 – <0.01
0.01 0.06 – 0.02
5 4 Δ lyin ng time in 75% v vs. 150% 3 treatment 2 (h h/24h) 1 0 -1 -2
0
20
40 0
60
80
100
Time lying (%) in stalls blocked tto create 150% treatment
Fig. 3. Difference in lying time () at 75% and 150% stocking densities in relation to amount of time spent lying down during the 75% treatment in the six stalls blocked to reduce freestall availability.
role in discrepancies between epidemiological and experimentalbased results. When access to freestalls was not limited, cows spent more time lying at night and less during the day. Lying behaviour was also more synchronous with increasing stall availability. On pasture, cattle behaviour can be highly synchronous with 70–80% of cows, on average, lying down at the same time (Stoye et al., 2012). In the current study providing more stalls than animals, cows are better able to achieve what seems to be a preferred, higher level of synchrony and timing of rest. Differences in the degree of synchrony between systems (e.g. pasture vs. freestall barns, Miller and Wood Gush, 1991), could be due to differences in other zeitgebers, such as artificial light or provision of feed. Cattle prefer some stalls over others and this could also affect the degree of synchrony. For example, some of the stalls we blocked off to manipulate freestall availability were less popular than the remaining stalls in the pen. Cows also directly competed for stalls by displacing others. This type of competitive interaction was more common when freestalls were overstocked, as found by others (Fregonesi et al., 2007; Friend et al., 1977; Wierenga, 1990). However, displacements from the stalls further decreased with understocking, indicating that competition for stalls exists even when one stall is available for every cow. Such competition may be explained by the fact that stalls are not all equally preferred. Competitive interactions at the feeder were unaffected by stocking density, likely because feeding space was unchanged by the treatments. The higher number of head butts at the feeder with low stocking density was unexpected and remains difficult to explain. Cows that were unable to displace others from stalls were more affected by overstocking than were more dominant animals. For example, subordinate animals spent more time lying during the day, possibly a less preferred time for this behaviour (Sambraus, 1971; Fregonesi et al., 2007). Despite the increase in use during the day by subordinate animals, time to lie down after return from milking was only slightly affected by stall availability. The effect size (2–3 min/day) was less than the 13 min/day found by Fregonesi et al. (2007) and may not be biologically relevant. In addition, in agreement with Fregonesi et al. (2007), there was no relationship
between the animals’ success in displacing others from the stall and total lying time, possibly because cattle are highly motivated to maintain lying time (Jensen et al., 2005; Metz, 1985; Munksgaard et al., 2005). Subordinate animals also spent more time standing in the stall with two legs, compared with more dominant animals, when provided at least 1 stall/cow. Others have reported a similar pattern in standing behaviour (Galindo and Broom, 2000) and suggested that, when space is available, animals may perform this behaviour to hide. Indeed, freestall dividers greatly reduce aggressive interactions when used as barriers between cows at the feedbunk (DeVries and von Keyserlingk, 2006). Standing partially in the stall decreased, when cows were overstocked, but this was more than compensated for by an increase in standing in the crossover alleyway. This change in behaviour may increase hoof exposure to concrete and manure, considered risk factors for claw injuries, infectious diseases and lameness (Gregory et al., 2006; Somers et al., 2005; Webster, 2001). 5. Conclusions Behavioural responses were most affected by overstocking; this reduced lying time, especially at night, and synchrony of lying behaviour and increased competition for stalls. Subordinate animals were particularly affected by overstocking. Understocking provided more modest benefits, mainly related to reduced competition for stalls and improved synchrony of lying behaviour. In combination with other experimental research, this work illustrates the importance of providing cattle with at least 1 stall/cow. Acknowledgments We thank the staff at University of British Columbia’s Dairy Education and Research Centre and the University’s Animal Welfare Program. CW was funded by a research fellowship provided by Deutsche Forschungsgemeinschaft. The Animal Welfare Program is funded by Canada’s NSERC Industrial Research Chair Program with industry contributions from the Dairy Farmers of Canada, the British Columbia Dairy Association, the Westgen Endowment Fund,
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model APPLAN-4085; No. of Pages 6
ARTICLE IN PRESS C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
6
