Thirty-six Hanwoo steers were randomly chosen in Gyeongbuk Livestock Research Institute from May, 2014 to November, 2015. Preliminary experiments were carried out for three weeks, such that the individuals were grouped by considering average weight. The average yearling weight of the individuals was 291.4±12.2 kg at 12 month of age. The steers were raised in the same rearing area of 50 m² per pen, whereas the stocking density varied (Figure 1). In the pen for the control group (C), a set of five steers of 12 to 30 months of age was assigned, so as to allow 10 m² per head of rearing space. For treatment 1 (T1) and treatment 2 (T2) groups, four and three steers were assigned in a pen, so as to give 12.6 m² and 16.7 m²/head rearing space, respectively. Each steer was randomly assigned to each treatment, and three replicates for each treatment were applied, totaling 36 samples (3×5 [C]+3×4 [T1]+3×3 [T2]).

The nutrient composition of feedstuffs is presented in Table 1 [10]. All animals were given the same diets, while forage was restricted to consumption of rice straw. The amount of feedstuffs was changed at three month intervals that were based on the mean body weight.

The fattening stage was divided into two phases: phase I (12 to 21 months of age) and phase II (22 to 30 months of age), and the average amount of concentrate feeding for a day was 5.5 to 10.5 kg per steer, which was approximately equal to 1.6% to 1.8% of body weight (Table 2). During the fattening period, rice straw diets were provided ad libitum, twice a day at 8 am and 5 pm, respectively. One water dish and mineral block trough was provided per group, allowing animals to be freely accessed to feed and water.

The total amount of feed intake per steer was recorded by measuring the differences of feed weight provided each day and residual quantity at the end of the next morning (Table 3). Body weight of each steer was measured between 12 and 30 months of age in a three month interval, i.e. before morning feeding at the beginning of each quarter (Table 4). Average daily gain (ADG) between three month intervals was also calculated (Table 5). To evaluate carcass quality traits, a 2 MHz real time ultrasound equipment was applied between the 13th rib and lumber vertebra (HS-2000; HONDA, Tokyo, Japan). The images reflecting ultrasound back fat thickness, longissimus muscle area, and marbling score were processed and analyzed using the Image-pro Express (Version 4.1, Media Cybernetics, Rockville, MD, USA) software. Also, all of the steers were slaughtered at a local municipal slaughterhouse at 30 months of age, and carcass quality traits were evaluated according to the Korean carcass grading standard [11].

Digital cameras (IR LED camera; Sony, Changwon, Korea) and video-recording devices (STLV-36D; Samsung, Korea) were utilized for precise monitoring of animal behaviors within each treatment group. Nine digital cameras were allocated in front of the feeding area of each pen at a height of approximately 3 m. Animal behavior for a given quarter interval was analyzed by selecting a month and by recording on every weekend of the month. Behavior characteristics were classified as six categories, i.e. standing, lying down, walking, self-grooming, leaning, and fighting (Figure 2), and each individual was observed for 12 hours a day from 7 am to 7 pm. Behavioral time was analyzed by sampling at a two minute interval of the observation records, after which the best behavioral category was determined for each individual [12].

The mean value of the analyzed traits was calculated for each treatment group. Statistical analysis was conducted using the general linear model Procedure, Statistical Analysis System [13] by fitting a linear model with one main effect of different stocking density. Duncan multiple range tests were used to determine statistical significance between treatment means at 0.05 level.

Table 3 shows the average of total feed intake per steer. The amounts of feed intake were not significantly different between different space density groups (p>0.05). However, the steers in more spacious groups had marginally more feed intakes, which may be due to more appetite with increasing activity.

Generally T2 group had heavier body weights than the control and T1 groups during the early fattening period (p<0.05) (Table 4). This result supports the report of Boe and Faerevik [14], in which the welfare of livestock was closely related to productivity. Also, Rind and Phillips [15] reported that it was important to provide sufficient space to ensure welfare and productivity of cattle. However, there was no statistical significance of difference of body weight at 30 months of age between the three treatment groups in this study (Table 4). At 15 to 18 months of age, the control group had greater ADG than T1 group (p<0.05), but T1 group had higher ADG than the control and T2 groups at 18 to 21 months of age. This result corresponds to the report of Li et al [16], in which the animals reared with a large group size had significantly greater ADG in an early fattening period. This result reflects strong competition on eating in a large group. As fattening period is extended, the relative space becomes more limited for the steers in denser pens and thus activity of the steers would decrease correspondingly.

In the late fattening period (21 to 30 month), the steers in T2 group, in general, had lowest ADG, followed by T1 and the control group, even if there was no statistical significance between the treatment groups (Table 5).

Li et al [16] reported that the animals reared in a spacious group had greater ADG in an early fattening period. However, our results in this study were not consistent with Li et al [16]. Further, the boy weight at 30 months of age was not significantly heavier in the T2 group (Table 4). This result might be partly due to the fact that the density space (10 m² per steer) in the control group was larger than the average space (7 to 8 m² per steer) in the Hanwoo farms, so that Hanwoo steers have been well adapted to low stocking density pens for the several decades. Also, the average live weight of steers in Hanwoo farms before slaughter was around 700 kg, while our results showed around 680 kg in the control group, i.e. about 3% lower, which might be due to more motility of the steers in spacious stocking density than in Hanwoo farms.

