If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The objective of this study was to investigate the effects of different stocking densities of 82 (0.82 cows per freestall and feed bin), 100, and 129% on behavior, productivity, and comfort indices of lactating Holstein dairy cows. Twenty-seven lactating cows (15 primiparous and 12 multiparous) were assigned to 1 of the 3 treatments, which were balanced for parity, milk yield, days in milk, and body weight in a 3 × 3 Latin square design with 14-d periods. After 7 d of adaptation to the treatments, lying time and bouts were recorded at 1-min intervals for 3 d, DMI and feeding time were monitored electronically by feed bins, and rumination time was quantified at 2-h periods for 5 d during each period. The cow comfort index, stall standing index, stall perching index, and stall use index (SUI) were calculated using 10-min scan samples of video recording from d 8 to 10 of each period. Milk yield was recorded from d 8 to 12 and milk composition was determined from composite samples on d 12 in each period. Daily lying time, lying bouts, and bout duration did not differ among the stocking densities. The ratio of lying time ≥12 h/d (the number of cows with daily lying time ≥12 h/d divided by number of cows per pen) was higher for cows housed at 82% stocking density compared with those housed at 100% stocking density, with stocking density of 129% intermediate. Hourly lying time was lower at 100% stocking density compared with 82 and 129% stocking densities during the peak period (2300–0400 h), determined based on diurnal pattern of lying time. Daily dry matter intake, feeding time, and feeding rate were not affected by stocking density. After morning milking, dry matter intake and feeding time was reduced at 129 versus 82% stocking density during peak feeding time (0600–0800 h), determined based on diurnal patterns of feeding behavior. Stocking density had no effect on rumination time, milk yield and milk composition. The ratio of SUI ≥85% (mean of the number of SUI ≥85% divided by the number of SUI at 10-min scan samples during a 24-h period) was lower at 129 versus 82% stocking density, with stocking density of 100% intermediate. During peak lying time after evening milking (2300–0400 h), both cow comfort index and SUI were higher at 129 than at 100% stocking density. The SUI was lower 2 h after morning milking (0800–0900 h) for cows housed at 129% compared with those housed at 82 and 100% stocking densities. In conclusion, when compared with 100% stocking density, understocking contributed to natural behaviors of cows that including lying, feeding, and rumination behavior, whereas overstocking did not cause negative effect on behavior, productivity, and comfort indices of cows in this study.
Dairy cow welfare has become a hot topic in recent years, especially considering the rise of large-scale dairy farms. The most important concerns regarding dairy cow welfare include whether the dairy cow is feeling well, functioning well, and performing natural behavior (
). Behavior and welfare of dairy cows is affected by the physical environment they are housed in (stall design, flooring type, feed bunk design, environmental quality, and so on) and by grouping strategy and stocking density (
). A practice employed by dairy farmers is to increase the herd size without extending housing facilities, which has resulted great farm to farm variation in stocking densities for stalls (number of cows per number of stalls × 100) and feed bunks (61cm/number of feed bunk spaces × 100) ranging between 71 to 197% and 58% to 228%, respectively, on North American dairy farms (
Benchmarking cow comfort on North America freestall dairies: Lameness, leg injuries, lying time, facility design and management for high-producing Holstein dairy cows.
The effect of stocking density on stall comfort, welfare, and natural behavior of cows can be evaluated by determining their resting, feeding, and rumination patterns (
), stall perching index (number of cows standing with 2 front feet in the stall and the rear feet in the alley per number of cows in contact with stalls;
found that SUI reduced as stocking density increased from 100 to 142% with no change in CCI or SSI during 24-h periods. Recent studies indicate that overstocking of stalls and headlocks to about 115% did not affect daily lying and rumination time (
To our knowledge the effects of stocking densities representative of on-farm situations on diurnal patterns of behavior and cow comfort indices have not been studied before. Therefore, the aim of the current study was to investigate effects of different stocking densities [82, 100, and 129% (i.e., 0.82, 1.00, and 1.29 cows per stall and feed bin, respectively)] on daily and diurnal patterns of behavior, productivity, and comfort indices in lactating dairy cows.