Intervet Canada Corporation, Zoetis, BC Cattle Industry Development Fund, Alberta Milk, Valacta and CanWest DHI. References Amory, J.R., Barker, Z.E., Wright, J.L., Mason, S.A., Blowey, R.W., Green, L.E., 2008. Associations between sole ulcer, white line disease and digital dermatitis and the milk yield of 1824 dairy cows on 30 dairy cow farms in England and Wales from February 2003–November 2004. Prev. Vet. Med. 83, 381–391. Archer, S.C., Green, M.J., Huxley, J.N., 2010. Association between milk yield and serial locomotion score assessments in UK dairy cows. J. Dairy Sci. 93, 4045–4053. Asher, L., Collins, L.M., 2012. Assessing synchrony in groups: are you measuring what you think you are measuring? Appl. Anim. Behav. Sci. 138, 162–169. Bach, A., Valls, N., Solans, A., Torrent, T., 2008. Associations between nondietary factors and dairy herd performance. J. Dairy Sci. 91, 3259–3267. Charlton, G.L., Haley, D.B., Rushen, J., de Passillé, A.M., 2014. Stocking density, milking duration, and lying times of lactating cows on Canadian freestall dairy farms. J. Dairy Sci. 97, 2694–2700. DeVries, T.J., von Keyserlingk, M.A.G., 2006. Feed stalls affect the social and feeding behavior of lactating dairy cows. J. Dairy Sci. 89, 3522–3531. Dippel, S., Dolezal, M., Brenninkmeyer, C., Brinkmann, J., March, S., Knierim, U., Winckler, C., 2009. Risk factors for lameness in freestall-housed dairy cows across two breeds, farming systems, and countries. J. Dairy Sci. 92, 5476–5486. Espejo, L.A., Endres, M.I., 2007. Herd-level risk factors for lameness in high-producing Holstein cows housed in freestall barns. J. Dairy Sci. 90, 306–314. Falk, A.C., Weary, D.M., Winckler, C., von Keyserlingk, M.A.G., 2012. Preference for pasture versus freestall housing by dairy cattle when stall availability indoors is reduced. J. Dairy Sci. 95, 6409–6415. Fregonesi, J.A., Tucker, C.B., Weary, D.M., 2007. Overstocking reduces lying time in dairy cows. J. Dairy Sci. 90, 3349–3354. Friend, T.H., Polan, C.E., McGilliard, M.L., 1977. Free stall and feed bunk requirements relative to behavior, production and individual feed intake in dairy cows. J. Dairy Sci. 60, 108–116. Galindo, F., Broom, D.M., 2000. The relationships between social behaviour of dairy cows and the occurrence of lameness in three herds. Res. Vet. Sci. 69, 75–79. Green, L.E., Hedges, V.J., Schukken, Y.H., Blowey, R.W., Packington, A.J., 2002. The impact of clinical lameness on the milk yield of dairy cows. J. Dairy Sci. 85, 2250–2256. Gregory, N., Craggs, L., Hobson, N., Krogh, C., 2006. Softening of cattle hoof soles and swelling of heel horn by environmental agents. Food Chem. Toxicol. 44, 1223–1227. Hill, C.T., Krawczel, P.D., Dann, H.M., Ballard, C.S., Hovey, R.C., Falls, W.A., Grant, R.J., 2009. Effect of stocking density on the short-term behavioural responses of dairy cows. Appl. Anim. Behav. Sci. 117, 144–149.
Ito, K., Weary, D.M., von Keyserlingk, M.A.G., 2009. Lying behavior: assessing within- and between-herd variation in free-stall-housed dairy cows. J. Dairy Sci. 92, 4412–4420. Ito, K., Chapinal, N., Weary, D.M., von Keyserlingk, M.A.G., 2014. Associations between herd-level factors and lying behavior of freestall-housed dairy cows. J. Dairy Sci. 97, 2081–2089. Jensen, M.B., Pedersen, L.J., Munksgaard, L., 2005. The effect of reward duration on demand functions for rest in dairy heifers and lying requirements as measured by demand functions. Appl. Anim. Behav. Sci. 90, 207–217. Krawczel, P.D., Hill, C.T., Dann, H.M., Grant, R.J., 2008. Effect of stocking density on indices of cow comfort. J. Dairy Sci. 91, 1903–1907. Krawczel, P.D., Mooney, C.S., Dann, H.M., Carter, M.P., Butzler, R.E., Ballard, C.S., Grant, R.J., 2012. Effect of alternative models for increasing stocking density on the short-term behavior and hygiene of Holstein dairy cows. J. Dairy Sci. 95, 2467–2475. Leonard, F.C., O’Connell, J.M., O’Farrell, K.J., 1996. Effect of overcrowding on claw health in first-calved Friesian heifers. Br. Vet. J. 152, 459–472. Lombard, J.E., Tucker, C.B., von Keyserlingk, M.A.G., Kopral, C.A., Weary, D.M., 2010. Associations between cow hygiene, hock injuries, and free stall usage on US dairy farms. J. Dairy Sci. 93, 4668–4676. Mendl, M., Zanella, A.J., Broom, D.M., 1992. Physiological and reproductive correlates of behavioural strategies in female domestic pigs. Anim. Behav. 44, 1107–1121. Metz, J.H.M., 1985. The reaction of cows to a short-term deprivation of lying. Appl. Anim. Behav. Sci. 13, 301–307. Miller, K., Wood Gush, D.G.M., 1991. Some effects of housing on the social behaviour of dairy cows. Anim. Prod. 53, 271–278. Munksgaard, L., Jensen, M.B., Pedersen, L.J., Hansen, S.W., Matthews, L., 2005. Quantifying behavioural priorities – effects of time constraints on behaviour of dairy cows, Bos taurus. Appl. Anim. Behav. Sci. 92, 3–14. Sambraus, H.H., 1971. Zum Liegeverhalten der Wiederkäuer. Züchtungskunde 43, 187–198. Somers, J.G.C.J., Frankena, K., Noordhuizen-Stassen, E.N., Metz, J.H.M., 2005. Risk factors for interdigital dermatitis and heel erosion in dairy cows kept in cubicle houses in The Netherlands. Prev. Vet. Med. 71, 23–34. Stoye, S., Porter, M.A., Dawkins, M.S., 2012. Synchronized lying in cattle in relation to time of day. Livest. Sci. 149, 70–73. Telezhenko, E., von Keyserlingk, M.A.G., Talebi, A., Weary, D.M., 2012. Effect of pen size, group size, and stocking density on activity in freestall-housed dairy cows. J. Dairy Sci. 95, 3064–3069. USDA, 2010. Facility Characteristics and Cow Comfort on U.S. Dairy Operations, 2007. USDA–APHIS–VS, CEAH, Fort Collins, CO, #524.1010. Webster, A.J.F., 2001. Effects of housing and two forage diets on the development of claw horn lesions in dairy cows at first calving and in first lactation. Vet. J. 162, 56–65. Wierenga, H.K., 1990. Social dominance in dairy cattle and the influences of housing and management. Appl. Anim. Behav. Sci. 27, 201–229.