Berg and Butterfield [17] reported that ADG was also affected by differences between individuals, nutritional level of feeds or other various environmental factors. Therefore, further investigation on the relationship between ADG and motility is needed.

Table 6 shows ultrasound carcass quality traits at different growth stages and at slaughter. In general, T2 group had lower back fat thickness across all growth stages, except at slaughter, even though there was no statistical significance (p>0.05). This may be partly due to greater motility environments in the T2 group compared to T1 and the control groups. Kang et al [18] reported a back fat thickness of 9.9 mm when the Hanwoo steers were slaughtered after rearing for 540 days under the similar management as in this study. They found that all treatment groups showed the greatest fat deposition after 27 months of age. Lee et al [19] reported that there was an increase in back fat thickness in response to an increased level of concentrated feeds during early and late fattening periods, and that back fat thickness increased with unlimited feeding concentrate at 24 months of age.

For the longissimus muscle area, T2 group was, in general, the greatest throughout the entire growth period and also at slaughter, i.e. 94.6 cm², followed by T1 group (92.8 cm²), and the control group was lowest (91.0 cm²), even if statistical significance was limited between the groups, partly due to the small sample size (Table 6). This result was comparable to the report of KOSIS [20], in which the average longissimus muscle area of steers that were raised in Hanwoo farms and slaughtered was about 94.3 cm², and are consistent with the report that an increase in motility leads to an increase in musculoskeletal system, thereby improving muscle quantity [21]. The longissimus muscle area in all treatment groups increased greatly from 18 to 21 months of age, when the ADG was the greatest. This result corresponds to Kim et al [22], who reported that increases in body weight led to increased longissimus muscle area.

The longissimus muscle area after slaughter ranged 91.0 to 94.6 cm², which was greater than the result (87.7 cm²) of Kang et al [18], for which Hanwoo steers were fed with concentrated feed ad libitum during late fattening period. The difference of results between Kang et al [18] and this study may be partly due to different genetics of individuals and rearing factors.

The marbling was nearly absent (score, 1) for all steers before 18 months of age, after which marbling deposition occurred in all treatment groups (Table 6). After 18 months of age, T2 had the greatest marbling score across growth stages, and also at slaughter, while the control group had the smallest score (p<0.05). This result indicates that there is a relationship between fat deposition and motility of steers, i.e. the more spacious in a pen would cause greater motility of steers, such that more deposition of fat between muscles would occur. From the report of Hiroyuki et al [23], it can be inferred that beef cattle consumes large amounts of concentrated feed to convert surplus energy to fat accumulation. During the process of digestion and absorption metabolism for the excess energy of fat, a great deal of oxidative stress occurs in the cell, leading to an increase of active oxygen and nitric oxide and to inhibition of fat cell adiposity.

Table 7 shows the carcass weight and dressing percentage (ratio [%] of carcass weight to final live weight before slaughter) for each stocking density. T1 (423 kg) and T2 (417 kg) groups had heavier carcass weights than the control groups (398 kg) (p<0.05). Also, the dressing percentages for the T1 (62%) and T2 (61%) steers were significantly greater than the control group (58%) (p<0.05). This result supports the report that increasing motility during the fattening periods increased muscle amount, causing greater carcass rate [21].

Table 8 shows the average time and frequency of the six behavior characteristics in different stocking densities. The time of standing behavior was the longest in the control group, but the shortest for T2 group, while for a resting behavior measure, lying-down, T2 group had the longest time, even if statistical significance was limited (p>0.05). Houpt et al [24] reported that bulls spent 12 hours a day lying down, which means that steers spend much time in a lying-down position. Also, our results showed that the lying-down behavior was opposed to standing behavior. T2 had greater amount of walking behavior than the other two treatment groups (p<0.05). This result makes sense in that more space in the pen enabled the steers to allow more chance of motility. Instead, less time of walking would be caused by smaller space of the pen [25].

Self-grooming behavior may be caused by sanitary purposes or adaptation into the environment [26]. T2 group had the lowest frequency of self-grooming behavior, while the control group had the greatest frequency (p<0.05). Ishi [27] reported that narrow rearing spaces resulted in increased excretion storage areas, causing increased contamination of the fleece and skin, resulting in increased frequency of self-grooming behaviors.

The frequency of fighting was the greatest in the control group, while the lowest in T2 group (p<0.05). According to Kondo et al [28], high stocking density increased fighting behavior patterns such as intimidation, assault or fighting between group members. Our results also revealed more frequent fighting behavior in the control group (Table 8), which was supported by Zeeb et al [29], who reported reduced frequency of fighting behavior as stocking density decreased. Jensen et al [30] also reported that fighting behavior was affected by stocking density.