Materials and Methods
Animals, Treatments, and Management
Twenty-seven Holstein cows, including 15 primiparous and 12 multiparous, were selected and assigned to 1 of 3 pens balanced for parity (1.80 ± 0.06; mean ± SD), milk yield (35.03 ± 5.19 kg/d), DIM (192.07 ± 23.15 d), and BW (493.47 ± 50.31 kg) at the beginning of the study. Three treatment stocking densities of 82 (11 stalls and feed bins for 9 cows), 100 (9 stalls and feed bins for 9 cows), and 129% (7 stalls and feed bins for 9 cows) were applied to the 3 pens, with cows in a balanced 3 × 3 Latin square design with 14-d periods that included 7 d of treatment acclimatization and 7 d of data collection. Each pen (20 m long, 12.75 m wide) contained 16 head-to-head waterbeds (2.40 m long, 1.20 m wide; Advanced Comfort Technology Inc., Sun Prairie, WI) covered with 2.5 to 4.0 cm of rice husk, 16 electronic feed bins (1.00 wide, 0.75 m high, 0.84 m depth; Roughage Intake Control system, Insentec B.V, Marknesse, the Netherlands; validated by
), and one box-type water bin. Feed alleys and back alleys were 4.25 and 3.70 m wide, respectively, and scraped with an automatic scraper system (GEA Farm Technologies, Düsseldorf, Germany). The TMR fed to all cows was formulated according to recommendations in
; Table 1) and was refilled twice daily at 0730 and 1430 h for ad libitum intake. Feed samples that were collected on d 8, 11, and 14 of each experimental period for particle size distribution determination using a Penn State Particle Separator (Nasco, Fort Atkinson, WI); fresh feed samples were collected from d 8 to d 12 of each experimental period for DM and chemical composition analysis by State Key Laboratory of Animal Nutrition of China Agricultural University. Cows were milked 3 times daily at 0600, 1300, and 2000 h in a double-48 parallel parlor.
Table 1Ingredient and chemical composition (% of DM) of the TMR
The data of daily temperature and relative humidity were recorded at 5-min intervals during the whole experimental period using a temperature data logger (HOBO Pendant G, Onset Computer Corp., Bourne, MA).
Lying Behavior
Data loggers (HOBO Pendant G, Onset Computer Corp.; validated by
) were attached to right hind leg along the metatarsus bone of all cows to record lying time and bouts at 1-min intervals from d 8 to 10 of each experimental period. To avoid any interference with lying behavior of cows, data loggers were attached on the night before starting data collection. Lying time was summarized in 1-h periods to draw the diurnal pattern of lying that was based on to determine peak periods of lying behavior. The recommended lying time for dairy cows by
). Therefore, the ratio of lying time ≥12 h/d was calculated as the number of cows with daily lying time ≥12 h/d divided by number of cows in each pen. Bout duration was calculated as total daily lying time divided by number of lying bouts per day.
Feeding Behavior
Individual feed intake and eating time were monitored by Insentec feed bins (Roughage Intake Control system, Insentec B.V; validated by
). Data were summed to calculate total DMI and feeding time per 24- and 2-h periods for 5 d. Diurnal patterns of feeding behavior that reflected the peak periods of feeding were drawn according to the data of DMI and feeding time per 2-h periods. Feeding rate was calculated as total daily DMI divided by total daily feeding time.
Rumination Behavior
Rumination time was quantified using HR-Tags (SCR Engineers Ltd., Netanya, Israel; validated by
) that were fitted to the left side of the neck of each cow from d 8 to 12 of each experimental period and averaged in 2-h periods.