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
ARTICLE IN PRESS
APPLAN-4085; No. of Pages 6
Applied Animal Behaviour Science xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim
Effects of under- and overstocking freestalls on dairy cattle behaviour Christoph Winckler a,∗ , Cassandra B. Tucker b , Daniel M. Weary c a Department of Sustainable Agricultural Systems, Division of Livestock Sciences, University of Natural Resources and Life Sciences, Vienna, Gregor-Mendel-Strasse 33, A-1180 Vienna, Austria b Department of Animal Science, University of California, Davis, One Shields Avenue, Davis, CA 95616, USA c Animal Welfare Program, University of British Columbia Vancouver, 2357 Main Mall, Vancouver, BC V6T 1Z4, Canada
a r t i c l e
i n f o
Article history: Received 28 November 2014 Received in revised form 10 June 2015 Accepted 15 June 2015 Available online xxx Keywords: Cubicle Animal welfare Stocking density Housing Agonistic interactions
a b s t r a c t Freestall availability affects cattle behaviour and most studies in this area have focused on overstocking. We studied the effects of three levels of stall availability, including both over- and understocking on the time budgets and agonistic interactions in 36 dairy cattle in four stable groups. Using a switch-back design, with treatment order balanced, groups of nine cows were given access to 6, 9, and 12 stalls for 1 week each, allowing for a within-cow test of stocking density of 150, 100 and 75% (cows/stalls). After 5 days of acclimatization at each density, time budgets and displacements from stalls were measured during the last 48 h of each treatment period using continuous video recording and direct observation. When animals had access to fewer stalls, they spent less time lying down (11.6, 12.6, 12.8 h/24 h in 150, 100 and 75% treatments, respectively; SE: 0.31, 0.31, 0.28 h/24 h), particularly at night (6.6, 7.5, 7.6 h; SE: 0.20, 0.20, 0.17 h). Lying behaviour was also more synchronous when more freestalls were available (Kappa coefficient of agreement 0.00, 0.13, 0.17 for lying time in 150, 100 and 75% treatments, respectively). Cows spent more time standing in the alleyways when overstocked (between two rows of stalls: 1.8, 0.8, 0.6 h/24 h in 150, 100 and 75%, respectively; SE: 0.09, 0.09, 0.06 h/24 h; between feeder and stalls: 1.5, 1.3, 1.3 h/24 h in 150, 100 and 75%, respectively; SE: 0.13, 0.13, 0.11 h/24 h), but did not alter the time they spent feeding. Moreover, cows were more likely to displace one another from stalls at greater stocking densities (2.9, 1.1, 0.6 displacements per cow/24 h in 150, 100 and 75% treatments, respectively; SE: 0.16, 0.16, 0.11 displacements per cow/24 h). Cows that were less successful at displacing others spent a higher proportion of their time lying during the day when overstocked, indicating that lying during this time is less preferred. For all variables, the magnitude of response was most affected by overstocking; this practice reduced lying time, especially at night, synchrony of lying behaviour and increased competition for stalls. Understocking provided benefits, but the degree of behavioural change was smaller than when stalls were limited. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Recent surveys suggest that in four out of every 10 USA freestall barns there are more cows than stalls available (USDA, 2010). Dairy producers overstock to save building costs, or because herd growth is not matched by expansion of barns. Epidemiological evidence indicates that this overstocking explains a significant portion of the non-dietary variation in milk production, with a lower milk yield per cow in overstocked farms (Bach et al., 2008). The mechanism associated with lower productivity is not understood, but could be related to increased lameness within the herd (e.g. Leonard et al., 1996), as lame cows or those with hoof injuries produce less milk
∗ Corresponding author. Tel.: +43 1 47 654 3261; fax: +43 1 47 654 3254. E-mail address: [email protected] (C. Winckler).
(Amory et al., 2008; Archer et al., 2010; Green et al., 2002). Indeed, the cow:stall ratio was eligible for inclusion in the final model evaluating lameness prevalence in two recent epidemiological studies (Dippel et al., 2009; Espejo and Endres, 2007). Increased lameness in overstocked barns is likely mediated by changes in behaviour. Cows spend less time lying down (Fregonesi et al., 2007; Hill et al., 2009; Krawczel et al., 2008, 2012) and less time standing partially in the stall (Hill et al., 2009; Lombard et al., 2010) when fewer freestalls are available. These changes in time budgets, particularly lying time, are robust; they are apparent regardless of the method used to experimentally overstocking freestalls, by either blocking stalls or adding new individuals to the group (Krawczel et al., 2012) or the method used to record stall usage, either through continuous monitoring over 24-h periods (Fregonesi et al., 2007; Hill et al., 2009; Krawczel et al., 2012) or less accurate single measures within a day
http://dx.doi.org/10.1016/j.applanim.2015.06.003 0168-1591/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model APPLAN-4085; No. of Pages 6
ARTICLE IN PRESS C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
2
(Krawczel et al., 2008; Ito et al., 2009). However, lying time is not consistently affected by stocking density in on-farm research. Some studies have failed to find a relation between stocking density and lying behaviour (Charlton et al., 2014; Ito et al., 2014; Lombard et al., 2010). Understocking has been included in the epidemiological comparisons (e.g. median stocking density was 96% in Charlton et al., 2014; range was 71–197% in Ito et al., 2014), but has been rarely included in experimental studies. Thirty-nine percent of US freestall farms understock (USDA, 2010) and recent research suggests that this practice could increase lying times beyond those seen when 1 stall is offered for every cow (Telezhenko et al., 2012). However, little is known about the other effects of this practice. For example, cows kept in overstocked conditions are more likely to both compete directly for freestalls and adopt non-competitive strategies, such as using the freestalls more at non-peak times, and thus reducing synchrony of lying behaviour (Fregonesi et al., 2007). It is unclear how understocking would affect competition and stall use. Our objective was to better understand how both under- and overstocking freestalls affects time budgets, social behaviour and synchrony of dairy cattle in stable groups.
2. Materials and methods 2.1. Animals and treatments The experiment was conducted at the University of British Columbia Dairy Education and Research Centre in Agassiz, British Columbia. Thirty-six dairy cows were randomly assigned to four groups of nine animals averaging (mean ± SD) 40.2 ± 8.07 kg/day milk production, 146 ± 25.6 days in milk, 2.4 ± 1.19 parity, 679 ± 70.2 kg body weight and 47 ± 16.3 months in age. Each experimental pen (width = 7.5 m, length = 13.5 m) contained 12 freestalls configured in three rows. In two rows, the stalls were open at the front (‘head-to-head’; two cows facing one another) and had a bed length of 2.4 m. The third row of freestalls faced a cement wall, and these stalls were 0.3 m longer to allow more space for the cow to lunge forward when getting up and lying down. Freestalls were separated by Dutch comfort style partitions and measured 1.2 m wide centre to centre and the neck rail was 1.14 m from the stall surface. Stalls were deep-bedded with 0.4 m of sand. The flooring throughout the pens was grooved concrete. The alley closest to the feed bunk measured 3.5 m and the floor of half of this alley was inlaid with a 2.5 cm-thick durable rubber mat. The rubber surface was level with the concrete flooring and had grooving similar to the concrete flooring. Alleys were cleaned six times per day with automatic scrapers. Each pen had 7.5 m of feed bunk space available through a post and rail barrier. Animals were fed for ad libitum consumption with a total mixed ration of corn silage, grass silage, barley, canola meal and soybean meal. Fresh feed was provided twice daily (at 06:00 h and 15:30 h) and feed was pushed up three times per day. Water was freely available from a self-filling trough. Cows were milked twice daily (approximately 06:00 h and 17:00 h) and spent 1.3 h/day (0.50 SD) away from the pen. We manipulated stocking density by providing 6, 9, or 12 freestalls to groups of nine animals, thus creating stocking densities of 150, 100, and 75% (cows/stalls), respectively. Using a switch-back design with treatment order balanced, all groups were exposed to all treatments for 1 week each. Each group returned to the 75% stocking density treatment after exposure to either the 150% or the 100% treatment. Stalls along one side of the pen were blocked to create the 100% and 150% treatments. For the 100% treatment, the three stalls along a wall and closest to the crossover/water trough were blocked. To create the 150% treatment, cows were denied
access to the three stalls adjacent to the stalls blocked for the 100% stocking density treatment. 2.2. Sampling and measurement Behaviour was recorded using eight Panasonic WV 330 cameras, positioned approximately 10 m above the experimental pens for 48 h during each of the 5 weeks (i.e. days 6 and 7) at 3 frames/s. The cameras were attached to a Panasonic video multiplexer (WV-FS416) and time-lapse recorder (AG-6540p; Panasonic; Mississauga, Ontario, Canada). Red lights (100 W) were hung approximately 10 m above the pens to facilitate video recording at night. Cows were marked with unique symbols using hair dye to identify individuals. Behaviour was recorded in two ways (Table 1). Time budgets were scored from video using instantaneous scan sampling once every 10 min. At each scan we scored if the cow was in the freestall (lying, standing with two or four legs in the stall), standing between the feeder and stalls, in the crossover, between the two rows of stalls, or at the feeder (Table 1). The number of displacements from the stalls was also recorded for each cow during 48 h of continuous observation from video, resulting in 240 h of observations per cow. Finally, we recorded social interactions at the feeder with continuous live observation in the 2 h after morning and afternoon milking. Each pen was observed for 30 min twice a day for 2 days during each week, for a total of 10 h observation/pen. The order of the 30-min observation period was balanced across pens and days. Both aggressive and positive interactions were recorded (Table 1). Displacements from the stalls were used Table 1 Definitions of behaviours used to evaluate the effects of freestall availability on dairy cattle. Behaviour In the stall Lyinga Standing with front two legsa Standing with four legsa Displacementa In the alley Standing between feeder and stallsa Standing in crossovera Standing between two rows of stallsa At the feeder Feedinga Displacementsb
Head buttingb
Pushingb
Total agonistic interactionsb Positive interactionsb
a b
Definition Flank in contact with ground Two front legs in contact with the stall surface, weight on legs Four legs in contact with the stall surface, weight on legs Contact (butt or push) immediately followed by leaving the stall Four legs in contact with the alley surface between the feeder and the row of stalls closest to the feeder, weight on legs, head not above the feeder Four legs in contact with the alley surface on the crossover between the alleys, weight on legs Four legs in contact with the alley surface on the alley between the two rows of stalls, weight on legs Presence at the feeder with the head above the feed bunk Contact (butt or push) immediately followed by leaving the feeding place by at least a cow width or half a cow length Contact with the forehead or horn base using a forceful movement, receiving animal does not give up its present position Applying steady force with any part of the head, receiving animal does not give up its present position Sum of displacements, head butting and pushing Sum of social licking (tactile oral contact directed to the body except the anal region, udder or claws) and horning (rubbing of foreheads or horn bases against the head or neck of another animal)
Scored from video recordings (48 h/week). Scored with live observation (2 h/week).