Milk Samples
During each experimental period, milk yield of each cow was recorded at each milking from d 8 to 12. On d 12 of each experimental period, milk samples were collected during all 3 milkings for milk composition analysis including fat, protein, lactose, and SCC (Dairy Products Quality Supervision and Inspection Center, Beijing, China). Fat-corrected milk (3.5% FCM) was calculated using the equation [0.432 × milk yield (kg/d) + 16.23 × milk fat yield (kg/d)] (
The number of cows lying in a stall, standing in a stall, perching in a stall, standing in the alley but not eating, and eating at feed bunks were counted by 10-min scan samples of video recording for 24 h starting from d 8 to 10 of each experimental period. The video data were collected using a digital video software (DS-7808, Hikvision, Zhejiang, China) with 6 cameras (AJ-86A, Aijia, Zheijiang, China) mounted to the periphery of the pen at 4.60 m over the floor. Data were analyzed to calculate CCI (
Behavior (lying, feeding, and rumination) and comfort indices collected at 24-h periods or peak periods (based on the diurnal patterns of behavior identified visually by plotting the data in graphs), and productivity data were analyzed as a 3 × 3 Latin square design with period and treatment (stocking density) as fixed effects and pen (experimental unit) as a random effect using the MIXED procedure of SPSS 22.0 (IBM Corp. Armonk, NY). All data were tested for normality and homogeneity of variances using SPSS 22.0. Extreme outliers (the difference value between figure and average value is 3 times higher than the standard deviation) were detected and removed using the descriptive statistics procedure of SPSS 22.0. Data of diurnal patterns of behavior and cow comfort indices were analyzed considering hourly periods (averaged in 2-h bins, expressed as 1-h values) as repeated measures using a repeated measures mixed model (MIXED procedure of SPSS 22.0), including fixed effects for period and treatment (stocking density) and a random effect of pen (experimental unit). The lowest Bayesian information criterion (fit statistic) level was used to select covariance structure of the model for each parameter. The least significant difference test was used for multiple treatment comparisons. Significance was declared at P ≤ 0.05.
Results
Environmental Conditions and Particle Size Distribution of TMR
During the trial, mean daily temperature and relative humidity were 23.1°C and 73.7%, ranging from 11.0 to 33.0°C and 22.9 to 97.7%, respectively. The long, medium, short, and fine particle size of TMR using Penn State Particle Separator (mean ± SD) were 4.2 ± 0.7, 41.4 ± 1.2, 37.9 ± 1.7, and 14.0 ± 0.8%, respectively.
Lying Behavior
Daily lying time (11.91 ± 0.39 h/d), lying bouts (13.13 ± 0.72 n/d), and bout duration (59.87 ± 3.27 min/bout) did not differ among the stocking densities (Table 2). The ratio of lying time ≥12 h/d was higher in 82% stocking density compared with 100% stocking density, with a stocking density of 129% being intermediate (P = 0.04; Table 2). The diurnal pattern of lying time indicated 4 distinct peak periods of lying activity: one from 2300 to 0400 h and other peak periods of lying activities beginning 2 h after each milking (0800–0900, 1500–1600, and 2200–2300 h; Figure 1). Hourly lying time was lower at 100% stocking density compared with 82 and 129% stocking densities during 2300 to 0400 h (P = 0.02; Table 2).
Table 2Mean lying behavior of dairy cows at different stocking densities over 3 d for 9 cows per pen (n = 3 pens/treatment)
Figure 1Diurnal pattern of mean lying time per hour of dairy cows (n = 9 cows/pen) within a 24-h period at stocking densities (n = 3 pens/treatment) of 82 (—), 100 (∙∙∙), and 129% (- -; i.e., 0.82, 1.00, and 1.29 cows per stall and feed bin, respectively). Cows were fed approximately at 0730 and 1430 h, and milked at 0600, 1300, and 2000 h (black arrows).
Daily DMI (23.02 ± 0.57 kg/d), feeding time (2.56 ± 0.13 h/d), and feeding rate (158.32 ± 8.96 g/min) were not affected by stocking density (Table 3). Cows spent more time on feeding activity during the day (0600–1800 h), with feeding rate per hour being numerically lower during the day (0600–1800 h) for cows housed at 82% stocking density compared with 100 or 129% stocking density (Figure 2C). Feeding activity reached its peak within 1 h after returning to their pens from milking (0600–0800, 1400–1600, and 2000–2200 h; Figures 2A), and cows housed at 82% stocking density had a higher DMI (P = 0.04) and a longer feeding time (P = 0.03) after morning feed delivery (0600–0800 h; Table 3).
Table 3Mean feeding behavior and rumination time of dairy cows at different times and stocking densities over 5 d for 9 cows per pen (n = 3 pens/treatment)
Figure 2Diurnal patterns of mean DMI per hour (A), mean minutes of feeding per hour (B), and mean feeding rate (C) in 2-h periods of dairy cows (n = 9 cows/pen) within a 24-h period at stocking densities (n = 3 pens/treatment) of 82 (—), 100 (∙∙∙), and 129% (- -; i.e., 0.82, 1.00, and 1.29 cows per stall and feed bin, respectively). Cows were fed approximately at 0730 and 1430 h, and milked at 0600, 1300, and 2000 h (black arrows).