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model APPLAN-4085; No. of Pages 6
ARTICLE IN PRESS C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
to calculate an index of social success (Mendl et al., 1992; Galindo and Broom, 2000). This index of social success could vary from 0 to 1 and equalled the number of times the cow displaced other cows divided by the total number of times she displaced other cows and was displaced herself.
3
affected by the 100% and 150% treatments. Pearson correlation coefficients were calculated between the time spent lying in stalls blocked to create the 150% and 100% treatments during the periods with 75% stocking density (%) and how much the cow’s lying time was influenced by the treatments (lying time in 75% treatmentlying time in 150% treatment).
2.3. Statistical analyses 3. Results Several additional dependent variables were calculated: time spent lying during the day (time between am and pm milking), proportion of lying time occurring during the day (time between am and pm milking) and time to lie down after return from milking. For all variables, data were averaged to create single values for each 48 h period, i.e. days 6 and 7 of the 5 weeks of treatment. A linear mixed model (procedure MIXED; SAS) was used to analyze differences between the three levels of freestall availability (2 df) for all response variables as well as the interaction between treatment and cows’ ability to displace other cows from the stalls (1 df). The model included week as a repeated measure and cow within pen as subject, using a compound symmetry covariance structure. If not significant, the interaction term was removed from the model. Specified contrasts were used to test the effects of stocking density (i.e. 150% vs. 100% and 100% vs. 75%; df 1). Distribution of residuals was graphically evaluated using normal-qq-plots to verify model assumptions. As a measure of synchrony (Asher and Collins, 2012), we also calculated generalized Kappa statistics for lying, standing and feeding in the three treatments. Using the data from instantaneous scan sampling, the %MAGREE macro in SAS computes estimates and tests of agreement with individual cows in the four groups treated as multiple raters and the behavioural categories lying, standing and feeding as responses. Effects of treatment (df 2) were tested using a general linear model (procedure GLM; SAS) that included pen. Contrasts (df 1) were specified as described above. In addition to the primary objective of the study, we also explored whether cows that used the blocked stalls during the 75% stocking density periods (12 stalls available) were differentially
Cows spent less time lying in the freestalls when fewer stalls were available, with a 1.2 h difference between the 75% and 150% treatments (Table 2). The change in lying time was more apparent at night (6.6, 7.5 and 7.6 h lying at night in 150%, 100%, and 75% treatments, SE: 0.20, 0.20 and 0.17 h, P < 0.01, Fig. 1), than during the day (5.0, 5.1, and 5.2 h lying during day in 150%, 100%, and 75% treatments, SE: 0.15, 0.15, 0.13 h, P = 0.04). The proportion of lying time during the day increased with overstocking, and time to lie down after return from milking was shorter (Table 2). Freestall availability also affected standing behaviour, with cows changing where they stood within the pen. Cows spent less time standing with only front legs in the stall when fewer freestalls were provided, spending on average 0.9 h/24 h at a stocking density of 150% compared with 1.3 h when stocking density was 100%. At higher densities cows spent more time standing around the stalls, both between the feeder and the stalls (P = 0.04) and in the area between two rows of freestalls (P < 0.01) as well as in the crossover area (P < 0.01, Table 2). There was no effect of density on time spent standing at the feeder. Cows were five times more likely to displace others from the stall when only six freestalls were available (150% stocking density), compared with the 75% stocking density treatment (Table 2). In total, we observed 426 displacements from the stall. Of these, 46%, 43% and 12% occurred when the displaced cow was lying, standing with the front two legs or all four legs in the stall, respectively. Social success, as measured with displacements from the stall, influenced the response to treatments in several situations. Cows that were less successful displacing others from the stall
Table 2 Mean and S.E. for durations and frequencies of behaviours at stocking densities of 150%, 100%, and 75% (i.e. 6, 9, and 12 stalls for 9 cows/group; n = 4 groups). P-values are for the test of stocking density and the interaction between density and a cow’s success at displacing other cows from the stall as well as for contrasts comparing specific densities. Variable
Stocking density 150% S.E.
In the stall Lying (h/24 h) Proportion of lying time during day (%) Latency to lie upon return from milking (min) Standing with front two legs (h/24 h) Standing with four legs (h/24 h) Displacements (no./24 h)
P
100%
75% S.E.
Stocking density
Intx with social success
S.E.
Specified contrasts 150% vs. 100%
100% vs. 75%
11.6 43.2 59 0.9 0.2 2.9
0.31 0.77 5.4 0.17 0.05 0.16
12.6 40.6 61 1.3 0.2 1.1
0.31 0.77 5.4 0.17 0.05 0.16
12.8 40.7 62 1.4 0.2 0.6
0.28 0.52 4.2 0.15 0.04 0.11
<0.01 <0.01 0.04 <0.01 <0.07 <0.01
excl.a <0.01 0.04 <0.01 excl.a n/ab
<0.01 <0.01 0.04 0.01 – <0.01
0.36 0.91 0.23 0.69 – 0.04
In the alley Standing between feeder and stalls (h/24 h) Standing in crossover (h/24 h) Standing between two rows of stalls (h/24 h)
1.5 1.3 1.8
0.13 0.11 0.09
1.3 1.1 0.8
0.13 0.11 0.09
1.2 1.1 0.6
0.11 0.09 0.06
0.04 <0.01 <0.01
excl.a 0.02 excl.a
0.17 0.20 <0.01
0.55 0.17 0.03
At the feeder Feeding (h/24 h) Displacements (no./2 h) Head butting (no./2 h) Pushing (no./2 h) Total agonistic interactions (no./2 h) Sociopositive interactions (no./2 h)
5.4 2.0 5.0 2.7 9.8 0.9
0.13 0.53 1.07 0.52 1.65 0.26
5.4 2.2 5.3 2.4 10.0 0.7
0.13 0.53 1.07 0.52 1.65 0.26
5.5 2.6 7.1 2.2 12.0 1.0
0.12 0.44 0.84 0.37 1.38 0.17
0.67 0.32 0.04 0.57 0.13 0.58
excl.a n/ac n/ac n/ac n/ac n/ac
– – 0.78 – – –
– – 0.11 – – –
a b c
Excluded from final model as not significant. Displacements were used to calculate social success. Not applicable to research question.
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model
ARTICLE IN PRESS
APPLAN-4085; No. of Pages 6
C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
4
100 90 80
% cows lying
70 60 75%
50
100%
40
150%
30 20
4:00
2:00
0:00
22:00
20:00
18:00
16:00
14:00
12:00
10:00
8:00
0
6:00
10
Fig. 1. Diurnal variation in lying time at stocking densities of 150%, 100% and 75% (i.e. 6, 9, and 12 stalls for 9 cows/group; n = 4 groups). Each point shows the average for all 36 cows at 10-min intervals.
A) Proportion of lying time during the day 70
- - - 75% 60
___100% … 150%
% 50 40 30
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
B) Standing time with two front legs in stall
Synchrony of lying, feeding and overall synchrony (when considering the behaviour categories lying, feeding and standing) was lower with greater stocking density (all P < 0.01, Table 3). The stalls blocked to create the 150% and 100% treatments were less popular during the 75% treatment periods compared with stalls available during the treatments with higher stocking density. The three stalls blocked to create the 100% treatment were used for 20.2 ± 18.4% (mean ± SD) of lying during the 75% treatment (expected: 25%), and stalls blocked to create the 150% treatment accounted for 46.1 ± 20.6% of lying during the 75% treatment (expected: 50%). However, cows that used the blocked stalls during the 75% treatment were not differentially affected by reduced stall availability (Fig. 3; rpearson ≤ 0.20).
h/24h
6 5
4. Discussion
4
Lying time was reduced when cows were overstocked. Cattle spent more time standing in alleyways when fewer stalls were provided, but feeding time was not affected by stall availability. Overstocking also reduced synchrony of lying behaviour, resulted in less lying at night and increased competition for stalls. Cows that were less successful at displacing others from stalls were more affected by overstocking than were socially dominant animals. Lying time was reduced by 1 h/day when stalls were overstocked from 100 to 150%; understocking provided relatively little benefit in terms of lying time (increasing just 0.2 h/24 h). Experimental studies of freestall availability report that lying time is always reduced when fewer stalls are provided. Most studies have focused on overstocking (Falk et al., 2012; Fregonesi et al., 2007; Hill et al., 2009; Krawczel et al., 2008, 2012). In agreement with the present study, Telezhenko et al. (2012) found a similar small increase in lying with understocking. Epidemiological studies, with both under- and overstocked farms included, have often not identified stocking density as an important risk factor for total lying time (Charlton et al., 2014; Ito et al., 2014; Lombard et al., 2010). The current finding that understocking has a much smaller effect on lying time may explain some of the discrepancy between these two types of evidence, as a focus solely on overstocking may make it easier to detect a clear pattern in cattle responses. Other factors, such as the individual variation of lying time and the power of within-animal tests (all experimental work to date uses this approach), also likely play a
3 2 1 0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Index of social success at freestalls
0.8
Fig. 2. Proportion of time lying during the day (Panel A) and standing with two front legs in the stalls (Panel B) in relation to the social success index as regards displacements from the stalls. Three stocking densities at the freestalls were compared: 150%, 100%, and 75% (i.e. 6, 9, or 12 freestalls for groups of nine cows).
spent more time lying during the day in the 150% treatment (Pintx social success = 0.01) and spent a higher proportion of their lying time during the day in the 150% treatment (Pintx social success < 0.01, Fig. 2A). Similarly, cows that were less successful changed their standing behaviour in the stall when there was less competition: they spent more time standing with two legs in the stall in the 75% and 100% treatments (Pintx social success < 0.01, Fig. 2B). A lower social success index was also associated with more time spent standing in the crossover in the 100% and 150% treatments (Pintx social success = 0.02).
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model
ARTICLE IN PRESS
APPLAN-4085; No. of Pages 6
C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
5
Table 3 Mean Kappa coefficients and SE for synchrony of lying, feeding, standing and overall synchrony at stocking densities of 150%, 100%, and 75% (i.e. 6, 9, and 12 stalls for 9 cows/group, n = 4 groups). P-values are for the overall test of density and for contrasts comparing specific densities. Behaviour
Lying Feeding Standing Overall
Stocking density
P
150%
100%
75%
SE
Stocking density
0.00 0.15 0.02 0.05
0.13 0.19 0.08 0.12
0.17 0.23 0.04 0.15
0.01 0.01 0.03 0.01
<0.01 0.01 0.34 <0.01
Specified contrasts 150% vs. 100%
100% vs. 75%
<0.01 0.04 – <0.01
0.01 0.06 – 0.02
5 4 Δ lyin ng time in 75% v vs. 150% 3 treatment 2 (h h/24h) 1 0 -1 -2
0
20
40 0
60
80
100
Time lying (%) in stalls blocked tto create 150% treatment
Fig. 3. Difference in lying time () at 75% and 150% stocking densities in relation to amount of time spent lying down during the 75% treatment in the six stalls blocked to reduce freestall availability.