Stocking density had no effect on daily rumination time (6.15 ± 0.30 h/d; Table 3). Compared with the 100 and 129% stocking densities, cows housed at 82% stocking density exhibited 2 more distinct rumination peaks which occurred at 0800 to 1000 h and 1200 to 1400 h (Figure 3).
Figure 3Diurnal patterns of mean rumination time per hour in 2-h periods of the dairy cows (n = 9 cows/pen) within a 24-h period at stocking densities (n = 3 pens/treatment) of 82 (—), 100 (∙∙∙), and 129% (- -; i.e., 0.82, 1.00, and 1.29 cows per stall and feed bin, respectively). Cows were fed approximately at 0730 and 1430 h, and milked at 0600, 1300, and 2000 h (black arrows).
The ratio of SUI ≥85% was reduced at 129 versus 82% stocking density, with 100% stocking density being intermediate (P = 0.048; Figure 4). During peak lying time, at 2300 to 0400 h, both CCI and SUI were higher at 129 than at 100% stocking density (P < 0.001; Table 5). During peak lying time after morning milking (0800–0900 h), SUI was lower for cows housed at 129% compared with those housed at 82 and 100% stocking densities (P < 0.001; Table 5).
Figure 4Ratio of cow comfort index (CCI) and stall use index (SUI) greater than or equal to 85% of dairy cows (n = 9 cows/pen) at each stocking density (82, 100, and 129%) for 3 d using 10-min scan samples during a 24-h period in each pen (n = 3 pens/treatment). Ratio of CCI ≥85% (black bars; P = 0.63; mean of the number of CCI ≥85% divided by the number of CCI at 10-min scan samples during a 24-h period), ratio of SUI ≥85% (gray bars; P = 0.048; mean of the number of SUI ≥85% divided by the number of SUI at 10-min scan samples during a 24-h period), and bars with different superscripts (a, b) differ (P < 0.05). Error bars indicate ± 1 SE.
Table 5Comparison of different stocking densities on cow comfort index (CCI) and stall use index (SUI) of dairy cows (n = 9 cows/pen) over 3 d at 10-min scan samples during peak lying periods in each pen (n = 3 pens/treatment)
The aim of the current study was to determine the effects of 3 stocking densities (82, 100, and 129%) on natural behavior of lactating dairy cows in terms of daily and diurnal patterns of lying, feeding and ruminating, productivity, and comfort indices. Our main findings were that varying stocking densities had no effect on daily behaviors, including time spent lying, feeding, and ruminating or on milk production and composition in this trial. However, lying time during peak periods (2300–0400 h) and feeding time during peak periods (0600–0800 h) were higher when cows were housed at 82% stocking density and coincided with increased SUI (especially compared with 100% stocking density), and CCI and SUI were enhanced at 129% stocking density during the peak lying period (2300–0400 h).
Lying Behavior
Daily lying time was not affected by decreasing the stocking density from 100 to 82%, which was similar to finding of
found the maximum stall occupancy to range from 62 to 88% when stocking density was 100%, which might explain that reducing stocking density from 100 to 82% did not increase lying time in our study. Overstocking to 129% in the current trial had little effect on daily lying time, which was different from a recent study where daily lying time declined when stocking density increased beyond 113% (
was the group size (9 vs. 34 cows, respectively), which might have affected the response in daily lying times to overstocking differently, because overcrowding may not only occur at stalls with increasing group size but may also occur at the feed bunks and walkways. Therefore, cows housed in a larger group may experience more reduced laying time in response to overstocking than cows in smaller groups (
), which was, on average, achieved for all stocking densities. However, the proportion of cows having lying time greater than 12 h/d was higher for cows housed at 82% than those housed at 100 and 129% stocking density. This suggests that increased stocking density may mainly decrease lying time of specific dairy cows within a pen, especially subordinate cows (
) found the peak lying periods were at midnight and 2 h after milking. During the peak lying time (2300–0400 h), cows housed at 129% stocking density had longer lying time (min/h) relative to those housed at 100% stocking density. Overstocking was previously found to result in rescheduling of daily time target, due to short-term deprivation of lying, to recover the lost lying time (
). However, deprivation from lying when stalls are not available might increase discomfort behaviors such as leg stomping, repositioning, and weight-shifting (
); this needs further long-term exploration in overstocked housing.