role in discrepancies between epidemiological and experimentalbased results. When access to freestalls was not limited, cows spent more time lying at night and less during the day. Lying behaviour was also more synchronous with increasing stall availability. On pasture, cattle behaviour can be highly synchronous with 70–80% of cows, on average, lying down at the same time (Stoye et al., 2012). In the current study providing more stalls than animals, cows are better able to achieve what seems to be a preferred, higher level of synchrony and timing of rest. Differences in the degree of synchrony between systems (e.g. pasture vs. freestall barns, Miller and Wood Gush, 1991), could be due to differences in other zeitgebers, such as artificial light or provision of feed. Cattle prefer some stalls over others and this could also affect the degree of synchrony. For example, some of the stalls we blocked off to manipulate freestall availability were less popular than the remaining stalls in the pen. Cows also directly competed for stalls by displacing others. This type of competitive interaction was more common when freestalls were overstocked, as found by others (Fregonesi et al., 2007; Friend et al., 1977; Wierenga, 1990). However, displacements from the stalls further decreased with understocking, indicating that competition for stalls exists even when one stall is available for every cow. Such competition may be explained by the fact that stalls are not all equally preferred. Competitive interactions at the feeder were unaffected by stocking density, likely because feeding space was unchanged by the treatments. The higher number of head butts at the feeder with low stocking density was unexpected and remains difficult to explain. Cows that were unable to displace others from stalls were more affected by overstocking than were more dominant animals. For example, subordinate animals spent more time lying during the day, possibly a less preferred time for this behaviour (Sambraus, 1971; Fregonesi et al., 2007). Despite the increase in use during the day by subordinate animals, time to lie down after return from milking was only slightly affected by stall availability. The effect size (2–3 min/day) was less than the 13 min/day found by Fregonesi et al. (2007) and may not be biologically relevant. In addition, in agreement with Fregonesi et al. (2007), there was no relationship
between the animals’ success in displacing others from the stall and total lying time, possibly because cattle are highly motivated to maintain lying time (Jensen et al., 2005; Metz, 1985; Munksgaard et al., 2005). Subordinate animals also spent more time standing in the stall with two legs, compared with more dominant animals, when provided at least 1 stall/cow. Others have reported a similar pattern in standing behaviour (Galindo and Broom, 2000) and suggested that, when space is available, animals may perform this behaviour to hide. Indeed, freestall dividers greatly reduce aggressive interactions when used as barriers between cows at the feedbunk (DeVries and von Keyserlingk, 2006). Standing partially in the stall decreased, when cows were overstocked, but this was more than compensated for by an increase in standing in the crossover alleyway. This change in behaviour may increase hoof exposure to concrete and manure, considered risk factors for claw injuries, infectious diseases and lameness (Gregory et al., 2006; Somers et al., 2005; Webster, 2001). 5. Conclusions Behavioural responses were most affected by overstocking; this reduced lying time, especially at night, and synchrony of lying behaviour and increased competition for stalls. Subordinate animals were particularly affected by overstocking. Understocking provided more modest benefits, mainly related to reduced competition for stalls and improved synchrony of lying behaviour. In combination with other experimental research, this work illustrates the importance of providing cattle with at least 1 stall/cow. Acknowledgments We thank the staff at University of British Columbia’s Dairy Education and Research Centre and the University’s Animal Welfare Program. CW was funded by a research fellowship provided by Deutsche Forschungsgemeinschaft. The Animal Welfare Program is funded by Canada’s NSERC Industrial Research Chair Program with industry contributions from the Dairy Farmers of Canada, the British Columbia Dairy Association, the Westgen Endowment Fund,
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003
G Model APPLAN-4085; No. of Pages 6
ARTICLE IN PRESS C. Winckler et al. / Applied Animal Behaviour Science xxx (2015) xxx–xxx
6
Intervet Canada Corporation, Zoetis, BC Cattle Industry Development Fund, Alberta Milk, Valacta and CanWest DHI. References Amory, J.R., Barker, Z.E., Wright, J.L., Mason, S.A., Blowey, R.W., Green, L.E., 2008. Associations between sole ulcer, white line disease and digital dermatitis and the milk yield of 1824 dairy cows on 30 dairy cow farms in England and Wales from February 2003–November 2004. Prev. Vet. Med. 83, 381–391. Archer, S.C., Green, M.J., Huxley, J.N., 2010. Association between milk yield and serial locomotion score assessments in UK dairy cows. J. Dairy Sci. 93, 4045–4053. Asher, L., Collins, L.M., 2012. Assessing synchrony in groups: are you measuring what you think you are measuring? Appl. Anim. Behav. Sci. 138, 162–169. Bach, A., Valls, N., Solans, A., Torrent, T., 2008. Associations between nondietary factors and dairy herd performance. J. Dairy Sci. 91, 3259–3267. Charlton, G.L., Haley, D.B., Rushen, J., de Passillé, A.M., 2014. Stocking density, milking duration, and lying times of lactating cows on Canadian freestall dairy farms. J. Dairy Sci. 97, 2694–2700. DeVries, T.J., von Keyserlingk, M.A.G., 2006. Feed stalls affect the social and feeding behavior of lactating dairy cows. J. Dairy Sci. 89, 3522–3531. Dippel, S., Dolezal, M., Brenninkmeyer, C., Brinkmann, J., March, S., Knierim, U., Winckler, C., 2009. Risk factors for lameness in freestall-housed dairy cows across two breeds, farming systems, and countries. J. Dairy Sci. 92, 5476–5486. Espejo, L.A., Endres, M.I., 2007. Herd-level risk factors for lameness in high-producing Holstein cows housed in freestall barns. J. Dairy Sci. 90, 306–314. Falk, A.C., Weary, D.M., Winckler, C., von Keyserlingk, M.A.G., 2012. Preference for pasture versus freestall housing by dairy cattle when stall availability indoors is reduced. J. Dairy Sci. 95, 6409–6415. Fregonesi, J.A., Tucker, C.B., Weary, D.M., 2007. Overstocking reduces lying time in dairy cows. J. Dairy Sci. 90, 3349–3354. Friend, T.H., Polan, C.E., McGilliard, M.L., 1977. Free stall and feed bunk requirements relative to behavior, production and individual feed intake in dairy cows. J. Dairy Sci. 60, 108–116. Galindo, F., Broom, D.M., 2000. The relationships between social behaviour of dairy cows and the occurrence of lameness in three herds. Res. Vet. Sci. 69, 75–79. Green, L.E., Hedges, V.J., Schukken, Y.H., Blowey, R.W., Packington, A.J., 2002. The impact of clinical lameness on the milk yield of dairy cows. J. Dairy Sci. 85, 2250–2256. Gregory, N., Craggs, L., Hobson, N., Krogh, C., 2006. Softening of cattle hoof soles and swelling of heel horn by environmental agents. Food Chem. Toxicol. 44, 1223–1227. Hill, C.T., Krawczel, P.D., Dann, H.M., Ballard, C.S., Hovey, R.C., Falls, W.A., Grant, R.J., 2009. Effect of stocking density on the short-term behavioural responses of dairy cows. Appl. Anim. Behav. Sci. 117, 144–149.
Ito, K., Weary, D.M., von Keyserlingk, M.A.G., 2009. Lying behavior: assessing within- and between-herd variation in free-stall-housed dairy cows. J. Dairy Sci. 92, 4412–4420. Ito, K., Chapinal, N., Weary, D.M., von Keyserlingk, M.A.G., 2014. Associations between herd-level factors and lying behavior of freestall-housed dairy cows. J. Dairy Sci. 97, 2081–2089. Jensen, M.B., Pedersen, L.J., Munksgaard, L., 2005. The effect of reward duration on demand functions for rest in dairy heifers and lying requirements as measured by demand functions. Appl. Anim. Behav. Sci. 90, 207–217. Krawczel, P.D., Hill, C.T., Dann, H.M., Grant, R.J., 2008. Effect of stocking density on indices of cow comfort. J. Dairy Sci. 91, 1903–1907. Krawczel, P.D., Mooney, C.S., Dann, H.M., Carter, M.P., Butzler, R.E., Ballard, C.S., Grant, R.J., 2012. Effect of alternative models for increasing stocking density on the short-term behavior and hygiene of Holstein dairy cows. J. Dairy Sci. 95, 2467–2475. Leonard, F.C., O’Connell, J.M., O’Farrell, K.J., 1996. Effect of overcrowding on claw health in first-calved Friesian heifers. Br. Vet. J. 152, 459–472. Lombard, J.E., Tucker, C.B., von Keyserlingk, M.A.G., Kopral, C.A., Weary, D.M., 2010. Associations between cow hygiene, hock injuries, and free stall usage on US dairy farms. J. Dairy Sci. 93, 4668–4676. Mendl, M., Zanella, A.J., Broom, D.M., 1992. Physiological and reproductive correlates of behavioural strategies in female domestic pigs. Anim. Behav. 44, 1107–1121. Metz, J.H.M., 1985. The reaction of cows to a short-term deprivation of lying. Appl. Anim. Behav. Sci. 13, 301–307. Miller, K., Wood Gush, D.G.M., 1991. Some effects of housing on the social behaviour of dairy cows. Anim. Prod. 53, 271–278. Munksgaard, L., Jensen, M.B., Pedersen, L.J., Hansen, S.W., Matthews, L., 2005. Quantifying behavioural priorities – effects of time constraints on behaviour of dairy cows, Bos taurus. Appl. Anim. Behav. Sci. 92, 3–14. Sambraus, H.H., 1971. Zum Liegeverhalten der Wiederkäuer. Züchtungskunde 43, 187–198. Somers, J.G.C.J., Frankena, K., Noordhuizen-Stassen, E.N., Metz, J.H.M., 2005. Risk factors for interdigital dermatitis and heel erosion in dairy cows kept in cubicle houses in The Netherlands. Prev. Vet. Med. 71, 23–34. Stoye, S., Porter, M.A., Dawkins, M.S., 2012. Synchronized lying in cattle in relation to time of day. Livest. Sci. 149, 70–73. Telezhenko, E., von Keyserlingk, M.A.G., Talebi, A., Weary, D.M., 2012. Effect of pen size, group size, and stocking density on activity in freestall-housed dairy cows. J. Dairy Sci. 95, 3064–3069. USDA, 2010. Facility Characteristics and Cow Comfort on U.S. Dairy Operations, 2007. USDA–APHIS–VS, CEAH, Fort Collins, CO, #524.1010. Webster, A.J.F., 2001. Effects of housing and two forage diets on the development of claw horn lesions in dairy cows at first calving and in first lactation. Vet. J. 162, 56–65. Wierenga, H.K., 1990. Social dominance in dairy cattle and the influences of housing and management. Appl. Anim. Behav. Sci. 27, 201–229.
Please cite this article in press as: Winckler, C., et al., Effects of under- and overstocking freestalls on dairy cattle behaviour. Appl. Anim. Behav. Sci. (2015), http://dx.doi.org/10.1016/j.applanim.2015.06.003