Feeding Behavior
We observed that understocked cows had greater DMI and longer feeding time during the peak feeding period in the morning (0600–0800 h), with a lower value of feeding rate during the daytime (0600–1800 h) compared with cows at higher stocking rates. This indicates that overstocked cows would have to increase their feeding rate during quite feeding periods to meet their needs, as overstocking may increase competition around and availability of feed bunks (
), which might reduce the chewing time, saliva production, and ruminal pH. We found that overstocking at the level of 129% had no effect on daily feeding time and DMI, similar to the findings of
. However, others found that daily feeding time increased linearly as the feed bunk space increased (0.21, 0.41, 0.61, and 0.81 m/cow, corresponding to 303, 149, 100, and 75% stocking densities, respectively;
) might be associated with the different types of feed troughs used, because feed bins with vertical head gates separating consecutive bins in a row, which were used in our trial, would reduce cow displacement at feed bins when they were exposed to overstocking. Daily feeding time was shorter in our study than those reported by others (
). This might be attributed to a high proportion of concentrate and low proportion of long particles in the ration used in the current study, which were previously both found to decrease feeding time (
No response was noted in daily rumination time among stacking densities in our study, consistent with previous findings in which the respective range of stocking densities varied from 67 to 142% (
Effect of feeding cows in early lactation with diets differing in roughage-neutral detergent fiber content on intake behavior, rumination, and milk production.
), cows spent more time ruminating at night, most likely because it corresponded to their highest lying time. During the day (0600–1800 h), the pattern of rumination time per hour fluctuated more for cows housed at 82% stocking density compared with those housed at 100 and 129% stocking densities (Figure 3A), which may be related to differences in feeding behavior of dairy cows as affected by different stocking densities (see feeding behavior section).
Productivity
Milk yield was not influenced by stocking density in our study, which was consistent with previous findings for dairy cows housed at either 100 or 142% (
). Therefore, the absence of an effect of stocking density on daily lying and feeding time in our study might explain why milk yield was not affected by the stocking density.
Comfort Indices
In our study, CCI was highest at stocking density of 129% during the peak lying period (230 –0400 h). This might be due to cows at higher stocking densities laying down sooner and occupying the stall for longer periods of time after the night milking (
observed no change in CCI with increasing stocking density from 100 to 142% during peak lying time (0000–0400 h). This was likely because, in their study, alley surfaces within the pen were covered with rubber mats which tended to diminish the effect of high-stocking density standing and perching in a stall by cows. The ideal CCI was suggested to be above 85% (
). The ratio of CCI ≥85% and SUI ≥85% did not differ between 100 and 129% stocking densities, but cows kept at 129% stocking density had a lower ratio of SUI ≥85% compared with those kept at 82% stocking density; this was in agreement with findings of
Varying stocking densities of 82, 100, and 129% per stall and per feed bin had no effect on daily time spent lying, feeding, and ruminating in this trial. Milk yield and composition of cows were not affected by stocking density. Compared with 100% stocking density, the peak lying time (2300–0400 h) and peak feeding time (0600–0800 h) were increased at 82% stocking density, which coincided with increased SUI. The behavioral index did not differ between 100 and 129% stocking densities, whereas CCI and SUI were enhanced at 129% stocking density during the peak lying period (2300–0400 h).
Acknowledgments
The authors are grateful for the financial support of Beijing Nova Program (China; Z121105002512067), Beijing Young Scientist Program (China; YETP0305), China Agriculture Research System (Beijing, China; CARS-37), and Key Technologies R&D Program of China (Beijing, China; 2012BAD12B02).
References
Adin G.
Solomon R.
Nikbachat M.
Zenou A.
Yosef E.
Brosh A.
Shabtay A.
Mabjeesh S.J.
Halachmi I.
Miron J.
Effect of feeding cows in early lactation with diets differing in roughage-neutral detergent fiber content on intake behavior, rumination, and milk production.
Benchmarking cow comfort on North America freestall dairies: Lameness, leg injuries, lying time, facility design and management for high-producing Holstein dairy cows.
30047-9&id=gr1.jpg){kind=link}
30047-9&id=gr2.jpg){kind=link}
30047-9&id=gr3.jpg){kind=link}
30047-9&id=gr4.jpg){kind=link}